dpriskorn/asseeibot

Reads the change flow of Wikipedia and output DOI, ISBN, etc. identifiers as they are found

asseeibot's Issues

Factor out the matcher to it’s own library

Create a simple interface

Collect statistics and show the user at runtime

As a user I would like to know how many subjects I have added over time to wikidata using the tool.
Save using pandas pickle with the columns

• Item
• Subject item
• Original subject
• Whether label or alias was used to match
• datetime.now()
• rolled_back False

One row for each successful upload

Show the progress to the user every x upload

This can be used for rollback of bad edits later also...

Support cite book template also

https://github.com/earwig/mwparserfromhell

Add introduction

Tell the user how the program works and how to stop

Use pydantic to improve strict typing of classes

Crashes regularly with response error

this is because of upstream error
aio-libs/aiohttp#3904
It should be ignored, if anyone know how to do that, please comment.

save also match with insufficient rating

maybe this warrants a rename of Matches -> MatchResults

Rewrite the output to use pandas dataframes

Advantages:

• standardized
• lots of output possibilities
• easy to get an overview/statistics
• quite portable

Factor out subject logic to own class

improve splitting of subject

'Renewable Energy, Sustainability and the Environment' -> split along both comma and " and "
clean "the" from beginning of subjects?

scrub parentheses like these from the end: Environmental Science (miscellaneous)

Switch to click

Create an event object and let it hold all the other objects

With a WikipediaPage object
Which has one or more Reference objects like CiteJournal or CiteBook
The Reference object has
methods like lookup-doi or lookup-isbn

Find source for CC0 ISBN data

Where does SourceMD get data from?

Delete match from the cache

As a user I want to delete a bad match from the cache so I don't end up adding garbage to wikidata

E. G. python asseeibot.py --delete-match qid

Create model for a WikipediaCitationsPage

It should be able to check whether a reference is missing in https://wikipediacitations.wiki.opencura.com/
It could also enrich the graph by adding missing references.

New Web UI?

Rewrite as a backend and new frontend with a Web UI.

The idea that is that the community together approve and decline matches in the UI.

If a match is approved by at least two users it is uploaded to all items.

This requires the backend to have a copy of the whole crossref database.

New statistics flag

As a user I would like to see pretty statistics about my uploaded contributions

For example in a table that shows uploads per month

I would like to know how the ratio between label/alias matches looked over time.

rewrite: After approving a match it is not in the list nor uploaded

yes was answered here

Setup CI

piggyback on the configuration here by @cclauss https://github.com/internetarchive/wcdimportbot/blob/main/.github/workflows/lint_python.yml

Add information to match about whether alias/label was used

update the model