Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

add elleux

This commit is contained in:
thjbbvlt 2024-03-28 16:57:25 +01:00
parent 0f1a82b24a
commit aee667a781
1 changed files with 2 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
spacy/lang/fr/punctuation.py
View File

@ -18,17 +18,16 @@ HYPHENS = r"- – — ‐ ‑".replace(" ", "")
_suffix_inversion = [ _suffix_inversion = [
"je", "tu", "on", "il", "elle", "iel", "je", "tu", "on", "il", "elle", "iel",
"nous", "vous", "elles", "ils", "iels", "nous", "vous", "elles", "ils", "iels",
"moi", "toi", "lui", "leur", "eux", "moi", "toi", "lui", "leur", "eux", "elleux",
"ce", "ici", "là",
# to avoid matching: Villar-le-bois # to avoid matching: Villar-le-bois
fr"la(?![{HYPHENS}])", fr"la(?![{HYPHENS}])",
fr"le(?![{HYPHENS}])", fr"le(?![{HYPHENS}])",
fr"les(?![{HYPHENS}])", fr"les(?![{HYPHENS}])",
fr"en(?![{HYPHENS}])", "y", fr"en(?![{HYPHENS}])", "y",
"ce",
# a-t-on, a-t'on # a-t-on, a-t'on
fr"t[{HYPHENS}]??[{ELISION}]?", fr"t[{HYPHENS}]??[{ELISION}]?",
fr"m[{ELISION}]?", fr"m[{ELISION}]?",
"là", "ici",
] ]
_prefix_elision = [ _prefix_elision = [
"n", "s", "c", "d", "j", "m", "t", "l", "qu", "n", "s", "c", "d", "j", "m", "t", "l", "qu",