KevinMidboe/linguist
1
0
Fork 0
mirror of https://github.com/KevinMidboe/linguist.git synced 2025-10-29 17:50:22 +00:00
Code Issues Packages Projects Releases Wiki Activity

Use classifier as primary method for disambiguation

Browse Source
This commit is contained in:
Joshua Peek
2012-06-21 10:55:26 -05:00
parent 540f2a0941
commit 076bf7d0c8
1 changed files with 5 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
lib/linguist/blob_helper.rb
View File

@@ -454,30 +454,18 @@ module Linguist
# Returns a Language or nil. # Returns a Language or nil.
def disambiguate_extension_language def disambiguate_extension_language
if Language.ambiguous?(extname) if Language.ambiguous?(extname)
# name = "guess_#{extname.sub(/^\./, '')}_language"
# send(name) if respond_to?(name)
possible_languages = Language.all.select { |l| l.extensions.include?(extname) } possible_languages = Language.all.select { |l| l.extensions.include?(extname) }
name = "guess_#{extname.sub(/^\./, '')}_language"
language = send(name) if respond_to?(name)
if possible_languages.any? if possible_languages.any?
results = Classifier.instance.classify(data, possible_languages) if result = Classifier.instance.classify(data, possible_languages).first
guessed_language, score = results.first result[0]
if guessed_language != language
report_classifier_incorrect_guess(language, guessed_language, score)
end end
end end
language
end end
end end
class LanguageClassifierError < StandardError; end
def report_classifier_incorrect_guess(expected, actual, score)
raise LanguageClassifierError, "Expected #{expected}, but was #{actual} scoring #{score}.\n#{name}\n#{data}"
end
# Internal: Guess language of .cls files # Internal: Guess language of .cls files
# #
# Returns a Language. # Returns a Language.