Fragment of a discussion from User talk:Yair rand
Jump to: navigation, search

Would it be possible to make the transliterations re-usable, so that they could also be used in the translation editor? It would be a nice feature if it could automatically transliterate, too.

23:12, 4 November 2012

For what it's worth, Yiddish is largely autotransliteratable, especially verbs, but many words (especially nouns and adverbs) are not, and it's impossible for a machine to predict every case.

Anyway, Yair rand, thanks again and I hope that you can put in the translit so that I can apply the same logic to the other Yiddish templates.

00:59, 5 November 2012

Just wondering: any leads? If you're too busy, can you recommend somebody else to ask? Thanks

22:00, 17 November 2012

I haven't had much time to look at it. I'm not really very familiar with JavaScript, but Ruakh seems to know more.

03:52, 18 November 2012

Sorry, I forgot about this one pretty soon after I sent my last message. Thanks for reminding me.

So, the only way I could see to make this work was to special case Yiddish in the script itself, but since yi-proper noun is already accelerated, that required setting up for both yi-verb and yi-proper noun. Working with the standard bunch of functions used in the script was something I was unable to figure out how to do (despite spending way too much time trying), so I just set get_preload_text.yi to be a less-than-standard function in the context of the rest of the script, to just return the contents of the entries for proper noun plurals and verb past participles. Seems to work, hopefully without bugs.

The functions are now inconsistently built, so I might have to rewrite the function, though...

07:52, 18 November 2012

Awesome! What more needs to be done before I can accelerate the rest of the Yiddish templates? (Specifically, I'm thinking about {{yi-adj-1}}, {{yi-adj-n}}, {{yi-adj-vowel}}, {{yi-adj-m}}, {{yi-noun}}, and {{yi-conj-head}} — I can give more particulars and formatting examples if you're curious which forms I mean.) Thanks again!

16:06, 18 November 2012

Conjugation and declension tables are typically filled by bots, not with ACCEL. Is the format for Yiddish noun plurals different than for proper noun plurals?

04:52, 19 November 2012

Re conj/decl: Yeah, I keep getting told that, but nobody wants to run a bot to do it. Sadly enough, it'd be easier for me to create the forms semi-manually as needed than learn Python. So ACCEL still seems like the best option.

Re noun plurals: מענטשן is a good example of the ideal format.

04:59, 19 November 2012

The script doesn't have anything for the original entry's transliteration, and {{plural of}} doesn't actually even support the tr= parameter. Would it be a problem if I left that bit out?

05:18, 19 November 2012

If I understand you correctly that (in this example) מענטשן gets transliterated but מענטש does not, then I'm fine with it.

05:23, 19 November 2012

Okay, done for noun plurals.

05:27, 19 November 2012

We have apparently had a misunderstanding... here is the difference between what it does and what it should do.

05:38, 19 November 2012

You need to clear your cache, that's the old version.

05:47, 19 November 2012

Did so, then switched devices and forgot to do it again. Sorry about that. Anyway, it's perfect.

Example for the adjectives, if you're up for it: בעסערע. Thanks so much.

05:52, 19 November 2012

By the way, many Yiddish plurals are formed regularly by adding -n or -s to the end. I wanted to add n and s as possible values for the argument pl= (to get the same effect as if I typed pl=ס{{subst:PAGENAME}}|pltr={{subst:PAGENAME}}s for pl=s). I'm not sure how to deal with the spans in {{yi-noun}} for that, however. Can you do it or explain it? Thanks.

18:09, 21 November 2012

The spans are just HTML, they don't interact with the wikitext or anything. The span's class's transliteration- bit needs to get the right transliteration for the script, but it doesn't really work any differently than getting things to output regular text. Replacing {{{pl}}} with {{#switch:{{{pl}}}|s={{PAGENAME}}ס|n={{PAGENAME}}|{{{pl}}}ן}} (pardon the rtl mess), {{{pltr}}} with {{#switch:{{{pl}}}|s={{{tr}}}s|n={{{tr}}}n|{{{pltr}}}}}, and so on will work fine.

18:37, 21 November 2012

Excellent. Thank you (although the exact code you provided above doesn't actually work... but I temporarily overcame my ineptitude).

19:27, 21 November 2012

For some reason, this edit to {{yi-adj-1}} didn't accelerate the topmost link in the leftmost column of the inflection table on a page like זאַפֿטיק. Am I doing something wrong? If so, could you tell me my mistake? If not, what more needs to be done to get this to work?

04:00, 12 December 2012

"inflected-form" isn't a valid form value. The only valid ones are plural, diminutive, genitive, diminutive-plural, genitive-and-plural, plural-definite, plural-indefinite, singular-definite, vocative, singular-vocative, plural-vocative, third-person-singular, present-participle, simple-past, past-participle, simple-past-and-participle, present, past, infinitive, imperative, positive, comparative, superlative, and exaggerated (pretty sure that's all).

04:14, 12 December 2012

How unfortunate. Can you add inflected-form as an option? (If you don't, I'll try, but we all know that there's a fair chance you'll have to revert me if I do it.)

04:21, 12 December 2012