def setUp(self):
self.app = self.layer['app']
self.portal = self.layer['portal']
self.request = self.layer['request']
with api.env.adopt_roles(['Manager']):
self.image = api.content.create(
self.portal,
'Image',
'test_image',
image=dummy_image(),
)
annotations = IAnnotations(self.image)
thumbnail = ImageField()
thumbnail.set(self.image, dummy_image().data)
annotations[THUMBNAIL_KEY] = list([thumbnail])
self.image_view = api.content.get_view(
'document_preview',
self.image,
self.request,
)
self.portal_view = api.content.get_view(
'document_preview',
self.portal,
self.request,
)
def __call__(self):
helperutilities.RUN_SHELL_COMMANDS = True
log = self.mklog()
log('converting preview images...')
IF = ImageField()
cat = self.context.portal_catalog
results = cat(Language='all', portal_type=TYPES, b_size=100000)
log(str(len(results))+'\n')
for result in results:
try:
ob = result.getObject()
except AttributeError:
continue
if not hasattr(ob, 'pagePictures') or len(ob.pagePictures)==0:
log("NOTFOUND: %s has no pagePictures\n" % result.getPath())
continue
first = ob.pagePictures[0]
if type(first) == type(IF):
continue
# wrap as blobfield
newpics = []
for pic in ob.pagePictures:
NF = ImageField()
NF.set(ob, pic)
newpics.append(NF)
ob.pagePictures = newpics
log("Success: %s converted\n" % result.getPath())
return
# Code to kill old pagePictures from history
# PH = portal.portal_historiesstorage
# REP = PH._getZVCRepo()
# cnt = 0
# for zvh in REP.objectValues():
# for vid in zvh.getVersionIds():
# data = zvh.getVersionById(vid)._data
# wob = data.getWrappedObject()
# if not hasattr(wob, 'object'):
# continue
# rob = wob.object
# if hasattr(rob, 'pagePictures'):
# pp = rob.pagePictures
# if len(pp)>0 and type(pp[0]) == type(''):
# print "found one"
# rob.pagePictures = []
# rob._p_changed = 1
#
#
def __call__(self):
annotations = IAnnotations(self.context)
# get the contents of the context
mimetype, payload = self.getPayload()
if mimetype:
basetype = mimetype.split('/')[0]
else:
basetype = None
if basetype in EXCLUDE_TYPES:
logger.warn('Type {0} is in excluded types, '
'skipping {1}'.format(
basetype,
self.context.getId())
)
annotations[PREVIEW_MESSAGE_KEY] = ('There is no preview for this '
'file type')
return
try:
converted = self.convert_on_server(payload, mimetype)
except ServerError as e:
if e.args and e.args[0].startswith("Error connecting"):
annotations[PREVIEW_MESSAGE_KEY] = (
'Could not contact conversion server')
else:
annotations[PREVIEW_MESSAGE_KEY] = (
'Sorry, this document type cannot be converted. There is '
'no preview available.')
return
except ConfigError:
converted = self.convert_locally(payload, mimetype)
pdfdata = FileField()
pdfdata.set(self.context, converted['pdfs'][0])
previewdata = []
thumbdata = []
for filedata in converted['previews'][:20]:
IF = ImageField()
IF.set(self.context, filedata)
previewdata.append(IF)
for filedata in converted['thumbnails'][:20]:
IF = ImageField()
IF.set(self.context, filedata)
thumbdata.append(IF)
annotations[PDF_VERSION_KEY] = pdfdata
annotations[PREVIEW_IMAGES_KEY] = previewdata
annotations[THUMBNAIL_KEY] = thumbdata
def run_conversion(self):
file_id = 'dump'
ext = '.html'
filename = file_id + ext
inputfilepath = self.html_dump(filename)
if not inputfilepath:
return
cmd = [
docsplit.binary, 'pdf', inputfilepath, '--output', self.storage_dir
]
docsplit._run_command(cmd)
pdf_path = os.path.join(self.storage_dir, file_id + '.pdf')
annotations = IAnnotations(self.context)
gsettings = self.gsettings
sizes = (
('large', gsettings.large_size),
('normal', gsettings.normal_size),
('small', gsettings.thumb_size),
)
outputfilepath = os.path.join(self.storage_dir, file_id + '.pdf')
with codecs.open(outputfilepath, 'r') as pdf_file:
annotations[PDF_VERSION_KEY] = pdf_file.read()
docsplit.dump_images(pdf_path,
self.storage_dir,
sizes=sizes,
format='gif')
thumb_dir = os.path.join(self.storage_dir, 'small')
thumbnails = sorted(os.listdir(thumb_dir))
preview_dir = os.path.join(self.storage_dir, 'large')
previews = sorted(os.listdir(preview_dir))
previewdata = []
thumbdata = []
for filename in previews[:20]:
IF = ImageField()
with open(os.path.join(preview_dir, filename)) as img:
IF.set(self.context, img.read())
previewdata.append(IF)
for filename in thumbnails[:20]:
IF = ImageField()
with open(os.path.join(thumb_dir, filename)) as img:
IF.set(self.context, img.read())
thumbdata.append(IF)
annotations[PREVIEW_IMAGES_KEY] = previewdata
annotations[THUMBNAIL_KEY] = thumbdata
def _getAllPageImages(context, size=(320,452)):
""" Generate the preview images for a pdf """
pdf = context.get_review_pdf()
# import pdb; pdb.set_trace()
if pdf:
pdf_data = pdf["blob"].open().read()
if not pdf or not pdf_data:
return "%s has no pdf" %(
context.absolute_url()), None
else:
# Split the pdf, one file per page
try:
split_pdf_pages = RunSubprocess(
"pdftk",
output_params="burst output")
except SubprocessException, e:
return e
split_pdf_pages.create_tmp_input(suffix=".pdf", data=pdf_data)
split_pdf_pages.create_tmp_output_dir()
split_pdf_pages.output_path = os.path.join(
split_pdf_pages.tmp_output_dir,
"%04d.pdf")
split_pdf_pages.run()
msg = tuple()
if split_pdf_pages.errors != "":
msg += ("Message from split_pdf_pages:"
"\n%s\n" % split_pdf_pages.errors,)
# Convert the pages to .gifs
# rewritten to have one converter step per page as we have seen process
# sizes larger than 2GB for 60 pages in a batch
for filename in glob.glob(split_pdf_pages.tmp_output_dir+"/*.pdf"):
pdf_to_image = RunSubprocess(
"convert",
input_params="-density 250",
input_path=filename,
output_params= "-resize %sx%s -background white -flatten" % (
size[0], size[1]))
outputname = '.'.join(filename.split("/")[-1].split('.')[:-1])+'.gif'
pdf_to_image.output_path = os.path.join(
split_pdf_pages.tmp_output_dir,
outputname)
pdf_to_image.run()
if pdf_to_image.errors != "":
msg += ("Message from pdfs_to_images:"
"\n%s\n" % pdf_to_image.errors,)
pdf_to_image.clean_up()
imgfiles = [gif for gif
in os.listdir(split_pdf_pages.tmp_output_dir)
if os.path.splitext(gif)[1] == '.gif']
imgfiles.sort()
pages = []
for img in imgfiles:
img = open(os.path.join(
split_pdf_pages.tmp_output_dir, img),
"r")
img_data = img.read()
pages.append(img_data)
img.close()
# Remove temporary files
split_pdf_pages.clean_up()
if pages:
imgfields = []
for img in pages:
IF = ImageField()
IF.set(context, img)
imgfields.append(IF)
setattr(context, 'pagePictures', imgfields)
return msg or "Successfully converted %s pages" % len(pages)
