Skip to content

Latest commit

 

History

History
70 lines (50 loc) · 2.36 KB

README-lxx.md

File metadata and controls

70 lines (50 loc) · 2.36 KB
Raw

LXX [WORK IN PROGRESS]

  • lxxmorph/ -- directory of patch files to correct CATSS lxxmorph
  • lxx_lexicon.yaml -- WIP stem dataset for LXX
  • lxxmorph_generate.py -- script that goes through all the verb forms in lxxmorph/ and validates that the code + dataset generates the correct form
  • generate_lxxmorph_lexicon.py -- similar to lxxmorph_generate.py but instead of just validating and showing unexplainable forms, builds the starting point for a lexicon file to explain the forms
  • lxxmorph_utils.py -- common code used by the above scripts

Using lxxmorph_generate.py

The particular books to be tested are configured in the MLXX_FILES variable in the script.

How to Extend Coverage of Lexicon

  1. set LXX_FILENAME in generate_lxxmorph_lexicon.py to the path of the lxxmorph file you want to work on
  2. add an entry in book_to_num in lxxmorph_utils.py to map book code used by file to book number
  3. run ./generate_lxxmorph_lexicon.py > tmp1
  4. cat lxx_lexicon.yaml tmp1 > tmp2
  5. ./sort_lexicon.py tmp2 > lxx_lexicon.yaml
  6. remove tmp1 and tmp2
  7. review all lines in lxx_lexicon.yaml that have # @ (you can review about 10 a minute once you get good at it)

@m means multiple possible stems, for example:

ἀγνοέω:
    stems:
        1-: {'ἀγνοου{athematic}', 'ἀγνοε', 'ἀγνοο'}  # @m

which should be manually corrected to:

ἀγνοέω:
    stems:
        1-: ἀγνοε

@1 means a single possible stem. Verify and make sure if the lemma already exists that the new stem is moved in with the others.

For example:

ἀποκόπτω:
    stems:
        3+: ἀπεκοψ
    stems:
        2-: ἀποκοψ  # @1

needs to be changed to

ἀποκόπτω:
    stems:
        2-: ἀποκοψ
        3+: ἀπεκοψ
    stems:

@0 means no stem could be guessed. This normally means a missing stemming.yaml rule.

At the end of all this, you can run ./generate_lxxmorph_lexicon.py again and if you haven't made any mistakes, it should return nothing.

  1. Update MLXX_FILES in lxxmorph_generate.py and run to test your new lexicon.

You will almost certainly get failures. These could just be mistakes you made in step 7, could be missing stemming.yaml rules, or (actually most likely at this stage in the project) could be mistakes in the .mlxx file that need to be corrected.