श्रियाः

[vvasuki]

श्रिया इति पदस्य प्रक्रियाम् कथम् उत्पादयानि?

[akprasad]

There is a subtle bug here. This is the Python code that should work:

from vidyut.prakriya import *

v = Vyakarana()
shri = Dhatu.mula("SriY", Gana.Bhvadi)
shrii = Pratipadika.krdanta(shri, Krt.kvip)
shriyaah = Pada.Subanta(shrii, vibhakti=Vibhakti.Sasthi, vacana=Vacana.Eka, linga=Linga.Stri)
for x in v.derive(shriyaah):
    print(x.text)
    for step in x.history:
        print(step)

But for various internal reasons it doesn't. Will update when the fix is merged.

[akprasad]

@neeleshb

Would love your help here. The issue is that वेरप्कृतस्य is being applied too late in the derivation so that we get *श्र्यः instead. I tried moving it earlier in the derivation, but then I had errors when deriving पिपठीः. This is not an issue in unit tests because the test currently first derives श्री and then uses that to derive श्रियः

What is the order of operations here? Currently the system does this:

1. Add krt and sup as necessary.
2. Then, derive the final pada

If I remember correctly, you suggested that this is preferable to the alternative (and also lower latency):

1. Add krt as necessary.
2. Then, derive the pratipadika
3. Add sup as necessary.
4. Then, derive the pada.

Assuming you suggest the first approach, when is the right time to run वेरप्कृतस्य and अतो लोपः ? Is it as soon as sup is added? As soon as krt is added? or at some other time?

[neeleshb]

वेरपृक्तस्य is to be added after all the अङ्गकार्य (caused due to क्विप्) is completed. Start with श्रि. Add क्विप् by क्विब्वचिप्रच्छिश्रिस्रुद्रुप्रुज्वां दीर्घोऽसंप्रसारणं च. The same sutra does दीर्घ as well. Then you can remove क्विप् and add टा (in any sequence, does not matter). Finally अचिश्नु... gives इयङ् । श्रिया ।

[tbdasap]

क्विब्वचिप्रच्छिश्रिस्रुद्रुप्रुज्वां दीर्घोऽसंप्रसारणं च (in 3.2.178) implemented in PR#200 based on review comments by @akprasad

[tbdasap]

The following snippet

from vidyut.prakriya import *

v = Vyakarana()
shri = Dhatu.mula("SriY", Gana.Bhvadi)
shrii = Pratipadika.krdanta(shri, Krt.kvip)
shriyaah = Pada.Subanta(shrii, vibhakti=Vibhakti.Sasthi, vacana=Vacana.Eka, linga=Linga.Stri)

for prakriya in v.derive(shriyaah):
    print('===================')
    for step in prakriya.history:
        result = ""
        if step.code.count(" ") != 4:
            result = ' + '.join([ term.text for term in step.result ])
            print("{:<10}: {}".format(step.code, result))
    print("-->",prakriya.text) 

generates SriyAH and SriyaH

===================
1.3.1     : SriY
1.3.3     : SriY
1.3.9     : Sri
3.2.178   : Sri + kvi~p
1.3.2     : Sri + kvi~p
1.3.3     : Sri + kvi~p
1.3.8     : Sri + kvi~p
1.3.9     : Sri + v
3.2.178.1 : SrI + v
3.4.114   : SrI + v
4.1.2     : SrI + v + Nas
1.3.4     : SrI + v + Nas
1.3.8     : SrI + v + Nas
1.3.9     : SrI + v + as
7.2.8     : SrI + v + as
1.4.13    : SrI + v + as
1.1.5     : SrI + v + as
6.1.67    : SrI +  + as
1.4.6     : SrI +  + as
1.4.18    : SrI +  + as
7.3.112   : SrI +  + Aw + as
1.3.3     : SrI +  + Aw + as
1.3.9     : SrI +  + A + as
6.4.77    : Sriy +  + A + as
6.1.90    : Sriy +  +  + As
6.1.90    : Sriy +  +  + As
8.2.66    : Sriy +  +  + Aru~
1.3.2     : Sriy +  +  + Ar
8.3.15    : Sriy +  +  + AH
1.4.14    : Sriy +  +  + AH
--> SriyAH
===================
1.3.1     : SriY
1.3.3     : SriY
1.3.9     : Sri
3.2.178   : Sri + kvi~p
1.3.2     : Sri + kvi~p
1.3.3     : Sri + kvi~p
1.3.8     : Sri + kvi~p
1.3.9     : Sri + v
3.2.178.1 : SrI + v
3.4.114   : SrI + v
4.1.2     : SrI + v + Nas
1.3.4     : SrI + v + Nas
1.3.8     : SrI + v + Nas
1.3.9     : SrI + v + as
7.2.8     : SrI + v + as
1.4.13    : SrI + v + as
1.1.5     : SrI + v + as
6.1.67    : SrI +  + as
1.4.4     : SrI +  + as
1.4.18    : SrI +  + as
6.4.77    : Sriy +  + as
8.2.66    : Sriy +  + aru~
1.3.2     : Sriy +  + ar
8.3.15    : Sriy +  + aH
1.4.14    : Sriy +  + aH
8.4.68    : Sriy +  + aH
--> SriyaH

Fixed: but awaiting official python release.