pdf_helper: multi-attachments

Replace code by odoo.tools.pdf

pdf_helper/models/helper.py

@@ -1,11 +1,11 @@
 # Copyright 2022 Camptocamp SA
 # @author: Simone Orsi <[email protected]>
+# Copyright 2023 Jacques-Etienne Baudoux (BCIM) <[email protected]>
 # License LGPL-3.0 or later (http://www.gnu.org/licenses/lgpl).
-import logging
 
-from PyPDF2.utils import PdfReadError
+import logging
 
-from odoo import models
+from odoo import api, models
 
 from ..utils import PDFParser
 
@@ -18,16 +18,16 @@ class PDFHelper(models.AbstractModel):
 
     _PDF_PARSER_KLASS = PDFParser
 
+    @api.model
     def pdf_get_xml_files(self, pdf_file):
+        """Extract XML attachments from pdf
+
+        :param pdf_file: binary PDF file content
+        :returns: a dict like {$filename: $parsed_xml_file_obj}.
+        """
         parser = self._PDF_PARSER_KLASS(pdf_file)
         try:
             return parser.get_xml_files()
-        except self._pdf_get_xml_files_swallable_exceptions() as err:
-            # TODO: can't we catch specific exceptions?
-            # This try/except block was added to reflect what done
-            # in base_business_document_import till now.
+        except parser.get_xml_files_swallable_exceptions() as err:
             _logger.error("PDF file parsing failed: %s", str(err))
             return {}
-
-    def _pdf_get_xml_files_swallable_exceptions(self):
-        return (KeyError, PdfReadError)

pdf_helper/readme/CONTRIBUTORS.rst

@@ -1,2 +1,3 @@
 * Simone Orsi <[email protected]>
 * Alexis de Lattre <[email protected]>
+* Jacques-Etienne Baudoux (BCIM) <[email protected]>

pdf_helper/tests/test_helper.py

@@ -28,14 +28,14 @@ def test_parse_xml(self):
 
 
 class TestPDFHelper(TransactionCase):
-    def test_parse_xml(self):
+    def test_get_xml(self):
         pdf_content = read_test_file("pdf_with_xml_test.pdf", mode="rb")
         res = self.env["pdf.helper"].pdf_get_xml_files(pdf_content)
         fname, xml_root = tuple(res.items())[0]
         self.assertEqual(fname, "factur-x.xml")
         self.assertTrue(isinstance(xml_root, etree._Element))
 
-    def test_parse_xml_fail(self):
+    def test_get_xml_fail(self):
         with self.assertLogs(
             "odoo.addons.pdf_helper.models.helper", level="ERROR"
         ) as log_catcher:

pdf_helper/utils.py

@@ -5,17 +5,19 @@
 # License LGPL-3.0 or later (http://www.gnu.org/licenses/lgpl).
 
 import logging
-import mimetypes
 from io import BytesIO
+from struct import error as StructError
 
 from lxml import etree
 
-_logger = logging.getLogger(__name__)
-
 try:
-    import PyPDF2
+    from PyPDF2.errors import PdfReadError
 except ImportError:
-    _logger.debug("Cannot import PyPDF2")
+    from PyPDF2.utils import PdfReadError
+
+from odoo.tools.pdf import OdooPdfFileReader
+
+_logger = logging.getLogger(__name__)
 
 
 class PDFParser:
@@ -29,42 +31,18 @@ def get_xml_files(self):
         :returns: a dict like {$filename: $parsed_xml_file_obj}.
         """
         res = {}
-        with BytesIO(self.pdf_file) as fd:
-            xmlfiles = self._extract_xml_files(fd)
-            for filename, xml_obj in xmlfiles.items():
-                root = self._extract_xml_root(xml_obj)
-                if root is None or not len(root):
-                    continue
-                res[filename] = root
-        if res:
-            _logger.debug("Valid XML files found in PDF: %s", list(res.keys()))
+        with BytesIO(self.pdf_file) as buffer:
+            pdf_reader = OdooPdfFileReader(buffer, strict=False)
+
+            # Process embedded files.
+            for xml_name, content in pdf_reader.getAttachments():
+                try:
+                    res[xml_name] = etree.fromstring(content)
+                except Exception:
+                    _logger.debug("Non XML file found in PDF")
+            if res:
+                _logger.debug("Valid XML files found in PDF: %s", list(res.keys()))
         return res
 
-    def _extract_xml_files(self, fd):
-        pdf = PyPDF2.PdfFileReader(fd)
-        _logger.debug("pdf.trailer=%s", pdf.trailer)
-        pdf_root = pdf.trailer["/Root"]
-        _logger.debug("pdf_root=%s", pdf_root)
-        # TODO add support for /Kids
-        embeddedfiles = pdf_root["/Names"]["/EmbeddedFiles"]["/Names"]
-        i = 0
-        xmlfiles = {}  # key = filename, value = PDF obj
-        for embeddedfile in embeddedfiles[:-1]:
-            mime_res = mimetypes.guess_type(embeddedfile)
-            if mime_res and mime_res[0] in ["application/xml", "text/xml"]:
-                xmlfiles[embeddedfile] = embeddedfiles[i + 1]
-            i += 1
-        _logger.debug("xmlfiles=%s", xmlfiles)
-        return xmlfiles
-
-    def _extract_xml_root(self, xml_obj):
-        xml_root = None
-        try:
-            xml_file_dict = xml_obj.getObject()
-            _logger.debug("xml_file_dict=%s", xml_file_dict)
-            xml_string = xml_file_dict["/EF"]["/F"].getData()
-            xml_root = etree.fromstring(xml_string)
-        except Exception as err:
-            # TODO: can't we catch specific exceptions?
-            _logger.debug("_pdf_extract_xml_root failed: %s", str(err))
-        return xml_root
+    def get_xml_files_swallable_exceptions(self):
+        return (NotImplementedError, StructError, PdfReadError)