caerp-sign-pdf 2024.1.3__tar.gz → 2024.2.0__tar.gz

caerp_sign_pdf-2024.2.0/CURRENT_VERSION

@@ -0,0 +1 @@
+2024.2.0

{caerp_sign_pdf-2024.1.3 → caerp_sign_pdf-2024.2.0}/PKG-INFO

@@ -1,12 +1,13 @@
 Metadata-Version: 2.1
 Name: caerp_sign_pdf
-Version: 2024.1.3
+Version: 2024.2.0
 Summary: caerp_sign_pdf
 Home-page: https://framagit.org/caerp/caerp_sign_pdf
 Author: Kilya
 Author-email: contact@kilya.net
 License: GPLv3
 Keywords: web wsgi bfg pylons pyramid caerp
+Platform: UNKNOWN
 Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
 Classifier: Programming Language :: Python
 Classifier: Framework :: Pyramid
@@ -69,3 +70,5 @@ Sous linux la signature d'un document PDF peut être vérifiée facilement en li
 
    pdfsig <monfichierpdf.pdf>
 
+
+

caerp_sign_pdf-2024.2.0/requirements.txt

@@ -0,0 +1 @@
+pyHanko[pkcs11,image-support,opentype,xmp]==0.25.1

{caerp_sign_pdf-2024.1.3 → caerp_sign_pdf-2024.2.0}/src/caerp_sign_pdf/public.py

@@ -1,13 +1,13 @@
 # -*- coding: utf-8 -*-
 import logging
 import hashlib
+import subprocess
 
 from io import BytesIO
 from pyhanko.pdf_utils.incremental_writer import IncrementalPdfFileWriter
 from pyhanko.sign import fields, signers
 from pyhanko.sign.general import SigningError
 from pyhanko.stamp import TextStampStyle
-from PyPDF4.pdf import PdfFileReader, PdfFileWriter
 from pyramid.httpexceptions import HTTPUnsupportedMediaType
 
 from caerp.views.files.controller import FileData
@@ -27,6 +27,10 @@ nouveau fichier PDF avant de le recharger."
 
 
 class SignPDFService(object):
+    """
+    The PDF digital signing service
+    """
+
     def __init__(self, context, request):
         self.context = context
         self.request = request
@@ -58,18 +62,44 @@ class SignPDFService(object):
         """
         return hashlib.md5(file_data.getvalue()).hexdigest()
 
-    def _get_safe_pdf(self, file_data: BytesIO) -> str:
+    def _get_clean_pdf(self, file_data: BytesIO) -> BytesIO:
         """
-        Create a new clean PDF buffer from the original
-        to avoid some format errors
+        Create a new clean PDF buffer from the original to avoid format errors
         """
-        writer = PdfFileWriter()
-        reader = PdfFileReader(file_data, strict=False)
-        num_pages = reader.getNumPages()
-        for page in range(num_pages):
-            writer.addPage(reader.getPage(page))
-        writer.write(file_data)
-        return file_data
+        logger.info(f"Cleaning PDF buffer...")
+        file_data.seek(0)
+        process = subprocess.Popen(
+            ["/usr/bin/pdftocairo", "-pdf", "-", "-"],
+            stdin=subprocess.PIPE,
+            stdout=subprocess.PIPE,
+        )
+        output = process.communicate(input=file_data.read())[0]
+        result = BytesIO()
+        result.write(output)
+        result.seek(0)
+        return result
+
+    def _get_pdf_stamp(self, display_stamp: bool) -> TextStampStyle:
+        """
+        Prepare physical stamp to apply on the PDF
+
+        :param bool display_stamp: Whether the stamp must be visible or not
+
+        :return TextStampStyle: The stamp object
+        """
+        stamp = None
+        if display_stamp:
+            stamp = TextStampStyle(
+                timestamp_format="%d/%m/%Y %H:%M:%S %Z",
+                stamp_text="Signé par: %(signer)s\nLe: %(ts)s",
+                border_width=2,
+            )
+        else:
+            stamp = TextStampStyle(
+                stamp_text="",
+                border_width=0,
+            )
+        return stamp
 
     def sign(
         self, pdf_data: FileData, node_id: int = None, with_stamp: bool = False
@@ -80,7 +110,8 @@ class SignPDFService(object):
 
         :param FileData pdf_data: The FileData object describing the PDF file
         :param int node_id: The related node's id if relevant
-        :param bool with_stamp: whether we want a stamp on result file or not
+        :param bool with_stamp: Whether we want a printed stamp on result file or not
+
         :return bool: either the PDF has been signed or not
         """
 
@@ -88,7 +119,7 @@ class SignPDFService(object):
 
         # Check mimetype
         if not pdf_data.mimetype == "application/pdf":
-            logger.warn("File is not a PDF, signing aborted")
+            logger.warning("File is not a PDF, signing aborted")
             return False
 
         # Load sign certificate
@@ -103,26 +134,18 @@ class SignPDFService(object):
             )
             raise Exception(ERROR_LOAD_CERTIFICATE_MSG)
 
-        # Prepare stamp (if needed)
-        stamp = None
-        if with_stamp:
-            stamp = TextStampStyle(
-                timestamp_format="%d/%m/%Y %H:%M:%S %Z",
-                stamp_text="Signé par: %(signer)s\nLe: %(ts)s",
-                border_width=2,
-            )
-        else:
-            stamp = TextStampStyle(
-                stamp_text="",
-                border_width=0,
-            )
+        # Clean PDF
+        try:
+            pdf_data.data = self._get_clean_pdf(pdf_data.data)
+        except Exception as error:
+            logger.error(f"Error while cleaning original PDF file : {error}")
+            self.sign(pdf_data, node_id, with_stamp, as_image=True)
 
-        # Open original PDF
+        # Open PDF
         try:
-            pdf_data.data = self._get_safe_pdf(pdf_data.data)
             pdf_writer = IncrementalPdfFileWriter(pdf_data.data, strict=False)
         except Exception as error:
-            logger.error(f"Error while opening original PDF file : {error}")
+            logger.error(f"Error while opening PDF file : {error}")
             raise HTTPUnsupportedMediaType(ERROR_PDF_FORMAT_MSG)
 
         # Add signature field on the PDF
@@ -134,8 +157,11 @@ class SignPDFService(object):
                 ),
             )
         except SigningError:
-            logger.warn("File is already signed, nothing to do")
+            logger.warning("File is already signed, nothing to do")
             return True
+        except Exception as error:
+            logger.error(f"Error while adding signature field : {error}")
+            raise HTTPUnsupportedMediaType(ERROR_PDF_FORMAT_MSG)
 
         # Sign PDF
         try:
@@ -143,12 +169,14 @@ class SignPDFService(object):
             pdf_signer = signers.PdfSigner(
                 meta,
                 signer=signer,
-                stamp_style=stamp,
+                stamp_style=self._get_pdf_stamp(with_stamp),
             )
             pdf_data.data = pdf_signer.sign_pdf(pdf_writer)
         except Exception as error:
             logger.error(f"Error while signing PDF file : {error}")
             raise HTTPUnsupportedMediaType(ERROR_PDF_FORMAT_MSG)
+
+        # Success
         logger.info(
             "File '{}' signed successfully by {} !".format(
                 pdf_data.name, signer.subject_name

{caerp_sign_pdf-2024.1.3 → caerp_sign_pdf-2024.2.0}/src/caerp_sign_pdf.egg-info/PKG-INFO

@@ -1,12 +1,13 @@
 Metadata-Version: 2.1
 Name: caerp-sign-pdf
-Version: 2024.1.3
+Version: 2024.2.0
 Summary: caerp_sign_pdf
 Home-page: https://framagit.org/caerp/caerp_sign_pdf
 Author: Kilya
 Author-email: contact@kilya.net
 License: GPLv3
 Keywords: web wsgi bfg pylons pyramid caerp
+Platform: UNKNOWN
 Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
 Classifier: Programming Language :: Python
 Classifier: Framework :: Pyramid
@@ -69,3 +70,5 @@ Sous linux la signature d'un document PDF peut être vérifiée facilement en li
 
    pdfsig <monfichierpdf.pdf>
 
+
+

caerp_sign_pdf-2024.2.0/src/caerp_sign_pdf.egg-info/requires.txt

@@ -0,0 +1 @@
+pyHanko[image-support,opentype,pkcs11,xmp]==0.25.1

caerp_sign_pdf-2024.1.3/CURRENT_VERSION

@@ -1 +0,0 @@
-2024.1.3

caerp_sign_pdf-2024.1.3/requirements.txt

@@ -1 +0,0 @@
-pyHanko[pkcs11,image-support,opentype,xmp]==0.25.0

caerp_sign_pdf-2024.1.3/src/caerp_sign_pdf.egg-info/requires.txt

@@ -1 +0,0 @@
-pyHanko[image-support,opentype,pkcs11,xmp]==0.25.0