aboutsummaryrefslogtreecommitdiffstats
path: root/libqpdf
diff options
context:
space:
mode:
Diffstat (limited to 'libqpdf')
-rw-r--r--libqpdf/JSON.cc8
-rw-r--r--libqpdf/QPDF.cc7
-rw-r--r--libqpdf/QPDFJob.cc50
-rw-r--r--libqpdf/QPDFTokenizer.cc48
-rw-r--r--libqpdf/QPDFWriter.cc39
-rw-r--r--libqpdf/QUtil.cc31
6 files changed, 73 insertions, 110 deletions
diff --git a/libqpdf/JSON.cc b/libqpdf/JSON.cc
index cb60eabc..fbf06f88 100644
--- a/libqpdf/JSON.cc
+++ b/libqpdf/JSON.cc
@@ -1121,12 +1121,8 @@ JSONParser::getToken()
case ls_u4:
using ui = unsigned int;
- if ('0' <= *p && *p <= '9') {
- u_value = 16 * u_value + (ui(*p) - ui('0'));
- } else if ('a' <= *p && *p <= 'f') {
- u_value = 16 * u_value + (10 + ui(*p) - ui('a'));
- } else if ('A' <= *p && *p <= 'F') {
- u_value = 16 * u_value + (10 + ui(*p) - ui('A'));
+ if (ui val = ui(QUtil::hex_decode_char(*p)); val < 16) {
+ u_value = 16 * u_value + val;
} else {
tokenError();
}
diff --git a/libqpdf/QPDF.cc b/libqpdf/QPDF.cc
index 063c0f55..83944be4 100644
--- a/libqpdf/QPDF.cc
+++ b/libqpdf/QPDF.cc
@@ -2461,6 +2461,13 @@ QPDF::getRoot()
QPDFObjectHandle root = this->m->trailer.getKey("/Root");
if (!root.isDictionary()) {
throw damagedPDF("", 0, "unable to find /Root dictionary");
+ } else if (
+ // Check_mode is an interim solution to request #810 pending a more
+ // comprehensive review of the approach to more extensive checks and
+ // warning levels.
+ m->check_mode && !root.getKey("/Type").isNameAndEquals("/Catalog")) {
+ warn(damagedPDF("", 0, "catalog /Type entry missing or invalid"));
+ root.replaceKey("/Type", "/Catalog"_qpdf);
}
return root;
}
diff --git a/libqpdf/QPDFJob.cc b/libqpdf/QPDFJob.cc
index a4b1a026..fbaa53e5 100644
--- a/libqpdf/QPDFJob.cc
+++ b/libqpdf/QPDFJob.cc
@@ -798,6 +798,7 @@ QPDFJob::doCheck(QPDF& pdf)
bool okay = true;
auto& cout = *this->m->log->getInfo();
cout << "checking " << m->infilename.get() << "\n";
+ QPDF::JobSetter::setCheckMode(pdf, true);
try {
int extension_level = pdf.getExtensionLevel();
cout << "PDF Version: " << pdf.getPDFVersion();
@@ -2363,31 +2364,30 @@ QPDFJob::handleTransformations(QPDF& pdf)
int pageno = 0;
for (auto& ph: dh.getAllPages()) {
++pageno;
- QPDFObjectHandle page = ph.getObjectHandle();
- for (auto& iter2: ph.getImages()) {
- std::string name = iter2.first;
- QPDFObjectHandle& image = iter2.second;
- ImageOptimizer* io = new ImageOptimizer(
- *this,
- m->oi_min_width,
- m->oi_min_height,
- m->oi_min_area,
- image);
- std::shared_ptr<QPDFObjectHandle::StreamDataProvider> sdp(io);
- if (io->evaluate(
- "image " + name + " on page " +
- std::to_string(pageno))) {
- QPDFObjectHandle new_image = pdf.newStream();
- new_image.replaceDict(image.getDict().shallowCopy());
- new_image.replaceStreamData(
- sdp,
- QPDFObjectHandle::newName("/DCTDecode"),
- QPDFObjectHandle::newNull());
- ph.getAttribute("/Resources", true)
- .getKey("/XObject")
- .replaceKey(name, new_image);
- }
- }
+ ph.forEachImage(
+ true,
+ [this, pageno, &pdf](
+ QPDFObjectHandle& obj,
+ QPDFObjectHandle& xobj_dict,
+ std::string const& key) {
+ auto io = std::make_unique<ImageOptimizer>(
+ *this,
+ m->oi_min_width,
+ m->oi_min_height,
+ m->oi_min_area,
+ obj);
+ if (io->evaluate(
+ "image " + key + " on page " +
+ std::to_string(pageno))) {
+ QPDFObjectHandle new_image = pdf.newStream();
+ new_image.replaceDict(obj.getDict().shallowCopy());
+ new_image.replaceStreamData(
+ std::move(io),
+ QPDFObjectHandle::newName("/DCTDecode"),
+ QPDFObjectHandle::newNull());
+ xobj_dict.replaceKey(key, new_image);
+ }
+ });
}
}
if (m->generate_appearances) {
diff --git a/libqpdf/QPDFTokenizer.cc b/libqpdf/QPDFTokenizer.cc
index 990d5b65..d8d457ab 100644
--- a/libqpdf/QPDFTokenizer.cc
+++ b/libqpdf/QPDFTokenizer.cc
@@ -449,18 +449,9 @@ QPDFTokenizer::inNameHex1(char ch)
{
this->hex_char = ch;
- if ('0' <= ch && ch <= '9') {
- this->char_code = 16 * (int(ch) - int('0'));
+ if (char hval = QUtil::hex_decode_char(ch); hval < '\20') {
+ this->char_code = int(hval) << 4;
this->state = st_name_hex2;
-
- } else if ('A' <= ch && ch <= 'F') {
- this->char_code = 16 * (10 + int(ch) - int('A'));
- this->state = st_name_hex2;
-
- } else if ('a' <= ch && ch <= 'f') {
- this->char_code = 16 * (10 + int(ch) - int('a'));
- this->state = st_name_hex2;
-
} else {
QTC::TC("qpdf", "QPDFTokenizer bad name 1");
this->error_message = "name with stray # will not work with PDF >= 1.2";
@@ -475,15 +466,8 @@ QPDFTokenizer::inNameHex1(char ch)
void
QPDFTokenizer::inNameHex2(char ch)
{
- if ('0' <= ch && ch <= '9') {
- this->char_code += int(ch) - int('0');
-
- } else if ('A' <= ch && ch <= 'F') {
- this->char_code += 10 + int(ch) - int('A');
-
- } else if ('a' <= ch && ch <= 'f') {
- this->char_code += 10 + int(ch) - int('a');
-
+ if (char hval = QUtil::hex_decode_char(ch); hval < '\20') {
+ this->char_code |= int(hval);
} else {
QTC::TC("qpdf", "QPDFTokenizer bad name 2");
this->error_message = "name with stray # will not work with PDF >= 1.2";
@@ -675,16 +659,8 @@ QPDFTokenizer::inLiteral(char ch)
void
QPDFTokenizer::inHexstring(char ch)
{
- if ('0' <= ch && ch <= '9') {
- this->char_code = 16 * (int(ch) - int('0'));
- this->state = st_in_hexstring_2nd;
-
- } else if ('A' <= ch && ch <= 'F') {
- this->char_code = 16 * (10 + int(ch) - int('A'));
- this->state = st_in_hexstring_2nd;
-
- } else if ('a' <= ch && ch <= 'f') {
- this->char_code = 16 * (10 + int(ch) - int('a'));
+ if (char hval = QUtil::hex_decode_char(ch); hval < '\20') {
+ this->char_code = int(hval) << 4;
this->state = st_in_hexstring_2nd;
} else if (ch == '>') {
@@ -706,16 +682,8 @@ QPDFTokenizer::inHexstring(char ch)
void
QPDFTokenizer::inHexstring2nd(char ch)
{
- if ('0' <= ch && ch <= '9') {
- this->val += char(this->char_code + int(ch) - int('0'));
- this->state = st_in_hexstring;
-
- } else if ('A' <= ch && ch <= 'F') {
- this->val += char(this->char_code + 10 + int(ch) - int('A'));
- this->state = st_in_hexstring;
-
- } else if ('a' <= ch && ch <= 'f') {
- this->val += char(this->char_code + 10 + int(ch) - int('a'));
+ if (char hval = QUtil::hex_decode_char(ch); hval < '\20') {
+ this->val += char(this->char_code) | hval;
this->state = st_in_hexstring;
} else if (ch == '>') {
diff --git a/libqpdf/QPDFWriter.cc b/libqpdf/QPDFWriter.cc
index 8287412c..de1aa45b 100644
--- a/libqpdf/QPDFWriter.cc
+++ b/libqpdf/QPDFWriter.cc
@@ -1441,8 +1441,13 @@ QPDFWriter::unparseObject(
if (level < 0) {
throw std::logic_error("invalid level in QPDFWriter::unparseObject");
}
-
- std::string const indent(static_cast<size_t>(2 * level), ' ');
+ // For non-qdf, "indent" is a single space between tokens.
+ // For qdf, indent includes the preceding newline.
+ std::string indent = " ";
+ if (m->qdf_mode) {
+ indent.append(static_cast<size_t>(2 * level), ' ');
+ indent[0] = '\n';
+ }
if (auto const tc = object.getTypeCode(); tc == ::ot_array) {
// Note: PDF spec 1.4 implementation note 121 states that
@@ -1451,16 +1456,12 @@ QPDFWriter::unparseObject(
// unconditionally for all arrays because it looks nicer and
// doesn't make the files that much bigger.
writeString("[");
- writeStringQDF("\n");
for (auto const& item: object.getArrayAsVector()) {
- writeStringQDF(indent);
+ writeString(indent);
writeStringQDF(" ");
- writeStringNoQDF(" ");
unparseChild(item, level + 1, child_flags);
- writeStringQDF("\n");
}
- writeStringQDF(indent);
- writeStringNoQDF(" ");
+ writeString(indent);
writeString("]");
} else if (tc == ::ot_dictionary) {
// Make a shallow copy of this object so we can modify it
@@ -1619,14 +1620,12 @@ QPDFWriter::unparseObject(
}
writeString("<<");
- writeStringQDF("\n");
for (auto& item: object.getDictAsMap()) {
if (!item.second.isNull()) {
auto const& key = item.first;
- writeStringQDF(indent);
+ writeString(indent);
writeStringQDF(" ");
- writeStringNoQDF(" ");
writeString(QPDF_Name::normalizeName(key));
writeString(" ");
if (key == "/Contents" && object.isDictionaryOfType("/Sig") &&
@@ -1639,14 +1638,13 @@ QPDFWriter::unparseObject(
} else {
unparseChild(item.second, level + 1, child_flags);
}
- writeStringQDF("\n");
}
}
if (flags & f_stream) {
- writeStringQDF(indent);
- writeStringQDF(" ");
- writeString(" /Length ");
+ writeString(indent);
+ writeStringQDF(" ");
+ writeString("/Length ");
if (this->m->direct_stream_lengths) {
writeString(std::to_string(stream_length));
@@ -1654,17 +1652,14 @@ QPDFWriter::unparseObject(
writeString(std::to_string(this->m->cur_stream_length_id));
writeString(" 0 R");
}
- writeStringQDF("\n");
if (compress && (flags & f_filtered)) {
- writeStringQDF(indent);
- writeStringQDF(" ");
- writeString(" /Filter /FlateDecode");
- writeStringQDF("\n");
+ writeString(indent);
+ writeStringQDF(" ");
+ writeString("/Filter /FlateDecode");
}
}
- writeStringQDF(indent);
- writeStringNoQDF(" ");
+ writeString(indent);
writeString(">>");
} else if (tc == ::ot_stream) {
// Write stream data to a buffer.
diff --git a/libqpdf/QUtil.cc b/libqpdf/QUtil.cc
index bae067b6..03301d9d 100644
--- a/libqpdf/QUtil.cc
+++ b/libqpdf/QUtil.cc
@@ -783,28 +783,25 @@ std::string
QUtil::hex_decode(std::string const& input)
{
std::string result;
- size_t pos = 0;
+ // We know result.size() <= 0.5 * input.size() + 1. However, reserving
+ // string space for this upper bound has a negative impact.
+ bool first = true;
+ char decoded;
for (auto ch: input) {
- bool skip = false;
- if ((ch >= 'A') && (ch <= 'F')) {
- ch = QIntC::to_char(ch - 'A' + 10);
- } else if ((ch >= 'a') && (ch <= 'f')) {
- ch = QIntC::to_char(ch - 'a' + 10);
- } else if ((ch >= '0') && (ch <= '9')) {
- ch = QIntC::to_char(ch - '0');
- } else {
- skip = true;
- }
- if (!skip) {
- if (pos == 0) {
- result.push_back(static_cast<char>(ch << 4));
- pos = 1;
+ ch = hex_decode_char(ch);
+ if (ch < '\20') {
+ if (first) {
+ decoded = static_cast<char>(ch << 4);
+ first = false;
} else {
- result[result.length() - 1] |= ch;
- pos = 0;
+ result.push_back(decoded | ch);
+ first = true;
}
}
}
+ if (!first) {
+ result.push_back(decoded);
+ }
return result;
}
generated by cgit v1.2.3-54-g00ecf (git 2.45.2) at 2024-07-06 01:10:09 +0200