Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

rerun generation of ATM operator list #3352

Merged
merged 2 commits into from Oct 9, 2021

Conversation

matkoniecz
Copy link
Member

note that countries like Australia and USA have no longer lists of ATM operators split by state - I think that it is fine. Autocomplete works and it seems that benefit of such splits is not huge.

But I can add also that - at cost of some coding.

BTW, if someone is interested in cleanup of operator tags on ATM in their area I can share Python code that I created to support this (usable in manually reviewed edits and in fully automatic bot edits).

If someone cleaned operator tags in their area - let me know and I can run metadata update in their country.

note that countries like Australia and USA have no longer lists of ATM operators split by state
@matkoniecz matkoniecz marked this pull request as draft October 8, 2021 23:13
@matkoniecz
Copy link
Member Author

AW, AX, BA - I need to investigate why atm data disappeared for them.

@mnalis
Copy link
Member

mnalis commented Oct 9, 2021

Uh, so what is preferred way to fix that list in SC? Editing every bank and ATMs in whole country and fixing tags manually does not sound very appealing (even if JOSM does offer some semi-automated ways to help). Because in HR there are many duplicates for all banks in this PR; just for example this all refers to the same bank (first is preferred notation):

  • Zagrebačka banka
  • Zagrebacka banka
  • Zagrebacka Banka
  • Zagrebačka Banka
  • Zagrebačka banka d.d.
  • Zaba
  • ZABA

I could manually clean and suggest correct list for app/src/main/assets/country_metadata/HR.yml, but if it is just going to be overwritten on next run, then that is not ideal solution either...

What does your python code do to help with cleanup, @matkoniecz? I might give it a try, if it does not generate too much work.

@kmpoppe
Copy link
Contributor

kmpoppe commented Oct 9, 2021

I mean, the best way to fix duplicates would be to use wikidata for the operators, but looking at the absolute bucketload of entries for DE this is going to be a crazy task to fix ... I think I'll get some peeps together to fix that 😉

@matkoniecz
Copy link
Member Author

matkoniecz commented Oct 9, 2021

What does your python code do to help with cleanup, @matkoniecz? I might give it a try, if it does not generate too much work.

It makes such retagging much easier to do by automating most of it while still allowing review. Let me know if something can be improved.

It is a template that allows specifying:

  • unwanted operator values
  • wanted operator value
  • where cache file should be stored

Then it runs an Overpass query and for each element it

  • opens given element in browser - for review
  • lists tags of the object, lists what wold be edited
  • asks user whether edit should be made or not
  • on accepting the edit program makes the edit, it is possible to skip given edit. Typical edit looks like

It handles some things like retagging also name - but please do not use it blindly. Let me know if something is buggy, but potential users take a full responsibility. It is best to not use without looking through the code (it is relatively simple).

It is possible to configure it to perform edits fully automatically, but that would require permission from local community (see https://wiki.openstreetmap.org/wiki/Automated_Edits_code_of_conduct and https://wiki.openstreetmap.org/wiki/Mechanical_Edits/Mateusz_Konieczny_-_bot_account/fix_overly_verbose_Euronet_Sp._z_o.o. for a template of documentation page).

Practical hint: keep script window open on one part of the screen, browser on another so opening objects for review will not require you to switch windows. I even keep them on separate desktops, with tags reviewed before potential edit, and location reviewed after edits (in some cases I edit also nearby objects or open notes, reverts to already made edits are basically never required).

Example edit: https://www.openstreetmap.org/changeset/112288731

Code: https://gitlab.com/matkoniecz/osm-automation/-/blob/master/script_assisted_cleanup/tag_migrator_normalize_bank_polskiej_sp%C3%B3%C5%82dzielczo%C5%9Bci.py (AGPL licensed)

If you open issue/PR on gitlab feel free to ping me here - not sure are their email notifications actually working.

I mean, the best way to fix duplicates would be to use wikidata for the operators

https://github.com/osmlab/name-suggestion-index is trying to do this, not sure about progress with ATMs. In principle their name matching could be used for eliminating unwanted names and/or providing suggestions "did you mean XYZ". But as far as I know the ATM data is not sufficient and not worth using (so far!). Let me know if it is an outdated info!

@matkoniecz matkoniecz marked this pull request as ready for review October 9, 2021 07:03
@matkoniecz
Copy link
Member Author

matkoniecz commented Oct 9, 2021

China, Canada, Australia, India would be now having one atm operator list per country.

I think that missing special support for Aruba (32 mapped ATMs in total - http://overpass-turbo.eu/s/1bTy ) and similar admin_level=3 areas is not worth effort.

Though I will not merge without @westnordost review and I can add this ATM data back (via expanding code that generates it to support also admin_level=3 states)

See https://www.openstreetmap.org/relation/1231749 https://en.wikipedia.org/wiki/Aruba

Though Iceland, Mongolia, Bosnia and Herzegovina should have this data...

@matkoniecz matkoniecz marked this pull request as draft October 9, 2021 07:14
@matkoniecz matkoniecz marked this pull request as ready for review October 9, 2021 07:39
Copy link
Member

@westnordost westnordost left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@westnordost westnordost merged commit 7f15d82 into streetcomplete:master Oct 9, 2021
@FloEdelmann FloEdelmann added the hacktoberfest-accepted pull request that should be treated as eligible for Hacktoberfest event label Oct 10, 2021
@matkoniecz matkoniecz deleted the atm_operation branch October 17, 2021 06:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
hacktoberfest-accepted pull request that should be treated as eligible for Hacktoberfest event
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

5 participants