[RU] Etymology is stored in a module... How do we deal with it ?

[lasconic]

• Wiktionary page:

Etymology is Этимология in russian.

Wikicode:

Происходит от {{этимология:лоскут|да}}

This will call the Etymology template, which call the module Etymology:лоскут, which store the etymology for the word... Crazy...

---

@victornove implemented it in his PR by making a get call to wiktionary. It's probably ok to do that for a single word in get-word but can we do better when we process the dump ? Are the modules stored in the wiktionary dump ?

Upvote & Fund

• We're using Polar.sh so you can upvote and help fund this issue.
• We receive the funding once the issue is completed & confirmed by you.
• Thank you in advance for helping prioritize & fund our backlog.

[victornove]

Hi,
this is also a problem for the pronunciation:
{{transcription-ru|произноше́нье} calls some ru-pron module which generates the pronunciation of the word. I'm not sure I could get the content otherwise.

[victornove]

I had a look in the dump, I found over 15000 etymology templates with "find-templates". The complete wiktionary list has 17000 of them: https://ru.wiktionary.org/wiki/Категория:Шаблоны_этимологии.
I can't find the the expanded text in the dump used here. I'm not sure all the namespaces are included? It must be somewhere in one of the dumps, these guys seem to find template definitions and execute them with Lua, but I haven't figured out how yet: https://github.com/tatuylonen/wikitextprocessor.

The only way I found so far to get the template content without querying the full html page is to call the special template parser page:

https://ru.wiktionary.org/wiki/Служебная:Развёртка_шаблонов?wpInput={{transcriptions-ru|страни́ца|страни́цы|Ru-страница.ogg}}

[lasconic]

Good to know the etymology modules are in the dump. We could extract them in a separate json file in the "parse" step and keep a second list to query when we "render".