OpenSystemsDevelopment / qpdf

Commit 1eb35a355f12f2edf69868b749f58c08ef58f1c6

Authored by Jay Berkenbilt
1 parent 2b6c79bc

Exclude space after ID in image data

Inline Side-by-side
Showing 7 changed files with 33 additions and 19 deletions
ChangeLog
  View file @1eb35a3
1 1 2019-01-30 Jay Berkenbilt <ejb@ql.org>
2 2  
  3 + * Do not include space after ID operator in inline image data. The
  4 + token now correctly contains the image data, the EI operator,
  5 + and the delimiter that precedes the EI operator.
  6 +
3 7 * Improve locating of an inline image's EI operator to correctly
4 8 handle the case of EI appearing inside the image data.
5 9  
... ...
include/qpdf/QPDFTokenizer.hh
  View file @1eb35a3
... ... @@ -174,17 +174,19 @@ class QPDFTokenizer
174 174 size_t max_len = 0);
175 175  
176 176 // Calling this method puts the tokenizer in a state for reading
177   - // inline images. In that state, it will return all data up to and
178   - // including the next EI token. After you call this method, the
179   - // next call to readToken (or the token created next time getToken
180   - // returns true) will either be tt_inline_image or tt_bad. This is
181   - // the only way readToken returns a tt_inline_image token. The
182   - // version of this method that takes a PointerHolder<InputSource>
183   - // does a better job of locating the end of the inline image and
184   - // should be used whenever the input source is available. It
185   - // preserves both tell() and getLastOffset(). The version without
186   - // the input source will always end the inline image the first
187   - // time it sees something that looks like an EI operator.
  177 + // inline images. You should call this method after reading the
  178 + // character following the ID operator. In that state, it will
  179 + // return all data up to and including the next EI token. After
  180 + // you call this method, the next call to readToken (or the token
  181 + // created next time getToken returns true) will either be
  182 + // tt_inline_image or tt_bad. This is the only way readToken
  183 + // returns a tt_inline_image token. The version of this method
  184 + // that takes a PointerHolder<InputSource> does a better job of
  185 + // locating the end of the inline image and should be used
  186 + // whenever the input source is available. It preserves both
  187 + // tell() and getLastOffset(). The version without the input
  188 + // source will always end the inline image the first time it sees
  189 + // something that looks like an EI operator.
188 190 QPDF_DLL
189 191 void expectInlineImage(PointerHolder<InputSource> input);
190 192 QPDF_DLL
... ...
libqpdf/Pl_QPDFTokenizer.cc
  View file @1eb35a3
... ... @@ -59,6 +59,12 @@ Pl_QPDFTokenizer::finish()
59 59 else if ((token.getType() == QPDFTokenizer::tt_word) &&
60 60 (token.getValue() == "ID"))
61 61 {
  62 + // Read the space after the ID.
  63 + char ch = ' ';
  64 + input->read(&ch, 1);
  65 + this->m->filter->handleToken(
  66 + QPDFTokenizer::Token(
  67 + QPDFTokenizer::tt_space, std::string(1, ch)));
62 68 QTC::TC("qpdf", "Pl_QPDFTokenizer found ID");
63 69 this->m->tokenizer.expectInlineImage(input);
64 70 }
... ...
qpdf/qtest/qpdf/tokens-maxlen.out
  View file @1eb35a3
... ... @@ -599,7 +599,7 @@ skipping to endstream
599 599 103: dict_close: >>
600 600 105: space: \x0a
601 601 106: word: ID
602   -108: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
  602 +109: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
603 603 354: space: \x0a
604 604 355: word: BT
605 605 357: space: \x0a
... ... @@ -759,7 +759,7 @@ skipping to endstream
759 759 --- BEGIN PAGE 6 ---
760 760 0: word: ID
761 761 EI not found; resuming normal scanning
762   -2: space: \x0a
  762 +3: space:
763 763 5: name: /F1
764 764 8: space:
765 765 9: integer: 24
... ... @@ -810,7 +810,7 @@ EI not found; resuming normal scanning
810 810 67: dict_close: >>
811 811 69: space: \x0a
812 812 70: word: ID
813   -72: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
  813 +73: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
814 814 318: eof
815 815 --- END PAGE 7 ---
816 816 --- BEGIN PAGE 8 ---
... ...
qpdf/qtest/qpdf/tokens-no-ignorable.out
  View file @1eb35a3
... ... @@ -291,7 +291,7 @@ skipping to endstream
291 291 101: integer: 66
292 292 103: dict_close: >>
293 293 106: word: ID
294   -108: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
  294 +109: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
295 295 355: word: BT
296 296 360: name: /F1
297 297 364: integer: 24
... ... @@ -414,7 +414,7 @@ EI not found; resuming normal scanning
414 414 65: integer: 66
415 415 67: dict_close: >>
416 416 70: word: ID
417   -72: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
  417 +73: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
418 418 318: eof
419 419 --- END PAGE 7 ---
420 420 --- BEGIN PAGE 8 ---
... ...
qpdf/qtest/qpdf/tokens.out
  View file @1eb35a3
... ... @@ -599,7 +599,7 @@ skipping to endstream
599 599 103: dict_close: >>
600 600 105: space: \x0a
601 601 106: word: ID
602   -108: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
  602 +109: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
603 603 354: space: \x0a
604 604 355: word: BT
605 605 357: space: \x0a
... ... @@ -759,7 +759,7 @@ skipping to endstream
759 759 --- BEGIN PAGE 6 ---
760 760 0: word: ID
761 761 EI not found; resuming normal scanning
762   -2: space: \x0a
  762 +3: space:
763 763 5: name: /F1
764 764 8: space:
765 765 9: integer: 24
... ... @@ -810,7 +810,7 @@ EI not found; resuming normal scanning
810 810 67: dict_close: >>
811 811 69: space: \x0a
812 812 70: word: ID
813   -72: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
  813 +73: inline-image: x\x9c\xc5\xd6I\x0e\xc3 \x0c\x05P|\xffC;U\xc8`\xc0\xd37\x91Z\xa9\x0b\xa6\x17\x02\xc4\x98\xda\xe6\x8f\x1b}D\xf0\xef_\xb4\xf8\x1c\xc9W\xa9\x84\x9c\xc4-\x94\x88>\xff\x87\xc0\x8d>\x94^\x01&\xae\xa1u\xe2]\x80"!\x87\x95\x08\x96\x05*\xac&\x8fE|Sy\xae \xf0d-\x80<\x9d\x19B\x010B\x05\xfa@N\x11\xea+<\x1fhl\xe8K\xd0\xee/56L\xa0\x89\x90\xe3\x19\x1e \xa3\x96\xb9\xa6>0\x06>\x15Y\x81\xf9!c\xec\x0eY\x0c\xd8\x0f%Y\xf0\x01\xa5\xd68?&\xa0\xd6\xeb\x88}j\x92\xfb\xe8\x1d;\xab\x8d3\x9d\xc2\xd6l\x14p\xdbsH\xf6\xfbt\xfa\x01Q\x02\xd8Tt*h\xccU\xfa\xe3w\x07\xcd\xd5\xd0%\xa8)p\x96\xb3"\x95DiRj\xb9\x96D\x18YNU\x11\xd3\xd9Av\x92F\xe0&\x0d\x90\xcd\xd4u#c\x95\xc6W\x09\xf4\xdf\x89\x03W\x93O\x0d\x0aEI
814 814 318: eof
815 815 --- END PAGE 7 ---
816 816 --- BEGIN PAGE 8 ---
... ...
qpdf/test_tokenizer.cc
  View file @1eb35a3
... ... @@ -183,6 +183,8 @@ dump_tokens(PointerHolder&lt;InputSource&gt; is, std::string const&amp; label,
183 183 else if (skip_inline_images &&
184 184 (token == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "ID")))
185 185 {
  186 + char ch;
  187 + is->read(&ch, 1);
186 188 if (old_ei)
187 189 {
188 190 tokenizer.expectInlineImage();
... ...