From ae9455bf449a22329507efd01455d5e9322cf134 Mon Sep 17 00:00:00 2001
From: Jay Berkenbilt <ejb@ql.org>
Date: Sat, 22 Dec 2018 10:52:06 -0500
Subject: Implement --json-objects

---
 qpdf/qpdf.cc | 79 +++++++++++++++++++++++++++++++++++++++++++++---------------
 1 file changed, 60 insertions(+), 19 deletions(-)

diff --git a/qpdf/qpdf.cc b/qpdf/qpdf.cc
index 6cae61f1..acb2a948 100644
--- a/qpdf/qpdf.cc
+++ b/qpdf/qpdf.cc
@@ -195,6 +195,7 @@ struct Options
     bool show_page_images;
     bool json;
     std::set<std::string> json_keys;
+    std::set<std::string> json_objects;
     bool check;
     std::vector<PageSpec> page_specs;
     std::map<std::string, RotationSpec> rotations;
@@ -344,6 +345,26 @@ static JSON json_schema(std::set<std::string>* keys = 0)
     return schema;
 }
 
+static void parse_object_id(std::string const& objspec,
+                            bool& trailer, int& obj, int& gen)
+{
+    if (objspec == "trailer")
+    {
+        trailer = true;
+    }
+    else
+    {
+        trailer = false;
+        obj = QUtil::string_to_int(objspec.c_str());
+        size_t comma = objspec.find(',');
+        if ((comma != std::string::npos) && (comma + 1 < objspec.length()))
+        {
+            gen = QUtil::string_to_int(
+                objspec.substr(1 + comma, std::string::npos).c_str());
+        }
+    }
+}
+
 // This is not a general-purpose argument parser. It is tightly
 // crafted to work with qpdf. qpdf's command-line syntax is very
 // complex because of its long history, and it doesn't really follow
@@ -437,6 +458,7 @@ class ArgParser
     void argWithImages();
     void argJson();
     void argJsonKey(char* parameter);
+    void argJsonObject(char* parameter);
     void argCheck();
     void arg40Print(char* parameter);
     void arg40Modify(char* parameter);
@@ -638,6 +660,8 @@ ArgParser::initOptionTable()
     char const* jsonKeyChoices[] = {"objects", "pages", "pagelabels", 0};
     (*t)["json-key"] = oe_requiredChoices(
         &ArgParser::argJsonKey, jsonKeyChoices);
+    (*t)["json-object"] = oe_requiredParameter(
+        &ArgParser::argJsonObject, "trailer|obj[,gen]");
     (*t)["check"] = oe_bare(&ArgParser::argCheck);
 
     t = &this->encrypt40_option_table;
@@ -1143,21 +1167,7 @@ ArgParser::argShowXref()
 void
 ArgParser::argShowObject(char* parameter)
 {
-    if (strcmp(parameter, "trailer") == 0)
-    {
-        o.show_trailer = true;
-    }
-    else
-    {
-        char* obj = parameter;
-        char* gen = obj;
-        if ((gen = strchr(obj, ',')) != 0)
-        {
-            *gen++ = 0;
-            o.show_gen = QUtil::string_to_int(gen);
-        }
-        o.show_obj = QUtil::string_to_int(obj);
-    }
+    parse_object_id(parameter, o.show_trailer, o.show_obj, o.show_gen);
     o.require_outfile = false;
 }
 
@@ -1206,6 +1216,12 @@ ArgParser::argJsonKey(char* parameter)
     o.json_keys.insert(parameter);
 }
 
+void
+ArgParser::argJsonObject(char* parameter)
+{
+    o.json_objects.insert(parameter);
+}
+
 void
 ArgParser::argCheck()
 {
@@ -1691,7 +1707,11 @@ automated test suites for software that uses the qpdf library.\n\
 --json                    generate a json representation of the file\n\
 --json-help               describe the format of the json representation\n\
 --json-key=key            repeatable; prune json structure to include only\n\
-                          specified keys\n\
+                          specified keys. If absent, all keys are shown\n\
+--json-object=trailer|[obj,gen]\n\
+                          repeatable; include only specified objects in the\n\
+                          \"objects\" section of the json. If absent, all\n\
+                          objects are shown\n\
 \n\
 The json representation generated by qpdf is designed to facilitate\n\
 processing of qpdf from other programming languages that have a hard\n\
@@ -2586,14 +2606,35 @@ static void do_json_objects(QPDF& pdf, Options& o, JSON& j)
     // Add all objects. Do this first before other code below modifies
     // things by doing stuff like calling
     // pushInheritedAttributesToPage.
+    bool all_objects = o.json_objects.empty();
+    std::set<QPDFObjGen> wanted_og;
+    for (std::set<std::string>::iterator iter = o.json_objects.begin();
+         iter != o.json_objects.end(); ++iter)
+    {
+        bool trailer;
+        int obj = 0;
+        int gen = 0;
+        parse_object_id(*iter, trailer, obj, gen);
+        if (obj)
+        {
+            wanted_og.insert(QPDFObjGen(obj, gen));
+        }
+    }
     JSON j_objects = j.addDictionaryMember("objects", JSON::makeDictionary());
-    j_objects.addDictionaryMember("trailer", pdf.getTrailer().getJSON(true));
+    if (all_objects || o.json_objects.count("trailer"))
+    {
+        j_objects.addDictionaryMember(
+            "trailer", pdf.getTrailer().getJSON(true));
+    }
     std::vector<QPDFObjectHandle> objects = pdf.getAllObjects();
     for (std::vector<QPDFObjectHandle>::iterator iter = objects.begin();
          iter != objects.end(); ++iter)
     {
-        j_objects.addDictionaryMember(
-            (*iter).unparse(), (*iter).getJSON(true));
+        if (all_objects || wanted_og.count((*iter).getObjGen()))
+        {
+            j_objects.addDictionaryMember(
+                (*iter).unparse(), (*iter).getJSON(true));
+        }
     }
 }
 
-- 
cgit v1.2.3-54-g00ecf