ScrapboxAutoTrans Development Diary 2022-01-17

Incremental Static Regeneration

https://gyazo.com/6dc186055d1a7478fdc2b3d4322fe330

I'd like to see the trial and error of the development process reflected right away, but maybe even 60x60x24 once it's in stable operation.

Either way, it's a different cycle of translation, I feel.

Fixed a buggy 2-hop link.

I was trying to put in Google Analytics and realized I couldn't write raw HTML.

Is the linked JSON fetched beforehand?

https://gyazo.com/d06ac322e1ff1e63372dae3704d6568b

Maybe I'll map it to mem.nhiro.org or something.

smell of memo, memory, memex, mememt mori, etc.

I'm aiming for translations, but I think I need a list page first...

I don't use it because I'm annoyed by the pages it creates, but I'm thinking it would be good if a fixed URL could be opened and this look-back list would appear.

https://scrapbox.io/api/pages/nishio/search/titles?followingId=5a893319c2a2f300142bde60

Admin Menu

Recognize that the user is me and bring up a menu with this and that.

translation

The translated word text must be obtained by using the translated English string of the title as a key

Record the date and time of the translation

No need to retranslate if it hasn't been changed since the previous translation.

I don't think the trigger for translation should be access.

First, it needs to be translated in bulk, and if the title is not finalized, it won't even be accessed.

After that, retranslate once a day, updated in the last 24 hours

What should we do if we want to translate something that was poured in at an old date and time, like the Hatena Diary?

When you "retranslate those updated in the last 24 hours" above, do you check to see if those linked from there have already been translated?

2022-01-18

---