We were using SGML entities for various non-ASCII characters so they
could convert properly for both HTML and print, but this is no longer
necessary as we move from docbook to RST, so just replace them. Note
that the conversions done by sphinx automatically handle "smart
quotes", so it works to just use regular quotes in place of “
and ”.
Pandoc docbook -> rst fails to convert the following elements, so
change them to @1@tag@1@ ... @2@tag@2@ for later processing. This way,
they will survive the conversion, and we can deal with them later.
<application>
<command>
<filename>
<firstterm>
<option>
<replaceable>
With docbook, this was not converted properly in the PDF version, but
since we are moving out of docbook, we can just put the Unicode
character in the source.
* Handle error conditions that occur when using the object handle
interfaces. In the past, some exceptions were not correctly
converted to errors or warnings.
* Add more detailed information to qpdf-c.h
* Make it possible to work more explicitly with uninitialized objects
The impact on the code would be extremely high, and using it would
clutter the code greatly because it would break chaining like
a.getKey("/B").getKey("/C"). There are better ways to deal with the
issue.
Don't assume endobj is at the beginning of the line. This means we are
looking at tokens for every line, but the odds of n n obj appearing in
the middle of the object are likely much lower than endobj not being
at the beginning of the line or missing entirely. This will probably
have a negative impact on recovery time for very large files.
Hopefully it will be worth it.
When making resources indirect in from_dr, the code was using the
wrong owning QPDF, forgetting that from_dr had already been copied
using CopyForeignObject.
When adding a QPDFObjectHandle to an array or dictionary, if possible,
check if the new object belongs to the same QPDF. This makes it much
easier to find incorrect code than waiting for the situation to be
detected when the file is written.
