paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-10-30 03:56:23 -05:00
Code Issues Packages Projects Releases Wiki Activity

Moved pyocr.get_available_tools() into a method

Browse Source
This commit is contained in:
Daniel Quinn
2016-02-21 02:24:05 +00:00
parent 5f0962bc3e
commit 3a7923e32d
2 changed files with 15 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
.travis.yml
View File

@@ -8,7 +8,8 @@ matrix:
env: TOXENV=py34 env: TOXENV=py34
- python: 3.5 - python: 3.5
env: TOXENV=py35 env: TOXENV=py35
- env: TOXENV=pep8 - python: 3.5
env: TOXENV=pep8
install: install:
- pip install --requirement requirements.txt - pip install --requirement requirements.txt

27
src/documents/consumer.py
View File

@@ -26,18 +26,6 @@ from .models import Sender, Tag, Document
from .languages import ISO639 from .languages import ISO639
def image_to_string(args):
self, png, lang = args
with Image.open(os.path.join(self.SCRATCH, png)) as f:
if self.OCR.can_detect_orientation():
try:
orientation = self.OCR.detect_orientation(f, lang=lang)
f = f.rotate(orientation["angle"], expand=1)
except TesseractError:
pass
return self.OCR.image_to_string(f, lang=lang)
class OCRError(Exception): class OCRError(Exception):
pass pass
@@ -61,7 +49,6 @@ class Consumer(object):
CONSUME = settings.CONSUMPTION_DIR CONSUME = settings.CONSUMPTION_DIR
THREADS = int(settings.OCR_THREADS) if settings.OCR_THREADS else None THREADS = int(settings.OCR_THREADS) if settings.OCR_THREADS else None
OCR = pyocr.get_available_tools()[0]
DEFAULT_OCR_LANGUAGE = settings.OCR_LANGUAGE DEFAULT_OCR_LANGUAGE = settings.OCR_LANGUAGE
REGEX_TITLE = re.compile( REGEX_TITLE = re.compile(
@@ -239,12 +226,24 @@ class Consumer(object):
with Pool(processes=self.THREADS) as pool: with Pool(processes=self.THREADS) as pool:
r = pool.map( r = pool.map(
image_to_string, itertools.product([self], pngs, [lang])) self.image_to_string, itertools.product(pngs, [lang]))
r = " ".join(r) r = " ".join(r)
# Strip out excess white space to allow matching to go smoother # Strip out excess white space to allow matching to go smoother
return re.sub(r"\s+", " ", r) return re.sub(r"\s+", " ", r)
def image_to_string(self, args):
png, lang = args
ocr = pyocr.get_available_tools()[0]
with Image.open(os.path.join(self.SCRATCH, png)) as f:
if ocr.can_detect_orientation():
try:
orientation = ocr.detect_orientation(f, lang=lang)
f = f.rotate(orientation["angle"], expand=1)
except TesseractError:
pass
return ocr.image_to_string(f, lang=lang)
def _guess_attributes_from_name(self, parseable): def _guess_attributes_from_name(self, parseable):
""" """
We use a crude naming convention to make handling the sender, title, We use a crude naming convention to make handling the sender, title,