papis-extract/papis_extract/__init__.py
Marty Oehme 5a6d672c76
refactor: Move formatting logic to formatters
Formatters (previously templates) were pure data containers before,
continating the 'template' for how things should be formatted using
mustache. The formatting would be done a) in the exporters and b) in the
annotations.

This spread of formatting has now been consolidated into the Formatter,
which fixes the overall spread of formatting code and now can coherently
format a whole output instead of just individual annotations.

A formatter contains references to all documents and contained
annotations and will format everything at once by default, but the
formatting function can be invoked with reference to a specific
annotated document to only format that.

This commit should put more separation into the concerns of exporter and
formatter and made formatting a concern purely of the formatters and
annotation objects.
2023-09-20 09:14:58 +02:00

98 lines
2.8 KiB
Python

import click
import papis.cli
import papis.config
import papis.document
import papis.logging
import papis.notes
import papis.strings
from papis.document import Document
from papis_extract import extractor, exporter
from papis_extract.formatter import MarkdownFormatter, Formatter
logger = papis.logging.get_logger(__name__)
DEFAULT_OPTIONS = {
"plugins.extract": {
"tags": {},
"on_import": False,
"minimum_similarity": 0.75, # for checking against existing annotations
"minimum_similarity_content": 0.9, # for checking if highlight or note
"minimum_similarity_color": 0.833, # for matching tag to color
}
}
papis.config.register_default_settings(DEFAULT_OPTIONS)
@click.command("extract")
@click.help_option("-h", "--help")
@papis.cli.query_argument()
@papis.cli.doc_folder_option()
@papis.cli.git_option(help="Commit changes made to the notes files.")
@papis.cli.all_option()
@click.option(
"--write/--no-write",
"-w",
help="Do not write annotations to notes only print results to stdout.",
)
@click.option(
"--manual/--no-manual",
"-m",
help="Open note in editor for manual editing after annotation extraction.",
)
@click.option(
"--template",
"-t",
type=click.Choice(["markdown", "csv"], case_sensitive=False),
help="Choose an output template to format annotations with.",
)
def main(
query: str,
# info: bool,
# _papis_id: bool,
# _file: bool,
# notes: bool,
# _dir: bool,
# _format: str,
_all: bool,
doc_folder: str,
manual: bool,
write: bool,
template: str,
git: bool,
) -> None:
"""Extract annotations from any pdf document.
The extract plugin allows manual or automatic extraction of all annotations
contained in the pdf documents belonging to entries of the papis library.
It can write those changes to stdout or directly create and update notes
for papis documents.
It adds a `papis extract` subcommand through which it is invoked, but can
optionally run whenever a new document is imported for a papis entry,
if set in the plugin configuration.
"""
documents = papis.cli.handle_doc_folder_query_all_sort(
query, doc_folder, sort_field=None, sort_reverse=False, _all=_all
)
if not documents:
logger.warning(papis.strings.no_documents_retrieved_message)
return
if template == "csv":
raise NotImplementedError
run(documents, edit=manual, write=write, git=git, template=MarkdownFormatter())
def run(
documents: list[Document],
template: Formatter,
edit: bool = False,
write: bool = False,
git: bool = False,
) -> None:
template.annotated_docs = extractor.start(documents)
if write:
exporter.to_notes(template, edit=edit, git=git)
else:
exporter.to_stdout(template)