paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-04-02 13:45:10 -05:00
Code Issues Packages Projects Releases Wiki Activity

fix parse error of some documents by using gs

Browse Source
This commit is contained in:
Jens Pfeifle 2019-01-29 09:02:33 +01:00 committed by jenspfeifle
parent 3db98faf5c
commit 0220199766
1 changed files with 11 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
src/paperless_tesseract/parsers.py
View File

@ -45,13 +45,23 @@ class RasterisedDocumentParser(DocumentParser):
""" """
out_path = os.path.join(self.tempdir, "convert.png") out_path = os.path.join(self.tempdir, "convert.png")
gs_out_path = os.path.join(self.tempdir, "gs_out.png")
# Run convert to get a decent thumbnail # Run convert to get a decent thumbnail
# https://github.com/danielquinn/paperless/issues/447
# call gs first
environment = os.environ.copy()
cmd = ["gs", "-q", "-sDEVICE=pngalpha",
"-o", gs_out_path, self.document_path]
if not subprocess.Popen(cmd, env=environment).wait() == 0:
raise ParseError("Thumbnail (gs) failed at {}".format(cmd))
# then run convert on the output from gs
run_convert( run_convert(
self.CONVERT, self.CONVERT,
"-scale", "500x5000", "-scale", "500x5000",
"-alpha", "remove", "-alpha", "remove",
"{}[0]".format(self.document_path), "gs_out_path",
out_path out_path
) )