Created
September 8, 2012 23:22
-
-
Save natew/3680949 to your computer and use it in GitHub Desktop.
Rails model for parsing artist information from a song
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
class Song | |
# Regular expressions | |
RE = { | |
:featured => /(featuring | ?ft\.? |feat\.? |f\. |w\/){1}/i, | |
:remixer => / remix| rmx| edit| bootleg| mix| remake| re-work| rework| extended remix| bootleg remix/i, | |
:mashup_split => / \+ | x | vs\.? /i, | |
:producer => /^(produced by|prod\.? by |prod\. )/i, | |
:cover => / cover/i, | |
:split => /([^,&]+)(& ?([^,&]+)|, ?([^,&]+))*/i, # Splits "one, two & three" | |
:open => /[\(\[\{]/, | |
:close => /[\)\]\}]/, | |
:containers => /[\{\[\(\)\]\}]/i, | |
:percents => /(% ?){2,10}/, | |
:remove => /(extended|vip|original|club|vocal) mix|(extended|vip|radio) edit|(on )?soundcloud|(exclusive )?free (download|d\/?l)/i, | |
:and => /, | & | and /i | |
} | |
SPLITS = { | |
:featured => /#{RE[:featured]}#{RE[:split]}/i, | |
:producer => /#{RE[:producer]}#{RE[:split]}/i, | |
:remixer => /#{RE[:split]}#{RE[:remixer]}/i, | |
:cover => /#{RE[:split]}#{RE[:cover]}/i, | |
:mashup => /#{RE[:mashup_split]}/i | |
} | |
STRIP = { | |
:remixer => /\'s.*| (vocal|instrumental|(summer|fall|spring|winter)( 2[0-9]{3})?) /i, | |
:producer => /^by /i | |
} | |
PREFIX = { | |
:featured => /( |\(|\[|\{)/ | |
} | |
def parse_artists | |
logger.info "#{id}: #{full_name}" | |
parse_name = (name || link_info[1]).gsub(RE[:remove], '') | |
parse_artist = artist_name || link_info[0] | |
split_and_find_artists(parse_name) | find_artists(parse_artist) | split_and_find_artists(parse_artist) | |
end | |
def find_artists(name) | |
matches = [] | |
search_name = name.gsub(/#{RE[:open]}.*|#{RE[:featured]}.*|#{RE[:producer]}.*/i, '') | |
if has_mashups(search_name) | |
find_mashups(search_name) do |artist| | |
matches.push [artist, :mashup] | |
end | |
else | |
search_name.clean_split(RE[:and]) do |artist| | |
matches.push [artist.strip, :original] | |
end | |
end | |
matches.push [search_name.strip, :original] if matches.empty? | |
matches.reject(&:blank?) | |
end | |
def split_and_find_artists(name) | |
matches = [] | |
name.clean_split(RE[:containers]) do |part| | |
if has_mashups(part) | |
find_mashups(part) do |artist| | |
artist.gsub!(RE[:remixer], '') | |
matches.push [artist, :mashup] | |
end | |
else | |
matches = matches + find_artists_types(part, true) | |
end | |
end | |
matches.reject(&:blank?) | |
end | |
def has_mashups(name) | |
name.scan(RE[:mashup_split]).empty? ? false : true | |
end | |
def find_mashups(name) | |
name.clean_split(RE[:mashup_split]) do |artist| | |
artist.clean_split(RE[:and]) do |split_artist| | |
yield split_artist | |
end | |
end | |
end | |
def find_artists_types(part, container = false) | |
matches = [] | |
types = [:producer, :featured] | |
types = types + [:remixer, :cover] if container | |
scan_for(part, types) do |match| | |
matches.push match | |
end | |
matches.reject(&:empty?) | |
end | |
def scan_for(part, types) | |
types.each do |type| | |
scan_artists(part, type) do |match| | |
yield match | |
end | |
end | |
end | |
def scan_artists(part, type) | |
return unless part | |
scan = /#{PREFIX[type]}#{SPLITS[type]}/i | |
part.clean_scan(scan, RE[type]) do |artist| | |
artist.gsub!(STRIP[type], '') if STRIP.has_key? type | |
artist.split(RE[:and]).each do |split| | |
yield [split, type] | |
end | |
end | |
end | |
def link_info | |
if link_text | |
split = link_text.split(/\s*(-|—|–)\s*/) | |
split.size >= 3 ? [split[0], split[2]] : [nil,nil] | |
else | |
['',''] | |
end | |
end | |
end |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
class String | |
def clean_split(regex) | |
self.split(regex).reject(&:blank?).collect(&:strip).each do |part| | |
yield part | |
end | |
end | |
def clean_scan(regex, reject) | |
self.scan(regex).flatten.compact.each do |string| | |
yield string.strip unless string =~ reject or string =~ /^[&,]/ | |
end | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment