bubblbu / zotnote Goto Github PK

Retrieve some basic statistics:

• Number of reading notes
• Number of written words
• Number of direct quotes
• Average age of articles

etc.

Currently the retrieved fieldValues from Zotero are hardcoded.

This is the feature roadmap for ZotNote.

At this stage of the project, this roadmap will change quite a bit. However, each feature release will have a project page associated.

All previous changes are logged in the changelog.

0.4 | Testing

• Testing
  • Implement tests with pytest
  • Use Coveralls
  • Setup Travis CI
• Documentation
  • Choose framework
  • Write super basic documentation
• Manage multiple results for citekey search
  • Find out if we can retrieve formatted citations through BBT
• Fundamentally rethink and rewrite the model of Zotero items <-> Notes including the role of citekeys and Zotero IDs
  • Add Item class for objects retrieved from BBT
  • Rework Note class

0.5 | Collections & Tags

• Development of collections
  • Manage multiple notes as one collection. e.g., zotnote create -c PhD creates a new PhD collection. zotnote add -c PhD [citekey] will create a new reading note for [citekey] in that collection
• Development of the tags
  • Add possibility to tag reading notes. Import might be provided from Zotero, but also provided through a command line option (e.g., zotnote add -t important [citekey]

0.6 | Exports

• Development of reading note export
  • Export individual papers to different formats using pandoc:
    • .pdf (priority)
    • .docx

0.7 | Reports

• Development of reports
  • Create reports on collections or sets of reading notes (selected by tags)
  • Use Jupyter notebooks to export report as PDF or HTML
  • Basic analytics of progress, wordcounts, etc.

0.8 | Advanced exports and collections

• Process collections/set of notes into note summaries such as annotated bibliographies

This is a list of CLIs that also manages a library of items through various levels of categorization.

• Taskwarrior: Each item can be part of a project (which themselves are hiararchical) and can be tagged (tags/items have a many2many relationship)

> zotnote

Traceback (most recent call last):
  File "/home/username/.local/bin/zotnote", line 5, in <module>
    from zotnote.application import main

ModuleNotFoundError: No module named 'zotnote.application'

This is a major issue that affects various aspects of the software.

The underlying model of Zotero items, citekeys, and notes will change how certain functions (add, edit notes) should behave in case of name conflicts or changes.

• We have to assume that citekeys can be duplicates as this will depend on the user settings of BBT
• Should we assume that each item in Zotero can have multiple notes
• Citekeys can change which would be an argument for note filenames based on Zotero item IDs, however, at the same time legible filenames are nice. Also, even though the citekey is a bad unique identifier it is the citekey which will serve as an anchor for users when referring to items.

All notes could be converted to an HTML page including use of various tags/collections.

• Explore and read notes in the browser with tagging and linking
• Provide some of the advanced analysis features for collections
• Is this actually the best way to access/use Zotnote?

Importing a video recording is currently failing because the author/creator labels are not the same as with text documents.

Citekeys managed by BBT are per default only unique within a single collection. Accessing your library with a citekey might therefore lead to mutliple results.

Suggested solution:

• Display individual retrieved items and let user confirm if the version is correct.

With the recent update that annotations are extracted properly, this feature is becoming a lot more interesting. A possible Zotnote workflow:

1. Import article into Zotero
2. Read and annotate using compatible PDF reader
3. Extract annotations into a Zotero internal note
4. Use Zotnote to import that article and run some basic processing of the note to get to a nice literature note
5. Use Zotnote to create reports, annotated bibliographies etc.

BBT (starting v5.2.22) is now also providing an option to directly import notes from Zotero. This could be interesting for anyone wanting to import notes and access advanced features with those notes.

Options:

• Joplin - provides an API to manage notes and folders
• Zettlr

Currently, ZotNote is accessing is opening the better-bibtex.sqlite.bak (the backup of the database) database because of locking issues once Zotero is running. This obviously comes with several drawbacks and issues.

I am hoping that a better approach might be the scripting API that better-bibtex provides. https://retorque.re/zotero-better-bibtex/exporting/scripting/

Bibtex doesn't dictate a single format of citekeys therefore we shouldn't be checking for a specific format.

Template files should probably have a few common fields. E.g., yaml headers

Candidates:

• Sphinx with autogeneration from docstrings
• Write a simple but clean doc with mkdocs

Maybe existing notes should never be overwritten.

No existing note => create note => open note to edit.
Existing note => open note to edit.

Should be for free?

ZotNote is already managing multiple templates. Currently the template that is used for the notes is hardcoded.

Make this an option with default value.