def get_document_pdf(request, pk):
"""
Retrieve a document by primary key, as a PDF file. Note that this
is not a JSON view.
"""
document = request.user.documents.get(pk = pk)
assets = document.assets.filter(
asset_class__name = models.AssetClass.DOCUMENT,
mime_type__name = models.MimeType.PDF )
if len(assets) != 0:
meta = operations.instantiate_asset(assets[0])
try:
response = HttpResponse(open(meta['Local-Path'], 'rb'))
response['Content-Disposition'] = \
'attachment; filename=doc-%d.pdf' % document.pk
response['Content-Type'] = meta['Content-Type']
response['Content-Length'] = os.stat(meta['Local-Path']).st_size
response['ETag' ] = meta['ETag' ]
response['Last-Modified'] = meta['Last-Modified']
finally:
shutil.rmtree(os.path.dirname(meta['Local-Path']))
else:
response = HttpResponse(content_type = 'application/pdf')
response['Content-Disposition'] = \
'attachment; filename=doc-%d.pdf' % document.pk
pdf.render_document(document, response, request.user.username,
str(document.pk))
return response
def handle_work_item(processor, item):
""" Process a work item. The work item will be provided and its local
temp directory will be cleaned up by the process driver
framework. If this method does not raise an exception the
work item will also be removed from the work queue.
"""
document = item['Asset-Instance'].related_document
num_ocr_pages = document.pages.filter(
assets__asset_class__name=models.AssetClass.PAGE_TEXT).count()
if document.num_pages != num_ocr_pages:
raise NotReadyException(
"Postponing PDF generation, OCR not complete for pages")
pdf_stream = StringIO(
pdf.render_document(document,
output_buffer=StringIO(),
username=document.owner.username,
title=document.title).getvalue())
# trial account handling -- send the PDF in attachment
# and delete all associated assets
if handle_trial_account(document, pdf_stream.getvalue()):
return
# classify the document based on the creation time of its PDF asset
tag_document(document, datetime.timedelta(0,
UPLOAD_AGGREGATE_TIME_TRESHOLD))
pdf_assets = document.assets.filter(
asset_class__name=models.AssetClass.DOCUMENT,
mime_type__name=models.MimeType.PDF)
if len(pdf_assets) != 0:
pdf_asset = pdf_assets[0]
pdf_asset.producer = processor
operations.upload_asset_stream(pdf_asset, pdf_stream)
else:
pdf_asset = operations.create_asset_from_stream(
data_stream=pdf_stream,
owner=item['Owner'],
producer=processor,
asset_class=models.AssetClass.DOCUMENT,
related_document=document,
file_name=document.title,
parent=item['Asset-Instance'],
child_number=1,
mime_type=models.MimeType.PDF)
return [pdf_asset]
def handle_work_item(processor, item):
""" Process a work item. The work item will be provided and its local
temp directory will be cleaned up by the process driver
framework. If this method does not raise an exception the
work item will also be removed from the work queue.
"""
document = item["Asset-Instance"].related_document
num_ocr_pages = document.pages.filter(assets__asset_class__name=models.AssetClass.PAGE_TEXT).count()
if document.num_pages != num_ocr_pages:
raise NotReadyException("Postponing PDF generation, OCR not complete for pages")
pdf_stream = StringIO(
pdf.render_document(
document, output_buffer=StringIO(), username=document.owner.username, title=document.title
).getvalue()
)
# trial account handling -- send the PDF in attachment
# and delete all associated assets
if handle_trial_account(document, pdf_stream.getvalue()):
return
# classify the document based on the creation time of its PDF asset
tag_document(document, datetime.timedelta(0, UPLOAD_AGGREGATE_TIME_TRESHOLD))
pdf_assets = document.assets.filter(
asset_class__name=models.AssetClass.DOCUMENT, mime_type__name=models.MimeType.PDF
)
if len(pdf_assets) != 0:
pdf_asset = pdf_assets[0]
pdf_asset.producer = processor
operations.upload_asset_stream(pdf_asset, pdf_stream)
else:
pdf_asset = operations.create_asset_from_stream(
data_stream=pdf_stream,
owner=item["Owner"],
producer=processor,
asset_class=models.AssetClass.DOCUMENT,
related_document=document,
file_name=document.title,
parent=item["Asset-Instance"],
child_number=1,
mime_type=models.MimeType.PDF,
)
return [pdf_asset]
