Merge remote-tracking branch 'py-pdf/main' into iss2138

.github/workflows/github-ci.yaml

@@ -22,11 +22,13 @@ jobs:
     runs-on: ubuntu-20.04
     strategy:
       matrix:
-        python-version: ["3.6", "3.7", "3.8", "3.9", "3.10", "3.11", "3.12"]
-        use-crypto-lib: ["pycryptodome"]
+        python-version: ["3.7", "3.8", "3.9", "3.10", "3.11", "3.12"]
+        use-crypto-lib: ["cryptography"]
         include:
+          - python-version: "3.6"
+            use-crypto-lib: "pycryptodome"
           - python-version: "3.9"
-            use-crypto-lib: "cryptography"
+            use-crypto-lib: "pycryptodome"
           - python-version: "3.10"
             use-crypto-lib: ""
     steps:
@@ -72,14 +74,17 @@ jobs:
       run: |
         pip install -r requirements/ci-3.11.txt
       if: matrix.python-version == '3.11' || matrix.python-version == '3.12'
-    - name: Remove pycryptodome
+    - name: Remove pycryptodome and cryptography
       run: |
-        pip uninstall pycryptodome -y
-      if: matrix.use-crypto-lib != 'pycryptodome'
+        pip uninstall pycryptodome cryptography -y
     - name: Install cryptography
       run: |
         pip install cryptography
       if: matrix.use-crypto-lib == 'cryptography'
+    - name: Install pycryptodome
+      run: |
+        pip install pycryptodome
+      if: matrix.use-crypto-lib == 'pycryptodome'
     - name: Install pypdf
       run: |
         pip install .

CHANGELOG.md

@@ -1,5 +1,26 @@
 # CHANGELOG
 
+## Version 3.16.0, 2023-09-10
+
+### Security (SEC)
+-  Infinite recursion caused by IndirectObject clone (#2156)
+
+### New Features (ENH)
+-  Ease access to ViewerPreferences (#2144)
+
+### Bug Fixes (BUG)
+-  Catch the case where w[0] is an IndirectObject instead of an int (#2154)
+-  Cope with indirect objects in filters and remove deprecated code (#2177)
+-  Accept tabs in cmaps (#2174) / cope with extra space (#2151)
+-  Merge pages without resources (#2150)
+-  getcontents() shall return None if contents is NullObject (#2161)
+-  Fix conversion from 1 to LA (#2175)
+
+### Robustness (ROB)
+-  Accept XYZ with no arguments (#2178)
+
+[Full Changelog](https://github.com/py-pdf/pypdf/compare/3.15.5...3.16.0)
+
 ## Version 3.15.5, 2023-09-03
 
 ### Bug Fixes (BUG)

CONTRIBUTORS.md

@@ -32,9 +32,9 @@ history and [GitHubs 'Contributors' feature](https://github.com/py-pdf/pypdf/gra
 * [Majumder, Jonah](https://github.com/jonahmajumder)
 * [Manini, Lorenzo](https://github.com/lorenzomanini)
 * [maxbeer99](https://github.com/maxbeer99)
-* [nalin-udhaar](https://github.com/nalin-udhaar)
 * [McNeil, Karen](https://github.com/karenlmcneil): Arabic Language Support
 * [Mérino, Antoine](https://github.com/Merinorus)
+* [nalin-udhaar](https://github.com/nalin-udhaar)
 * [Paternault, Louis](https://framagit.org/spalax)
 * [Perrensen, Olsen](https://github.com/olsonperrensen)
 * [Pinheiro, Arthur](https://github.com/xilopaint)
@@ -43,11 +43,12 @@ history and [GitHubs 'Contributors' feature](https://github.com/py-pdf/pypdf/gra
 * [pubpub-zz](https://github.com/pubpub-zz): involved in community development
 * [Ramos, Leodanis Pozo](https://github.com/lpozo)
 * [RitchieP](https://github.com/RitchieP) | [LinkedIn](https://www.linkedin.com/in/ritchie-p-892b31115/) | [StackOverflow](https://stackoverflow.com/users/13328625/casual-r?tab=profile)
-* [Rogmann, Sascha](https://github.com/srogmann)
 * [robbiebusinessacc](https://github.com/robbiebusinessacc)
 * [Roder, Thomas](https://github.com/MrTomRod)
+* [Rogmann, Sascha](https://github.com/srogmann)
 * [Röthenbacher, Thomas](https://github.com/troethe)
 * [sietzeberends](https://github.com/sietzeberends)
+* [Stober, Marc](https://github.com/marcstober)
 * [Stüber, Timo](https://github.com/omit66)
 * [Thoma, Martin](https://github.com/MartinThoma): Maintainer of pypdf since April 2022. I hope to build a great community with many awesome contributors. [LinkedIn](https://www.linkedin.com/in/martin-thoma/) | [StackOverflow](https://stackoverflow.com/users/562769/martin-thoma) | [Blog](https://martin-thoma.com/)
 * [Thomas, Reuben](https://github.com/rrthomas)

docs/index.rst

@@ -35,6 +35,7 @@ You can contribute to `pypdf on GitHub <https://github.com/py-pdf/pypdf>`_.
    user/add-watermark
    user/reading-pdf-annotations
    user/adding-pdf-annotations
+   user/viewer-preferences
    user/forms
    user/streaming-data
    user/file-size

docs/user/viewer-preferences.md

@@ -0,0 +1,79 @@
+# Adding Viewer Preferences
+
+It is possible to set viewer preferences of the PDF file.
+These properties are described in Section 12.2 of the [PDF 1.7 specification](https://opensource.adobe.com/dc-acrobat-sdk-docs/pdfstandards/PDF32000_2008.pdf).
+
+Note that the `\ViewerPreferences` dictionary does not exist by default.
+If it's not already present, it must be created by calling the `create_viewer_preferences` method
+of the `PdfWriter` object.
+
+If viewer preferences exist in a PDF file being read with `PdfReader`,
+you can access them as properties of the `viewer_preferences` properties.
+Otherwise, the `viewer_preferences` property will be set to `None`.
+
+## Example
+
+```python
+from pypdf import PdfWriter
+from pypdf.generic import ArrayObject, NumberObject
+
+writer = PdfWriter()
+
+writer.create_viewer_preferences()
+
+# /HideToolbar
+writer.viewer_preferences.hide_toolbar = True
+# /HideMenubar
+writer.viewer_preferences.hide_menubar = True
+# /HideWindowUI
+writer.viewer_preferences.hide_windowui = True
+# /FitWindow
+writer.viewer_preferences.fit_window = True
+# /CenterWindow
+writer.viewer_preferences.center_window = True
+# /DisplayDocTitle
+writer.viewer_preferences.display_doctitle = True
+
+# /NonFullScreenPageMode
+writer.viewer_preferences.non_fullscreen_pagemode = "/UseNone"  # default
+writer.viewer_preferences.non_fullscreen_pagemode = "/UseOutlines"
+writer.viewer_preferences.non_fullscreen_pagemode = "/UseThumbs"
+writer.viewer_preferences.non_fullscreen_pagemode = "/UseOC"
+
+# /Direction
+writer.viewer_preferences.direction = "/L2R"  # default,
+writer.viewer_preferences.direction = "/R2L"
+
+# /ViewArea
+writer.viewer_preferences.view_area = "/CropBox"
+# /ViewClip
+writer.viewer_preferences.view_clip = "/CropBox"
+# /PrintArea
+writer.viewer_preferences.print_area = "/CropBox"
+# /PrintClip
+writer.viewer_preferences.print_clip = "/CropBox"
+
+# /PrintScaling
+writer.viewer_preferences.print_scaling = "/None"
+writer.viewer_preferences.print_scaling = "/AppDefault"  # default according to PDF spec
+
+# /Duplex
+writer.viewer_preferences.duplex = "/Simplex"
+writer.viewer_preferences.duplex = "/DuplexFlipShortEdge"
+writer.viewer_preferences.duplex = "/DuplexFlipLongEdge"
+
+# /PickTrayByPDFSize
+writer.viewer_preferences.pick_tray_by_pdfsize = True
+# /PrintPageRange
+writer.viewer_preferences.print_pagerange = ArrayObject(
+    [NumberObject("1"), NumberObject("10"), NumberObject("20"), NumberObject("30")]
+)
+# /NumCopies
+writer.viewer_preferences.num_copies = 2
+
+for i in range(40):
+    writer.add_blank_page(10, 10)
+
+with open("output.pdf", "wb") as output_stream:
+    writer.write(output_stream)
+```

pypdf/_cmap.py

@@ -410,7 +410,7 @@ def compute_space_width(
         else:
             w = []
         while len(w) > 0:
-            st = w[0]
+            st = w[0] if isinstance(w[0], int) else w[0].get_object()
             second = w[1].get_object()
             if isinstance(second, int):
                 for x in range(st, second):

pypdf/_codecs/adobe_glyphs.py

@@ -48,7 +48,6 @@
 #   (2) Unicode scalar value--four uppercase hexadecimal digits
 #
 adobe_glyphs = {
-    "/.notdef": "\u0000",
     "/A": "\u0041",
     "/AA": "\uA732",
     "/AE": "\u00C6",

pypdf/_crypt_providers/_fallback.py

@@ -28,7 +28,7 @@
 from pypdf._crypt_providers._base import CryptBase
 from pypdf.errors import DependencyError
 
-_DEPENDENCY_ERROR_STR = "PyCryptodome is required for AES algorithm"
+_DEPENDENCY_ERROR_STR = "cryptography>=3.1 is required for AES algorithm"
 
 
 crypt_provider = ("local_crypt_fallback", "0.0.0")

pypdf/_page.py

@@ -1183,8 +1183,13 @@ def _merge_page_writer(
         pdf = self.indirect_reference.pdf
 
         rename = {}
+        if PG.RESOURCES not in self:
+            self[NameObject(PG.RESOURCES)] = DictionaryObject()
         original_resources = cast(DictionaryObject, self[PG.RESOURCES].get_object())
-        page2resources = cast(DictionaryObject, page2[PG.RESOURCES].get_object())
+        if PG.RESOURCES not in page2:
+            page2resources = DictionaryObject()
+        else:
+            page2resources = cast(DictionaryObject, page2[PG.RESOURCES].get_object())
 
         for res in (
             RES.EXT_G_STATE,

pypdf/_version.py

@@ -1 +1 @@
-__version__ = "3.15.5"
+__version__ = "3.16.0"

pypdf/_writer.py

@@ -384,7 +384,7 @@ def viewer_preferences(self) -> Optional[ViewerPreferences]:
                 self._root_object[NameObject(CD.VIEWER_PREFERENCES)] = o
         return o
 
-    def create_viewer_preference(self) -> ViewerPreferences:
+    def create_viewer_preferences(self) -> ViewerPreferences:
         o = ViewerPreferences()
         self._root_object[NameObject(CD.VIEWER_PREFERENCES)] = self._add_object(o)
         return o
@@ -862,6 +862,7 @@ def _update_text_field(self, field: DictionaryObject) -> None:
         # Extract font information
         da = cast(str, field[AA.DA])
         font_properties = da.replace("\n", " ").replace("\r", " ").split(" ")
+        font_properties = [x for x in font_properties if x != ""]
         font_name = font_properties[font_properties.index("Tf") - 2]
         font_height = float(font_properties[font_properties.index("Tf") - 1])
         if font_height == 0: