Files
mayan-edms/mayan/apps/ocr/api_views.py
Roberto Rosario 0699ad0556 Add support for new document page structure
Documents now have their own dedicated DocumentPage
submodel. The old DocumentPage is now called DocumentVersionPage.
This allows mappings between document pages and document version
pages, allowing renumbering, appending pages.
DocumentPages have a content_object to map them to any other
object. For now they only map to DocumentVersionPages.
New option added to the version upload form to append the
pages of the new version.
A new view was added to just append new pages with wraps the
new document version upload form and hides the append pages
checkbox set to True.
Add a new action, reset_pages to reset the pages of the
document to those of the latest version.

Missing: appending tests, checks for proper content_object in OCR and
document parsing.

Author: Roberto Rosario <roberto.rosario@mayan-edms.com>
Date:   Thu Oct 11 12:00:25 2019 -0400
2019-10-10 11:55:42 -04:00

98 lines
2.9 KiB
Python

from __future__ import absolute_import, unicode_literals
from django.shortcuts import get_object_or_404
from rest_framework import generics, status
from rest_framework.response import Response
from mayan.apps.documents.models import Document, DocumentVersion
from mayan.apps.rest_api.permissions import MayanPermission
from .models import DocumentVersionPageOCRContent
from .permissions import permission_ocr_content_view, permission_ocr_document
from .serializers import DocumentPageOCRContentSerializer
class APIDocumentOCRView(generics.GenericAPIView):
"""
post: Submit a document for OCR.
"""
mayan_object_permissions = {
'POST': (permission_ocr_document,)
}
permission_classes = (MayanPermission,)
queryset = Document.objects.all()
def get_serializer(self, *args, **kwargs):
return None
def get_serializer_class(self):
return None
def post(self, request, *args, **kwargs):
self.get_object().submit_for_ocr()
return Response(status=status.HTTP_202_ACCEPTED)
class APIDocumentVersionOCRView(generics.GenericAPIView):
"""
post: Submit a document version for OCR.
"""
lookup_url_kwarg = 'version_pk'
mayan_object_permissions = {
'POST': (permission_ocr_document,)
}
permission_classes = (MayanPermission,)
queryset = DocumentVersion.objects.all()
def get_document(self):
return get_object_or_404(klass=Document, pk=self.kwargs['document_pk'])
def get_queryset(self):
return self.get_document().versions.all()
def get_serializer(self, *args, **kwargs):
return None
def get_serializer_class(self):
return None
def post(self, request, *args, **kwargs):
self.get_object().submit_for_ocr()
return Response(status=status.HTTP_202_ACCEPTED)
class APIDocumentPageOCRContentView(generics.RetrieveAPIView):
"""
get: Returns the OCR content of the selected document page.
"""
lookup_url_kwarg = 'page_pk'
mayan_object_permissions = {
'GET': (permission_ocr_content_view,),
}
permission_classes = (MayanPermission,)
serializer_class = DocumentPageOCRContentSerializer
def get_document(self):
return get_object_or_404(klass=Document, pk=self.kwargs['document_pk'])
def get_document_version(self):
return get_object_or_404(
klass=self.get_document().versions.all(),
pk=self.kwargs['version_pk']
)
def get_queryset(self):
return self.get_document_version().pages.all()
def retrieve(self, request, *args, **kwargs):
instance = self.get_object()
try:
ocr_content = instance.ocr_content
except DocumentVersionPageOCRContent.DoesNotExist:
ocr_content = DocumentVersionPageOCRContent.objects.none()
serializer = self.get_serializer(ocr_content)
return Response(serializer.data)