[FIX] tools.pdf: do not crash when PDF has no attachments

In the PDF, `DictionaryObject` can be wrapped in `IndirectObject`. For nested
dictionaries, using `get` on a `DictionaryObject` will unwrap the result if it's
an `IndirectObject`, but not `__getitem__` so when trying to futher call `get`
on the result will cause an error: we need to unwrap the object first.
Instead, we patch `DictionaryObject.get()` for it to unwrap the object in case
it's an `IndirectObject`.

This is a follow-up commit of fece5ab1bf2fef043b131f1bd0886f0841116103

closes odoo/odoo#53269

X-original-commit: 189a0b28ec7fdb7472f936450cfd7b4bfd6912ed
Signed-off-by: Olivier Dony (odo) <odo@openerp.com>
Signed-off-by: bfr-o <bfr-o@users.noreply.github.com>

odoo/addons/base/tests/__init__.py

@@ -45,3 +45,4 @@ from . import test_tests_tags
 from . import test_base_document_layout
 from . import test_form_create
 from . import test_cloc
+from . import test_pdf

odoo/addons/base/tests/minimal.pdf

@@ -0,0 +1,44 @@
+%PDF-1.3
+%µ¶
+
+1 0 obj
+<</Type/Catalog/Pages 2 0 R/Names 5 0 R>>
+endobj
+
+2 0 obj
+<</Type/Pages/Kids[3 0 R]/Count 1>>
+endobj
+
+3 0 obj
+<</Type/Page/Parent 2 0 R/Resources<</Font<</F1<</Type/Font/Subtype/Type1/BaseFont/Arial>>>>>>/Contents 4 0 R>>
+endobj
+
+4 0 obj
+<</Length 44>>
+stream
+BT
+/F1 100 Tf
+10 400 Td
+(!Odoo Test!) Tj
+ET
+endstream
+endobj
+
+5 0 obj
+<</Foo/Bar>>
+endobj
+
+xref
+0 6
+0000000000 00001 f 
+0000000016 00000 n 
+0000000074 00000 n 
+0000000126 00000 n 
+0000000254 00000 n 
+0000000346 00000 n 
+
+trailer
+<</Size 6/Root 1 0 R>>
+startxref
+376
+%%EOF

odoo/addons/base/tests/test_pdf.py

@@ -0,0 +1,75 @@
+# -*- coding: utf-8 -*-
+# Part of Odoo. See LICENSE file for full copyright and licensing details.
+
+from odoo.tests.common import TransactionCase
+from odoo.tools import pdf
+from odoo.modules.module import get_module_resource
+import io
+
+
+class TestPdf(TransactionCase):
+    """ Tests on pdf. """
+
+    def setUp(self):
+        super().setUp()
+        file_path = get_module_resource('base', 'tests', 'minimal.pdf')
+        self.file = open(file_path, 'rb').read()
+        self.minimal_reader_buffer = io.BytesIO(self.file)
+        self.minimal_pdf_reader = pdf.OdooPdfFileReader(self.minimal_reader_buffer)
+
+    def test_odoo_pdf_file_reader(self):
+        attachments = list(self.minimal_pdf_reader.getAttachments())
+        self.assertEqual(len(attachments), 0)
+
+        pdf_writer = pdf.PdfFileWriter()
+        pdf_writer.cloneReaderDocumentRoot(self.minimal_pdf_reader)
+        pdf_writer.addAttachment('test_attachment.txt', b'My awesome attachment')
+
+        attachments = list(self.minimal_pdf_reader.getAttachments())
+        self.assertEqual(len(attachments), 1)
+
+    def test_odoo_pdf_file_writer(self):
+        attachments = list(self.minimal_pdf_reader.getAttachments())
+        self.assertEqual(len(attachments), 0)
+
+        pdf_writer = pdf.OdooPdfFileWriter()
+        pdf_writer.cloneReaderDocumentRoot(self.minimal_pdf_reader)
+
+        pdf_writer.addAttachment('test_attachment.txt', b'My awesome attachment')
+        attachments = list(self.minimal_pdf_reader.getAttachments())
+        self.assertEqual(len(attachments), 1)
+
+        pdf_writer.addAttachment('another_attachment.txt', b'My awesome OTHER attachment')
+        attachments = list(self.minimal_pdf_reader.getAttachments())
+        self.assertEqual(len(attachments), 2)
+
+    def test_merge_pdf(self):
+        self.assertEqual(self.minimal_pdf_reader.getNumPages(), 1)
+        page = self.minimal_pdf_reader.getPage(0)
+
+        merged_pdf = pdf.merge_pdf([self.file, self.file])
+        merged_reader_buffer = io.BytesIO(merged_pdf)
+        merged_pdf_reader = pdf.OdooPdfFileReader(merged_reader_buffer)
+        self.assertEqual(merged_pdf_reader.getNumPages(), 2)
+        merged_reader_buffer.close()
+
+    def test_branded_file_writer(self):
+        # It's not easy to create a PDF with PyPDF2, so instead we copy minimal.pdf with our custom pdf writer
+        pdf_writer = pdf.PdfFileWriter()  # BrandedFileWriter
+        pdf_writer.cloneReaderDocumentRoot(self.minimal_pdf_reader)
+        writer_buffer = io.BytesIO()
+        pdf_writer.write(writer_buffer)
+        branded_content = writer_buffer.getvalue()
+        writer_buffer.close()
+
+        # Read the metadata of the newly created pdf.
+        reader_buffer = io.BytesIO(branded_content)
+        pdf_reader = pdf.PdfFileReader(reader_buffer)
+        pdf_info = pdf_reader.getDocumentInfo()
+        self.assertEqual(pdf_info['/Producer'], 'Odoo')
+        self.assertEqual(pdf_info['/Creator'], 'Odoo')
+        reader_buffer.close()
+
+    def tearDown(self):
+        super().tearDown()
+        self.minimal_reader_buffer.close()

odoo/tools/pdf.py

@@ -12,6 +12,17 @@ import hashlib
 DEFAULT_PDF_DATETIME_FORMAT = "D:%Y%m%d%H%M%S+00'00'"
 
 
+# make sure values are unwrapped by calling the specialized __getitem__
+def _unwrapping_get(self, key, default=None):
+    try:
+        return self[key]
+    except KeyError:
+        return default
+
+
+DictionaryObject.get = _unwrapping_get
+
+
 class BrandedFileWriter(PdfFileWriter):
     def __init__(self):
         super().__init__()
@@ -25,7 +36,8 @@ PdfFileWriter = BrandedFileWriter
 
 
 def merge_pdf(pdf_data):
-    ''' Merge a collection of PDF documents in one
+    ''' Merge a collection of PDF documents in one.
+    Note that the attachments are not merged.
     :param list pdf_data: a list of PDF datastrings
     :return: a unique merged PDF datastring
     '''
@@ -40,7 +52,8 @@ def merge_pdf(pdf_data):
 
 
 def rotate_pdf(pdf):
-    ''' Rotate clockwise PDF (90°)
+    ''' Rotate clockwise PDF (90°) into a new PDF.
+    Note that the attachments are not copied.
     :param pdf: a PDF to rotate
     :return: a PDF rotated
     '''