diff --git a/ChangeLog b/ChangeLog index 861ff64c..44396f55 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,5 +1,9 @@ 2019-01-29 Jay Berkenbilt + * Bug fix: when returning an inline image token, the tokenizer no + longer includes the delimiter that follows EI. The + QPDFObjectHandle created from the token was correct. + * Handle files with direct page objects, which is not allowed by the PDF spec but has been seen in the wild. Fixes #164. diff --git a/libqpdf/QPDFObjectHandle.cc b/libqpdf/QPDFObjectHandle.cc index 4e77bbaa..ecaa49bd 100644 --- a/libqpdf/QPDFObjectHandle.cc +++ b/libqpdf/QPDFObjectHandle.cc @@ -1571,9 +1571,9 @@ QPDFObjectHandle::parseContentStream_data( else { // Skip back over EI - input->seek(-3, SEEK_CUR); + input->seek(-2, SEEK_CUR); std::string inline_image = t.getRawValue(); - for (int i = 0; i < 4; ++i) + for (int i = 0; i < 3; ++i) { if (inline_image.length() > 0) { diff --git a/libqpdf/QPDFTokenizer.cc b/libqpdf/QPDFTokenizer.cc index e1399d82..c11c8218 100644 --- a/libqpdf/QPDFTokenizer.cc +++ b/libqpdf/QPDFTokenizer.cc @@ -468,6 +468,7 @@ QPDFTokenizer::presentCharacter(char ch) } else if (this->m->state == st_inline_image) { + this->m->val += ch; size_t len = this->m->val.length(); if ((len >= 4) && isDelimiter(this->m->val.at(len-4)) && @@ -475,22 +476,18 @@ QPDFTokenizer::presentCharacter(char ch) (this->m->val.at(len-2) == 'I') && isDelimiter(this->m->val.at(len-1))) { + this->m->val.erase(len - 1); this->m->type = tt_inline_image; this->m->unread_char = true; this->m->char_to_unread = ch; this->m->state = st_token_ready; } - else - { - this->m->val += ch; - } } else { handled = false; } - if (handled) { // okay diff --git a/qpdf/qtest/qpdf/tokens-maxlen.out b/qpdf/qtest/qpdf/tokens-maxlen.out index eb5dd8d7..e7942b58 100644 --- a/qpdf/qtest/qpdf/tokens-maxlen.out +++ b/qpdf/qtest/qpdf/tokens-maxlen.out @@ -599,7 +599,8 @@ skipping to endstream 103: dict_close: >> 105: space: \x0a 106: word: ID -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\x0a +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 +354: space: \x0a 355: word: BT 357: space: \x0a 360: name: /F1 diff --git a/qpdf/qtest/qpdf/tokens-no-ignorable.out b/qpdf/qtest/qpdf/tokens-no-ignorable.out index ca977b8c..3216cab5 100644 --- a/qpdf/qtest/qpdf/tokens-no-ignorable.out +++ b/qpdf/qtest/qpdf/tokens-no-ignorable.out @@ -291,7 +291,7 @@ skipping to endstream 101: integer: 66 103: dict_close: >> 106: word: ID -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\x0a +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 355: word: BT 360: name: /F1 364: integer: 24 diff --git a/qpdf/qtest/qpdf/tokens.out b/qpdf/qtest/qpdf/tokens.out index 9151f593..6ea927d4 100644 --- a/qpdf/qtest/qpdf/tokens.out +++ b/qpdf/qtest/qpdf/tokens.out @@ -599,7 +599,8 @@ skipping to endstream 103: dict_close: >> 105: space: \x0a 106: word: ID -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\x0a +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 +354: space: \x0a 355: word: BT 357: space: \x0a 360: name: /F1