gaṇadehā (Shreevatsa) vs. upacitrā (Kosambi) शृङ्गारशतकम् 90

[gasyoun]

I have an issue to report about the results for the following input:

तरुणीवेषोद्दीपितकामा विकसज्जातीपुष्पसुगन्धिः ।
उन्नतपीनपयोधरभारा प्रावृट् तनुते कस्य न हर्षम् ॥ ९० ॥

Can Kosambi be wrong?

Parts:
  full pattern LLGGGGLLGGLLGGGGLLGLGLLGLLGLLGGGGLLGGLLGG (41 syllables, 63 mātras)
  half_1 pattern LLGGGGLLGGLLGGGGLLGL (20 syllables, 31 mātras)
                                                                 partial match for: gaṇadehā set([1, 2])
  half_1 pattern LLGGGGLLGGLLGGGGLLGLG (21 syllables, 33 mātras)
  half_2 pattern GLLGLLGLLGGGGLLGGLLGG (21 syllables, 32 mātras)
  half_2 pattern LLGLLGLLGGGGLLGGLLGG (20 syllables, 30 mātras)
  pada_1 pattern LLGGGGLLGG (10 syllables, 16 mātras)
                                                       partial match for: gaṇadehā set([1, 2, 3, 4])
  pada_1 pattern LLGGGGLLGGL (11 syllables, 17 mātras)
  pada_2 pattern LLGGGGLLGLG (11 syllables, 17 mātras)
  pada_2 pattern LGGGGLLGLG (10 syllables, 16 mātras)
  pada_2 pattern LLGGGGLLGL (10 syllables, 15 mātras)
                                                       partial match for: gaṇadehā set([1, 2, 3, 4])
  pada_3 pattern GLLGLLGLLGG (11 syllables, 16 mātras)
                                                        partial match for: dodhakam set([1, 2, 3, 4])
                                                        partial match for: vargavatī set([1, 3])
                                                        accidental match for: akusumacaram set([2, 4])
                                                        partial match for: dodhaka set([1, 2, 3, 4])
                                                        partial match for: avahitrā set([1, 3])
                                                        partial match for: korakitā set([1, 3])
                                                        partial match for: kamalākarā set([1, 3])
                                                        accidental match for: vegavatī set([2, 4])
                                                        accidental match for: īhā set([2, 4])
                                                        accidental match for: pāṭalikā set([2, 4])
                                                        partial match for: drutamadhyā set([1, 3])
                                                        accidental match for: upacitrā set([2, 4])
  pada_3 pattern GLLGLLGLLG (10 syllables, 14 mātras)
                                                       partial match for: viśvamukhī set([1, 2, 3, 4])
  pada_3 pattern LLGLLGLLGG (10 syllables, 14 mātras)
                                                       partial match for: lulitam set([1, 2, 3, 4])
                                                       accidental match for: vargavatī set([2, 4])
                                                       partial match for: uditam set([1, 2, 3, 4])
                                                       partial match for: vegavatī set([1, 3])
  pada_3 pattern GGLLGGLLGG (10 syllables, 16 mātras)
  pada_3 pattern GGGLLGGLLGG (11 syllables, 18 mātras)

upacitrā is mentioned only in 1 place

[shreevatsa]

The input verse scans as:

• taruṇīveṣoddīpitakāmā = LLG-GG-GLL-GG
• vikasajjātīpuṣpasugandhiḥ | = LLG-GG-GLL-GG (scanned as GL but that's a separate bug and doesn't affect the output)
• unnatapīnapayodharabhārā = GLL-GLL-GLL-GG
• prāvṛṭ tanute kasya na harṣam || 90 || GG-LLG-GLL-GG

That is:

LLGGGGLLGG
LLGGGGLLGG
GLLGLLGLLGG
GGLLGGLLGG

This does not match the definition of Upacitrā we have:

sanskrit/data/vrttaratnakara.json

Line 133 in 6748493

["upacitrā", ["LLGLLGLLGLG", "GLLGLLGLLGG"]],

LLGLLGLLGLG
GLLGLLGLLGG
LLGLLGLLGLG
GLLGLLGLLGG

Maybe Kosambi is using a different definition.

The first two lines match gaṇadeha which is 4 lines each LLGGGGLLGG.

We could probably highlight all matches ranked, but I suspect gaṇadeha would still end up on top.

In any case, ideally what we should highlight here is the "4-ness" ("laya" / "madhyagati"), because that's what will help the reader to read this.