def factory(resource_type, html, casebook=None, ordinals=None):
# create casebook, resource, resource_target, and annotations
if not casebook:
casebook = CasebookFactory()
SectionFactory(casebook=casebook,
ordinals=[1],
display_ordinals=[1])
ordinals = [1, 1]
resource_target = {
'TextBlock': TextBlockFactory,
'LegalDocument': LegalDocumentFactory
}[resource_type](content=html)
resource_type = resource_type
resource = ResourceFactory(casebook=casebook,
ordinals=ordinals,
display_ordinals=ordinals,
resource_type=resource_type,
resource=resource_target)
# retrieve the processed, cleansed html of the saved resource
processed_html = resource.resource.content
# strip html tags, break apart the text, and get annotation brackets and offsets
text = re.sub(r'<[^>]+?>', '', processed_html)
html_strs, annotation_offsets, annotation_strs = re_split_offsets(
r'\[/?.+?\]', text)
# resave the resource's content with the annotating brackets stripped
content = re.sub(r'\[.*?\]', '', processed_html)
resource.resource.content = content
resource.resource.save()
# create each annotation. to support overlapping annotations, pop off each starting annotation and then find
# the nearest ending annotation:
annotations = list(zip(annotation_strs, annotation_offsets))
while annotations:
annotation_str, annotation_offset = annotations.pop(0)
kind, content = (annotation_str[1:-1].split(" ", 1) + [None])[:2]
closing_annotation_str = f'[/{kind}]'
try:
closing_index = next(
i for i in range(len(annotations))
if annotations[i][0] == closing_annotation_str)
except StopIteration:
raise Exception(
f"Closing annotation {closing_annotation_str} not found.")
_, closing_annotation_offset = annotations.pop(closing_index)
ContentAnnotationFactory(
resource=resource,
kind=kind,
content=content,
global_start_offset=annotation_offset,
global_end_offset=closing_annotation_offset)
return casebook, resource
def factory(resource_type, html, casebook=None, ordinals=None):
# break apart provided html and get annotation brackets and offsets
content = re.sub(r'\[.*?\]', '', html) # strip brackets
html = re.sub(r'<[^>]+?>', '', html) # strip html tags
html_strs, annotation_offsets, annotation_strs = re_split_offsets(
r'\[/?.+?\]', html)
# create casebook, resource, resource_target, and annotations
if not casebook:
casebook = CasebookFactory()
SectionFactory(casebook=casebook, ordinals=[1])
ordinals = [1, 1]
resource_target = {
'Case': CaseFactory,
'TextBlock': TextBlockFactory
}[resource_type](content=content)
resource = ResourceFactory(casebook=casebook,
ordinals=ordinals,
resource_type=resource_type,
resource_id=resource_target.id)
# create each annotation. to support overlapping annotations, pop off each starting annotation and then find
# the nearest ending annotation:
annotations = list(zip(annotation_strs, annotation_offsets))
while annotations:
annotation_str, annotation_offset = annotations.pop(0)
kind, content = (annotation_str[1:-1].split(" ", 1) + [None])[:2]
closing_annotation_str = '[/%s]' % kind
try:
closing_index = next(
i for i in range(len(annotations))
if annotations[i][0] == closing_annotation_str)
except StopIteration:
raise Exception("Closing annotation %s not found." %
closing_annotation_str)
_, closing_annotation_offset = annotations.pop(closing_index)
ContentAnnotationFactory(
resource=resource,
kind=kind,
content=content,
global_start_offset=annotation_offset,
global_end_offset=closing_annotation_offset)
return casebook, resource
def dump_annotated_text(case_or_textblock):
"""
Return an annotated Case or TextBlock as html with annotation [brackets]. Example:
>>> annotations_factory, *_ = [getfixture(f) for f in ['annotations_factory']]
>>> html = '<p>[replace]This[/replace] [highlight]is[/highlight] [elide]a[/elide] [note]case[/note].</p>'
>>> casebook, case = annotations_factory('Case', html)
>>> assert dump_annotated_text(case) == html
"""
text_strs, offsets, tags = re_split_offsets(r'<[^>]+?>', case_or_textblock.resource.content)
to_insert = list(zip(offsets, tags))
for annotation in case_or_textblock.annotations.filter(global_start_offset__gte=0):
to_insert.extend([
(annotation.global_start_offset, '[%s]' % annotation.kind),
(annotation.global_end_offset, '[/%s]' % annotation.kind),
])
content = "".join(text_strs)
for offset, text in sorted(to_insert, reverse=True):
content = content[:offset] + text + content[offset:]
return content
