Add attachment information to the json output

17 files changed, 99 insertions, 2 deletions

diff --git a/ChangeLog b/ChangeLog
index be8d2564..f587b967 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,8 @@
 2021-02-10  Jay Berkenbilt  <ejb@ql.org>
 
+	* Add "attachments" as an additional json key, and add some
+	information about attachments to the json output.
+
 	* Add new command-line arguments for operating on attachments:
 	--list-attachments, --add-attachment, --remove-attachment,
 	--copy-attachments-from. See --help and manual for details.
diff --git a/manual/qpdf-manual.xml b/manual/qpdf-manual.xml
index 5205028d..6885aff6 100644
--- a/manual/qpdf-manual.xml
+++ b/manual/qpdf-manual.xml
@@ -5104,6 +5104,19 @@ print "\n";
           than using <option>@file</option> for this purpose.
          </para>
         </listitem>
+        <listitem>
+         <para>
+          Add some information about attachments to the json output,
+          and added <literal>attachments</literal> as an additional
+          json key. The information included here is limited to the
+          preferred name and content stream and a reference to the
+          file spec object. This is enough detail for clients to avoid
+          the hassle of navigating a name tree and provides what is
+          needed for basic enumeration and extraction of attachments.
+          More detailed information can be obtained by following the
+          reference to the file spec object.
+         </para>
+        </listitem>
        </itemizedlist>
       </listitem>
       <listitem>
diff --git a/qpdf/qpdf.cc b/qpdf/qpdf.cc
index d23fad09..e41be64b 100644
--- a/qpdf/qpdf.cc
+++ b/qpdf/qpdf.cc
@@ -699,6 +699,22 @@ static JSON json_schema(std::set<std::string>* keys = 0)
             "filemethod",
             JSON::makeString("encryption method for attachments"));
     }
+    if (all_keys || keys->count("attachments"))
+    {
+        JSON attachments = schema.addDictionaryMember(
+            "attachments", JSON::makeDictionary());
+        JSON details = attachments.addDictionaryMember(
+            "<attachment-key>", JSON::makeDictionary());
+        details.addDictionaryMember(
+            "filespec",
+            JSON::makeString("object containing the file spec"));
+        details.addDictionaryMember(
+            "preferredname",
+            JSON::makeString("most preferred file name"));
+        details.addDictionaryMember(
+            "preferredcontents",
+            JSON::makeString("most preferred embedded file stream"));
+    }
     return schema;
 }
 
@@ -1114,7 +1130,7 @@ ArgParser::initOptionTable()
     // places: json_schema, do_json, and initOptionTable.
     char const* json_key_choices[] = {
         "objects", "objectinfo", "pages", "pagelabels", "outlines",
-        "acroform", "encrypt", 0};
+        "acroform", "encrypt", "attachments", 0};
     (*t)["json-key"] = oe_requiredChoices(
         &ArgParser::argJsonKey, json_key_choices);
     (*t)["json-object"] = oe_requiredParameter(
@@ -4568,6 +4584,28 @@ static void do_json_encrypt(QPDF& pdf, Options& o, JSON& j)
         "filemethod", JSON::makeString(s_file_method));
 }
 
+static void do_json_attachments(QPDF& pdf, Options& o, JSON& j)
+{
+    JSON j_attachments = j.addDictionaryMember(
+        "attachments", JSON::makeDictionary());
+    QPDFEmbeddedFileDocumentHelper efdh(pdf);
+    for (auto const& iter: efdh.getEmbeddedFiles())
+    {
+        std::string const& key = iter.first;
+        auto fsoh = iter.second;
+        auto j_details = j_attachments.addDictionaryMember(
+            key, JSON::makeDictionary());
+        j_details.addDictionaryMember(
+            "filespec",
+            JSON::makeString(fsoh->getObjectHandle().unparse()));
+        j_details.addDictionaryMember(
+            "preferredname", JSON::makeString(fsoh->getFilename()));
+        j_details.addDictionaryMember(
+            "preferredcontents",
+            JSON::makeString(fsoh->getEmbeddedFileStream().unparse()));
+    }
+}
+
 static void do_json(QPDF& pdf, Options& o)
 {
     JSON j = JSON::makeDictionary();
@@ -4628,6 +4666,10 @@ static void do_json(QPDF& pdf, Options& o)
     {
         do_json_encrypt(pdf, o, j);
     }
+    if (all_keys || o.json_keys.count("attachments"))
+    {
+        do_json_attachments(pdf, o, j);
+    }
 
     // Check against schema
 
diff --git a/qpdf/qtest/qpdf.test b/qpdf/qtest/qpdf.test
index 348d3948..dba10181 100644
--- a/qpdf/qtest/qpdf.test
+++ b/qpdf/qtest/qpdf.test
@@ -523,7 +523,7 @@ $td->runtest("page operations on form xobject",
 show_ntests();
 # ----------
 $td->notify("--- File Attachments ---");
-$n_tests += 33;
+$n_tests += 34;
 
 open(F, ">auto-txt") or die;
 print F "from file";
@@ -547,6 +547,10 @@ $td->runtest("list attachments verbose",
 	     {$td->COMMAND => "qpdf --list-attachments --verbose a.pdf"},
 	     {$td->FILE => "test76-list-verbose.out", $td->EXIT_STATUS => 0},
 	     $td->NORMALIZE_NEWLINES);
+$td->runtest("attachments json",
+	     {$td->COMMAND => "qpdf --json --json-key=attachments a.pdf"},
+	     {$td->FILE => "test76-json.out", $td->EXIT_STATUS => 0},
+	     $td->NORMALIZE_NEWLINES);
 $td->runtest("remove attachment (test_driver)",
 	     {$td->COMMAND => "test_driver 77 test76.pdf"},
 	     {$td->STRING => "test 77 done\n", $td->EXIT_STATUS => 0},
diff --git a/qpdf/qtest/qpdf/direct-pages-json.out b/qpdf/qtest/qpdf/direct-pages-json.out
index 0bd43563..52e5e2dd 100644
--- a/qpdf/qtest/qpdf/direct-pages-json.out
+++ b/qpdf/qtest/qpdf/direct-pages-json.out
@@ -4,6 +4,7 @@
     "hasacroform": false,
     "needappearances": false
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/json-field-types---show-encryption-key.out b/qpdf/qtest/qpdf/json-field-types---show-encryption-key.out
index 7bc40bdb..ad9c2003 100644
--- a/qpdf/qtest/qpdf/json-field-types---show-encryption-key.out
+++ b/qpdf/qtest/qpdf/json-field-types---show-encryption-key.out
@@ -385,6 +385,7 @@
     "hasacroform": true,
     "needappearances": true
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/json-field-types.out b/qpdf/qtest/qpdf/json-field-types.out
index 7bc40bdb..ad9c2003 100644
--- a/qpdf/qtest/qpdf/json-field-types.out
+++ b/qpdf/qtest/qpdf/json-field-types.out
@@ -385,6 +385,7 @@
     "hasacroform": true,
     "needappearances": true
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/json-image-streams-all.out b/qpdf/qtest/qpdf/json-image-streams-all.out
index 8eaa4583..3dea8852 100644
--- a/qpdf/qtest/qpdf/json-image-streams-all.out
+++ b/qpdf/qtest/qpdf/json-image-streams-all.out
@@ -4,6 +4,7 @@
     "hasacroform": false,
     "needappearances": false
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/json-image-streams-small.out b/qpdf/qtest/qpdf/json-image-streams-small.out
index dd7935ee..92d0c4f3 100644
--- a/qpdf/qtest/qpdf/json-image-streams-small.out
+++ b/qpdf/qtest/qpdf/json-image-streams-small.out
@@ -4,6 +4,7 @@
     "hasacroform": false,
     "needappearances": false
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/json-image-streams-specialized.out b/qpdf/qtest/qpdf/json-image-streams-specialized.out
index 9c6567f5..c342f9e6 100644
--- a/qpdf/qtest/qpdf/json-image-streams-specialized.out
+++ b/qpdf/qtest/qpdf/json-image-streams-specialized.out
@@ -4,6 +4,7 @@
     "hasacroform": false,
     "needappearances": false
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/json-image-streams.out b/qpdf/qtest/qpdf/json-image-streams.out
index 734868a5..2cfbd531 100644
--- a/qpdf/qtest/qpdf/json-image-streams.out
+++ b/qpdf/qtest/qpdf/json-image-streams.out
@@ -4,6 +4,7 @@
     "hasacroform": false,
     "needappearances": false
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/json-outlines-with-actions.out b/qpdf/qtest/qpdf/json-outlines-with-actions.out
index 25776abe..9755d0b8 100644
--- a/qpdf/qtest/qpdf/json-outlines-with-actions.out
+++ b/qpdf/qtest/qpdf/json-outlines-with-actions.out
@@ -4,6 +4,7 @@
     "hasacroform": false,
     "needappearances": false
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/json-outlines-with-old-root-dests.out b/qpdf/qtest/qpdf/json-outlines-with-old-root-dests.out
index 11735197..af3ce99c 100644
--- a/qpdf/qtest/qpdf/json-outlines-with-old-root-dests.out
+++ b/qpdf/qtest/qpdf/json-outlines-with-old-root-dests.out
@@ -4,6 +4,7 @@
     "hasacroform": false,
     "needappearances": false
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/json-page-labels-and-outlines.out b/qpdf/qtest/qpdf/json-page-labels-and-outlines.out
index b0de1c1d..e7d702f6 100644
--- a/qpdf/qtest/qpdf/json-page-labels-and-outlines.out
+++ b/qpdf/qtest/qpdf/json-page-labels-and-outlines.out
@@ -4,6 +4,7 @@
     "hasacroform": false,
     "needappearances": false
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/json-page-labels-num-tree.out b/qpdf/qtest/qpdf/json-page-labels-num-tree.out
index 497f428c..d0f73a61 100644
--- a/qpdf/qtest/qpdf/json-page-labels-num-tree.out
+++ b/qpdf/qtest/qpdf/json-page-labels-num-tree.out
@@ -4,6 +4,7 @@
     "hasacroform": false,
     "needappearances": false
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/page_api_2-json.out b/qpdf/qtest/qpdf/page_api_2-json.out
index b8ce4f38..172ce1c1 100644
--- a/qpdf/qtest/qpdf/page_api_2-json.out
+++ b/qpdf/qtest/qpdf/page_api_2-json.out
@@ -4,6 +4,7 @@
     "hasacroform": false,
     "needappearances": false
   },
+  "attachments": {},
   "encrypt": {
     "capabilities": {
       "accessibility": true,
diff --git a/qpdf/qtest/qpdf/test76-json.out b/qpdf/qtest/qpdf/test76-json.out
new file mode 100644
index 00000000..d1c88cf8
--- /dev/null
+++ b/qpdf/qtest/qpdf/test76-json.out
@@ -0,0 +1,23 @@
+{
+  "attachments": {
+    "att1": {
+      "filespec": "4 0 R",
+      "preferredcontents": "8 0 R",
+      "preferredname": "att1.txt"
+    },
+    "att2": {
+      "filespec": "5 0 R",
+      "preferredcontents": "10 0 R",
+      "preferredname": "att2.txt"
+    },
+    "att3": {
+      "filespec": "6 0 R",
+      "preferredcontents": "12 0 R",
+      "preferredname": "π.txt"
+    }
+  },
+  "parameters": {
+    "decodelevel": "generalized"
+  },
+  "version": 1
+}