from flexpy.Corpus import Corpus
from flexpy.tags.RtLexEntry import RtLexEntry
from flexpy.LexEntry import LexEntry
from flexpy.FlexPyUtil import get_tone_letters_from_string
project_dir = "/home/wesley/.local/share/fieldworks/Projects/"
project_name = "IxpantepecMixtec"
corpus = Corpus(project_dir, project_name, include_punctuation=False)
tag_dict = corpus.tag_dict
# desired result:
# list of lexical items, each place it occurs in 1st position
# and each place it occurs in 2nd position
# show baseline tone pattern and perturbed tone pattern of both
# look at this big list to see what jumps out
by_lex_guid = {}
for text in corpus.texts:
# print(f"current text is {text}")
# iterate over MORPHEMES in the text (not across paragraph boundaries)
# know the LexEntry that each of them belongs to
if text.paragraphs is None:
# print(f"skipping text {text} due to lack of paragraphs")
continue
for pg in text.paragraphs:
# print("\n---- new paragraph ----\n")
# print(f"current paragraph is: {pg}")
# print(f"run texts is: {pg.run_texts}")
"Yangulam",
]
print("found {} languages".format(len(language_names)))
project_dir = "/home/wesley/.local/share/fieldworks/Projects/"
lexicons = {}
wordlist_dir = "/home/wesley/flexpy/flexpy/language_data/MadangWordlists/"
for ln in language_names:
print("-- loading language {}".format(ln))
filename = "{}.tsv".format(ln)
fp = os.path.join(wordlist_dir, filename)
if os.path.exists(fp):
lexicon = load_lexicon_from_tsv(fp, ln)
lexicons[ln] = lexicon
else:
print("creating tsv for lexicon of {}".format(ln))
corpus = Corpus(project_dir, ln, include_punctuation=False)
write_lexicon_tsv(corpus, fp)
assert os.path.exists(fp) # should be there now that we wrote it
lexicon = load_lexicon_from_tsv(fp, ln)
lexicons[ln] = lexicon
print("{} has {} lexeme entries".format(ln, len(lexicon.lexemes)))
test_show_similarity(lexicons)
lexicon1, lexicon2 = random.sample(list(lexicons.values()), 2)
glosses = [
"man",
"woman",
"canoe",
"sun",
"red",