octoleo/qpdf
2
1
Fork 0
mirror of https://github.com/qpdf/qpdf.git synced 2024-12-22 10:58:58 +00:00
Code Issues Releases Wiki Activity

Add private method QPDFTokenizer::inString

Browse Source
This commit is contained in:
m-holger 2022-08-18 19:16:51 +01:00
parent 2697ba49bc
commit d26b537a7c
2 changed files with 78 additions and 64 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
include/qpdf/QPDFTokenizer.hh
View File

@ -213,6 +213,7 @@ class QPDFTokenizer
void handleCharacter(char);
void inHexstring(char);
void inString(char, size_t);
void reset();
// Lexer state

141
libqpdf/QPDFTokenizer.cc
View File

@ -368,70 +368,7 @@ QPDFTokenizer::handleCharacter(char ch)
bs_num_count = 0;
}
if (this->string_ignoring_newline && (ch == '\n')) {
// ignore
this->string_ignoring_newline = false;
} else if (
ch_is_octal && (this->last_char_was_bs || (bs_num_count > 0))) {
this->bs_num_register[bs_num_count++] = ch;
} else if (this->last_char_was_bs) {
switch (ch) {
case 'n':
this->val += '\n';
break;
case 'r':
this->val += '\r';
break;
case 't':
this->val += '\t';
break;
case 'b':
this->val += '\b';
break;
case 'f':
this->val += '\f';
break;
case '\n':
break;
case '\r':
this->string_ignoring_newline = true;
break;
default:
// PDF spec says backslash is ignored before anything else
this->val += ch;
break;
}
} else if (ch == '\\') {
// last_char_was_bs is set/cleared below as appropriate
if (bs_num_count) {
throw std::logic_error(
"INTERNAL ERROR: QPDFTokenizer: bs_num_count != 0 "
"when ch == '\\'");
}
} else if (ch == '(') {
this->val += ch;
++this->string_depth;
} else if ((ch == ')') && (--this->string_depth == 0)) {
this->type = tt_string;
this->state = st_token_ready;
} else if (ch == '\r') {
// CR by itself is converted to LF
this->val += '\n';
} else if (ch == '\n') {
// CR LF is converted to LF
if (!this->last_char_was_cr) {
this->val += ch;
}
} else {
this->val += ch;
}
inString(ch, bs_num_count);
this->last_char_was_cr =
((!this->string_ignoring_newline) && (ch == '\r'));
@ -514,6 +451,82 @@ QPDFTokenizer::inHexstring(char ch)
}
}
void
QPDFTokenizer::inString(char ch, size_t bs_num_count)
{
bool ch_is_octal = ((ch >= '0') && (ch <= '7'));
if (this->string_ignoring_newline && (ch == '\n')) {
// ignore
this->string_ignoring_newline = false;
return;
} else if (ch_is_octal && (this->last_char_was_bs || (bs_num_count > 0))) {
this->bs_num_register[bs_num_count++] = ch;
return;
} else if (this->last_char_was_bs) {
switch (ch) {
case 'n':
this->val += '\n';
return;
case 'r':
this->val += '\r';
return;
case 't':
this->val += '\t';
return;
case 'b':
this->val += '\b';
return;
case 'f':
this->val += '\f';
return;
case '\n':
return;
case '\r':
this->string_ignoring_newline = true;
return;
default:
// PDF spec says backslash is ignored before anything else
this->val += ch;
return;
}
} else if (ch == '\\') {
// last_char_was_bs is set/cleared below as appropriate
if (bs_num_count) {
throw std::logic_error(
"INTERNAL ERROR: QPDFTokenizer: bs_num_count != 0 "
"when ch == '\\'");
}
} else if (ch == '(') {
this->val += ch;
++this->string_depth;
return;
} else if ((ch == ')') && (--this->string_depth == 0)) {
this->type = tt_string;
this->state = st_token_ready;
return;
} else if (ch == '\r') {
// CR by itself is converted to LF
this->val += '\n';
return;
} else if (ch == '\n') {
// CR LF is converted to LF
if (!this->last_char_was_cr) {
this->val += ch;
}
return;
} else {
this->val += ch;
return;
}
}
void
QPDFTokenizer::presentEOF()
{