Merge branch 'main' into mslichao/capifreebuf

.idea/cmake.xml

@@ -2,7 +2,6 @@
 <project version="4">
   <component name="CMakeSharedSettings">
     <configurations>
-      <configuration PROFILE_NAME="Debug" ENABLED="true" CONFIG_NAME="Debug" />
       <configuration PROFILE_NAME="Maintainer" ENABLED="true" CONFIG_NAME="RelWithDebInfo" GENERATION_OPTIONS="-DMAINTAINER_MODE=ON -DBUILD_STATIC_LIBS=OFF" />
       <configuration PROFILE_NAME="Windows" ENABLED="true" CONFIG_NAME="RelWithDebInfo" TOOLCHAIN_NAME="Visual Studio" GENERATION_OPTIONS="-DBUILD_SHARED_LIBS=OFF" />
     </configurations>

ChangeLog

@@ -1,7 +1,13 @@
 2024-09-20  Chao Li  <mslichao@outlook.com>
 
 	* Add C API qpdf_oh_free_buffer to release memory allocated by
-	stream data functions
+	stream data functions.
+
+2024-08-25  M Holger  <m.holger@qpdf.org>
+
+	* Add new command-line arguments --remove-metadata and --remove-info
+	to exclude document metadata and information from the output PDF
+	file. Patially fixes #1145.
 
 2024-08-06  M Holger  <m.holger@qpdf.org>
 

fuzz/CMakeLists.txt

@@ -142,6 +142,10 @@ set(CORPUS_OTHER
   70306b.fuzz
   71624.fuzz
   71689.fuzz
+  99999a.fuzz
+  99999b.fuzz
+  99999c.fuzz
+  99999d.fuzz
 )
 
 set(CORPUS_DIR ${CMAKE_CURRENT_BINARY_DIR}/qpdf_corpus)

fuzz/qpdf_extra/99999a.fuzz

@@ -0,0 +1,63 @@
+%PDF-1.5
+%€€€€
+1 0 obj
+<<
+	/Type /Catalog
+	/Pages 2 0 R
+>>
+endobj
+2 0 obj
+<<
+	/Count 6  Ri
+	0K/ds [3 0 R]
+	/Type /Pages
+>>
+endobj
+3 0 obj
+<<
+	/Resources <<
+		/Font <<
+			/F1 5 0 R
+		>>
+	>>
+	/MediaBox [0 0 795 842]
+	/Parent 2 0 R
+	/Contents 4 0 R
+	/Type /Page
+=>
+endobj
+4 0 obj
+<<444444444444444444444444 1 Tr /F1 30 Tf 350 750 Td (foobar) Tj ET
+endstream
+endobj
+5 0 obj
+<<
+	/Name /F1
+	/BaseFont /Helvetica
+	/Type /Font
+	/Subtype /Type1
+>>
+e„dobj
+6 0 obj
+<< /Length 6 0 R >>
+stre444444444444444444444444444444<<>>
+endobj
+xref
+0 8
+0000000000 65535 f 
+0000000015 00000 n 
+0000000066 00000 n 
+0000000130 00000 n 
+0000000269 00000 n 
+0000000362 00000 n 
+000000ÎËËÉßÏÏÏ00 n 
+0000000500 00000 n 
+trailer
+<<
+	/Size 713115528178535
+	/Root 1 0 R
+	/Info 7 0 R
+>>
+startxref
+520
+%%EOF

fuzz/qtest/fuzz.test

@@ -11,7 +11,7 @@ my $td = new TestDriver('fuzz');
 
 my $qpdf_corpus = $ENV{'QPDF_FUZZ_CORPUS'} || die "must set QPDF_FUZZ_CORPUS";
 
-my $n_qpdf_files = 79;       # increment when adding new files
+my $n_qpdf_files = 83;       # increment when adding new files
 
 my @fuzzers = (
     ['ascii85' => 1],

include/qpdf/QPDFJob.hh

@@ -692,6 +692,8 @@ class QPDFJob
         bool optimize_images{false};
         bool externalize_inline_images{false};
         bool keep_inline_images{false};
+        bool remove_info{false};
+        bool remove_metadata{false};
         bool remove_page_labels{false};
         size_t oi_min_width{DEFAULT_OI_MIN_WIDTH};
         size_t oi_min_height{DEFAULT_OI_MIN_HEIGHT};

include/qpdf/auto_job_c_main.hh

@@ -32,6 +32,8 @@ QPDF_DLL Config* progress();
 QPDF_DLL Config* qdf();
 QPDF_DLL Config* rawStreamData();
 QPDF_DLL Config* recompressFlate();
+QPDF_DLL Config* removeInfo();
+QPDF_DLL Config* removeMetadata();
 QPDF_DLL Config* removePageLabels();
 QPDF_DLL Config* reportMemoryUsage();
 QPDF_DLL Config* requiresPassword();

job.sums

@@ -4,17 +4,17 @@ generate_auto_job f64733b79dcee5a0e3e8ccc6976448e8ddf0e8b6529987a66a7d3ab2ebc10a
 include/qpdf/auto_job_c_att.hh 4c2b171ea00531db54720bf49a43f8b34481586ae7fb6cbf225099ee42bc5bb4
 include/qpdf/auto_job_c_copy_att.hh 50609012bff14fd82f0649185940d617d05d530cdc522185c7f3920a561ccb42
 include/qpdf/auto_job_c_enc.hh 28446f3c32153a52afa239ea40503e6cc8ac2c026813526a349e0cd4ae17ddd5
-include/qpdf/auto_job_c_main.hh dbfc221d1533120d1aa9c361d8d2483dea5fcb1c0fd95144d98d305e64ed32a6
+include/qpdf/auto_job_c_main.hh 84f463237235b2c095b747a4f5dd00f109ee596a1c207b944efb296c0c568cae
 include/qpdf/auto_job_c_pages.hh 09ca15649cc94fdaf6d9bdae28a20723f2a66616bf15aa86d83df31051d82506
 include/qpdf/auto_job_c_uo.hh 9c2f98a355858dd54d0bba444b73177a59c9e56833e02fa6406f429c07f39e62
-job.yml 53cad86659db6722e8f415aacb19fc51ab81bb1589c3cb8f65ec893bb4bf5566
+job.yml 31935064eca625af7657b23f2f12c614d14751ec0b12702482b1768a04905d22
 libqpdf/qpdf/auto_job_decl.hh 20d6affe1e260f5a1af4f1d82a820b933835440ff03020e877382da2e8dac6c6
-libqpdf/qpdf/auto_job_help.hh 74b2982771720927ce7be8f1690720ec65cb9989620493a0c154f50ba2c254e4
-libqpdf/qpdf/auto_job_init.hh 19d1da7c4c0c635bd1c5db8d5f17df8edad3442f8eba006adb075cec295fa158
+libqpdf/qpdf/auto_job_help.hh 1e9181f4729a22ff91ab54e2b4a82e6af0c57a8327efb222a4196adb609c1ade
+libqpdf/qpdf/auto_job_init.hh e2a6bb87870c5522a01b15461c9fe909e360f5c7fed06e41acf13a125bd1d03e
 libqpdf/qpdf/auto_job_json_decl.hh 843892c8e8652a86b7eb573893ef24050b7f36fe313f7251874be5cd4cdbe3fd
-libqpdf/qpdf/auto_job_json_init.hh a87256c082427ec0318223762472970b2eced535c0c8b0288d45c8cdaaf62f74
-libqpdf/qpdf/auto_job_schema.hh 5dac568dff39614e161a0af59a0f328f1e28edf69b96f08bb76fd592d51bb053
+libqpdf/qpdf/auto_job_json_init.hh 344c2fb473f88fe829c93b1efe6c70a0e4796537b8eb35e421d955fff481ba7d
+libqpdf/qpdf/auto_job_schema.hh 6d3eef5137b8828eaa301a1b3cf75cb7bb812aa6e2d8301de865b42d238d7a7c
 manual/_ext/qpdf.py 6add6321666031d55ed4aedf7c00e5662bba856dfcd66ccb526563bffefbb580
-manual/cli.rst 94057baba9ecffb4ce19ae61c8fa507ef07209c280fccae97b283c3dfce834e0
-manual/qpdf.1 0ec05f1392c160165cdf6adada4de84c0de75bd2fb5762caff4e1372aacada4c
+manual/cli.rst b7f37995f13346518ae7b2ea84836fba13b4da4e1f55be5f2a861f20dea0ccdb
+manual/qpdf.1 59c26635017cba5d142ec3fcc4aebcb91e0cf1355d51365db84f48b21585ad8d
 manual/qpdf.1.in 436ecc85d45c4c9e2dbd1725fb7f0177fb627179469f114561adf3cb6cbb677b

job.yml

@@ -130,6 +130,8 @@ options:
       - qdf
       - raw-stream-data
       - recompress-flate
+      - remove-info
+      - remove-metadata
       - remove-page-labels
       - replace-input
       - report-memory-usage
@@ -440,6 +442,8 @@ json:
     - Pages.file:
       Pages.password:
       range:
+  remove-info:
+  remove-metadata:
   remove-page-labels:
   report-memory-usage:
   rotate:

libqpdf/QPDF.cc

@@ -832,10 +832,6 @@ std::vector<QPDF::Xref_table::Subsection>
 QPDF::Xref_table::bad_subsections(std::string& line, qpdf_offset_t start)
 {
     std::vector<QPDF::Xref_table::Subsection> result;
-    qpdf_offset_t f1 = 0;
-    int f2 = 0;
-    char type = '\0';
-
     file->seek(start, SEEK_SET);
 
     while (true) {
@@ -844,7 +840,7 @@ QPDF::Xref_table::bad_subsections(std::string& line, qpdf_offset_t start)
         auto [obj, num, offset] = result.emplace_back(subsection(line));
         file->seek(offset, SEEK_SET);
         for (qpdf_offset_t i = obj; i - num < obj; ++i) {
-            if (!read_entry(f1, f2, type)) {
+            if (!std::get<0>(read_entry())) {
                 QTC::TC("qpdf", "QPDF invalid xref entry");
                 throw damaged_table("invalid xref entry (obj=" + std::to_string(i) + ")");
             }
@@ -890,9 +886,13 @@ QPDF::Xref_table::subsections(std::string& line)
     }
 }
 
-bool
-QPDF::Xref_table::read_bad_entry(qpdf_offset_t& f1, int& f2, char& type)
+// Returns (success, f1, f2, type).
+std::tuple<bool, qpdf_offset_t, int, char>
+QPDF::Xref_table::read_bad_entry()
 {
+    qpdf_offset_t f1{0};
+    int f2{0};
+    char type{'\0'};
     // Reposition after initial read attempt and reread.
     file->seek(file->getLastOffset(), SEEK_SET);
     auto line = file->readLine(30);
@@ -910,7 +910,7 @@ QPDF::Xref_table::read_bad_entry(qpdf_offset_t& f1, int& f2, char& type)
     }
     // Require digit
     if (!QUtil::is_digit(*p)) {
-        return false;
+        return {false, 0, 0, '\0'};
     }
     // Gather digits
     std::string f1_str;
@@ -919,7 +919,7 @@ QPDF::Xref_table::read_bad_entry(qpdf_offset_t& f1, int& f2, char& type)
     }
     // Require space
     if (!QUtil::is_space(*p)) {
-        return false;
+        return {false, 0, 0, '\0'};
     }
     if (QUtil::is_space(*(p + 1))) {
         QTC::TC("qpdf", "QPDF ignore first extra space in xref entry");
@@ -931,7 +931,7 @@ QPDF::Xref_table::read_bad_entry(qpdf_offset_t& f1, int& f2, char& type)
     }
     // Require digit
     if (!QUtil::is_digit(*p)) {
-        return false;
+        return {false, 0, 0, '\0'};
     }
     // Gather digits
     std::string f2_str;
@@ -940,7 +940,7 @@ QPDF::Xref_table::read_bad_entry(qpdf_offset_t& f1, int& f2, char& type)
     }
     // Require space
     if (!QUtil::is_space(*p)) {
-        return false;
+        return {false, 0, 0, '\0'};
     }
     if (QUtil::is_space(*(p + 1))) {
         QTC::TC("qpdf", "QPDF ignore second extra space in xref entry");
@@ -953,7 +953,7 @@ QPDF::Xref_table::read_bad_entry(qpdf_offset_t& f1, int& f2, char& type)
     if ((*p == 'f') || (*p == 'n')) {
         type = *p;
     } else {
-        return false;
+        return {false, 0, 0, '\0'};
     }
     if ((f1_str.length() != 10) || (f2_str.length() != 5)) {
         QTC::TC("qpdf", "QPDF ignore length error xref entry");
@@ -967,18 +967,23 @@ QPDF::Xref_table::read_bad_entry(qpdf_offset_t& f1, int& f2, char& type)
     f1 = QUtil::string_to_ll(f1_str.c_str());
     f2 = QUtil::string_to_int(f2_str.c_str());
 
-    return true;
+    return {true, f1, f2, type};
 }
 
 // Optimistically read and parse xref entry. If entry is bad, call read_bad_xrefEntry and return
-// result.
-bool
-QPDF::Xref_table::read_entry(qpdf_offset_t& f1, int& f2, char& type)
+// result. Returns (success, f1, f2, type).
+std::tuple<bool, qpdf_offset_t, int, char>
+QPDF::Xref_table::read_entry()
 {
+    qpdf_offset_t f1{0};
+    int f2{0};
+    char type{'\0'};
     std::array<char, 21> line;
+    f1 = 0;
+    f2 = 0;
     if (file->read(line.data(), 20) != 20) {
         // C++20: [[unlikely]]
-        return false;
+        return {false, 0, 0, '\0'};
     }
     line[20] = '\0';
     char const* p = line.data();
@@ -1002,7 +1007,7 @@ QPDF::Xref_table::read_entry(qpdf_offset_t& f1, int& f2, char& type)
     if (!QUtil::is_space(*p++)) {
         // Entry doesn't start with space or digit.
         // C++20: [[unlikely]]
-        return false;
+        return {false, 0, 0, '\0'};
     }
     // Gather digits. NB No risk of overflow as 99'999 < max int.
     while (*p == '0') {
@@ -1019,10 +1024,10 @@ QPDF::Xref_table::read_entry(qpdf_offset_t& f1, int& f2, char& type)
         // No test for valid line[19].
         if (*(++p) && *(++p) && (*p == '\n' || *p == '\r') && f1_len == 10 && f2_len == 5) {
             // C++20: [[likely]]
-            return true;
+            return {true, f1, f2, type};
         }
     }
-    return read_bad_entry(f1, f2, type);
+    return read_bad_entry();
 }
 
 // Read a single cross-reference table section and associated trailer.
@@ -1052,7 +1057,10 @@ QPDF::Xref_table::process_section(qpdf_offset_t xref_offset)
             QTC::TC("qpdf", "QPDF trailer size not integer");
             throw qpdf.damagedPDF("trailer", "/Size key in trailer dictionary is not an integer");
         }
-
+        if (sz >= static_cast<unsigned int>(max_id_)) {
+            QTC::TC("qpdf", "QPDF trailer size impossibly large");
+            throw qpdf.damagedPDF("trailer", "/Size key in trailer dictionary is impossibly large");
+        }
         table.resize(sz);
     }
 
@@ -1064,10 +1072,8 @@ QPDF::Xref_table::process_section(qpdf_offset_t xref_offset)
                 first_item_offset_ = file->tell();
             }
             // For xref_table, these will always be small enough to be ints
-            qpdf_offset_t f1 = 0;
-            int f2 = 0;
-            char type = '\0';
-            if (!read_entry(f1, f2, type)) {
+            auto [success, f1, f2, type] = read_entry();
+            if (!success) {
                 throw damaged_table("invalid xref entry (obj=" + std::to_string(i) + ")");
             }
             if (type == 'f') {
@@ -1585,8 +1591,7 @@ QPDF::Xref_table::read_trailer()
 {
     qpdf_offset_t offset = file->tell();
     bool empty = false;
-    auto object =
-        QPDFParser(*file, "trailer", tokenizer, nullptr, &qpdf, true).parse(empty, false);
+    auto object = QPDFParser(*file, "trailer", tokenizer, nullptr, &qpdf, true).parse(empty, false);
     if (empty) {
         // Nothing in the PDF spec appears to allow empty objects, but they have been encountered in
         // actual PDF files and Adobe Reader appears to ignore them.

libqpdf/QPDFJob.cc

@@ -471,6 +471,21 @@ QPDFJob::createQPDF()
     }
     handleUnderOverlay(pdf);
     handleTransformations(pdf);
+    if (m->remove_info) {
+        auto trailer = pdf.getTrailer();
+        auto mod_date = trailer.getKey("/Info").getKeyIfDict("/ModDate");
+        if (mod_date.isNull()) {
+            trailer.removeKey("/Info");
+        } else {
+            auto info = trailer.replaceKeyAndGetNew(
+                "/Info", pdf.makeIndirectObject(QPDFObjectHandle::newDictionary()));
+            info.replaceKey("/ModDate", mod_date);
+        }
+        pdf.getRoot().removeKey("/Metadata");
+    }
+    if (m->remove_metadata) {
+        pdf.getRoot().removeKey("/Metadata");
+    }
 
     for (auto& foreign: page_heap) {
         if (foreign->anyWarnings()) {

libqpdf/QPDFJob_config.cc

@@ -510,6 +510,20 @@ QPDFJob::Config::removeAttachment(std::string const& parameter)
     return this;
 }
 
+QPDFJob::Config*
+QPDFJob::Config::removeInfo()
+{
+    o.m->remove_info = true;
+    return this;
+}
+
+QPDFJob::Config*
+QPDFJob::Config::removeMetadata()
+{
+    o.m->remove_metadata = true;
+    return this;
+}
+
 QPDFJob::Config*
 QPDFJob::Config::removePageLabels()
 {

libqpdf/QPDFParser.cc

@@ -469,14 +469,15 @@ QPDFParser::fixMissingKeys()
 bool
 QPDFParser::tooManyBadTokens()
 {
-    if (good_count <= 4) {
+    if (--max_bad_count > 0 && good_count > 4) {
+        good_count = 0;
+        bad_count = 1;
+        return false;
+    }
     if (++bad_count > 5) {
         warn("too many errors; giving up on reading object");
         return true;
     }
-    } else {
-        bad_count = 1;
-    }
     good_count = 0;
     return false;
 }

libqpdf/QPDFTokenizer.cc

@@ -47,7 +47,7 @@ QPDFWordTokenFinder::check()
     // Find a word token matching the given string, preceded by a delimiter, and followed by a
     // delimiter or EOF.
     QPDFTokenizer tokenizer;
-    QPDFTokenizer::Token t = tokenizer.readToken(is, "finder", true);
+    QPDFTokenizer::Token t = tokenizer.readToken(is, "finder", true, str.size() + 2);
     qpdf_offset_t pos = is.tell();
     if (!(t == QPDFTokenizer::Token(QPDFTokenizer::tt_word, str))) {
         QTC::TC("qpdf", "QPDFTokenizer finder found wrong word");

libqpdf/qpdf/QPDFParser.hh

@@ -83,9 +83,11 @@ class QPDFParser
     std::vector<StackFrame> stack;
     StackFrame* frame;
     // Number of recent bad tokens.
-    int bad_count = 0;
+    int bad_count{0};
+    // Number of bad tokens (remaining) before giving up.
+    int max_bad_count{15};
     // Number of good tokens since last bad token. Irrelevant if bad_count == 0.
-    int good_count = 0;
+    int good_count{0};
     // Start offset including any leading whitespace.
     qpdf_offset_t start;
     // Number of successive integer tokens.

libqpdf/qpdf/QPDF_private.hh

@@ -292,8 +292,8 @@ class QPDF::Xref_table
     std::vector<Subsection> subsections(std::string& line);
     std::vector<Subsection> bad_subsections(std::string& line, qpdf_offset_t offset);
     Subsection subsection(std::string const& line);
-    bool read_entry(qpdf_offset_t& f1, int& f2, char& type);
-    bool read_bad_entry(qpdf_offset_t& f1, int& f2, char& type);
+    std::tuple<bool, qpdf_offset_t, int, char> read_entry();
+    std::tuple<bool, qpdf_offset_t, int, char> read_bad_entry();
 
     // Methods to parse streams
     qpdf_offset_t read_stream(qpdf_offset_t offset);

libqpdf/qpdf/auto_job_help.hh

@@ -414,6 +414,13 @@ Don't optimize images whose area in pixels is below the specified value.
 )");
 ap.addOptionHelp("--keep-inline-images", "modification", "exclude inline images from optimization", R"(Prevent inline images from being considered by --optimize-images.
 )");
+ap.addOptionHelp("--remove-info", "modification", "remove file information", R"(Exclude file information (except modification date) from the output file.
+)");
+ap.addOptionHelp("--remove-metadata", "modification", "remove metadata", R"(Exclude metadata from the output file.
+)");
+}
+static void add_help_5(QPDFArgParser& ap)
+{
 ap.addOptionHelp("--remove-page-labels", "modification", "remove explicit page numbers", R"(Exclude page labels (explicit page numbers) from the output file.
 )");
 ap.addOptionHelp("--set-page-labels", "modification", "number pages for the entire document", R"(--set-page-labels label-spec ... --
@@ -460,9 +467,6 @@ iv, then the remaining pages with Arabic numerals starting with
 1 and continuing sequentially until the end of the document. For
 additional examples, please consult the manual.
 )");
-}
-static void add_help_5(QPDFArgParser& ap)
-{
 ap.addHelpTopic("encryption", "create encrypted files", R"(Create encrypted files. Usage:
 
 --encrypt \
@@ -641,6 +645,9 @@ ap.addOptionHelp("--force-R5", "encryption", "use unsupported R=5 encryption", R
 algorithm that existed only in Acrobat version IX. This option
 should not be used except for compatibility testing.
 )");
+}
+static void add_help_6(QPDFArgParser& ap)
+{
 ap.addHelpTopic("page-selection", "select pages from one or more files", R"(Use the --pages option to select pages from multiple files. Usage:
 
 qpdf in.pdf --pages --file=input-file \
@@ -725,9 +732,6 @@ appearance: first underlays, then the original page, then overlays.
 
 Run qpdf --help=page-ranges for help with page ranges.
 )");
-}
-static void add_help_6(QPDFArgParser& ap)
-{
 ap.addOptionHelp("--to", "overlay-underlay", "destination pages for underlay/overlay", R"(--to=page-range
 
 Specify the range of pages in the primary output to apply
@@ -829,6 +833,9 @@ its terminating "--".
 To copy attachments from a password-protected file, use
 the --password option after the file name.
 )");
+}
+static void add_help_7(QPDFArgParser& ap)
+{
 ap.addOptionHelp("--prefix", "copy-attachments", "key prefix for copying attachments", R"(--prefix=prefix
 
 Prepend a prefix to each key; may be needed if there are
@@ -839,9 +846,6 @@ ap.addHelpTopic("inspection", "inspect PDF files", R"(These options provide tool
 the options in this section are specified, no output file may be
 given.
 )");
-}
-static void add_help_7(QPDFArgParser& ap)
-{
 ap.addOptionHelp("--is-encrypted", "inspection", "silently test whether a file is encrypted", R"(Silently exit with a code indicating the file's encryption status:
 
 0: the file is encrypted
@@ -919,6 +923,9 @@ output as binary data. Get the key with --list-attachments.
 ap.addHelpTopic("json", "JSON output for PDF information", R"(Show information about the PDF file in JSON format. Please see the
 JSON chapter in the qpdf manual for details.
 )");
+}
+static void add_help_8(QPDFArgParser& ap)
+{
 ap.addOptionHelp("--json", "json", "show file in JSON format", R"(--json[=version]
 
 Generate a JSON representation of the file. This is described in
@@ -932,9 +939,6 @@ Describe the format of the JSON output by writing to standard
 output a JSON object with the same keys and with values
 containing descriptive text.
 )");
-}
-static void add_help_8(QPDFArgParser& ap)
-{
 ap.addOptionHelp("--json-key", "json", "limit which keys are in JSON output", R"(--json-key=key
 
 This option is repeatable. If given, only the specified

libqpdf/qpdf/auto_job_init.hh

@@ -68,6 +68,8 @@ this->ap.addBare("progress", [this](){c_main->progress();});
 this->ap.addBare("qdf", [this](){c_main->qdf();});
 this->ap.addBare("raw-stream-data", [this](){c_main->rawStreamData();});
 this->ap.addBare("recompress-flate", [this](){c_main->recompressFlate();});
+this->ap.addBare("remove-info", [this](){c_main->removeInfo();});
+this->ap.addBare("remove-metadata", [this](){c_main->removeMetadata();});
 this->ap.addBare("remove-page-labels", [this](){c_main->removePageLabels();});
 this->ap.addBare("replace-input", b(&ArgParser::argReplaceInput));
 this->ap.addBare("report-memory-usage", [this](){c_main->reportMemoryUsage();});

libqpdf/qpdf/auto_job_json_init.hh

@@ -412,6 +412,12 @@ addParameter([this](std::string const& p) { c_pages->range(p); });
 popHandler(); // key: range
 popHandler(); // array: .pages[]
 popHandler(); // key: pages
+pushKey("removeInfo");
+addBare([this]() { c_main->removeInfo(); });
+popHandler(); // key: removeInfo
+pushKey("removeMetadata");
+addBare([this]() { c_main->removeMetadata(); });
+popHandler(); // key: removeMetadata
 pushKey("removePageLabels");
 addBare([this]() { c_main->removePageLabels(); });
 popHandler(); // key: removePageLabels

libqpdf/qpdf/auto_job_schema.hh

@@ -145,6 +145,8 @@ static constexpr char const* JOB_SCHEMA_DATA = R"({
       "range": "page range"
     }
   ],
+  "removeInfo": "remove file information",
+  "removeMetadata": "remove metadata",
   "removePageLabels": "remove explicit page numbers",
   "reportMemoryUsage": "best effort report of memory usage",
   "rotate": "rotate pages",

manual/cli.rst

@@ -1773,6 +1773,27 @@ Related Options
    Prevent inline images from being included in image optimization
    done by :qpdf:ref:`--optimize-images`.
 
+.. qpdf:option:: --remove-info
+
+   .. help: remove file information
+
+      Exclude file information (except modification date) from the output file.
+
+   Exclude file information (except modification date) from the output file by
+   omitting all entries (except ``/ModDate``) from the ``/Info`` dictionary in
+   the document trailer.
+   See also :qpdf:ref:`--remove-metadata`.
+
+.. qpdf:option:: --remove-metadata
+
+   .. help: remove metadata
+
+      Exclude metadata from the output file.
+
+   Exclude metadata from the output file by omitting the ``/Metadata``
+   dictionary in the document catalog.
+   See also :qpdf:ref:`--remove-info`.
+
 .. qpdf:option:: --remove-page-labels
 
    .. help: remove explicit page numbers

manual/qpdf.1

@@ -530,6 +530,12 @@ Don't optimize images whose area in pixels is below the specified value.
 .B --keep-inline-images \-\- exclude inline images from optimization
 Prevent inline images from being considered by --optimize-images.
 .TP
+.B --remove-info \-\- remove file information
+Exclude file information (except modification date) from the output file.
+.TP
+.B --remove-metadata \-\- remove metadata
+Exclude metadata from the output file.
+.TP
 .B --remove-page-labels \-\- remove explicit page numbers
 Exclude page labels (explicit page numbers) from the output file.
 .TP

qpdf/qpdf.testcov

@@ -55,6 +55,7 @@ QPDF invalid xref entry 0
 QPDF missing trailer 0
 QPDF trailer lacks size 0
 QPDF trailer size not integer 0
+QPDF trailer size impossibly large 0
 QPDF trailer prev not integer 0
 QPDFParser bad brace 0
 QPDFParser bad brace in parseRemainder 0

qpdf/qtest/merge-and-split.test

@@ -14,7 +14,7 @@ cleanup();
 
 my $td = new TestDriver('merge-and-split');
 
-my $n_tests = 28;
+my $n_tests = 34;
 
 # Select pages from the same file multiple times including selecting
 # twice from an encrypted file and specifying the password only the
@@ -103,6 +103,39 @@ $td->runtest("check output",
              {$td->COMMAND => "qpdf-test-compare a.pdf remove-labels.pdf"},
              {$td->FILE => "remove-labels.pdf", $td->EXIT_STATUS => 0});
 
+$td->runtest("remove metadata",
+             {$td->COMMAND =>
+                  "qpdf metadata-crypt-filter.pdf a.pdf" .
+                  " --remove-metadata" .
+                  " --decrypt" .
+                  " --static-id"},
+             {$td->STRING => "", $td->EXIT_STATUS => 0});
+$td->runtest("check output",
+             {$td->FILE => "a.pdf"},
+             {$td->FILE => "remove-metadata.pdf"});
+
+$td->runtest("remove info (with moddate)",
+             {$td->COMMAND =>
+                  "qpdf remove-metadata.pdf a.pdf" .
+                  " --remove-info" .
+                  " --decrypt" .
+                  " --static-id"},
+             {$td->STRING => "", $td->EXIT_STATUS => 0});
+$td->runtest("check output",
+             {$td->FILE => "a.pdf"},
+             {$td->FILE => "remove-info.pdf"});
+
+$td->runtest("remove info (without moddate)",
+             {$td->COMMAND =>
+                  "qpdf remove-metadata-no-moddate.pdf a.pdf" .
+                  " --remove-info" .
+                  " --decrypt" .
+                  " --static-id"},
+             {$td->STRING => "", $td->EXIT_STATUS => 0});
+$td->runtest("check output",
+             {$td->FILE => "a.pdf"},
+             {$td->FILE => "remove-info-no-moddate.pdf"});
+
 $td->runtest("split with shared resources",
              {$td->COMMAND =>
                   "qpdf --qdf --static-id" .

qpdf/qtest/qpdf/issue-fuzz.out

@@ -0,0 +1,19 @@
+WARNING: issue-fuzz.pdf: can't find PDF header
+WARNING: issue-fuzz.pdf (xref table, offset 19): accepting invalid xref table entry
+WARNING: issue-fuzz.pdf (trailer, offset 36): unknown token while reading object; treating as string
+WARNING: issue-fuzz.pdf (trailer, offset 53): unexpected >
+WARNING: issue-fuzz.pdf (trailer, offset 54): unknown token while reading object; treating as string
+WARNING: issue-fuzz.pdf (trailer, offset 58): unknown token while reading object; treating as string
+WARNING: issue-fuzz.pdf (trailer, offset 72): unknown token while reading object; treating as string
+WARNING: issue-fuzz.pdf (trailer, offset 36): dictionary ended prematurely; using null as value for last key
+WARNING: issue-fuzz.pdf (trailer, offset 36): expected dictionary key but found non-name object; inserting key /QPDFFake1
+WARNING: issue-fuzz.pdf (trailer, offset 36): expected dictionary key but found non-name object; inserting key /QPDFFake2
+WARNING: issue-fuzz.pdf (trailer, offset 36): expected dictionary key but found non-name object; inserting key /QPDFFake3
+WARNING: issue-fuzz.pdf (trailer, offset 36): expected dictionary key but found non-name object; inserting key /QPDFFake4
+WARNING: issue-fuzz.pdf (trailer, offset 36): expected dictionary key but found non-name object; inserting key /QPDFFake5
+WARNING: issue-fuzz.pdf (trailer, offset 36): expected dictionary key but found non-name object; inserting key /QPDFFake6
+WARNING: issue-fuzz.pdf (trailer, offset 36): expected dictionary key but found non-name object; inserting key /QPDFFake7
+WARNING: issue-fuzz.pdf: file is damaged
+WARNING: issue-fuzz.pdf (trailer, offset 32): /Size key in trailer dictionary is impossibly large
+WARNING: issue-fuzz.pdf: Attempting to reconstruct cross-reference table
+qpdf: issue-fuzz.pdf: unable to find /Root dictionary

qpdf/qtest/specific-bugs.test

@@ -38,6 +38,7 @@ my @bug_tests = (
     ["263", "empty xref stream", 2],
     ["335a", "ozz-fuzz-12152", 2],
     ["335b", "ozz-fuzz-14845", 2],
+    ["fuzz", "impossibly large trailer /Size"],
 #    ["fuzz-16214", "stream in object stream", 3, "--preserve-unreferenced"],
     # When adding to this list, consider adding to CORPUS_FROM_TEST in
     # fuzz/CMakeLists.txt and updating the count in