qpdf_ruby 0.1.0 → 0.2.0

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: qpdf_ruby
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.2.0
 platform: ruby
 authors:
 - Dieter S.
@@ -130,35 +130,35 @@ extensions:
 - ext/qpdf_ruby/extconf.rb
 extra_rdoc_files: []
 files:
-- ".clang-format"
-- ".rspec"
-- ".rubocop.yml"
-- ".ruby-gemset"
-- ".ruby-version"
 - CHANGELOG.md
 - LICENSE.txt
 - README.md
-- Rakefile
-- docker/Dockerfile
 - exe/qpdf_ruby
 - ext/qpdf_ruby/array_node.cpp
 - ext/qpdf_ruby/document_handle.cpp
 - ext/qpdf_ruby/document_handle.hpp
 - ext/qpdf_ruby/extconf.rb
 - ext/qpdf_ruby/figure_node.cpp
+- ext/qpdf_ruby/mcid_bounds.cpp
+- ext/qpdf_ruby/mcid_bounds.hpp
 - ext/qpdf_ruby/mcid_node.cpp
 - ext/qpdf_ruby/mcr_node.cpp
-- ext/qpdf_ruby/pdf_image_mapper.cpp
-- ext/qpdf_ruby/pdf_image_mapper.hpp
 - ext/qpdf_ruby/pdf_struct_walker.cpp
 - ext/qpdf_ruby/pdf_struct_walker.hpp
+- ext/qpdf_ruby/pdfua.cpp
+- ext/qpdf_ruby/pdfua.hpp
 - ext/qpdf_ruby/qpdf_ruby.cpp
 - ext/qpdf_ruby/qpdf_ruby.hpp
+- ext/qpdf_ruby/ruby_guard.hpp
 - ext/qpdf_ruby/stream_node.cpp
+- ext/qpdf_ruby/stroke_bounds.cpp
+- ext/qpdf_ruby/stroke_bounds.hpp
 - ext/qpdf_ruby/struct_elem_node.cpp
 - ext/qpdf_ruby/struct_node.cpp
 - ext/qpdf_ruby/struct_node.hpp
 - ext/qpdf_ruby/unknown_node.cpp
+- ext/qpdf_ruby/xmp_scan.cpp
+- ext/qpdf_ruby/xmp_scan.hpp
 - lib/qpdf_ruby.rb
 - lib/qpdf_ruby/version.rb
 - sig/qpdf_ruby.rbs
@@ -185,7 +185,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.6.8
+rubygems_version: 4.0.20
 specification_version: 4
 summary: A Ruby interface .
 test_files: []

data/.clang-format

@@ -1,4 +0,0 @@
-BasedOnStyle: Google
-SortIncludes: false
-IndentWidth: 2
-ColumnLimit: 120

data/.rspec

@@ -1,3 +0,0 @@
---format documentation
---color
---require spec_helper

data/.rubocop.yml

@@ -1,88 +0,0 @@
-AllCops:
-  NewCops: enable
-  TargetRubyVersion: 3.3
-  Exclude:
-    - 'ext/qpdf_ruby/extconf.rb'
-    - 'tmp/**/*'
-    - 'vendor/**/*'
-
-Style/StringLiterals:
-  EnforcedStyle: double_quotes
-
-Style/StringLiteralsInInterpolation:
-  EnforcedStyle: double_quotes
-
-Style/Documentation:
-  Enabled: false
-
-RSpec/SubjectStub:
-  Enabled: false
-
-RSpec/ExampleLength:
-  Enabled: false
-
-Layout/MultilineMethodCallIndentation:
-  Enabled: false
-
-Layout/FirstArgumentIndentation:
-  Enabled: false
-
-Layout/ClosingParenthesisIndentation:
-  Enabled: false
-
-Layout/FirstHashElementIndentation:
-  EnforcedStyle: consistent
-
-Layout/FirstArrayElementIndentation:
-  Enabled: false
-
-Layout/MultilineOperationIndentation:
-  Enabled: false
-
-Layout/BeginEndAlignment:
-  Enabled: false
-
-Layout/ArrayAlignment:
-  Enabled: false
-
-Layout/LineLength:
-  Enabled: false
-
-Layout/LineEndStringConcatenationIndentation:
-  Enabled: false
-
-RSpec/MultipleMemoizedHelpers:
-  Max: 10
-
-Metrics/MethodLength:
-  Enabled: false
-
-Metrics/ClassLength:
-  Enabled: false
-
-Metrics/ParameterLists:
-  Max: 10
-
-
-Gemspec/DevelopmentDependencies:
-  EnforcedStyle: gemspec
-
-RSpec/InstanceVariable:
-  Enabled: false
-
-RSpec/BeforeAfterAll:
-  Enabled: false
-
-RSpec/SpecFilePathFormat:
-  Enabled: true
-  Exclude:
-    - 'spec/acceptance/**/*_spec.rb'
-
-RSpec/DescribeClass:
-  Enabled: true
-  Exclude:
-    - 'spec/acceptance/**/*_spec.rb'
-
-plugins:
-  - rubocop-rake
-  - rubocop-rspec

data/.ruby-gemset

@@ -1 +0,0 @@
-qpdf_ruby

data/.ruby-version

@@ -1 +0,0 @@
-ruby-3.3.4

data/Rakefile

@@ -1,25 +0,0 @@
-# frozen_string_literal: true
-
-require "bundler/gem_tasks"
-require "rspec/core/rake_task"
-
-RSpec::Core::RakeTask.new(:spec)
-
-require "rubocop/rake_task"
-
-RuboCop::RakeTask.new
-
-require "rake/extensiontask"
-
-desc "Build the gem including native extensions"
-task build: :compile
-
-GEMSPEC = Gem::Specification.load("qpdf_ruby.gemspec")
-
-Rake::ExtensionTask.new("qpdf_ruby", GEMSPEC) do |ext|
-  ext.lib_dir = "lib/qpdf_ruby"
-end
-
-Dir.glob("tasks/*.rake").each { |r| load r }
-
-task default: %i[clobber compile spec rubocop]

data/docker/Dockerfile

@@ -1,46 +0,0 @@
-FROM ruby:3.3
-
-ENV DEBIAN_FRONTEND=noninteractive
-
-RUN set -eux; \
-    echo "deb http://deb.debian.org/debian testing main"   >  /etc/apt/sources.list.d/extra.list; \
-    echo "deb http://deb.debian.org/debian unstable main" >> /etc/apt/sources.list.d/extra.list; \
-    \
-    printf "Package: *\nPin: release a=testing\nPin-Priority: 100\n\n"   >  /etc/apt/preferences.d/99-pin-testing; \
-    printf "Package: *\nPin: release a=unstable\nPin-Priority: 100\n"  >> /etc/apt/preferences.d/99-pin-testing; \
-    \
-    apt-get update; \
-    apt-get -y --no-install-recommends install qpdf libqpdf-dev -t testing; \
-    apt-mark hold qpdf libqpdf-dev; \
-    \
-    apt-get clean; rm -rf /var/lib/apt/lists/*;
-
-# Smoke‑test (shown in build log)
-RUN qpdf --version  # should print "qpdf version 12.x"
-
-
-COPY ./pkg/qpdf_ruby-*.gem ./
-COPY ./spec/fixtures/example_accessibility.pdf /tmp/dummy.pdf
-
-RUN gem install  ./qpdf_ruby-*.gem -- --with-qpdf-include=/usr/local/include/qpdf --with-qpdf-lib=/usr/local/lib
-
-RUN ruby -r qpdf_ruby -e 'puts "QpdfRuby version: #{QpdfRuby::VERSION}";doc = QpdfRuby::Document.new "/tmp/dummy.pdf"; puts doc.show_structure'
-
-# Create a non-root user
-RUN set -eux; \
-    groupadd -r appuser && useradd -r -g appuser -m -d /home/appuser appuser
-
-
-RUN set -eux; \
-    mkdir -p /workspace
-
-# Set working directory
-WORKDIR /workspace
-
-RUN chown -R appuser:appuser /workspace
-
-# Switch to non-root user
-USER appuser
-
-CMD ["/usr/bin/bash"]
-

data/ext/qpdf_ruby/pdf_image_mapper.cpp

@@ -1,294 +0,0 @@
-#include "pdf_image_mapper.hpp"
-#include <qpdf/BufferInputSource.hh>
-#include <qpdf/QPDFTokenizer.hh>
-
-#include <algorithm>
-#include <optional>
-
-using Matrix = std::array<double, 6>;
-
-static std::pair<double, double> apply_matrix(const Matrix& m, double x, double y) {
-  double x_new = m[0] * x + m[2] * y + m[4];
-  double y_new = m[1] * x + m[3] * y + m[5];
-  return {x_new, y_new};
-}
-
-static std::array<double, 4> compute_bbox(double width, double height, const Matrix& matrix) {
-  std::array<std::pair<double, double>, 4> corners = {apply_matrix(matrix, 0, 0), apply_matrix(matrix, width, 0),
-                                                      apply_matrix(matrix, width, height),
-                                                      apply_matrix(matrix, 0, height)};
-
-  double min_x = corners[0].first;
-  double max_x = corners[0].first;
-  double min_y = corners[0].second;
-  double max_y = corners[0].second;
-
-  for (int i = 1; i < 4; ++i) {
-    min_x = std::min(min_x, corners[i].first);
-    max_x = std::max(max_x, corners[i].first);
-    min_y = std::min(min_y, corners[i].second);
-    max_y = std::max(max_y, corners[i].second);
-  }
-
-  return {min_x, min_y, max_x, max_y};
-}
-
-static char const* tokenTypeName(QPDFTokenizer::token_type_e ttype) {
-  // Do this is a case statement instead of a lookup so the compiler
-  // will warn if we miss any.
-  switch (ttype) {
-    case QPDFTokenizer::tt_bad:
-      return "bad";
-    case QPDFTokenizer::tt_array_close:
-      return "array_close";
-    case QPDFTokenizer::tt_array_open:
-      return "array_open";
-    case QPDFTokenizer::tt_brace_close:
-      return "brace_close";
-    case QPDFTokenizer::tt_brace_open:
-      return "brace_open";
-    case QPDFTokenizer::tt_dict_close:
-      return "dict_close";
-    case QPDFTokenizer::tt_dict_open:
-      return "dict_open";
-    case QPDFTokenizer::tt_integer:
-      return "integer";
-    case QPDFTokenizer::tt_name:
-      return "name";
-    case QPDFTokenizer::tt_real:
-      return "real";
-    case QPDFTokenizer::tt_string:
-      return "string";
-    case QPDFTokenizer::tt_null:
-      return "null";
-    case QPDFTokenizer::tt_bool:
-      return "bool";
-    case QPDFTokenizer::tt_word:
-      return "word";
-    case QPDFTokenizer::tt_eof:
-      return "eof";
-    case QPDFTokenizer::tt_space:
-      return "space";
-    case QPDFTokenizer::tt_comment:
-      return "comment";
-    case QPDFTokenizer::tt_inline_image:
-      return "inline-image";
-  }
-  return nullptr;
-}
-
-static std::optional<ImageInfo> get_image_info(QPDFObjectHandle resources, const std::string& name) {
-  if (resources.isNull() || !resources.isDictionary()) {
-    return std::nullopt;
-  }
-
-  QPDFObjectHandle xobjects = resources.getKey("/XObject");
-  if (xobjects.isNull() || !xobjects.isDictionary()) {
-    return std::nullopt;
-  }
-
-  if (!xobjects.hasKey(name)) {
-    return std::nullopt;
-  }
-
-  QPDFObjectHandle xobject = xobjects.getKey(name);
-  if (!xobject.isStream()) {
-    return std::nullopt;
-  }
-
-  QPDFObjectHandle dict = xobject.getDict();
-  bool is_image =
-      (dict.hasKey("/Subtype") && dict.getKey("/Subtype").isName() && dict.getKey("/Subtype").getName() == "/Image");
-
-  if (is_image) {
-    ImageInfo image_info;
-
-    if (dict.hasKey("/Height") && dict.getKey("/Height").isInteger()) {
-      image_info.height = dict.getKey("/Height").getIntValue();
-    }
-
-    if (dict.hasKey("/Width") && dict.getKey("/Width").isInteger()) {
-      image_info.width = dict.getKey("/Width").getIntValue();
-    }
-
-    return image_info;
-  }
-
-  return std::nullopt;
-  ;
-}
-
-PDFImageMapper::PDFImageMapper(int target_mcid) : target_mcid(target_mcid), in_target_mcid(false) {}
-
-void PDFImageMapper::push_cm(double value) {
-  if (cm_fixed_size_queue.size() == 6) {
-    cm_fixed_size_queue.pop_front();  // Remove oldest element
-  }
-  cm_fixed_size_queue.push_back(value);
-}
-
-void PDFImageMapper::find(QPDF& pdf) {
-  QPDFPageDocumentHelper doc_helper(pdf);
-  std::vector<QPDFPageObjectHelper> pages = doc_helper.getAllPages();
-  for (auto& page : pages) {
-    QPDFPageObjectHelper poh(page);
-
-    auto crop_box = poh.getCropBox();
-
-    this->find(poh);
-  }
-}
-
-struct OperatorInfo {
-  const char* name;
-  const char* description;
-  int operand_count;
-};
-
-static constexpr OperatorInfo OPERATOR_CM = {"cm", "Concatenate matrix (set CTM)", 6};
-static constexpr OperatorInfo OPERATOR_Q = {"q", "Save graphics state", 0};
-static constexpr OperatorInfo OPERATOR_Q_UPPER = {"Q", "Restore graphics state", 0};
-static constexpr OperatorInfo OPERATOR_DO = {"Do", "Invoke named XObject (draw image/form)", 1};
-static constexpr OperatorInfo OPERATOR_BDC = {"BDC", "Begin Marked Content sequence with property list", 2};
-static constexpr OperatorInfo OPERATOR_EMC = {"EMC", "End Marked Content sequence", 0};
-
-class CMDoExtractor : public QPDFObjectHandle::ParserCallbacks {
- public:
-  explicit CMDoExtractor(QPDFPageObjectHelper& page_ref) : page(page_ref) {
-    // You can add any other initialization logic here if needed
-  }
-
-  std::vector<double> current_matrix = {1, 0, 0, 1, 0, 0};
-  std::stack<std::vector<double>> matrix_stack;
-  std::vector<QPDFObjectHandle> operand_stack;
-
-  std::stack<int> mcid_stack;
-  int current_mcid;
-
-  void handleEOF() override {
-    // No action needed for this example
-  }
-
-  std::optional<ImageInfo> image_info(std::string imgName) {
-    QPDFObjectHandle resources = page.getObjectHandle().getKey("/Resources");
-
-    return get_image_info(resources, imgName);
-  }
-
-  void handleObject(QPDFObjectHandle obj, size_t offset, size_t length) override {
-    if (obj.isOperator()) {
-      std::string op = obj.getOperatorValue();
-
-      if (op == OPERATOR_CM.name && operand_stack.size() >= OPERATOR_CM.operand_count) {
-        auto numeric_at = [&](std::size_t distance_from_top) -> double {
-          auto& obj = operand_stack[operand_stack.size() - distance_from_top];
-          auto t = obj.getTypeName();  // "integer", "real", …
-          if (std::strcmp(t, "integer") != 0 && std::strcmp(t, "real") != 0) {
-            std::stringstream ss;
-            ss << "numeric operand expected (got " << t << ")";
-            throw std::runtime_error(ss.str());
-          }
-          return obj.getNumericValue();
-        };
-
-        // pull operands (top of stack is distance 1)
-        double f = numeric_at(1);
-        double e = numeric_at(2);
-        double d = numeric_at(3);
-        double c = numeric_at(4);
-        double b = numeric_at(5);
-        double a = numeric_at(6);
-        current_matrix = {a, b, c, d, e, f};
-
-        operand_stack.resize(operand_stack.size() - 6);
-      } else if (op == OPERATOR_Q.name) {
-        matrix_stack.push(current_matrix);
-      } else if (op == OPERATOR_Q_UPPER.name) {
-        if (!matrix_stack.empty()) {
-          current_matrix = matrix_stack.top();
-          matrix_stack.pop();
-        }
-      } else if (op == OPERATOR_BDC.name && operand_stack.size() >= OPERATOR_BDC.operand_count) {
-        mcid_stack.push(current_mcid);  // Save current MCID
-
-        QPDFObjectHandle properties_obj = operand_stack.back();  // Properties operand
-        // QPDFObjectHandle tag_obj = operand_stack[operand_stack.size() - 2]; // Tag operand
-
-        int mcid_val = -1;  // Default if MCID not found or not an integer
-        if (properties_obj.isDictionary()) {
-          if (properties_obj.hasKey("/MCID") && properties_obj.getKey("/MCID").isInteger()) {
-            mcid_val = properties_obj.getKey("/MCID").getIntValue();
-          }
-        } else if (properties_obj.isName()) {
-          // Properties is a name, look it up in Resources /Properties dictionary
-          QPDFObjectHandle resources = page.getObjectHandle().getKey("/Resources");
-          if (resources.isDictionary() && resources.hasKey("/Properties")) {
-            QPDFObjectHandle properties_dict = resources.getKey("/Properties");
-            if (properties_dict.isDictionary() && properties_dict.hasKey(properties_obj.getName())) {
-              QPDFObjectHandle actual_props = properties_dict.getKey(properties_obj.getName());
-              if (actual_props.isDictionary() && actual_props.hasKey("/MCID") &&
-                  actual_props.getKey("/MCID").isInteger()) {
-                mcid_val = actual_props.getKey("/MCID").getIntValue();
-              }
-            }
-          }
-        }
-        current_mcid = mcid_val;
-        operand_stack.resize(operand_stack.size() - OPERATOR_BDC.operand_count);
-      } else if (op == OPERATOR_EMC.name) {  // EMC takes 0 operands
-        if (!mcid_stack.empty()) {
-          current_mcid = mcid_stack.top();
-          mcid_stack.pop();
-        } else {
-          current_mcid = -1;  // Reset to default if stack underflow (should be balanced)
-        }
-        // EMC takes 0 operands, so no change to operand_stack based on operand_count
-      } else if (op == OPERATOR_DO.name && !operand_stack.empty()) {
-        std::string imgName = operand_stack.back().getName();
-        auto image_info = this->image_info(imgName);
-
-        std::copy(current_matrix.begin(), current_matrix.end(), image_info->cm_matrix.begin());
-
-        image_info->mcid = current_mcid;
-
-        image_info->bbox = compute_bbox(image_info->width, image_info->height, image_info->cm_matrix);
-
-        QPDFObjectHandle mb = page.getObjectHandle().getKey("/MediaBox");
-
-        double px0 = mb.getArrayItem(0).getNumericValue();
-        double py0 = mb.getArrayItem(1).getNumericValue();
-        double px1 = mb.getArrayItem(2).getNumericValue();
-        double py1 = mb.getArrayItem(3).getNumericValue();
-
-        auto& bb = image_info->bbox;   // shorthand reference
-        bb[0] = std::max(bb[0], px0);  // left  edge
-        bb[1] = std::max(bb[1], py0);  // bottom edge
-        bb[2] = std::min(bb[2], px1);  // right edge
-        bb[3] = std::min(bb[3], py1);  // top   edge
-
-        image_to_mcid[imgName] = *image_info;
-
-        operand_stack.pop_back();
-      } else {
-        // For other operators, just clear the operand stack
-        operand_stack.clear();
-      }
-    } else {
-      operand_stack.push_back(obj);
-    }
-  }
-
-  const std::map<std::string, ImageInfo>& getImageMap() const { return image_to_mcid; }
-
- private:
-  QPDFPageObjectHelper& page;
-  std::map<std::string, ImageInfo> image_to_mcid;
-};
-
-void PDFImageMapper::find(QPDFPageObjectHelper& page) {
-  CMDoExtractor cb(page);
-  page.parseContents(&cb);
-
-  const auto& extracted_map = cb.getImageMap();
-  image_to_mcid.insert(extracted_map.begin(), extracted_map.end());
-}

data/ext/qpdf_ruby/pdf_image_mapper.hpp

@@ -1,66 +0,0 @@
-// x_object_finder.hpp
-#pragma once
-
-#include <qpdf/QPDF.hh>
-#include <qpdf/QPDFPageDocumentHelper.hh>
-#include <qpdf/QPDFPageObjectHelper.hh>
-#include <qpdf/QPDFObjectHandle.hh>
-#include <qpdf/QIntC.hh>
-#include <qpdf/QPDFExc.hh>
-#include <qpdf/QTC.hh>
-#include <qpdf/QUtil.hh>
-#include <qpdf/BufferInputSource.hh>
-#include <qpdf/QPDFTokenizer.hh>
-#include <qpdf/Buffer.hh>
-
-#include <string>
-#include <vector>
-#include <iostream>
-#include <map>
-#include <stack>
-#include <regex>
-#include <sstream>
-#include <deque>
-
-struct ImageInfo {
-  int mcid;
-  double width;
-  double height;
-  std::array<double, 6> cm_matrix;            // stores the transformation matrix values
-  std::array<double, 4> bbox = {0, 0, 0, 0};  // user-space bounding box [llx, lly, urx, ury]
-
-  ImageInfo() : mcid(-1), width(0), height(0), cm_matrix{0, 0, 0, 0, 0, 0} {}
-
-  ImageInfo(int m, double w, double h, const std::array<double, 6>& cm) : mcid(m), width(w), height(h), cm_matrix(cm) {}
-
-  std::string to_string() const {
-    std::ostringstream oss;
-    oss << "ImageInfo(mcid=" << mcid << ", width=" << width << ", height=" << height << ", cm_matrix=[";
-    for (size_t i = 0; i < cm_matrix.size(); ++i) {
-      oss << cm_matrix[i];
-      if (i != cm_matrix.size() - 1) oss << ", ";
-    }
-    oss << "])";
-    return oss.str();
-  }
-};
-
-class PDFImageMapper {
- public:
-  explicit PDFImageMapper(int target_mcid);
-
-  void find(QPDF& pdf);
-  // Parses the page's content stream to find the XObject name for the MCID
-  void find(QPDFPageObjectHelper& page);
-
-  void push_cm(double value);
-
-  // Expose internal map for external use
-  const std::map<std::string, ImageInfo>& getImageMap() const { return image_to_mcid; }
-
- private:
-  int target_mcid;
-  bool in_target_mcid;
-  std::map<std::string, ImageInfo> image_to_mcid;
-  std::deque<double> cm_fixed_size_queue;
-};