Module talk:pi-conj/verb
Thai Script Second Person Plural Middle without Implicit vowels[edit]
@Octahedron80: Could someone please check the second person plural middle. The forms corresponding to pacavhe and pacavho come out as ปะจะวเห and ปะจะวโห (or maybe they should be ปะจัวเห and ปะจัวโห) - neither set looks right to me. The conjugation table is set up in ปะจะติ (pacati). -- RichardW57 (talk) 23:55, 28 May 2019 (UTC)
I compare with pacati (ปจติ / ปะจะติ) and I have some comments:
- pacāmhe = ปจามฺเห / ปะจามเห
- pacavhe = ปจวฺเห / ปะจัวเห
- apacavhaṃ or pacavhaṃ = อปจวฺหํ or ปจวฺหํ / อะปะจัวหัง or ปะจัวหัง (อ- must all become อะ- either.)
- pacavho = ปจวฺโห / ปะจัวโห
- paceyyavho = ปเจยฺยวฺโห / ปะเจยยัวโห
Other than these are incorrect. See if you can adjust these. However -ัว reads like เ-า in Thai but I never see เ-า in Pali text. Just leave it like that. --Octahedron80 (talk) 02:32, 29 May 2019 (UTC)
- Thanks for the positive information, and thanks for catching the error with the alphabetic augment. I'll check ปจาเมฺห against a Thai CD of the Tipitaka; I expect to find both ปจาเมฺห and ปจามฺเห on it. We'll see how long a quotation from it lasts. For the alphabetic forms, I will assume that ปะจะเวฺห and similar do not exist This is easy to implement. -- RichardW57 (talk) 06:13, 29 May 2019 (UTC)
- The changes for the spelling without implicit vowels have now been implemented. However, irregular optatives will not yet work with all vowels explicit. I first want to create a formal set of test cases to protect against regression errors. -- RichardW57m (talk) 11:51, 29 May 2019 (UTC)
- You may want to see this: http://www.palidict.com/content/%E0%B8%AD%E0%B8%B2%E0%B8%82%E0%B8%A2%E0%B8%B2%E0%B8%95 --Octahedron80 (talk) 08:22, 29 May 2019 (UTC)
- I don't think the alternations in the aorist are as free as the guide suggests. Different words use different endings. I suppose there may be an issue with poetic licence, in which case we (I?) may need to address adding footnotes for particular endings. I don't think we can do it as is done for Latin inflections. -- RichardW57m (talk) 11:51, 29 May 2019 (UTC)
- @Octahedron80: Searching through the Tiptaka (CD received in 2006 - published 2003, vaguely identified as "ฉบับสยามรัฏฐ") is harder than I remembered it being. I have found my analysis from June 2011 (on my hard drive in ~/tpk_notes/ana). Anyway, what transliterates to 'vho' occurred 33 times. When it was at the end of a word (31 examples), it was written 'โวฺห' every time. The other two examples were in the middle of a word; one was written 'โวฺห' and the other was written 'วฺโห'. There were 4 examples for 'vhe'. All four were written เวฺห, but they are all in the middle of a word and so probably not useful for discussing the inflection.
- There is a similar pattern for 'mhe'. At the end of a word (987 instances), it was written 'เมฺห' every time. In the middle of a word, it was written 'เมฺห' 257 times and มฺเห 16 times. It did not occur at the start of a word. -- RichardW57 (talk) 07:13, 30 May 2019 (UTC)
Other Tenses[edit]
I think you are developing on conjugation right? There are 8 types of conjugations but I only see 4 here. --Octahedron80 (talk) 03:05, 29 May 2019 (UTC)
- @Octahedron80: Correct - only four tenses are covered. I've only coded up the present tense system, i.e. those formed on what is called the 'special' base - the present, the imperfect, the imperative and the optative, which is why the template is called
{{pi-conj-special}}. The future and conditional will together have their own template, probably{{pi-conj-future}}. The perfect will also have its own template.
- The aorist is a problem, as we should not record forms that do not exist. There are aorists in -i that are formed on the special base; perhaps I will extend
{{pi-conj-special}}to include them. I fear I will need several templates to cover the aorist. Perhaps one form will just be a catalogue of what forms actually exist - I have been using{{pi-conj-pres}}for aorists, but that only covers active forms. The aorists in -esi are also regular enough to be covered by a conjugation template that will by default give forms for all 6 combinations. -- RichardW57 (talk) 06:13, 29 May 2019 (UTC)
{{pi-conj-future}}is now just about usable, though I've only tested it for the Latin script. Other scripts should 'just work', but that is to be checked. For progress, see the template. The best sign of suitability for use will be the appearance of the documentation page. --RichardW57 (talk) 16:43, 26 May 2020 (UTC)
Issues with Control of the Imperfect[edit]
Some obscure combinations are not working well. I need to review the documentation, fix bugs and deal with booby traps in the combinations of parameter values. --RichardW57 (talk) 16:43, 26 May 2020 (UTC)
The concepts behind the design is that a verb may have different active and middle stems, and possibly no active stems at all. (I've a feeling this is not the case, though there are works around that wrongly claim that passive verbs have no active stems.) An important modification to this idea is that the middle of the optative is formed from the active stem when active and middle differ. The verb receives an extra element of complication because adding the augment is not mechanically simple - it occurs between a prefix and the rest of the vowel, and may restore the initial letter to being a geminate when it is the reduction of an earlier cluster, as with the verb kamati. --RichardW57 (talk) 16:43, 26 May 2020 (UTC)
Case 1[edit]
{{pi-conj-special|kamati|augment=both_given|impf=akkama|impf_voice=both|impr_voice=none|opta_voice=none|part_voice=none}} shows the usual case for handling a slightly 'irregular' augmented imperfect. It works as wanted and as documented.
| Active | Middle | |||
|---|---|---|---|---|
| Singular | Plural | Singular | Plural | |
| Present | ||||
| 1st | kamāmi | kamāma | kame | kamāmhe |
| 2nd | kamasi | kamatha | kamase | kamavhe |
| 3rd | kamati | kamanti | kamate | kamante or kamare |
| Imperfect | ||||
| 1st | akkama or akkamaṃ or kama or kamaṃ | akkamamhā or kamamhā | akkamiṃ or kamiṃ | akkamāmhase or akkamamhase or kamāmhase or kamamhase |
| 2nd | akkamo or kamo | akkamattha or kamattha | akkamase or kamase | akkamavhaṃ or kamavhaṃ |
| 3rd | akkama or akkamā or kama or kamā | akkamu or akkamū or kamu or kamū | akkamattha or kamattha | akkamatthuṃ or kamatthuṃ |
--RichardW57 (talk) 16:43, 26 May 2020 (UTC)
Case 2[edit]
{{pi-conj-special|kamati|augment=both_given|impf_midl=apacattha|impf_voice=both|impr_voice=none|opta_voice=none|part_voice=none}} does not include apacattha in the table. This is in accordance with the documentation, but may be regarded as a 'gotcha'.
Lua error in Module:pi-conj/verb at line 719: Argument impf_midl may only be used if argument midl is.
--RichardW57 (talk) 16:43, 26 May 2020 (UTC)
I've now made the use of impf_midl without midl generate an error message. --RichardW57 (talk) 17:20, 29 May 2020 (UTC)
Case 3[edit]
{{pi-conj-special|kamate|augment=both_given|impf_midl=akkamattha|impf_voice=both|impr_voice=none|opta_voice=none|part_voice=none}} includes apacattha despite the documentation. It might be better to declare this to be as it should be.
Lua error in Module:pi-conj/verb at line 719: Argument impf_midl may only be used if argument midl is.
--RichardW57 (talk) 16:43, 26 May 2020 (UTC)
I've now made the use of impf_midl without midl generate an error message. --RichardW57 (talk) 17:20, 29 May 2020 (UTC)
Case 4[edit]
{{pi-conj-special|kamate|augment=both_given|impf=akkamattha|impf_voice=both|impr_voice=none|opta_voice=none|part_voice=none}} does not record akkamattha, though the documentation says it will.
| Active | Middle | |||
|---|---|---|---|---|
| Singular | Plural | Singular | Plural | |
| Present | ||||
| 1st | kame | kamāmhe | ||
| 2nd | kamase | kamavhe | ||
| 3rd | kamate | kamante or kamare | ||
| Imperfect | ||||
| 1st | akkamiṃ | akkamāmhase or akkamamhase | akkamiṃ or kamiṃ | akkamāmhase or akkamamhase or kamāmhase or kamamhase |
| 2nd | akkamase | akkamavhaṃ | akkamase or kamase | akkamavhaṃ or kamavhaṃ |
| 3rd | akkamattha | akkamatthuṃ | akkamattha or kamattha | akkamatthuṃ or kamatthuṃ |
--RichardW57 (talk) 16:43, 26 May 2020 (UTC)
Case 5[edit]
{{pi-conj-special|kamati|augment=both_given|impf=akkama|impf_midl=apacattha|impf_voice=both|impr_voice=none|opta_voice=none|part_voice=none}} includes the augmented form akkama in both voices, but not the explicit middle form. This is as documented, but may be a surprise.
Lua error in Module:pi-conj/verb at line 719: Argument impf_midl may only be used if argument midl is.
--RichardW57 (talk) 16:43, 26 May 2020 (UTC)
I've now made the use of impf_midl without midl generate an error message. --RichardW57 (talk) 17:20, 29 May 2020 (UTC)
Case 6[edit]
{{pi-conj-special|kamati|augment=both_given|impf=akkama|impf_midl=apacattha|impf_voice=both|impr_voice=none|midl=desete|opta_voice=none|part_voice=none}} takes separate inputs for the input of active and middle imperfect forms with augments, as documented. The warning that the author of the template does not know the middle imperfect of deseti is reasonable, though a bit of an irritant.
| Active | Middle | |||
|---|---|---|---|---|
| Singular | Plural | Singular | Plural | |
| Present | ||||
| 1st | kamāmi | kamāma | desaye | desayāmhe |
| 2nd | kamasi | kamatha | desayase | desayavhe |
| 3rd | kamati | kamanti | desayate | desayante or desayare |
| Imperfect | ||||
| 1st | akkama or akkamaṃ or kama or kamaṃ | akkamamhā or kamamhā | apaciṃ | apacāmhase or apacamhase |
| 2nd | akkamo or kamo | akkamattha or kamattha | apacase | apacavhaṃ |
| 3rd | akkama or akkamā or kama or kamā | akkamu or akkamū or kamu or kamū | apacattha | apacatthuṃ |
--RichardW57 (talk) 16:43, 26 May 2020 (UTC)
The warning about deseti not have a clear imperfect middle has now been removed. --RichardW57 (talk) 17:20, 29 May 2020 (UTC)
Case 7[edit]
{{pi-conj-special|kamati|augment=both_given|impf=akkama|impf_midl=apacattha|impf_voice=both|impr_voice=none|midl=passate|opta_voice=none|part_voice=none}} shows a nice clean separation of augmented imperfect forms by voice.
| Active | Middle | |||
|---|---|---|---|---|
| Singular | Plural | Singular | Plural | |
| Present | ||||
| 1st | kamāmi | kamāma | passe | passāmhe |
| 2nd | kamasi | kamatha | passase | passavhe |
| 3rd | kamati | kamanti | passate | passante or passare |
| Imperfect | ||||
| 1st | akkama or akkamaṃ or kama or kamaṃ | akkamamhā or kamamhā | apaciṃ or passiṃ | apacāmhase or apacamhase or passāmhase or passamhase |
| 2nd | akkamo or kamo | akkamattha or kamattha | apacase or passase | apacavhaṃ or passavhaṃ |
| 3rd | akkama or akkamā or kama or kamā | akkamu or akkamū or kamu or kamū | apacattha or passattha | apacatthuṃ or passatthuṃ |
--RichardW57 (talk) 16:43, 26 May 2020 (UTC)
Case 8[edit]
{{pi-conj-special|kamati|augment=both_given|impf=akkamā|impf_voice=both|impr_voice=none|opta_voice=none|part_voice=none}} shows a minor issue. Because the 3s active imperfect form is the one ending in a long vowel, the logic for verbs in -āti is invoked, and I don't know their imperfect middle, no augmented imperfect middle is provided. This is an undocumented feature, and arguably a bug.
| Active | Middle | |||
|---|---|---|---|---|
| Singular | Plural | Singular | Plural | |
| Present | ||||
| 1st | kamāmi | kamāma | kame | kamāmhe |
| 2nd | kamasi | kamatha | kamase | kamavhe |
| 3rd | kamati | kamanti | kamate | kamante or kamare |
| Imperfect | ||||
| 1st | akkama or akkamaṃ or kama or kamaṃ | akkamamhā or kamamhā | akkamiṃ or kamiṃ | akkamāmhase or akkamamhase or kamāmhase or kamamhase |
| 2nd | akkamo or kamo | akkamattha or kamattha | akkamase or kamase | akkamavhaṃ or kamavhaṃ |
| 3rd | akkama or akkamā or kama or kamā | akkamu or akkamū or kamu or kamū | akkamattha or kamattha | akkamatthuṃ or kamatthuṃ |
--RichardW57 (talk) 16:43, 26 May 2020 (UTC)
|impf=akkamā now has the same meaning as |impf=akkama. --RichardW57 (talk) 17:20, 29 May 2020 (UTC)
Chakma Future System Conjugation[edit]
When User:Theknightwho added Chakma to the list of scripts, I had to add the data for ending detection for the Chakma scripts. Unfortunately:
- I overlooked
future_endings. I need to move it up to next to the other endings so it doesn't get forgotten next time, and populate it for the Cakma script.
Done - The module's test cases don't include the future system. Its testing needs to be added. Done
- It would be good to document what needs to be done to properly add a simple new script with one writing system just in case we need to add more scripts.
In the meantime, a quick but limited test for the Chakma script future system is:
{{pi-conj-special|𑄛𑄌𑄖𑄨|cond=𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄖𑄨|futu=𑄛𑄌𑄨𑄥𑄴𑄥𑄖𑄨|sc=Cakm}}:
| Active | Middle | |||
|---|---|---|---|---|
| Singular | Plural | Singular | Plural | |
| Present | ||||
| 1st | 𑄛𑄌𑄂𑄟𑄨 (pacāmi) | 𑄛𑄌𑄂𑄟 (pacāma) | 𑄛𑄌𑄬 (pace) | 𑄛𑄌𑄂𑄟𑄴𑄦𑄬 (pacāmhe) |
| 2nd | 𑄛𑄌𑄥𑄨 (pacasi) | 𑄛𑄌𑄗 (pacatha) | 𑄛𑄌𑄥𑄬 (pacase) | 𑄛𑄌𑅇𑄴𑄦𑄬 (pacavhe) |
| 3rd | 𑄛𑄌𑄖𑄨 (pacati) | 𑄛𑄌𑄚𑄴𑄖𑄨 (pacanti) | 𑄛𑄌𑄖𑄬 (pacate) | 𑄛𑄌𑄚𑄴𑄖𑄬 (pacante) or 𑄛𑄌𑄢𑄬 (pacare) |
| Imperative | ||||
| 1st | 𑄛𑄌𑄂𑄟𑄨 (pacāmi) | 𑄛𑄌𑄂𑄟 (pacāma) | 𑄛𑄌𑄬 (pace) | 𑄛𑄌𑄂𑄟𑄥𑄬 (pacāmase) |
| 2nd | 𑄛𑄌 (paca) or 𑄛𑄌𑄂𑄦𑄨 (pacāhi) | 𑄛𑄌𑄗 (pacatha) | 𑄛𑄌𑄥𑄴𑄥𑄪 (pacassu) | 𑄛𑄌𑅇𑄴𑄦𑄮 (pacavho) |
| 3rd | 𑄛𑄌𑄖𑄪 (pacatu) | 𑄛𑄌𑄚𑄴𑄖𑄪 (pacantu) | 𑄛𑄌𑄖𑄁 (pacataṃ) | 𑄛𑄌𑄚𑄴𑄖𑄁 (pacantaṃ) |
| Optative | ||||
| 1st | 𑄛𑄌𑄬𑄠𑄳𑄠𑄂𑄟𑄨 (paceyyāmi) or 𑄛𑄌𑄬𑄠𑄳𑄠𑄁 (paceyyaṃ) or 𑄛𑄌𑄬 (pace) | 𑄛𑄌𑄬𑄠𑄳𑄠𑄂𑄟 (paceyyāma) | 𑄛𑄌𑄬𑄠𑄳𑄠𑄁 (paceyyaṃ) | 𑄛𑄌𑄬𑄠𑄳𑄠𑄂𑄟𑄴𑄦𑄬 (paceyyāmhe) |
| 2nd | 𑄛𑄌𑄬𑄠𑄳𑄠𑄂𑄥𑄨 (paceyyāsi) or 𑄛𑄌𑄬 (pace) | 𑄛𑄌𑄬𑄠𑄳𑄠𑄂𑄗 (paceyyātha) | 𑄛𑄌𑄬𑄗𑄮 (pacetho) | 𑄛𑄌𑄬𑄠𑄳𑄠𑅇𑄴𑄦𑄮 (paceyyavho) |
| 3rd | 𑄛𑄌𑄬𑄠𑄳𑄠 (paceyya) or 𑄛𑄌𑄬 (pace) | 𑄛𑄌𑄬𑄠𑄳𑄠𑄪𑄁 (paceyyuṃ) | 𑄛𑄌𑄬𑄗 (pacetha) | 𑄛𑄌𑄬𑄢𑄁 (paceraṃ) |
| Future | ||||
| 1st | 𑄛𑄌𑄨𑄥𑄴𑄥𑄂𑄟𑄨 (pacissāmi) | 𑄛𑄌𑄨𑄥𑄴𑄥𑄂𑄟 (pacissāma) | 𑄛𑄌𑄨𑄥𑄴𑄥𑄁 (pacissaṃ) | 𑄛𑄌𑄨𑄥𑄴𑄥𑄂𑄟𑄴𑄦𑄬 (pacissāmhe) |
| 2nd | 𑄛𑄌𑄨𑄥𑄴𑄥𑄥𑄨 (pacissasi) | 𑄛𑄌𑄨𑄥𑄴𑄥𑄗 (pacissatha) | 𑄛𑄌𑄨𑄥𑄴𑄥𑄥𑄬 (pacissase) | 𑄛𑄌𑄨𑄥𑄴𑄥𑅇𑄴𑄦𑄬 (pacissavhe) |
| 3rd | 𑄛𑄌𑄨𑄥𑄴𑄥𑄖𑄨 (pacissati) | 𑄛𑄌𑄨𑄥𑄴𑄥𑄚𑄴𑄖𑄨 (pacissanti) | 𑄛𑄌𑄨𑄥𑄴𑄥𑄖𑄬 (pacissate) | 𑄛𑄌𑄨𑄥𑄴𑄥𑄚𑄴𑄖𑄬 (pacissante) |
| Conditional | ||||
| 1st | 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄁 (apacissaṃ) | 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄁 (apacissaṃ) | 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄂𑄟𑄴𑄦𑄥𑄬 (apacissāmhase) | |
| 2nd | 𑄃𑄛𑄌𑄨𑄥𑄴𑄥 (apacissa) or 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄨 (apacissi) | 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄗 (apacissatha) | 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄬 (apacisse) or 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄥𑄬 (apacissase) | 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑅇𑄴𑄦𑄬 (apacissavhe) |
| 3rd | 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄂 (apacissā) or 𑄃𑄛𑄌𑄨𑄥𑄴𑄥 (apacissa) | 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄁𑄥𑄪 (apacissaṃsu) | 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄗 (apacissatha) | 𑄃𑄛𑄌𑄨𑄥𑄴𑄥𑄨𑄁𑄥𑄪 (apacissiṃsu) |
- Present active participle: 𑄛𑄌𑄚𑄴𑄖𑄴 (pacant), which see for forms and usage
- Present middle participle: 𑄛𑄌𑄟𑄂𑄚 (pacamāna), which see for forms and usage
- Future active participle: 𑄛𑄌𑄨𑄥𑄴𑄥𑄚𑄴𑄖𑄴 (pacissant), which see for forms and usage
- Future middle participle: 𑄛𑄌𑄨𑄥𑄴𑄥𑄟𑄂𑄚 (pacissamāna) or 𑄛𑄌𑄨𑄥𑄴𑄥𑄂𑄚 (pacissāna), which see for forms and usage
--RichardW57m (talk) 12:48, 6 March 2023 (UTC)
- As the above now shows, the input of a future in -ati works - in fact futures in -ati and -ate now confirmed to work and populate the future and conditional, the aorist is confirmed to work, and the tenses etc. formed on the special base are now confirmed by testing to work. I am, however, having trouble getting the system to accept the input of conditionals. However, the following two commands demonstrate the conditional input being accepted:
{{pi-conj-special|gacchate|cond=agamissatha|cond_augment=with_given|futu=pacissate}}{{pi-conj-special|gacchati|cond=agamissa|cond_augment=with_given|futu=pacissati}}
--RichardW57 (talk) 22:42, 6 March 2023 (UTC)
| I clearly hadn't tested inflection from conditional tense stems for non-Roman scripts. That is now implemented and the tests have been passed, with the 4 errors for -ute middles in the writing systems that distinguish /a/ in open and closed syllables. Such stem modification is an aberration in the context of Pali. --RichardW57m (talk) 11:52, 7 March 2023 (UTC)