Module talk:be-verb

Automatic conjugation is live

@Atitarev {{be-conj}} now exists and supports classes 1 through 6. The remaining classes are coming. You can see various examples at User:Benwing2/test-be-conj; if you have questions about what things mean, let me know. Benwing2 (talk) 22:52, 11 July 2020 (UTC)

I converted most manual declensions for classes 1 through 6 to use {{be-conj}}. Benwing2 (talk) 03:28, 12 July 2020 (UTC)

@Benwing2: Brilliant, man! --Anatoli T. ^{(обсудить}/^вклад) 04:02, 12 July 2020 (UTC)

@Atitarev Thanks. It may take a little longer to get the remaining classes. I have to rethink and restructure the way I handle vowel alternations and haven't completely figured out the best way. Belarusian vowel alternations are annoying, esp. since there are exceptions where at least э and ё can occur in unstressed syllables. (Maybe о as well, in syllables with secondary stress?) Benwing2 (talk) 20:55, 12 July 2020 (UTC)

@Benwing2: Take your time, mate. Your efforts are appreciated. I am not sure about verbs but borrowed nouns may have unstressed о and ё and not just in final positions in Taraškievica orthography, e.g. біёлё́гія (bijoljóhija) or біолё́гія (bioljóhija), normally біяло́гія (bijalóhija). These are rare though.

"э" might happen a lot in unstressed positions but does it happen a lot in pre-tonal positions? Probably just loanwords? --Anatoli T. ^{(обсудить}/^вклад) 00:44, 13 July 2020 (UTC)

@Atitarev I have rewritten the handling of vowel alternations in the module and added class 7 support. The remaining classes should be fairly easy now. What I do now is first reconstruct (as much as is possible) the underlying vowels and do all the manipulations/stress movements on this form, and then at the very end convert to the surface representation. Benwing2 (talk) 04:08, 14 July 2020 (UTC)

@Benwing2: Aha, I see. You're using the original Proto-Slavic vowels and then applying akanye, akanye and de-iotification (using a noun analogy - roughly рѣка́->река́->ряка́->рака́) rules? --Anatoli T. ^{(обсудить}/^вклад) 04:16, 14 July 2020 (UTC)

@Atitarev Yes, that's basically what I'm doing. What I was doing previously was trying to account for vowel changes when I moved the stress from one place to another, which was a lot more complicated and didn't work very well. Benwing2 (talk) 04:19, 14 July 2020 (UTC)

@Benwing2: Thank you. If you use this logic, you may find it a little bit easier to deal with Belarusian nouns if you want to tackle them in the future. In a way, they are simpler than Ukrainian, there are similarities, like you'll have to use |genu= occasionally and you're already familiar with some some other alternations like нага́/назе́/но́гі. I don't know if your commons cover weird changes like брыво́->бро́вы, кроў->крыві́. --Anatoli T. ^{(обсудить}/^вклад) 04:41, 14 July 2020 (UTC)

@Atitarev Not yet, I'll have to add a flag for those cases. Benwing2 (talk) 04:47, 14 July 2020 (UTC)

@Benwing2: Yes, that would require a new param, IMO, just like you you use io, ie, etc. in Ukrainian. Also, just occurred to me, that there's a rule you could use. Native Belarusian words can't start with "о", that's why you get акно́->во́кны, во́зера->азёры, аўца́->во́ўцы. I have just fixed аўца́ (aŭcá), which I had to Google for forms before. аве́чка (avjéčka) is the regular modern term. --Anatoli T. ^{(обсудить}/^вклад) 04:58, 14 July 2020 (UTC)

@Atitarev All the remaining classes should be done, and all irregular verbs except быць (compounds of быць like забы́ць are correctly handled). Benwing2 (talk) 03:28, 15 July 2020 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ @Benwing2: Thank you very much!

I noticed that бе́гчы (bjéhčy) has two present tense adverbials and бегучы́ and бягу́чы. бягу́чы is also a present tense participle. --Anatoli T. ^{(обсудить}/^вклад) 03:33, 15 July 2020 (UTC)

@Atitarev Fixed the present adverbial participles. I'm not sure what to do with the present active participle бягу́чы. The module doesn't currently support present active participles at all. I could add them although they must be rare. Do you know of any other examples of present active participles? Benwing2 (talk) 03:44, 15 July 2020 (UTC)

@Benwing2: There's little info on this, unfortunately. Perhaps this can be optional, parameterised, other examples: ве́раваць->ве́руючы, утвара́ць->утвара́ючы. Like in Ukrainian, there is a perception that present active participles are not used, so that Russian "делающий" should be translated as "які робіць". Note that the candidate "ро́бячы" coincides with the adverbial. I can try and find more examples. --Anatoli T. ^{(обсудить}/^вклад) 04:01, 15 July 2020 (UTC)

@Benwing2: Present active participles are often perceived as adjectives, despite the origin (same in Ukrainian) e.g. пі́шучы (writing), hence пі́шучая машы́нка (píšučaja mašýnka) (typewriter). Pure adjectives are пісо́ўны, піса́льны. --Anatoli T. ^{(обсудить}/^вклад) 04:05, 15 July 2020 (UTC)

@Atitarev I added support for present active participles. You can add them using a parameter |pres_actv_part=, e.g. {{be-conj|бе́гчы<irreg.impf.intr>|pres_actv_part=бягу́чы}}. Benwing2 (talk) 05:58, 15 July 2020 (UTC)

@Atitarev I did another run converting verbs to use {{be-conj}}. There are only 22 verbs left that use {{be-conj-manual}}, and another 55 that use {{rfinfl|be|verb}} (of these, 53 have a comment next to them indicating the automatic parameters to supply to {{be-conj}}; they need to be checked as some could be wrong). Benwing2 (talk) 07:08, 15 July 2020 (UTC)

@Benwing2: Thanks. I'll work with them.

быць (bycʹ) inflections can used as is now but modularised and ёсць (joscʹ) has an alternative ё (jo) form. Including Taraškievica forms (such as ёсьць (josʹcʹ)) in the tables might be a mess. --Anatoli T. ^{(обсудить}/^вклад) 07:18, 15 July 2020 (UTC)

валачы

@Benwing2: Hi. At User:Benwing2/test-be-conj in the case for валачы́ (valačý) the present tense forms валаче́ш, валаче́, валачём, валачяце́ are incorrect and should be валачэ́ш, валачэ́, валачо́м, валачаце́.

ч [t͡ʂ] is always hard in Belarusian. The impossible combinations are any of 1) [чшжр][еёіьюя], 2) final -в, 3) probably ў + vowel (apostrophe ' is OK).

I've created валачы́ (valačý) anyway since I hope it's a quick fix for you :)

BTW (unrelated), if you're not doing it already, we want any words with ґ tracked. In Taraškievica this letter is sometimes "allowed" but not mandatory. (It's only standard in Ukrainian and Rusyn) There are cases where a regular г can be pronounced as [ɡ] in loanwords with no change in spellings and there are positions where it's considered correct to pronounce, pls see Module_talk:be-pronunciation#Questions. --Anatoli T. ^{(обсудить}/^вклад) 00:40, 15 July 2020 (UTC)

@Atitarev Fixed the issue with валачы́. ў + vowel does occur in the present-tense forms of вы́віць. I can add a category for tracking Belarusian words with ґ. Benwing2 (talk) 00:48, 15 July 2020 (UTC)

@Benwing2: Thanks! Ah, yes ў + vowel - should only be iotated е, ё, ю, я, not а, о, і, у, ы. --Anatoli T. ^{(обсудить}/^вклад) 00:53, 15 July 2020 (UTC)

Special case for ілгаць / лгаць

@Benwing2: Hi. Thank you for creating the complex module for Belarusian verbs. It's really good!

This verb ілгаць / лгаць is a special case and maybe merits more attention.

1. As you labelled in the comments, it's irregular in the present tense, like the Russian equivalent лгать (lgatʹ).
2. It doesn't use і/й alternation but drops "і" after vowels instead. So it should show "бу́дзе лга́ць" (not йлга́ць).
3. There may be some other examples but the same alternations is in the noun лёд (ljod) when the vowel is reduced, e.g. "under the ice" is "пад ільдо́м" (or "пад лёдам").
4. Are you interested in adding a special treatment for this verb? If not, I will keep manual but the future tense needs an override to remove "й" where it's not correct. --Anatoli T. ^{(обсудить}/^вклад) 23:58, 15 July 2020 (UTC)

@Atitarev Yup, I'll add some special casing for this verb. Benwing2 (talk) 02:17, 16 July 2020 (UTC)

@Benwing2: Thanks, please let me know if the usage notes or examples need tweaking. --Anatoli T. ^{(обсудить}/^вклад) 02:25, 16 July 2020 (UTC)

@Benwing2: Almost there, thanks! Just need that й/0 to work. --Anatoli T. ^{(обсудить}/^вклад) 03:41, 16 July 2020 (UTC)

@Atitarev It's working now. I changed the і/й alternation code to drop і before л/р + consonant after a vowel, and conversely to add і before л/р + consonant after a consonant. This should handle (і)рваць as well. Benwing2 (talk) 03:43, 16 July 2020 (UTC)

@Benwing2: Thank you for addressing this and another case! --Anatoli T. ^{(обсудить}/^вклад) 03:46, 16 July 2020 (UTC)

@Benwing2: Another case is (і)ржаць (an noun (і)ржа)! I will add any missing entries soon. --Anatoli T. ^{(обсудить}/^вклад) 03:52, 16 July 2020 (UTC)

@Benwing2: And yet another one is (і)граць. --Anatoli T. ^{(обсудить}/^вклад) 03:55, 16 July 2020 (UTC)

@Benwing2: I've created missing entries here - just as normal and alt forms. If you decide to add handling for this alternation, I will add usage notes as in ілгаць / лгаць. I don't know if there are other examples. --Anatoli T. ^{(обсудить}/^вклад) 00:29, 18 July 2020 (UTC)

@Atitarev (і)граць is strange; it doesn't fit the phonological pattern of (і)лгаць/(і)ржаць/etc. There are examples in Slounik where іграць occurs after a vowel and граць after a consonant so I'm not sure this alternation is real:

Гаворыць як у скрыпку іграе, а робіць як у разбіты збан бразгае.

Эх, ды каб пайсці са сваёй скрыпкай у родную вёску, а там граць, граць да ахмялення, да адурэння, да сёмага поту. Бядуля.

цяпер, паводле Маркса, пачынаюць граць самастойную ролю, робяцца нібы незалежнымі фактарамі тых дачыненняў і вызначаюць стасункі паміж асобамі;

Benwing2 (talk) 00:35, 18 July 2020 (UTC)

@Benwing2: Good findings, thanks. You can add only to those you think are appropriate, if there are any. On the other hand, a preferred way to use these alternatives, IMO, would be following the same rule for sound harmony. I'm sure it's easy to find examples where ў/у rule is broken. --Anatoli T. ^{(обсудить}/^вклад) 00:47, 18 July 2020 (UTC)

@Benwing2: A good test for my theory would be "ён іграе" and "яна грае" but it fails. The reverse usages are common as well. You're right about (і)граць. --Anatoli T. ^{(обсудить}/^вклад) 00:53, 18 July 2020 (UTC)

раіць - 4b (etym #2)

@Benwing2: Hi. Per edit summary in раі́ць (raícʹ) etym #2: Is "longimp" supposed to make long imperatives? Can't get раі́/раі́це on etym #2. Another example is даі́ць (daícʹ, “to milk”) (*dojiti) where imperatives can be дой or даі́. Please assist if you can. --Anatoli T. ^{(обсудить}/^вклад) 11:25, 4 August 2020 (UTC)

@Atitarev Fixed. The logic was ignoring longimp in verbs whose stem ends in a vowel. Benwing2 (talk) 00:43, 5 August 2020 (UTC)

@Benwing2: Thank you! I will try даі́ць (daícʹ) and паі́ць (paícʹ). Interesting that Ukrainian дої́ти (dojíty) and пої́ти (pojíty) don't have long imperatives. --Anatoli T. ^{(обсудить}/^вклад) 00:51, 5 August 2020 (UTC)

@Atitarev It looks like some such Ukrainian verbs, like рої́ти, строї́ти, труї́ти, do have long imperatives. For Ukrainian I implemented a special flag 'ї' to signal these imperatives. Probably just using 'longimp' would be better. Benwing2 (talk) 00:55, 5 August 2020 (UTC)

@Benwing2: Perhaps yes. 'longimp' doesn't seem to work with Ukrainian. I've made рої́ти (rojíty) using {{uk-conj|рої́ти<4b.impf.tr.-ppp.ї>}}. (not sure about the stress at ppp)--Anatoli T. ^{(обсудить}/^вклад) 01:41, 5 August 2020 (UTC)