Convert scientific notation in JSON to fixed point (fixes #1079)

JSON accepts scientific notation, but PDF doesn't.

5 files changed, 113 insertions, 2 deletions

diff --git a/ChangeLog b/ChangeLog
index 7450ddbc..b267f4f6 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,9 @@
 2023-12-21  Jay Berkenbilt  <ejb@ql.org>
 
+	* Fix to QPDF JSON: a floating point number that appears in
+	scientific notation will be converted to fixed-point notation,
+	rounded to six digits after the decimal point. Fixes #1079.
+
 	* Fix to QPDF JSON: the syntax "n:/pdf-syntax" is now accepted as
 	an alternative way to represent names. This can be used for any
 	name (e.g. "n:/text#2fplain"), but it is necessary when the name
diff --git a/libqpdf/QPDF_json.cc b/libqpdf/QPDF_json.cc
index 864e1a56..7951b1e4 100644
--- a/libqpdf/QPDF_json.cc
+++ b/libqpdf/QPDF_json.cc
@@ -732,6 +732,15 @@ QPDF::JSONReactor::makeObject(JSON const& value)
         if (QUtil::is_long_long(str_v.c_str())) {
             result = QPDFObjectHandle::newInteger(QUtil::string_to_ll(str_v.c_str()));
         } else {
+            // JSON allows scientific notation, but PDF does not.
+            if (str_v.find('e') != std::string::npos || str_v.find('E') != std::string::npos) {
+                try {
+                    auto v = std::stod(str_v);
+                    str_v = QUtil::double_to_string(v);
+                } catch (std::exception&) {
+                    // Keep it as it was
+                }
+            }
             result = QPDFObjectHandle::newReal(str_v);
         }
     } else if (value.getString(str_v)) {
diff --git a/manual/release-notes.rst b/manual/release-notes.rst
index f720f99e..2fdd197f 100644
--- a/manual/release-notes.rst
+++ b/manual/release-notes.rst
@@ -52,6 +52,10 @@ Planned changes for future 12.x (subject to change):
       must be represented as ``"n:/one#a0two"`` since the single byte
       ``a0`` is not valid in JSON.
 
+    - QPDF JSON will convert floating numbers that appear in the JSON
+      in scientific notation to fixed-point notation since PDF doesn't
+      accept scientific notation.
+
   - Build Enhancements:
 
     - The qpdf test suite now passes when qpdf is linked with an
diff --git a/qpdf/qtest/qpdf-json.test b/qpdf/qtest/qpdf-json.test
index 9691d995..0ea126ec 100644
--- a/qpdf/qtest/qpdf-json.test
+++ b/qpdf/qtest/qpdf-json.test
@@ -347,16 +347,27 @@ $td->runtest("check C API write to JSON stream",
 # (using #xx) would generate invalid JSON, even though qpdf's own JSON
 # parser would accept it. Also, the JSON spec allows real numbers in
 # scientific notation, but the PDF spec does not.
-$n_tests += 2;
+$n_tests += 4;
 $td->runtest("handle binary names",
              {$td->COMMAND =>
                   "qpdf --json-output weird-tokens.pdf a.json"},
              {$td->STRING => "", $td->EXIT_STATUS => 0});
-# Round-trip is tested above.
+# Round-trip back to PDF is tested above.
 $td->runtest("check json",
              {$td->FILE => "a.json"},
              {$td->FILE => "weird-tokens.json"},
              $td->NORMALIZE_NEWLINES);
+# Make sure we can properly handle JSON with scientific notation.
+$td->runtest("weird tokens round trip json",
+             {$td->COMMAND =>
+                  "qpdf --json-input --json-output weird-tokens.json -"},
+             {$td->FILE => "weird-tokens.json", $td->EXIT_STATUS => 0},
+             $td->NORMALIZE_NEWLINES);
+$td->runtest("weird tokens with scientific notation",
+             {$td->COMMAND =>
+                  "qpdf --json-input --json-output weird-tokens-alt.json -"},
+             {$td->FILE => "weird-tokens.json", $td->EXIT_STATUS => 0},
+             $td->NORMALIZE_NEWLINES);
 
 cleanup();
 $td->report($n_tests);
diff --git a/qpdf/qtest/qpdf/weird-tokens-alt.json b/qpdf/qtest/qpdf/weird-tokens-alt.json
new file mode 100644
index 00000000..5a9f8ff8
--- /dev/null
+++ b/qpdf/qtest/qpdf/weird-tokens-alt.json
@@ -0,0 +1,83 @@
+{
+  "qpdf": [
+    {
+      "jsonversion": 2,
+      "pdfversion": "2.0",
+      "pushedinheritedpageresources": false,
+      "calledgetallpages": false,
+      "maxobjectid": 6
+    },
+    {
+      "obj:1 0 R": {
+        "value": {
+          "/Extra": [
+            "u:Names with binary data",
+            "n:/ABCDEF+#ba#da#cc#e5",
+            "/ABCEDEF+π",
+            "n:/one+#a0two",
+            "n:/text#2fplain",
+            "u:Very small/large reals",
+            1e-05,
+            1e12
+          ],
+          "/Pages": "2 0 R",
+          "/Type": "/Catalog"
+        }
+      },
+      "obj:2 0 R": {
+        "value": {
+          "/Count": 1,
+          "/Kids": [
+            "3 0 R"
+          ],
+          "/Type": "/Pages"
+        }
+      },
+      "obj:3 0 R": {
+        "value": {
+          "/Contents": "4 0 R",
+          "/MediaBox": [
+            0,
+            0,
+            612,
+            792
+          ],
+          "/Parent": "2 0 R",
+          "/Resources": {
+            "/Font": {
+              "/F1": "6 0 R"
+            }
+          },
+          "/Type": "/Page"
+        }
+      },
+      "obj:4 0 R": {
+        "stream": {
+          "data": "QlQKICAvRjEgMjQgVGYKICA3MiA3MjAgVGQKICAoUG90YXRvKSBUagpFVAo=",
+          "dict": {}
+        }
+      },
+      "obj:5 0 R": {
+        "value": 44
+      },
+      "obj:6 0 R": {
+        "value": {
+          "/BaseFont": "/Helvetica",
+          "/Encoding": "/WinAnsiEncoding",
+          "/Subtype": "/Type1",
+          "/Type": "/Font"
+        }
+      },
+      "trailer": {
+        "value": {
+          "/ID": [
+            "b:42841c13bbf709d79a200fa1691836f8",
+            "b:728c020f464c3cf7e02c12605fa7d88b"
+          ],
+          "/Root": "1 0 R",
+          "/Size": 7
+        }
+      }
+    }
+  ]
+}