Marty/pubs-extract
1
0
Fork 0
Code Issues Pull requests Projects Releases 2 Packages Activity

Add tags to formatted annotations

Browse source
This commit is contained in:
Marty Oehme 2022-12-24 18:37:36 +01:00
parent 6bb7ed9c9f
commit 5bddf97e58
Signed by: Marty
GPG key ID: 73BA40D5AFAF49C9
1 changed files with 15 additions and 12 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

15
extract/extract.py
View file

@ -40,6 +40,7 @@ class Annotation:
content: str = ""
page: int = 1
colors: Tuple = (0.0, 0.0, 0.0)
tag: str = None
def formatted(self, formatting):
output = formatting
@ -48,6 +49,7 @@ class Annotation:
r"{note}": self.content,
r"{page}": str(self.page),
r"{newline}": "\n",
r"{tag}": self.tag,
}
if self.text == "":
output = re.sub(r"{quote_begin}.*{quote_end}", "", output)
@ -77,6 +79,7 @@ class Annotation:
nearest = name
return nearest
class ExtractPlugin(PapersPlugin):
"""Extract annotations from any pdf document.
@ -105,7 +108,7 @@ class ExtractPlugin(PapersPlugin):
self.minimum_similarity = float(settings.get("minimum_similarity", 0.75))
self.formatting = settings.get(
"formatting",
"{newline}{quote_begin}> {quote} {quote_end}[{page}]{note_begin}{newline}Note: {note}{note_end}",
"{newline}{quote_begin}> {quote} {quote_end}[{page}]{note_begin}{newline}Note: {note} {note_end} #{tag}",
)
self.color_mapping = settings.get("color_mapping", {})
@ -189,7 +192,7 @@ class ExtractPlugin(PapersPlugin):
self.ui.error(f"Document {file} is broken: {e}")
return papers_annotated
def mapped_tag(self, colorname):
def tag_from_colorname(self, colorname):
return self.color_mapping.get(colorname)
def _gather_papers(self, conf, args):
@ -256,17 +259,17 @@ class ExtractPlugin(PapersPlugin):
for page in doc:
for annot in page.annots():
quote, note = self._retrieve_annotation_content(page, annot)
annotations.append(
Annotation(
a = Annotation(
file=filename,
paper=paper,
text=quote,
content=note,
colors=annot.colors,
type=annot.type,
type=annot.type[1],
page=(page.number or 0) + 1,
)
)
a.tag = self.tag_from_colorname(a.colorname)
annotations.append(a)
return annotations
def _retrieve_annotation_content(self, page, annotation):