edstratford
commented
2 years ago STILL DRAFTING STEP 2 (This will only work for strong finite verbs) Declare the following variables: Morphological Form Stem Tense Person Gender Grammatical NUmber Clitic 1 (for ventive) Clitic 2 (for suffix pronouns) Clitic 2 Person Clitic 2 Grammatical Number Clitic 2 Gender Clitic 2 Case Clitic 3 (for second suffix pronoun) Clitic 3 Person Clitic 3 Grammatical Number Clitic 3 Gender Clitic 3 Case Clitic 4 (for -man/min/man) Pass
STEP 3 (Also - any form that is less than 5 characters will automatically fail) Start the sorting through analysis of the the pseudo-transcription (PT). **Note assignment of some variables is temporary, for example assigning Stem= 'G' before final determination of Gt or Gtn - this is revised through the process. --First loop If PT REGEXP '^i' THEN Stem = 'G', Person = 'Third Person', Gender = 'Masculine', Grammatical Number = 'Singular' AND PT = PT minus first character If PT REGEXP '^ta' THEN Stem = 'G', Person = 'Second Person', Gender = 'Masculine', Grammatical Number = 'Singular' AND PT = PT minus first two characters If PT REGEXP '^a' THEN Stem = 'G', Person = 'First Person', Gender = 'Common', Grammatical Number = 'Singular' AND PT = PT minus first character If PT REGEXP '^ni' THEN Stem = 'G', Person = 'First Person', Gender = 'Common', Grammatical Number = 'Plural' AND PT = PT minus first character If PT REGEXP '^u' THEN Stem = 'D', Person = 'Third Person', Gender = 'Masculine', Grammatical Number = 'Singular' AND PT = PT minus first character If PT REGEXP '^ta' THEN Stem = 'D', Person = 'Second Person', Gender = 'Masculine', Grammatical Number = 'Singular' AND PT = PT minus first two characters If PT REGEXP '^nu' THEN Stem = 'D', Person = 'First Person', Gender = 'Common', Grammatical Number = 'Plural' AND PT = PT minus first character ELSE DROP form (Pass = "Fail")
--Second loop IF PT REGEXP 'ma$' THEN Clitic 4 = 'ma' AND PT = PT minus last character IF PT REGEXP 'man$' THEN Clitic 4 = 'man' AND PT = PT minus last two characters IF PT REGEXP 'min$' THEN Clitic 4 = 'min' AND PT = PT minus last two characters
--Third loop IF PT REGEXP 'am$' THEN Clitic 4 = 'man' AND PT = PT minus last two characters --ni (for subjunctive?)
Fourth loop IF PR REGEXP 'ka$' THEN Clitc 2 Person = 'Second Person', Clitic 2 Grammatical Number = 'Singular', Clitic 2 Gender = 'Masculine', Clitic 2 Case = 'Accusative' AND PT = PT minus last 2 characters IF PR REGEXP 'ki$' THEN Clitc 2 Person = 'Second Person', Clitic 2 Grammatical Number = 'Singular', Clitic 2 Gender = 'Feminine', Clitic 2 Case = 'Accusative' AND PT = PT minus last 2 characters IF PR REGEXP 'kunuti$' THEN Clitc 2 Person = 'Second Person', Clitic 2 Grammatical Number = 'Plural', Clitic 2 Gender = 'Masculine', Clitic 2 Case = 'Dative' AND PT = PT minus last 6 characters IF PR REGEXP 'kunu$' THEN Clitc 2 Person = 'Second Person', Clitic 2 Grammatical Number = 'Plural', Clitic 2 Gender = 'Masculine', Clitic 2 Case = 'Accusative' AND PT = PT minus last 4 characters
Gertrudius
Have lots of explicit_spellings that need to be added to dictionary. Many are verbal forms - these can (to some degree) be recognized and analyzed through a script. Will try to create the rules for such analysis here.
STEP 1 Change the cuneiform spelling into a pseudo-bound transcription. To do this: 1) convert all á,à to just a, and same for i,e,u 2) convert any dash that has a consonant (bdgḫklmnpqrsṣštṭwz) before it and a vowel after into a single quote: (i.e t-a > t'a) 3) eliminate any remaining dashes 4) convert any instance of three vowels in a row to vowel-'-vowel (i.e. aaa > a'a) 5) delete the second instance of any remaining two vowels in a row (i.e aa > a)
This will not be a final form. It ignores the possibility of a doubled consonant unless it's already in the spelling (rare), and doubled alephs also won't show up, and some instances where there should be an aleph between vowels will be lost. But with this pseudo-transcription, can start guesses.