def test_get_images_from_pdf():
from lembrar import image
from StringIO import StringIO
test_file_stream = resource_stream(__name__, 'test.pdf')
images = image.get_images_from_stream(test_file_stream)
assert len(images) == 2
# abuse the fact that get_images_from_stream would fail
# if it's own returned images aren't jpeg
assert len(image.get_images_from_stream(StringIO(images[0]))) == 1
assert len(image.get_images_from_stream(StringIO(images[1]))) == 1
def test_get_images_from_image():
from lembrar import image
test_file_stream = resource_stream(__name__, 'test.jpg')
images = image.get_images_from_stream(test_file_stream)
test_file_stream.seek(0)
assert len(images) == 1
assert images[0] == test_file_stream.read()
def add(request):
title = request.params['title']
created = request.params.get('created', datetime.utcnow()) or \
datetime.utcnow()
description = request.params.get('description', '')
force_detection = request.params.get("force_detection",
'true').lower() == 'true'
accepted_languages = request.registry.settings['accepted_languages']
imgstream = request.params['file'].file
doc_list = request.db.docs
for image in get_images_from_stream(imgstream):
try:
lang, img, text = recognize(image, accepted_languages,
force_detection)
except TypeError, e:
err_msg = "Error: " + str(e)
request.session.flash(err_msg, 'failure')
return HTTPServerError(explanation=err_msg,
detail='Go back, unselect force '\
'detection, and try again')
text += " " + description + " " + title
if text:
search_terms = list(index(text, [lang]))
else:
search_terms = ''
thumb = get_thumbnail(image)
doc_list.insert({
'img': Binary(image),
'thumb': Binary(thumb),
'created': created,
'version': 4,
'forced_detection': force_detection,
'language': lang,
'keywords': [],
'search_terms': search_terms,
'title': title
})
def add(request):
title = request.params['title']
created = request.params.get('created', datetime.utcnow()) or \
datetime.utcnow()
description = request.params.get('description', '')
force_detection = request.params.get("force_detection",
'true').lower() == 'true'
accepted_languages = request.registry.settings['accepted_languages']
imgstream = request.params['file'].file
doc_list = request.db.docs
for image in get_images_from_stream(imgstream):
try:
lang, img, text = recognize(image, accepted_languages,
force_detection)
except TypeError, e:
err_msg = "Error: " + str(e)
request.session.flash(err_msg, 'failure')
return HTTPServerError(explanation=err_msg,
detail='Go back, unselect force '\
'detection, and try again')
text += " " + description + " " + title
if text:
search_terms = list(index(text, [lang]))
else:
search_terms = ''
thumb = get_thumbnail(image)
doc_list.insert({'img': Binary(image),
'thumb': Binary(thumb),
'created': created,
'version': 4,
'forced_detection': force_detection,
'language': lang,
'keywords': [],
'search_terms': search_terms,
'title': title})
def test_get_images_from_bad():
from lembrar import image
test_file_stream = resource_stream(__name__, 'test_image.py')
images = image.get_images_from_stream(test_file_stream)
