def language_percentage(file_paths):
"""
Computes the percentage composition of each language, with unknown
extensions tagged with the ``Unknown`` key.
:param file_paths: A list of file paths.
:return: A dict with file name as key and the percentage
of occurences as the value.
"""
if file_paths:
delta = 100 / len(file_paths)
results = defaultdict(lambda: 0)
for file_path in file_paths:
ext = os.path.splitext(file_path)[1]
if ext in exts:
for lang in exts[ext]:
results[lang] += delta
continue
hashbang = get_hashbang(file_path)
if not hashbang:
continue
language = get_language_from_hashbang(hashbang).lower()
for ext in exts:
for lang in exts[ext]:
if language == lang.lower():
results[lang.lower()] += delta
return results
def language_percentage(file_paths):
"""
Computes the percentage composition of each language, with unknown
extensions tagged with the ``Unknown`` key.
:param file_paths: A list of file paths.
:return: A dict with file name as key and the percentage
of occurences as the value.
"""
if file_paths:
delta = 100 / len(file_paths)
results = defaultdict(lambda: 0)
for file_path in file_paths:
ext = os.path.splitext(file_path)[1]
if ext in exts:
for lang in exts[ext]:
results[lang] += delta
continue
hashbang = get_hashbang(file_path)
if not hashbang:
continue
language = get_language_from_hashbang(hashbang).lower()
for ext in exts:
for lang in exts[ext]:
if language == lang.lower():
results[lang.lower()] += delta
return results
def language_percentage(file_paths):
"""
Computes the percentage composition of each language, with unknown
extensions tagged with the ``Unknown`` key.
:param file_paths: A list of file paths.
:return: A dict with file name as key and the percentage
of occurences as the value.
"""
if file_paths:
delta = 100 / len(file_paths)
results = defaultdict(lambda: 0)
for file_path in file_paths:
ext = os.path.splitext(file_path)[1]
if ext in exts:
for lang in exts[ext]:
results[lang] += delta
elif os.path.exists(file_path):
with open(file_path, 'r') as data:
hashbang = data.readline()
if re.match(HASHBANG_REGEX, hashbang):
language = get_language_from_hashbang(hashbang).lower()
for ext in exts:
for lang in exts[ext]:
if language == lang.lower():
results[lang.lower()] += delta
return results
def test_get_language_from_hashbang(self):
self.assertEqual(get_language_from_hashbang('#!/usr/bin/env python'),
'python')
self.assertEqual(get_language_from_hashbang('#!bin/bash'), 'bash')