Relax checking of Forms, assume Form XObject if in doubt.

Previously we had handled Form XObjects which were missing a Subtype
by checking for a FormType, and only throwing an error if the FormType
was also missing.

It appears we have several files in the test suite where both keys
are missing (!!) and Acrobat simply assumes these are Form XObjects. So
here we relax the checking and assume that anything missing a Subtype
is in fact a From XObject. Several files now work.

1 files changed, 7 insertions, 9 deletions

diff --git a/pdf/pdf_image.c b/pdf/pdf_image.c
index 81544afe8..6d6bb502e 100644
--- a/pdf/pdf_image.c
+++ b/pdf/pdf_image.c
@@ -2185,15 +2185,13 @@ int pdfi_do_image_or_form(pdf_context *ctx, pdf_dict *stream_dict,
     code = pdfi_dict_get(ctx, xobject_dict, "Subtype", (pdf_obj **)&n);
     if (code < 0) {
         if (code == gs_error_undefined) {
-            code = pdfi_dict_get(ctx, xobject_dict, "FormType", (pdf_obj **)&n);
-            if (code >= 0) {
-                pdfi_countdown(n);
-                code = pdfi_name_alloc(ctx, (byte *)"Form", 4, (pdf_obj **)&n);
-                pdfi_countup(n);
-                ctx->pdf_errors |= E_PDF_NO_SUBTYPE;
-            }
-            else
-                goto exit;
+            /* This is illegal, because we have no way to tell is an XObject is a Form
+             * or Image object. However it seems Acrobat just assumes that it's a Form!
+             * See test file /tests_private/pdf/PDFIA1.7_SUBSET/CATX2063.pdf
+             */
+            code = pdfi_name_alloc(ctx, (byte *)"Form", 4, (pdf_obj **)&n);
+            pdfi_countup(n);
+            ctx->pdf_errors |= E_PDF_NO_SUBTYPE;
         }
         else
             goto exit;