explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-07-10 16:22:29 +03:00
Code Issues Packages Projects Releases Wiki Activity

Move functions to deprecated

Browse Source
This commit is contained in:
ines 2017-04-15 12:12:31 +02:00
parent c05ec4b89a
commit 956dc36785
2 changed files with 19 additions and 35 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
spacy/deprecated.py
View File

@ -53,6 +53,23 @@ def detokenize(token_rules, words): # Deprecated?
return positions return positions
def match_best_version(target_name, target_version, path):
path = util.ensure_path(path)
if path is None or not path.exists():
return None
matches = []
for data_name in path.iterdir():
name, version = split_data_name(data_name.parts[-1])
if name == target_name:
matches.append((tuple(float(v) for v in version.split('.')), data_name))
if matches:
return Path(max(matches)[1])
else:
return None
def split_data_name(name):
return name.split('-', 1) if '-' in name else (name, '')
def fix_glove_vectors_loading(overrides): def fix_glove_vectors_loading(overrides):
@ -72,11 +89,11 @@ def fix_glove_vectors_loading(overrides):
vec_path = None vec_path = None
if 'add_vectors' not in overrides: if 'add_vectors' not in overrides:
if 'vectors' in overrides: if 'vectors' in overrides:
vec_path = util.match_best_version(overrides['vectors'], None, data_path) vec_path = match_best_version(overrides['vectors'], None, data_path)
if vec_path is None: if vec_path is None:
return overrides return overrides
else: else:
vec_path = util.match_best_version('en_glove_cc_300_1m_vectors', None, data_path) vec_path = match_best_version('en_glove_cc_300_1m_vectors', None, data_path)
if vec_path is not None: if vec_path is not None:
vec_path = vec_path / 'vocab' / 'vec.bin' vec_path = vec_path / 'vocab' / 'vec.bin'
if vec_path is not None: if vec_path is not None:

33
spacy/util.py
View File

@ -55,39 +55,6 @@ def or_(val1, val2):
return val2 return val2
def match_best_version(target_name, target_version, path):
path = path if not isinstance(path, basestring) else pathlib.Path(path)
if path is None or not path.exists():
return None
matches = []
for data_name in path.iterdir():
name, version = split_data_name(data_name.parts[-1])
if name == target_name and constraint_match(target_version, version):
matches.append((tuple(float(v) for v in version.split('.')), data_name))
if matches:
return pathlib.Path(max(matches)[1])
else:
return None
def split_data_name(name):
return name.split('-', 1) if '-' in name else (name, '')
def constraint_match(constraint_string, version):
# From http://github.com/spacy-io/sputnik
if not constraint_string:
return True
constraints = [c.strip() for c in constraint_string.split(',') if c.strip()]
for c in constraints:
if not re.match(r'[><=][=]?\d+(\.\d+)*', c):
raise ValueError('invalid constraint: %s' % c)
return all(semver.match(version, c) for c in constraints)
def read_regex(path): def read_regex(path):
path = ensure_path(path) path = ensure_path(path)
with path.open() as file_: with path.open() as file_: