almet/dangerzone
1
0
Fork 0
mirror of https://github.com/freedomofpress/dangerzone.git synced 2025-04-28 18:02:38 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 11

Make PyMuPDF always log to stderr

Browse source 
PyMUPDF logs to stdout by default, which is problematic because we use
the stdout of the conversion process to read the pixel stream of a
document.

Make PyMuPDF always log to stderr, by setting the following environment
variables: PYMUPDF_MESSAGE and PYMUPDF_LOG.

Fixes #877
This commit is contained in:
Alex Pyrgiotis 2024-07-25 12:41:23 +03:00
parent 08f03b4bb4
commit 3f86e7b465
No known key found for this signature in database
GPG key ID: B6C15EBA0357C9AA
2 changed files with 28 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

11
dangerzone/conversion/doc_to_pixels.py
View file

@ -3,6 +3,17 @@ import os
import sys import sys
from typing import Dict, Optional from typing import Dict, Optional
# XXX: PyMUPDF logs to stdout by default [1]. The PyMuPDF devs provide a way [2] to log to
# stderr, but it's based on environment variables. These envvars are consulted at import
# time [3], so we have to set them here, before we import `fitz`.
#
# [1] https://github.com/freedomofpress/dangerzone/issues/877
# [2] https://github.com/pymupdf/PyMuPDF/issues/3135#issuecomment-1992625724
# [3] https://github.com/pymupdf/PyMuPDF/blob/9717935eeb2d50d15440d62575878214226795f9/src/__init__.py#L62-L63
os.environ["PYMUPDF_MESSAGE"] = "fd:2"
os.environ["PYMUPDF_LOG"] = "fd:2"
import fitz import fitz
import magic import magic

25
dangerzone/conversion/pixels_to_pdf.py
View file

@ -16,6 +16,16 @@ from typing import Optional
from .common import DEFAULT_DPI, DangerzoneConverter, get_tessdata_dir, running_on_qubes from .common import DEFAULT_DPI, DangerzoneConverter, get_tessdata_dir, running_on_qubes
# XXX: PyMUPDF logs to stdout by default [1]. The PyMuPDF devs provide a way [2] to log to
# stderr, but it's based on environment variables. These envvars are consulted at import
# time [3], so we have to set them here, before we import `fitz`.
#
# [1] https://github.com/freedomofpress/dangerzone/issues/877
# [2] https://github.com/pymupdf/PyMuPDF/issues/3135#issuecomment-1992625724
# [3] https://github.com/pymupdf/PyMuPDF/blob/9717935eeb2d50d15440d62575878214226795f9/src/__init__.py#L62-L63
os.environ["PYMUPDF_MESSAGE"] = "fd:2"
os.environ["PYMUPDF_LOG"] = "fd:2"
class PixelsToPDF(DangerzoneConverter): class PixelsToPDF(DangerzoneConverter):
async def convert( async def convert(
@ -50,14 +60,13 @@ class PixelsToPDF(DangerzoneConverter):
# The first few operations happen on a per-page basis. # The first few operations happen on a per-page basis.
page_size = len(untrusted_rgb_data) page_size = len(untrusted_rgb_data)
total_size += page_size total_size += page_size
with contextlib.redirect_stdout(io.StringIO()): pixmap = fitz.Pixmap(
pixmap = fitz.Pixmap( fitz.Colorspace(fitz.CS_RGB),
fitz.Colorspace(fitz.CS_RGB), width,
width, height,
height, untrusted_rgb_data,
untrusted_rgb_data, False,
False, )
)
pixmap.set_dpi(DEFAULT_DPI, DEFAULT_DPI) pixmap.set_dpi(DEFAULT_DPI, DEFAULT_DPI)
if ocr_lang: # OCR the document if ocr_lang: # OCR the document
self.update_progress( self.update_progress(