2015-09-03 17:48:53 +00:00
|
|
|
|
//! The **Details** output view displays each file as a row in a table.
|
|
|
|
|
//!
|
|
|
|
|
//! It's used in the following situations:
|
|
|
|
|
//!
|
|
|
|
|
//! - Most commonly, when using the `--long` command-line argument to display the
|
|
|
|
|
//! details of each file, which requires using a table view to hold all the data;
|
|
|
|
|
//! - When using the `--tree` argument, which uses the same table view to display
|
|
|
|
|
//! each file on its own line, with the table providing the tree characters;
|
|
|
|
|
//! - When using both the `--long` and `--grid` arguments, which constructs a
|
|
|
|
|
//! series of tables to fit all the data on the screen.
|
|
|
|
|
//!
|
|
|
|
|
//! You will probably recognise it from the `ls --long` command. It looks like
|
|
|
|
|
//! this:
|
|
|
|
|
//!
|
2016-04-19 06:48:41 +00:00
|
|
|
|
//! ```text
|
2015-09-03 17:48:53 +00:00
|
|
|
|
//! .rw-r--r-- 9.6k ben 29 Jun 16:16 Cargo.lock
|
|
|
|
|
//! .rw-r--r-- 547 ben 23 Jun 10:54 Cargo.toml
|
|
|
|
|
//! .rw-r--r-- 1.1k ben 23 Nov 2014 LICENCE
|
|
|
|
|
//! .rw-r--r-- 2.5k ben 21 May 14:38 README.md
|
|
|
|
|
//! .rw-r--r-- 382k ben 8 Jun 21:00 screenshot.png
|
|
|
|
|
//! drwxr-xr-x - ben 29 Jun 14:50 src
|
|
|
|
|
//! drwxr-xr-x - ben 28 Jun 19:53 target
|
2016-04-19 06:48:41 +00:00
|
|
|
|
//! ```
|
2015-09-03 17:48:53 +00:00
|
|
|
|
//!
|
|
|
|
|
//! The table is constructed by creating a `Table` value, which produces a `Row`
|
|
|
|
|
//! value for each file. These rows can contain a vector of `Cell`s, or they can
|
|
|
|
|
//! contain depth information for the tree view, or both. These are described
|
|
|
|
|
//! below.
|
|
|
|
|
//!
|
|
|
|
|
//!
|
|
|
|
|
//! ## Constructing Detail Views
|
|
|
|
|
//!
|
|
|
|
|
//! When using the `--long` command-line argument, the details of each file are
|
|
|
|
|
//! displayed next to its name.
|
|
|
|
|
//!
|
|
|
|
|
//! The table holds a vector of all the column types. For each file and column, a
|
|
|
|
|
//! `Cell` value containing the ANSI-coloured text and Unicode width of each cell
|
|
|
|
|
//! is generated, with the row and column determined by indexing into both arrays.
|
|
|
|
|
//!
|
|
|
|
|
//! The column types vector does not actually include the filename. This is
|
|
|
|
|
//! because the filename is always the rightmost field, and as such, it does not
|
|
|
|
|
//! need to have its width queried or be padded with spaces.
|
|
|
|
|
//!
|
|
|
|
|
//! To illustrate the above:
|
|
|
|
|
//!
|
2016-04-19 06:48:41 +00:00
|
|
|
|
//! ```text
|
2015-09-03 17:48:53 +00:00
|
|
|
|
//! ┌─────────────────────────────────────────────────────────────────────────┐
|
|
|
|
|
//! │ columns: [ Permissions, Size, User, Date(Modified) ] │
|
|
|
|
|
//! ├─────────────────────────────────────────────────────────────────────────┤
|
|
|
|
|
//! │ rows: cells: filename: │
|
|
|
|
|
//! │ row 1: [ ".rw-r--r--", "9.6k", "ben", "29 Jun 16:16" ] Cargo.lock │
|
|
|
|
|
//! │ row 2: [ ".rw-r--r--", "547", "ben", "23 Jun 10:54" ] Cargo.toml │
|
|
|
|
|
//! │ row 3: [ "drwxr-xr-x", "-", "ben", "29 Jun 14:50" ] src │
|
|
|
|
|
//! │ row 4: [ "drwxr-xr-x", "-", "ben", "28 Jun 19:53" ] target │
|
|
|
|
|
//! └─────────────────────────────────────────────────────────────────────────┘
|
2016-04-19 06:48:41 +00:00
|
|
|
|
//! ```
|
2015-09-03 17:48:53 +00:00
|
|
|
|
//!
|
|
|
|
|
//! Each column in the table needs to be resized to fit its widest argument. This
|
|
|
|
|
//! means that we must wait until every row has been added to the table before it
|
|
|
|
|
//! can be displayed, in order to make sure that every column is wide enough.
|
|
|
|
|
//!
|
|
|
|
|
//!
|
|
|
|
|
//! ## Extended Attributes and Errors
|
|
|
|
|
//!
|
|
|
|
|
//! Finally, files' extended attributes and any errors that occur while statting
|
|
|
|
|
//! them can also be displayed as their children. It looks like this:
|
|
|
|
|
//!
|
2016-04-19 06:48:41 +00:00
|
|
|
|
//! ```text
|
2015-09-03 17:48:53 +00:00
|
|
|
|
//! .rw-r--r-- 0 ben 3 Sep 13:26 forbidden
|
|
|
|
|
//! └── <Permission denied (os error 13)>
|
|
|
|
|
//! .rw-r--r--@ 0 ben 3 Sep 13:26 file_with_xattrs
|
|
|
|
|
//! ├── another_greeting (len 2)
|
|
|
|
|
//! └── greeting (len 5)
|
2016-04-19 06:48:41 +00:00
|
|
|
|
//! ```
|
2015-09-03 17:48:53 +00:00
|
|
|
|
//!
|
|
|
|
|
//! These lines also have `None` cells, and the error string or attribute details
|
|
|
|
|
//! are used in place of the filename.
|
|
|
|
|
|
|
|
|
|
|
2016-04-18 17:39:32 +00:00
|
|
|
|
use std::io::{Write, Error as IOError, Result as IOResult};
|
2015-12-22 01:15:59 +00:00
|
|
|
|
use std::ops::Add;
|
2015-08-26 10:36:10 +00:00
|
|
|
|
use std::path::PathBuf;
|
2015-06-28 12:21:21 +00:00
|
|
|
|
use std::string::ToString;
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
use std::sync::{Arc, Mutex};
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
use ansi_term::Style;
|
2015-02-05 14:39:56 +00:00
|
|
|
|
|
2016-02-10 15:22:43 +00:00
|
|
|
|
use datetime::fmt::DateFormat;
|
|
|
|
|
use datetime::{LocalDateTime, DatePiece};
|
|
|
|
|
use datetime::TimeZone;
|
2016-02-10 19:11:10 +00:00
|
|
|
|
use zoneinfo_compiled::{CompiledData, Result as TZResult};
|
2015-06-08 20:33:39 +00:00
|
|
|
|
|
2015-02-10 16:08:10 +00:00
|
|
|
|
use locale;
|
2015-02-05 14:39:56 +00:00
|
|
|
|
|
2016-02-10 15:22:43 +00:00
|
|
|
|
use users::{Users, Groups, UsersCache};
|
2015-06-08 20:33:39 +00:00
|
|
|
|
|
2016-04-16 17:59:25 +00:00
|
|
|
|
use fs::{Dir, File, fields as f};
|
|
|
|
|
use fs::feature::xattr::{Attribute, FileAttributes};
|
2015-12-22 01:15:59 +00:00
|
|
|
|
use options::{FileFilter, RecurseOptions};
|
|
|
|
|
use output::colours::Colours;
|
|
|
|
|
use output::column::{Alignment, Column, Columns, SizeFormat};
|
|
|
|
|
use output::cell::{TextCell, DisplayWidth};
|
2015-12-22 04:44:51 +00:00
|
|
|
|
use output::tree::TreeTrunk;
|
2015-06-08 20:33:39 +00:00
|
|
|
|
use super::filename;
|
2015-06-04 14:15:39 +00:00
|
|
|
|
|
2015-05-11 22:28:01 +00:00
|
|
|
|
|
2015-02-26 08:27:29 +00:00
|
|
|
|
/// With the **Details** view, the output gets formatted into columns, with
|
|
|
|
|
/// each `Column` object showing some piece of information about the file,
|
|
|
|
|
/// such as its size, or its permissions.
|
|
|
|
|
///
|
|
|
|
|
/// To do this, the results have to be written to a table, instead of
|
|
|
|
|
/// displaying each file immediately. Then, the width of each column can be
|
|
|
|
|
/// calculated based on the individual results, and the fields are padded
|
|
|
|
|
/// during output.
|
|
|
|
|
///
|
|
|
|
|
/// Almost all the heavy lifting is done in a Table object, which handles the
|
|
|
|
|
/// columns for each row.
|
2016-10-30 14:31:25 +00:00
|
|
|
|
#[derive(PartialEq, Debug, Clone, Default)]
|
2015-02-05 14:39:56 +00:00
|
|
|
|
pub struct Details {
|
2015-02-26 08:27:29 +00:00
|
|
|
|
|
|
|
|
|
/// A Columns object that says which columns should be included in the
|
|
|
|
|
/// output in the general case. Directories themselves can pick which
|
|
|
|
|
/// columns are *added* to this list, such as the Git column.
|
2015-08-03 12:54:25 +00:00
|
|
|
|
pub columns: Option<Columns>,
|
2015-02-26 08:27:29 +00:00
|
|
|
|
|
|
|
|
|
/// Whether to recurse through directories with a tree view, and if so,
|
|
|
|
|
/// which options to use. This field is only relevant here if the `tree`
|
|
|
|
|
/// field of the RecurseOptions is `true`.
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
pub recurse: Option<RecurseOptions>,
|
|
|
|
|
|
|
|
|
|
/// How to sort and filter the files after getting their details.
|
|
|
|
|
pub filter: FileFilter,
|
2015-02-26 08:27:29 +00:00
|
|
|
|
|
|
|
|
|
/// Whether to show a header line or not.
|
|
|
|
|
pub header: bool,
|
|
|
|
|
|
|
|
|
|
/// Whether to show each file's extended attributes.
|
2015-02-22 12:55:13 +00:00
|
|
|
|
pub xattr: bool,
|
2015-05-09 22:57:18 +00:00
|
|
|
|
|
2015-05-16 15:10:58 +00:00
|
|
|
|
/// The colours to use to display information in the table, including the
|
|
|
|
|
/// colour of the tree view symbols.
|
2015-05-09 22:57:18 +00:00
|
|
|
|
pub colours: Colours,
|
2015-02-05 14:39:56 +00:00
|
|
|
|
}
|
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
/// The **environment** struct contains any data that could change between
|
|
|
|
|
/// running instances of exa, depending on the user's computer's configuration.
|
|
|
|
|
///
|
|
|
|
|
/// Any environment field should be able to be mocked up for test runs.
|
|
|
|
|
pub struct Environment<U: Users+Groups> {
|
|
|
|
|
|
|
|
|
|
/// The year of the current time. This gets used to determine which date
|
|
|
|
|
/// format to use.
|
|
|
|
|
current_year: i64,
|
|
|
|
|
|
|
|
|
|
/// Localisation rules for formatting numbers.
|
|
|
|
|
numeric: locale::Numeric,
|
|
|
|
|
|
|
|
|
|
/// Localisation rules for formatting timestamps.
|
|
|
|
|
time: locale::Time,
|
|
|
|
|
|
|
|
|
|
/// The computer's current time zone. This gets used to determine how to
|
|
|
|
|
/// offset files' timestamps.
|
2016-03-31 20:19:29 +00:00
|
|
|
|
tz: Option<TimeZone>,
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
|
|
|
|
|
/// Mapping cache of user IDs to usernames.
|
|
|
|
|
users: Mutex<U>,
|
|
|
|
|
}
|
|
|
|
|
|
2016-02-10 15:22:43 +00:00
|
|
|
|
impl Default for Environment<UsersCache> {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
fn default() -> Self {
|
2016-03-31 20:19:29 +00:00
|
|
|
|
let tz = determine_time_zone();
|
2016-02-11 15:52:40 +00:00
|
|
|
|
|
2016-03-31 20:19:29 +00:00
|
|
|
|
if let Err(ref e) = tz {
|
|
|
|
|
println!("Unable to determine time zone: {}", e);
|
|
|
|
|
}
|
2016-02-11 15:52:40 +00:00
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
Environment {
|
|
|
|
|
current_year: LocalDateTime::now().year(),
|
|
|
|
|
numeric: locale::Numeric::load_user_locale().unwrap_or_else(|_| locale::Numeric::english()),
|
|
|
|
|
time: locale::Time::load_user_locale().unwrap_or_else(|_| locale::Time::english()),
|
2016-03-31 20:19:29 +00:00
|
|
|
|
tz: tz.ok(),
|
2016-02-10 15:22:43 +00:00
|
|
|
|
users: Mutex::new(UsersCache::new()),
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2016-02-10 19:11:10 +00:00
|
|
|
|
fn determine_time_zone() -> TZResult<TimeZone> {
|
2016-03-31 20:19:29 +00:00
|
|
|
|
TimeZone::from_file("/etc/localtime")
|
2016-02-10 19:02:20 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-02-05 14:39:56 +00:00
|
|
|
|
impl Details {
|
2015-09-03 17:48:53 +00:00
|
|
|
|
|
|
|
|
|
/// Print the details of the given vector of files -- all of which will
|
|
|
|
|
/// have been read from the given directory, if present -- to stdout.
|
2016-04-18 17:39:32 +00:00
|
|
|
|
pub fn view<W: Write>(&self, dir: Option<&Dir>, files: Vec<File>, w: &mut W) -> IOResult<()> {
|
2015-09-03 17:48:53 +00:00
|
|
|
|
|
2015-02-26 08:27:29 +00:00
|
|
|
|
// First, transform the Columns object into a vector of columns for
|
|
|
|
|
// the current directory.
|
2015-08-03 12:54:25 +00:00
|
|
|
|
let columns_for_dir = match self.columns {
|
|
|
|
|
Some(cols) => cols.for_dir(dir),
|
|
|
|
|
None => Vec::new(),
|
|
|
|
|
};
|
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
// Then, retrieve various environment variables.
|
2016-02-10 15:22:43 +00:00
|
|
|
|
let env = Arc::new(Environment::<UsersCache>::default());
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
|
|
|
|
|
// Build the table to put rows in.
|
|
|
|
|
let mut table = Table {
|
|
|
|
|
columns: &*columns_for_dir,
|
|
|
|
|
opts: &self,
|
|
|
|
|
env: env,
|
|
|
|
|
rows: Vec::new(),
|
|
|
|
|
};
|
|
|
|
|
|
2015-09-03 17:48:53 +00:00
|
|
|
|
// Next, add a header if the user requests it.
|
2015-02-26 07:18:08 +00:00
|
|
|
|
if self.header { table.add_header() }
|
|
|
|
|
|
2015-02-26 08:27:29 +00:00
|
|
|
|
// Then add files to the table and print it out.
|
2015-02-26 07:18:08 +00:00
|
|
|
|
self.add_files_to_table(&mut table, files, 0);
|
2015-08-25 20:08:25 +00:00
|
|
|
|
for cell in table.print_table() {
|
2016-04-18 17:39:32 +00:00
|
|
|
|
try!(writeln!(w, "{}", cell.strings()));
|
2015-06-28 12:21:21 +00:00
|
|
|
|
}
|
2016-04-18 17:39:32 +00:00
|
|
|
|
|
|
|
|
|
Ok(())
|
2015-02-26 07:18:08 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-09-03 17:48:53 +00:00
|
|
|
|
/// Adds files to the table, possibly recursively. This is easily
|
|
|
|
|
/// parallelisable, and uses a pool of threads.
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
fn add_files_to_table<'dir, U: Users+Groups+Send>(&self, mut table: &mut Table<U>, src: Vec<File<'dir>>, depth: usize) {
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
use num_cpus;
|
|
|
|
|
use scoped_threadpool::Pool;
|
|
|
|
|
use std::sync::{Arc, Mutex};
|
2016-04-16 17:59:25 +00:00
|
|
|
|
use fs::feature::xattr;
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
|
|
|
|
|
let mut pool = Pool::new(num_cpus::get() as u32);
|
|
|
|
|
let mut file_eggs = Vec::new();
|
|
|
|
|
|
2016-03-31 22:13:15 +00:00
|
|
|
|
struct Egg<'a> {
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
cells: Vec<TextCell>,
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
xattrs: Vec<Attribute>,
|
2016-04-18 17:39:32 +00:00
|
|
|
|
errors: Vec<(IOError, Option<PathBuf>)>,
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
dir: Option<Dir>,
|
2016-03-31 22:13:15 +00:00
|
|
|
|
file: File<'a>,
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
impl<'a> AsRef<File<'a>> for Egg<'a> {
|
|
|
|
|
fn as_ref(&self) -> &File<'a> {
|
|
|
|
|
&self.file
|
|
|
|
|
}
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
}
|
2015-08-26 11:00:31 +00:00
|
|
|
|
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
pool.scoped(|scoped| {
|
|
|
|
|
let file_eggs = Arc::new(Mutex::new(&mut file_eggs));
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let table = Arc::new(&mut table);
|
2015-08-25 14:04:15 +00:00
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
for file in src {
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
let file_eggs = file_eggs.clone();
|
|
|
|
|
let table = table.clone();
|
2015-08-26 10:36:10 +00:00
|
|
|
|
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
scoped.execute(move || {
|
|
|
|
|
let mut errors = Vec::new();
|
|
|
|
|
let mut xattrs = Vec::new();
|
2016-01-16 22:19:00 +00:00
|
|
|
|
|
|
|
|
|
if xattr::ENABLED {
|
|
|
|
|
match file.path.attributes() {
|
|
|
|
|
Ok(xs) => xattrs.extend(xs),
|
|
|
|
|
Err(e) => errors.push((e, None)),
|
|
|
|
|
};
|
|
|
|
|
}
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let cells = table.cells_for_file(&file, !xattrs.is_empty());
|
2015-09-03 17:48:53 +00:00
|
|
|
|
|
2016-01-16 22:19:00 +00:00
|
|
|
|
if !table.opts.xattr {
|
|
|
|
|
xattrs.clear();
|
|
|
|
|
}
|
|
|
|
|
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
let mut dir = None;
|
|
|
|
|
|
|
|
|
|
if let Some(r) = self.recurse {
|
|
|
|
|
if file.is_directory() && r.tree && !r.is_too_deep(depth) {
|
|
|
|
|
if let Ok(d) = file.to_dir(false) {
|
|
|
|
|
dir = Some(d);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
let egg = Egg {
|
|
|
|
|
cells: cells,
|
|
|
|
|
xattrs: xattrs,
|
|
|
|
|
errors: errors,
|
|
|
|
|
dir: dir,
|
|
|
|
|
file: file,
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
file_eggs.lock().unwrap().push(egg);
|
|
|
|
|
});
|
|
|
|
|
}
|
|
|
|
|
});
|
|
|
|
|
|
2016-03-31 22:13:15 +00:00
|
|
|
|
self.filter.sort_files(&mut file_eggs);
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
|
|
|
|
|
let num_eggs = file_eggs.len();
|
|
|
|
|
for (index, egg) in file_eggs.into_iter().enumerate() {
|
|
|
|
|
let mut files = Vec::new();
|
|
|
|
|
let mut errors = egg.errors;
|
Print the parent path for passed-in files
This commit changes all the views to accommodate printing each path's prefix, if it has one.
Previously, each file was stripped of its ancestry, leaving only its file name to be displayed. So running "exa /usr/bin/*" would display only filenames, while running "ls /usr/bin/*" would display each file prefixed with "/usr/bin/". But running "ls /usr/bin/" -- without the glob -- would run ls on just the directory, printing out the file names with no prefix or anything.
This functionality turned out to be useful in quite a few situations: firstly, if the user passes in files from different directories, it would be hard to tell where they came from (especially if they have the same name, such as find | xargs). Secondly, this also applied when following symlinks, making it unclear exactly which file a symlink would be pointing to.
The reason that it did it this way beforehand was that I didn't think of these use-cases, rather than for any technical reason; this new method should not have any drawbacks save making the output slightly wider in a few cases. Compatibility with ls is also a big plus.
Fixes #104, and relates to #88 and #92.
2016-04-11 18:10:55 +00:00
|
|
|
|
let mut width = DisplayWidth::from(&*egg.file.name);
|
|
|
|
|
|
|
|
|
|
if egg.file.dir.is_none() {
|
|
|
|
|
if let Some(ref parent) = egg.file.path.parent() {
|
|
|
|
|
width = width + 1 + DisplayWidth::from(parent.to_string_lossy().as_ref());
|
|
|
|
|
}
|
|
|
|
|
}
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let name = TextCell {
|
Print the parent path for passed-in files
This commit changes all the views to accommodate printing each path's prefix, if it has one.
Previously, each file was stripped of its ancestry, leaving only its file name to be displayed. So running "exa /usr/bin/*" would display only filenames, while running "ls /usr/bin/*" would display each file prefixed with "/usr/bin/". But running "ls /usr/bin/" -- without the glob -- would run ls on just the directory, printing out the file names with no prefix or anything.
This functionality turned out to be useful in quite a few situations: firstly, if the user passes in files from different directories, it would be hard to tell where they came from (especially if they have the same name, such as find | xargs). Secondly, this also applied when following symlinks, making it unclear exactly which file a symlink would be pointing to.
The reason that it did it this way beforehand was that I didn't think of these use-cases, rather than for any technical reason; this new method should not have any drawbacks save making the output slightly wider in a few cases. Compatibility with ls is also a big plus.
Fixes #104, and relates to #88 and #92.
2016-04-11 18:10:55 +00:00
|
|
|
|
contents: filename(&egg.file, &self.colours, true),
|
2015-12-17 02:34:11 +00:00
|
|
|
|
width: width,
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
};
|
|
|
|
|
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
let row = Row {
|
|
|
|
|
depth: depth,
|
|
|
|
|
cells: Some(egg.cells),
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
name: name,
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
last: index == num_eggs - 1,
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
table.rows.push(row);
|
|
|
|
|
|
|
|
|
|
if let Some(ref dir) = egg.dir {
|
|
|
|
|
for file_to_add in dir.files() {
|
|
|
|
|
match file_to_add {
|
|
|
|
|
Ok(f) => files.push(f),
|
|
|
|
|
Err((path, e)) => errors.push((e, Some(path)))
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2016-10-30 14:43:33 +00:00
|
|
|
|
self.filter.filter_child_files(&mut files);
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
|
|
|
|
|
if !files.is_empty() {
|
|
|
|
|
for xattr in egg.xattrs {
|
|
|
|
|
table.add_xattr(xattr, depth + 1, false);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
for (error, path) in errors {
|
|
|
|
|
table.add_error(&error, depth + 1, false, path);
|
2015-08-26 10:36:10 +00:00
|
|
|
|
}
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
|
|
|
|
|
self.add_files_to_table(table, files, depth + 1);
|
|
|
|
|
continue;
|
2015-02-26 07:18:08 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
2015-08-26 10:36:10 +00:00
|
|
|
|
|
Parallelise the details view!
This commit removes the threadpool in `main.rs` that stats each command-line argument separately, and replaces it with a *scoped* threadpool in `options/details.rs` that builds the table in parallel! Running this on my machine halves the execution time when tree-ing my entire home directory (which isn't exactly a common occurrence, but it's the only way to give exa a large running time)
The statting will be added back in parallel at a later stage. This was facilitated by the previous changes to recursion that made it easier to deal with.
There's a lot of large sweeping architectural changes. Here's a smattering of them:
- In `main.rs`, the files are now passed around as vectors of files rather than array slices of files. This is because `File`s aren't `Clone`, and the `Vec` is necessary to give away ownership of the files at the appropriate point.
- In the details view, files are now sorted *all* the time, rather than obeying the command-line order. As they're run in parallel, they have no guaranteed order anyway, so we *have* to sort them again. (I'm not sure if this should be the intended behaviour or not!) This means that the `Details` struct has to have the filter *all* the time, not only while recursing, so it's been moved out of the `recurse` field.
- We use `scoped_threadpool` over `threadpool`, a recent addition. It's only safely used on Nightly, which we're using anyway, so that's OK!
- Removed a bunch of out-of-date comments.
This also fixes #77, mainly by accident :)
2015-09-02 22:19:10 +00:00
|
|
|
|
let count = egg.xattrs.len();
|
|
|
|
|
for (index, xattr) in egg.xattrs.into_iter().enumerate() {
|
2015-08-26 10:36:10 +00:00
|
|
|
|
table.add_xattr(xattr, depth + 1, errors.is_empty() && index == count - 1);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
let count = errors.len();
|
|
|
|
|
for (index, (error, path)) in errors.into_iter().enumerate() {
|
|
|
|
|
table.add_error(&error, depth + 1, index == count - 1, path);
|
|
|
|
|
}
|
2015-02-05 14:39:56 +00:00
|
|
|
|
}
|
2015-02-26 07:18:08 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
2015-02-05 14:39:56 +00:00
|
|
|
|
|
2015-06-08 20:33:39 +00:00
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
pub struct Row {
|
2015-02-26 08:27:29 +00:00
|
|
|
|
|
|
|
|
|
/// Vector of cells to display.
|
2015-08-25 10:50:07 +00:00
|
|
|
|
///
|
2015-09-03 17:48:53 +00:00
|
|
|
|
/// Most of the rows will be used to display files' metadata, so this will
|
|
|
|
|
/// almost always be `Some`, containing a vector of cells. It will only be
|
|
|
|
|
/// `None` for a row displaying an attribute or error, neither of which
|
|
|
|
|
/// have cells.
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
cells: Option<Vec<TextCell>>,
|
2015-02-26 08:27:29 +00:00
|
|
|
|
|
|
|
|
|
/// This file's name, in coloured output. The name is treated separately
|
|
|
|
|
/// from the other cells, as it never requires padding.
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
name: TextCell,
|
2015-02-26 08:27:29 +00:00
|
|
|
|
|
|
|
|
|
/// How many directories deep into the tree structure this is. Directories
|
|
|
|
|
/// on top have depth 0.
|
2015-08-25 10:50:07 +00:00
|
|
|
|
depth: usize,
|
2015-02-26 08:27:29 +00:00
|
|
|
|
|
|
|
|
|
/// Whether this is the last entry in the directory. This flag is used
|
|
|
|
|
/// when calculating the tree view.
|
2015-08-25 10:50:07 +00:00
|
|
|
|
last: bool,
|
2015-02-26 07:18:08 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-08-25 10:45:27 +00:00
|
|
|
|
impl Row {
|
2015-08-25 10:50:07 +00:00
|
|
|
|
|
2015-09-03 17:48:53 +00:00
|
|
|
|
/// Gets the Unicode display width of the indexed column, if present. If
|
|
|
|
|
/// not, returns 0.
|
2015-08-25 10:45:27 +00:00
|
|
|
|
fn column_width(&self, index: usize) -> usize {
|
|
|
|
|
match self.cells {
|
2015-12-17 02:34:11 +00:00
|
|
|
|
Some(ref cells) => *cells[index].width,
|
2015-08-25 10:45:27 +00:00
|
|
|
|
None => 0,
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-06-08 20:33:39 +00:00
|
|
|
|
|
2015-02-26 08:27:29 +00:00
|
|
|
|
/// A **Table** object gets built up by the view as it lists files and
|
|
|
|
|
/// directories.
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
pub struct Table<'a, U: Users+Groups+'a> {
|
|
|
|
|
pub rows: Vec<Row>,
|
2015-02-26 07:18:08 +00:00
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
pub columns: &'a [Column],
|
|
|
|
|
pub opts: &'a Details,
|
|
|
|
|
pub env: Arc<Environment<U>>,
|
2015-05-16 20:02:28 +00:00
|
|
|
|
}
|
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
impl<'a, U: Users+Groups+'a> Table<'a, U> {
|
2015-02-05 14:39:56 +00:00
|
|
|
|
|
2015-02-26 08:27:29 +00:00
|
|
|
|
/// Add a dummy "header" row to the table, which contains the names of all
|
|
|
|
|
/// the columns, underlined. This has dummy data for the cases that aren't
|
|
|
|
|
/// actually used, such as the depth or list of attributes.
|
2015-06-28 12:21:21 +00:00
|
|
|
|
pub fn add_header(&mut self) {
|
2015-02-26 07:18:08 +00:00
|
|
|
|
let row = Row {
|
2015-02-26 08:27:29 +00:00
|
|
|
|
depth: 0,
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
cells: Some(self.columns.iter().map(|c| TextCell::paint_str(self.opts.colours.header, c.header())).collect()),
|
|
|
|
|
name: TextCell::paint_str(self.opts.colours.header, "Name"),
|
2015-02-26 08:27:29 +00:00
|
|
|
|
last: false,
|
2015-02-26 07:18:08 +00:00
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
self.rows.push(row);
|
|
|
|
|
}
|
|
|
|
|
|
2016-04-18 17:39:32 +00:00
|
|
|
|
fn add_error(&mut self, error: &IOError, depth: usize, last: bool, path: Option<PathBuf>) {
|
2015-08-25 14:04:15 +00:00
|
|
|
|
let error_message = match path {
|
|
|
|
|
Some(path) => format!("<{}: {}>", path.display(), error),
|
|
|
|
|
None => format!("<{}>", error),
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
let row = Row {
|
|
|
|
|
depth: depth,
|
|
|
|
|
cells: None,
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
name: TextCell::paint(self.opts.colours.broken_arrow, error_message),
|
2015-08-25 14:04:15 +00:00
|
|
|
|
last: last,
|
2015-08-25 20:08:25 +00:00
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
self.rows.push(row);
|
|
|
|
|
}
|
|
|
|
|
|
2015-08-26 10:36:10 +00:00
|
|
|
|
fn add_xattr(&mut self, xattr: Attribute, depth: usize, last: bool) {
|
2015-08-25 20:08:25 +00:00
|
|
|
|
let row = Row {
|
|
|
|
|
depth: depth,
|
|
|
|
|
cells: None,
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
name: TextCell::paint(self.opts.colours.perms.attribute, format!("{} (len {})", xattr.name, xattr.size)),
|
2015-08-25 20:08:25 +00:00
|
|
|
|
last: last,
|
2015-08-25 14:04:15 +00:00
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
self.rows.push(row);
|
|
|
|
|
}
|
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
pub fn filename_cell(&self, file: File, links: bool) -> TextCell {
|
Print the parent path for passed-in files
This commit changes all the views to accommodate printing each path's prefix, if it has one.
Previously, each file was stripped of its ancestry, leaving only its file name to be displayed. So running "exa /usr/bin/*" would display only filenames, while running "ls /usr/bin/*" would display each file prefixed with "/usr/bin/". But running "ls /usr/bin/" -- without the glob -- would run ls on just the directory, printing out the file names with no prefix or anything.
This functionality turned out to be useful in quite a few situations: firstly, if the user passes in files from different directories, it would be hard to tell where they came from (especially if they have the same name, such as find | xargs). Secondly, this also applied when following symlinks, making it unclear exactly which file a symlink would be pointing to.
The reason that it did it this way beforehand was that I didn't think of these use-cases, rather than for any technical reason; this new method should not have any drawbacks save making the output slightly wider in a few cases. Compatibility with ls is also a big plus.
Fixes #104, and relates to #88 and #92.
2016-04-11 18:10:55 +00:00
|
|
|
|
let mut width = DisplayWidth::from(&*file.name);
|
|
|
|
|
|
|
|
|
|
if file.dir.is_none() {
|
|
|
|
|
if let Some(ref parent) = file.path.parent() {
|
|
|
|
|
width = width + 1 + DisplayWidth::from(parent.to_string_lossy().as_ref());
|
|
|
|
|
}
|
|
|
|
|
}
|
2015-12-17 02:34:11 +00:00
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
TextCell {
|
Print the parent path for passed-in files
This commit changes all the views to accommodate printing each path's prefix, if it has one.
Previously, each file was stripped of its ancestry, leaving only its file name to be displayed. So running "exa /usr/bin/*" would display only filenames, while running "ls /usr/bin/*" would display each file prefixed with "/usr/bin/". But running "ls /usr/bin/" -- without the glob -- would run ls on just the directory, printing out the file names with no prefix or anything.
This functionality turned out to be useful in quite a few situations: firstly, if the user passes in files from different directories, it would be hard to tell where they came from (especially if they have the same name, such as find | xargs). Secondly, this also applied when following symlinks, making it unclear exactly which file a symlink would be pointing to.
The reason that it did it this way beforehand was that I didn't think of these use-cases, rather than for any technical reason; this new method should not have any drawbacks save making the output slightly wider in a few cases. Compatibility with ls is also a big plus.
Fixes #104, and relates to #88 and #92.
2016-04-11 18:10:55 +00:00
|
|
|
|
contents: filename(&file, &self.opts.colours, links),
|
2015-12-17 02:34:11 +00:00
|
|
|
|
width: width,
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
pub fn add_file_with_cells(&mut self, cells: Vec<TextCell>, name_cell: TextCell, depth: usize, last: bool) {
|
2015-05-12 01:54:34 +00:00
|
|
|
|
let row = Row {
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
depth: depth,
|
|
|
|
|
cells: Some(cells),
|
|
|
|
|
name: name_cell,
|
|
|
|
|
last: last,
|
2015-05-12 01:54:34 +00:00
|
|
|
|
};
|
|
|
|
|
|
2015-05-12 02:14:56 +00:00
|
|
|
|
self.rows.push(row);
|
2015-05-12 01:54:34 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-02-26 08:27:29 +00:00
|
|
|
|
/// Use the list of columns to find which cells should be produced for
|
|
|
|
|
/// this file, per-column.
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
pub fn cells_for_file(&self, file: &File, xattrs: bool) -> Vec<TextCell> {
|
2015-02-26 07:18:08 +00:00
|
|
|
|
self.columns.clone().iter()
|
2015-08-26 11:00:31 +00:00
|
|
|
|
.map(|c| self.display(file, c, xattrs))
|
2015-02-26 07:18:08 +00:00
|
|
|
|
.collect()
|
|
|
|
|
}
|
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
fn display(&self, file: &File, column: &Column, xattrs: bool) -> TextCell {
|
2015-11-15 16:12:16 +00:00
|
|
|
|
use output::column::TimeType::*;
|
|
|
|
|
|
2015-05-11 22:28:01 +00:00
|
|
|
|
match *column {
|
2016-04-16 19:01:45 +00:00
|
|
|
|
Column::Permissions => self.render_permissions(file.type_char(), file.permissions(), xattrs),
|
2015-11-15 16:12:16 +00:00
|
|
|
|
Column::FileSize(fmt) => self.render_size(file.size(), fmt),
|
|
|
|
|
Column::Timestamp(Modified) => self.render_time(file.modified_time()),
|
|
|
|
|
Column::Timestamp(Created) => self.render_time(file.created_time()),
|
|
|
|
|
Column::Timestamp(Accessed) => self.render_time(file.accessed_time()),
|
|
|
|
|
Column::HardLinks => self.render_links(file.links()),
|
|
|
|
|
Column::Inode => self.render_inode(file.inode()),
|
|
|
|
|
Column::Blocks => self.render_blocks(file.blocks()),
|
|
|
|
|
Column::User => self.render_user(file.user()),
|
|
|
|
|
Column::Group => self.render_group(file.group()),
|
|
|
|
|
Column::GitStatus => self.render_git_status(file.git_status()),
|
2015-05-11 22:28:01 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2016-04-16 19:01:45 +00:00
|
|
|
|
fn render_permissions(&self, file_type: f::Type, permissions: f::Permissions, xattrs: bool) -> TextCell {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let perms = self.opts.colours.perms;
|
|
|
|
|
let types = self.opts.colours.filetypes;
|
|
|
|
|
|
2015-05-12 01:54:34 +00:00
|
|
|
|
let bit = |bit, chr: &'static str, style: Style| {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
if bit { style.paint(chr) } else { self.opts.colours.punctuation.paint("-") }
|
2015-02-26 07:18:08 +00:00
|
|
|
|
};
|
|
|
|
|
|
2016-04-16 19:01:45 +00:00
|
|
|
|
let type_char = match file_type {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
f::Type::File => types.normal.paint("."),
|
|
|
|
|
f::Type::Directory => types.directory.paint("d"),
|
2016-06-14 03:14:36 +00:00
|
|
|
|
f::Type::Pipe => types.pipe.paint("|"),
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
f::Type::Link => types.symlink.paint("l"),
|
2016-06-14 03:14:36 +00:00
|
|
|
|
f::Type::CharDevice => types.device.paint("c"),
|
|
|
|
|
f::Type::BlockDevice => types.device.paint("b"),
|
|
|
|
|
f::Type::Socket => types.socket.paint("s"),
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
f::Type::Special => types.special.paint("?"),
|
2015-05-12 01:54:34 +00:00
|
|
|
|
};
|
|
|
|
|
|
2016-04-16 19:01:45 +00:00
|
|
|
|
let x_colour = if file_type.is_regular_file() { perms.user_execute_file }
|
|
|
|
|
else { perms.user_execute_other };
|
2015-05-12 01:54:34 +00:00
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let mut chars = vec![
|
2016-04-16 19:01:45 +00:00
|
|
|
|
type_char,
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
bit(permissions.user_read, "r", perms.user_read),
|
|
|
|
|
bit(permissions.user_write, "w", perms.user_write),
|
2015-05-12 01:54:34 +00:00
|
|
|
|
bit(permissions.user_execute, "x", x_colour),
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
bit(permissions.group_read, "r", perms.group_read),
|
|
|
|
|
bit(permissions.group_write, "w", perms.group_write),
|
|
|
|
|
bit(permissions.group_execute, "x", perms.group_execute),
|
|
|
|
|
bit(permissions.other_read, "r", perms.other_read),
|
|
|
|
|
bit(permissions.other_write, "w", perms.other_write),
|
|
|
|
|
bit(permissions.other_execute, "x", perms.other_execute),
|
2015-05-16 13:30:01 +00:00
|
|
|
|
];
|
|
|
|
|
|
2015-08-26 11:00:31 +00:00
|
|
|
|
if xattrs {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
chars.push(perms.attribute.paint("@"));
|
2015-05-16 13:30:01 +00:00
|
|
|
|
}
|
2015-05-12 01:54:34 +00:00
|
|
|
|
|
2015-12-17 02:15:09 +00:00
|
|
|
|
// As these are all ASCII characters, we can guarantee that they’re
|
|
|
|
|
// all going to be one character wide, and don’t need to compute the
|
|
|
|
|
// cell’s display width.
|
|
|
|
|
let width = DisplayWidth::from(chars.len());
|
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
TextCell {
|
2015-12-17 09:51:42 +00:00
|
|
|
|
contents: chars.into(),
|
2015-12-17 02:34:11 +00:00
|
|
|
|
width: width,
|
2015-05-12 01:54:34 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
fn render_links(&self, links: f::Links) -> TextCell {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let style = if links.multiple { self.opts.colours.links.multi_link_file }
|
|
|
|
|
else { self.opts.colours.links.normal };
|
2015-05-12 01:54:34 +00:00
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
TextCell::paint(style, self.env.numeric.format_int(links.count))
|
2015-05-12 01:54:34 +00:00
|
|
|
|
}
|
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
fn render_blocks(&self, blocks: f::Blocks) -> TextCell {
|
2015-05-12 01:54:34 +00:00
|
|
|
|
match blocks {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
f::Blocks::Some(blk) => TextCell::paint(self.opts.colours.blocks, blk.to_string()),
|
|
|
|
|
f::Blocks::None => TextCell::blank(self.opts.colours.punctuation),
|
2015-05-12 01:54:34 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
fn render_inode(&self, inode: f::Inode) -> TextCell {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
TextCell::paint(self.opts.colours.inode, inode.0.to_string())
|
2015-05-12 01:54:34 +00:00
|
|
|
|
}
|
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
fn render_size(&self, size: f::Size, size_format: SizeFormat) -> TextCell {
|
|
|
|
|
use number_prefix::{binary_prefix, decimal_prefix};
|
|
|
|
|
use number_prefix::{Prefixed, Standalone, PrefixNames};
|
2015-11-15 19:26:58 +00:00
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let size = match size {
|
|
|
|
|
f::Size::Some(s) => s,
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
f::Size::None => return TextCell::blank(self.opts.colours.punctuation),
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
};
|
2015-05-12 01:54:34 +00:00
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let result = match size_format {
|
|
|
|
|
SizeFormat::DecimalBytes => decimal_prefix(size as f64),
|
|
|
|
|
SizeFormat::BinaryBytes => binary_prefix(size as f64),
|
|
|
|
|
SizeFormat::JustBytes => {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let string = self.env.numeric.format_int(size);
|
|
|
|
|
return TextCell::paint(self.opts.colours.size.numbers, string);
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
},
|
|
|
|
|
};
|
2015-05-12 01:54:34 +00:00
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let (prefix, n) = match result {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
Standalone(b) => return TextCell::paint(self.opts.colours.size.numbers, b.to_string()),
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
Prefixed(p, n) => (p, n)
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
let symbol = prefix.symbol();
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let number = if n < 10f64 { self.env.numeric.format_float(n, 1) }
|
|
|
|
|
else { self.env.numeric.format_int(n as isize) };
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
|
2015-12-17 02:15:09 +00:00
|
|
|
|
// The numbers and symbols are guaranteed to be written in ASCII, so
|
|
|
|
|
// we can skip the display width calculation.
|
|
|
|
|
let width = DisplayWidth::from(number.len() + symbol.len());
|
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
TextCell {
|
2015-12-17 02:34:11 +00:00
|
|
|
|
width: width,
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
contents: vec![
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
self.opts.colours.size.numbers.paint(number),
|
|
|
|
|
self.opts.colours.size.unit.paint(symbol),
|
2015-12-17 09:51:42 +00:00
|
|
|
|
].into(),
|
2015-05-12 01:54:34 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-10-01 23:56:09 +00:00
|
|
|
|
#[allow(trivial_numeric_casts)]
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
fn render_time(&self, timestamp: f::Time) -> TextCell {
|
2016-03-31 20:19:29 +00:00
|
|
|
|
// TODO(ogham): This method needs some serious de-duping!
|
|
|
|
|
// zoned and local times have different types at the moment,
|
|
|
|
|
// so it's tricky.
|
|
|
|
|
|
|
|
|
|
if let Some(ref tz) = self.env.tz {
|
|
|
|
|
let date = tz.to_zoned(LocalDateTime::at(timestamp.0 as i64));
|
2015-05-12 01:54:34 +00:00
|
|
|
|
|
2016-03-31 20:19:29 +00:00
|
|
|
|
let datestamp = if date.year() == self.env.current_year {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
DATE_AND_TIME.format(&date, &self.env.time)
|
2015-05-12 01:54:34 +00:00
|
|
|
|
}
|
|
|
|
|
else {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
DATE_AND_YEAR.format(&date, &self.env.time)
|
2015-05-12 01:54:34 +00:00
|
|
|
|
};
|
|
|
|
|
|
2016-03-31 20:19:29 +00:00
|
|
|
|
TextCell::paint(self.opts.colours.date, datestamp)
|
|
|
|
|
}
|
|
|
|
|
else {
|
|
|
|
|
let date = LocalDateTime::at(timestamp.0 as i64);
|
|
|
|
|
|
|
|
|
|
let datestamp = if date.year() == self.env.current_year {
|
|
|
|
|
DATE_AND_TIME.format(&date, &self.env.time)
|
|
|
|
|
}
|
|
|
|
|
else {
|
|
|
|
|
DATE_AND_YEAR.format(&date, &self.env.time)
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
TextCell::paint(self.opts.colours.date, datestamp)
|
|
|
|
|
}
|
2015-05-12 01:54:34 +00:00
|
|
|
|
}
|
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
fn render_git_status(&self, git: f::Git) -> TextCell {
|
|
|
|
|
let git_char = |status| match status {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
f::GitStatus::NotModified => self.opts.colours.punctuation.paint("-"),
|
|
|
|
|
f::GitStatus::New => self.opts.colours.git.new.paint("N"),
|
|
|
|
|
f::GitStatus::Modified => self.opts.colours.git.modified.paint("M"),
|
|
|
|
|
f::GitStatus::Deleted => self.opts.colours.git.deleted.paint("D"),
|
|
|
|
|
f::GitStatus::Renamed => self.opts.colours.git.renamed.paint("R"),
|
|
|
|
|
f::GitStatus::TypeChange => self.opts.colours.git.typechange.paint("T"),
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
TextCell {
|
2015-12-17 02:34:11 +00:00
|
|
|
|
width: DisplayWidth::from(2),
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
contents: vec![
|
|
|
|
|
git_char(git.staged),
|
|
|
|
|
git_char(git.unstaged)
|
2015-12-17 09:51:42 +00:00
|
|
|
|
].into(),
|
2015-05-12 02:36:47 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
fn render_user(&self, user: f::User) -> TextCell {
|
|
|
|
|
let users = self.env.users.lock().unwrap();
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
let user_name = match users.get_user_by_uid(user.0) {
|
2016-02-10 15:22:43 +00:00
|
|
|
|
Some(user) => user.name().to_owned(),
|
2015-05-12 02:14:56 +00:00
|
|
|
|
None => user.0.to_string(),
|
2015-05-12 01:54:34 +00:00
|
|
|
|
};
|
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let style = if users.get_current_uid() == user.0 { self.opts.colours.users.user_you }
|
|
|
|
|
else { self.opts.colours.users.user_someone_else };
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
TextCell::paint(style, user_name)
|
2015-05-12 01:54:34 +00:00
|
|
|
|
}
|
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
fn render_group(&self, group: f::Group) -> TextCell {
|
2016-02-10 15:22:43 +00:00
|
|
|
|
use users::os::unix::GroupExt;
|
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let mut style = self.opts.colours.users.group_not_yours;
|
2015-05-12 01:54:34 +00:00
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let users = self.env.users.lock().unwrap();
|
|
|
|
|
let group = match users.get_group_by_gid(group.0) {
|
|
|
|
|
Some(g) => (*g).clone(),
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
None => return TextCell::paint(style, group.0.to_string()),
|
2015-05-12 01:54:34 +00:00
|
|
|
|
};
|
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let current_uid = users.get_current_uid();
|
|
|
|
|
if let Some(current_user) = users.get_user_by_uid(current_uid) {
|
2016-02-10 15:22:43 +00:00
|
|
|
|
if current_user.primary_group_id() == group.gid()
|
|
|
|
|
|| group.members().contains(¤t_user.name().to_owned()) {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
style = self.opts.colours.users.group_yours;
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2016-02-10 15:22:43 +00:00
|
|
|
|
TextCell::paint(style, group.name().to_owned())
|
2015-02-26 07:18:08 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-08-25 10:50:07 +00:00
|
|
|
|
/// Render the table as a vector of Cells, to be displayed on standard output.
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
pub fn print_table(self) -> Vec<TextCell> {
|
2015-12-22 04:44:51 +00:00
|
|
|
|
let mut tree_trunk = TreeTrunk::default();
|
2015-06-28 12:21:21 +00:00
|
|
|
|
let mut cells = Vec::new();
|
2015-02-05 14:39:56 +00:00
|
|
|
|
|
2015-02-26 08:27:29 +00:00
|
|
|
|
// Work out the list of column widths by finding the longest cell for
|
|
|
|
|
// each column, then formatting each cell in that column to be the
|
|
|
|
|
// width of that one.
|
2015-03-22 19:46:45 +00:00
|
|
|
|
let column_widths: Vec<usize> = (0 .. self.columns.len())
|
2015-08-25 10:45:27 +00:00
|
|
|
|
.map(|n| self.rows.iter().map(|row| row.column_width(n)).max().unwrap_or(0))
|
2015-02-26 07:18:08 +00:00
|
|
|
|
.collect();
|
|
|
|
|
|
2015-11-14 23:32:57 +00:00
|
|
|
|
let total_width: usize = self.columns.len() + column_widths.iter().fold(0, Add::add);
|
2015-08-25 14:04:15 +00:00
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
for row in self.rows {
|
|
|
|
|
let mut cell = TextCell::default();
|
|
|
|
|
|
|
|
|
|
if let Some(cells) = row.cells {
|
|
|
|
|
for (n, (this_cell, width)) in cells.into_iter().zip(column_widths.iter()).enumerate() {
|
2015-12-17 02:34:11 +00:00
|
|
|
|
let padding = width - *this_cell.width;
|
2015-06-28 12:21:21 +00:00
|
|
|
|
|
2015-08-25 10:45:27 +00:00
|
|
|
|
match self.columns[n].alignment() {
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
Alignment::Left => { cell.append(this_cell); cell.add_spaces(padding); }
|
|
|
|
|
Alignment::Right => { cell.add_spaces(padding); cell.append(this_cell); }
|
2015-08-25 10:45:27 +00:00
|
|
|
|
}
|
2015-06-28 12:21:21 +00:00
|
|
|
|
|
2015-08-25 10:45:27 +00:00
|
|
|
|
cell.add_spaces(1);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
else {
|
2015-08-25 14:04:15 +00:00
|
|
|
|
cell.add_spaces(total_width)
|
2015-02-05 14:39:56 +00:00
|
|
|
|
}
|
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let mut filename = TextCell::default();
|
2015-06-28 12:21:21 +00:00
|
|
|
|
|
2015-12-22 04:44:51 +00:00
|
|
|
|
for tree_part in tree_trunk.new_row(row.depth, row.last) {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
filename.push(self.opts.colours.punctuation.paint(tree_part.ascii_art()), 4);
|
2015-08-25 20:08:25 +00:00
|
|
|
|
}
|
2015-02-05 14:39:56 +00:00
|
|
|
|
|
2015-08-25 20:08:25 +00:00
|
|
|
|
// If any tree characters have been printed, then add an extra
|
|
|
|
|
// space, which makes the output look much better.
|
|
|
|
|
if row.depth != 0 {
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
filename.add_spaces(1);
|
2015-02-05 14:39:56 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-02-26 08:27:29 +00:00
|
|
|
|
// Print the name without worrying about padding.
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
filename.append(row.name);
|
2015-02-24 16:05:25 +00:00
|
|
|
|
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
cell.append(filename);
|
2015-06-28 12:21:21 +00:00
|
|
|
|
cells.push(cell);
|
2015-02-05 14:39:56 +00:00
|
|
|
|
}
|
2015-06-28 12:21:21 +00:00
|
|
|
|
|
|
|
|
|
cells
|
2015-02-05 14:39:56 +00:00
|
|
|
|
}
|
2015-02-26 07:18:08 +00:00
|
|
|
|
}
|
2015-02-05 14:39:56 +00:00
|
|
|
|
|
2015-06-08 20:33:39 +00:00
|
|
|
|
|
2015-11-15 19:26:58 +00:00
|
|
|
|
lazy_static! {
|
|
|
|
|
static ref DATE_AND_TIME: DateFormat<'static> =
|
|
|
|
|
DateFormat::parse("{2>:D} {:M} {2>:h}:{02>:m}").unwrap();
|
|
|
|
|
|
|
|
|
|
static ref DATE_AND_YEAR: DateFormat<'static> =
|
|
|
|
|
DateFormat::parse("{2>:D} {:M} {5>:Y}").unwrap();
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
2015-05-16 20:02:28 +00:00
|
|
|
|
#[cfg(test)]
|
|
|
|
|
pub mod test {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
pub use super::{Table, Environment, Details};
|
|
|
|
|
pub use std::sync::Mutex;
|
|
|
|
|
|
2016-04-16 17:59:25 +00:00
|
|
|
|
pub use fs::{File, fields as f};
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
pub use output::column::{Column, Columns};
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
pub use output::cell::TextCell;
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
|
|
|
|
pub use users::{User, Group, uid_t, gid_t};
|
|
|
|
|
pub use users::mock::MockUsers;
|
2016-02-10 15:22:43 +00:00
|
|
|
|
pub use users::os::unix::{UserExt, GroupExt};
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
2015-05-29 19:39:45 +00:00
|
|
|
|
pub use ansi_term::Style;
|
2015-05-16 20:02:28 +00:00
|
|
|
|
pub use ansi_term::Colour::*;
|
|
|
|
|
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
impl Default for Environment<MockUsers> {
|
|
|
|
|
fn default() -> Self {
|
|
|
|
|
use locale;
|
|
|
|
|
use users::mock::MockUsers;
|
|
|
|
|
use std::sync::Mutex;
|
|
|
|
|
|
|
|
|
|
Environment {
|
|
|
|
|
current_year: 1234,
|
|
|
|
|
numeric: locale::Numeric::english(),
|
|
|
|
|
time: locale::Time::english(),
|
2016-03-31 20:19:29 +00:00
|
|
|
|
tz: None,
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
users: Mutex::new(MockUsers::with_current_uid(0)),
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2016-02-10 16:08:15 +00:00
|
|
|
|
pub fn new_table<'a>(columns: &'a [Column], details: &'a Details, users: MockUsers) -> Table<'a, MockUsers> {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
use std::sync::Arc;
|
|
|
|
|
|
|
|
|
|
Table {
|
|
|
|
|
columns: columns,
|
|
|
|
|
opts: details,
|
2016-02-10 16:08:15 +00:00
|
|
|
|
env: Arc::new(Environment { users: Mutex::new(users), ..Environment::default() }),
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
rows: Vec::new(),
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-05-16 20:02:28 +00:00
|
|
|
|
mod users {
|
2015-09-15 19:05:27 +00:00
|
|
|
|
#![allow(unused_results)]
|
2015-05-16 20:02:28 +00:00
|
|
|
|
use super::*;
|
|
|
|
|
|
|
|
|
|
#[test]
|
|
|
|
|
fn named() {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let columns = Columns::default().for_dir(None);
|
|
|
|
|
let mut details = Details::default();
|
|
|
|
|
details.colours.users.user_you = Red.bold();
|
|
|
|
|
|
2015-05-16 20:02:28 +00:00
|
|
|
|
let mut users = MockUsers::with_current_uid(1000);
|
2016-02-10 15:22:43 +00:00
|
|
|
|
users.add_user(User::new(1000, "enoch", 100));
|
2016-02-10 16:08:15 +00:00
|
|
|
|
|
|
|
|
|
let table = new_table(&columns, &details, users);
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
|
|
|
|
let user = f::User(1000);
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let expected = TextCell::paint_str(Red.bold(), "enoch");
|
2015-05-16 20:02:28 +00:00
|
|
|
|
assert_eq!(expected, table.render_user(user))
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[test]
|
|
|
|
|
fn unnamed() {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let columns = Columns::default().for_dir(None);
|
|
|
|
|
let mut details = Details::default();
|
|
|
|
|
details.colours.users.user_you = Cyan.bold();
|
|
|
|
|
|
2015-05-16 20:02:28 +00:00
|
|
|
|
let users = MockUsers::with_current_uid(1000);
|
2016-02-10 16:08:15 +00:00
|
|
|
|
|
|
|
|
|
let table = new_table(&columns, &details, users);
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
|
|
|
|
let user = f::User(1000);
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let expected = TextCell::paint_str(Cyan.bold(), "1000");
|
2015-05-16 20:02:28 +00:00
|
|
|
|
assert_eq!(expected, table.render_user(user));
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[test]
|
|
|
|
|
fn different_named() {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let columns = Columns::default().for_dir(None);
|
|
|
|
|
let mut details = Details::default();
|
|
|
|
|
details.colours.users.user_someone_else = Green.bold();
|
|
|
|
|
|
2016-02-10 16:08:15 +00:00
|
|
|
|
let table = new_table(&columns, &details, MockUsers::with_current_uid(0));
|
2016-02-10 15:22:43 +00:00
|
|
|
|
table.env.users.lock().unwrap().add_user(User::new(1000, "enoch", 100));
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
|
|
|
|
let user = f::User(1000);
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let expected = TextCell::paint_str(Green.bold(), "enoch");
|
2015-05-16 20:02:28 +00:00
|
|
|
|
assert_eq!(expected, table.render_user(user));
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[test]
|
|
|
|
|
fn different_unnamed() {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let columns = Columns::default().for_dir(None);
|
|
|
|
|
let mut details = Details::default();
|
|
|
|
|
details.colours.users.user_someone_else = Red.normal();
|
|
|
|
|
|
2016-02-10 16:08:15 +00:00
|
|
|
|
let table = new_table(&columns, &details, MockUsers::with_current_uid(0));
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
|
|
|
|
let user = f::User(1000);
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let expected = TextCell::paint_str(Red.normal(), "1000");
|
2015-05-16 20:02:28 +00:00
|
|
|
|
assert_eq!(expected, table.render_user(user));
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[test]
|
|
|
|
|
fn overflow() {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let columns = Columns::default().for_dir(None);
|
|
|
|
|
let mut details = Details::default();
|
|
|
|
|
details.colours.users.user_someone_else = Blue.underline();
|
|
|
|
|
|
2016-02-10 16:08:15 +00:00
|
|
|
|
let table = new_table(&columns, &details, MockUsers::with_current_uid(0));
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
|
|
|
|
let user = f::User(2_147_483_648);
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let expected = TextCell::paint_str(Blue.underline(), "2147483648");
|
2015-05-16 20:02:28 +00:00
|
|
|
|
assert_eq!(expected, table.render_user(user));
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
mod groups {
|
2015-09-15 19:05:27 +00:00
|
|
|
|
#![allow(unused_results)]
|
2015-05-16 20:02:28 +00:00
|
|
|
|
use super::*;
|
|
|
|
|
|
|
|
|
|
#[test]
|
|
|
|
|
fn named() {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let columns = Columns::default().for_dir(None);
|
|
|
|
|
let mut details = Details::default();
|
|
|
|
|
details.colours.users.group_not_yours = Fixed(101).normal();
|
|
|
|
|
|
2015-05-16 20:02:28 +00:00
|
|
|
|
let mut users = MockUsers::with_current_uid(1000);
|
2016-02-10 15:22:43 +00:00
|
|
|
|
users.add_group(Group::new(100, "folk"));
|
2016-02-10 16:08:15 +00:00
|
|
|
|
let table = new_table(&columns, &details, users);
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
|
|
|
|
let group = f::Group(100);
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let expected = TextCell::paint_str(Fixed(101).normal(), "folk");
|
2015-05-16 20:02:28 +00:00
|
|
|
|
assert_eq!(expected, table.render_group(group))
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[test]
|
|
|
|
|
fn unnamed() {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let columns = Columns::default().for_dir(None);
|
|
|
|
|
let mut details = Details::default();
|
|
|
|
|
details.colours.users.group_not_yours = Fixed(87).normal();
|
|
|
|
|
|
2015-05-16 20:02:28 +00:00
|
|
|
|
let users = MockUsers::with_current_uid(1000);
|
2016-02-10 16:08:15 +00:00
|
|
|
|
let table = new_table(&columns, &details, users);
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
|
|
|
|
let group = f::Group(100);
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let expected = TextCell::paint_str(Fixed(87).normal(), "100");
|
2015-05-16 20:02:28 +00:00
|
|
|
|
assert_eq!(expected, table.render_group(group));
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[test]
|
|
|
|
|
fn primary() {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let columns = Columns::default().for_dir(None);
|
|
|
|
|
let mut details = Details::default();
|
|
|
|
|
details.colours.users.group_yours = Fixed(64).normal();
|
|
|
|
|
|
2015-05-16 20:02:28 +00:00
|
|
|
|
let mut users = MockUsers::with_current_uid(2);
|
2016-02-10 15:22:43 +00:00
|
|
|
|
users.add_user(User::new(2, "eve", 100));
|
|
|
|
|
users.add_group(Group::new(100, "folk"));
|
2016-02-10 16:08:15 +00:00
|
|
|
|
|
|
|
|
|
let table = new_table(&columns, &details, users);
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
|
|
|
|
let group = f::Group(100);
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let expected = TextCell::paint_str(Fixed(64).normal(), "folk");
|
2015-05-16 20:02:28 +00:00
|
|
|
|
assert_eq!(expected, table.render_group(group))
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[test]
|
|
|
|
|
fn secondary() {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let columns = Columns::default().for_dir(None);
|
|
|
|
|
let mut details = Details::default();
|
|
|
|
|
details.colours.users.group_yours = Fixed(31).normal();
|
|
|
|
|
|
2015-05-16 20:02:28 +00:00
|
|
|
|
let mut users = MockUsers::with_current_uid(2);
|
2016-02-10 15:22:43 +00:00
|
|
|
|
users.add_user(User::new(2, "eve", 666));
|
|
|
|
|
|
|
|
|
|
let test_group = Group::new(100, "folk").add_member("eve");
|
|
|
|
|
users.add_group(test_group);
|
2016-02-10 16:08:15 +00:00
|
|
|
|
|
|
|
|
|
let table = new_table(&columns, &details, users);
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
|
|
|
|
let group = f::Group(100);
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let expected = TextCell::paint_str(Fixed(31).normal(), "folk");
|
2015-05-16 20:02:28 +00:00
|
|
|
|
assert_eq!(expected, table.render_group(group))
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[test]
|
|
|
|
|
fn overflow() {
|
Use Mutex lock on only the users columns
This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.
The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)
Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).
2016-01-16 21:56:37 +00:00
|
|
|
|
let columns = Columns::default().for_dir(None);
|
|
|
|
|
let mut details = Details::default();
|
|
|
|
|
details.colours.users.group_not_yours = Blue.underline();
|
|
|
|
|
|
2016-02-10 16:08:15 +00:00
|
|
|
|
let table = new_table(&columns, &details, MockUsers::with_current_uid(0));
|
2015-05-16 20:02:28 +00:00
|
|
|
|
|
|
|
|
|
let group = f::Group(2_147_483_648);
|
Replace Cells with growable TextCells
A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:
1. Replace the `Cell` type with a `TextCell` type that holds a vector of
`ANSIString` values instead of a formatted string. It still does the
width tracking.
2. Rework the details module's `render` functions to emit values of this
type.
3. Similarly, rework the functions that produce cells containing filenames
to use a `File` value's `name` field, which is an owned `String` that
can now be re-used.
4. Update the printing, formatting, and width-calculating code in the
details and grid-details views to produce a table by adding vectors
together instead of adding strings together, delaying the formatting as
long as it can.
This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).
This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.
Casualties include:
- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
`ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
than a borrowed slice, so the filename cells produced now own the filename
strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
file produced refers to the same directory, rather than some phantom
directory with the same lifetime as the file. (This was wrong from the
start, but it broke nothing until now)
References:
[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0
2015-12-17 00:25:20 +00:00
|
|
|
|
let expected = TextCell::paint_str(Blue.underline(), "2147483648");
|
2015-05-16 20:02:28 +00:00
|
|
|
|
assert_eq!(expected, table.render_group(group));
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|