BUG : Invalid font pointed during merge_resources (#1641)

Fixed #1640
py-pdf · Feb 25, 2023 · 5173238 · 5173238
1 parent 2e4b657
commit 5173238
Show file tree

Hide file tree

Showing 2 changed files with 21 additions and 8 deletions.
diff --git a/pypdf/_page.py b/pypdf/_page.py
@@ -608,6 +608,9 @@ def compute_unique_key(base_key: str) -> Tuple[str, bool]:
                 value.
             """
             value = page2res.raw_get(base_key)
+            # TODO : possible improvement : in case of writer, the indirect_reference
+            # can not be found because translated : this may be improved
+
             # try the current key first (e.g. "foo"), but otherwise iterate
             # through "foo-0", "foo-1", etc. new_res can contain only finitely
             # many keys, thus this'll eventually end, even if it's been crafted
@@ -643,7 +646,7 @@ def compute_unique_key(base_key: str) -> Tuple[str, bool]:
                 if is_pdf_writer:
                     new_res[newname] = page2res.raw_get(key).clone(pdf)
                     try:
-                        new_res[newname] = new_res[key].indirect_reference
+                        new_res[newname] = new_res[newname].indirect_reference
                     except AttributeError:
                         pass
                 else:

diff --git a/tests/test_page.py b/tests/test_page.py
@@ -1092,17 +1092,27 @@ def test_merge_page_resources_smoke_test():
 
 @pytest.mark.external
 def test_merge_transformed_page_into_blank():
-    url = "https://github.com/py-pdf/pypdf/files/10540507/visitcard.pdf"
-    name = "visitcard.pdf"
-    r = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
+    url = "https://github.com/py-pdf/pypdf/files/10768334/badges_3vjrh_7LXDZ_1-1.pdf"
+    name = "badges_3vjrh_7LXDZ_1.pdf"
+    r1 = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
+    url = "https://github.com/py-pdf/pypdf/files/10768335/badges_3vjrh_7LXDZ_2-1.pdf"
+    name = "badges_3vjrh_7LXDZ_2.pdf"
+    r2 = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
     w = PdfWriter()
     w.add_blank_page(100, 100)
+    w.pages[0].merge_translated_page(r1.pages[0], 0, 0, True, True)
+    w.pages[0].merge_translated_page(r2.pages[0], 1000, 1000, True, True)
+    assert (
+        w.pages[0]["/Resources"]["/Font"].raw_get("/F2+0").idnum
+        != w.pages[0]["/Resources"]["/Font"].raw_get("/F2+0-0").idnum
+    )
+    w.add_blank_page(100, 100)
     for x in range(4):
         for y in range(7):
-            w.pages[0].merge_translated_page(
-                r.pages[0],
-                x * r.pages[0].trimbox[2],
-                y * r.pages[0].trimbox[3],
+            w.pages[1].merge_translated_page(
+                r1.pages[0],
+                x * r1.pages[0].trimbox[2],
+                y * r1.pages[0].trimbox[3],
                 True,
                 True,
             )