Upload 4 files

__init__.py

@@ -0,0 +1,8 @@
+from azapi import AZlyrics
+
+__name__ = 'azapi.azapi.AZlyrics'
+__author__ = 'Khaled H. El-Morshedy'
+__url__ = 'https://github.com/elmoiv/azapi'
+__description__ = 'Get Lyrics from AZLyrics.com like a Boss ~(0_0)~'
+__license__ = 'GPL-v3.0'
+__version__ = '3.0.7'

azapi.py

@@ -1,5 +1,7 @@
-from .requester import Requester
-from .tools import *
+# from .requester import Requester
+# from .tools import *
+from requester import Requester
+from tools import *
 
 class AZlyrics(Requester):
     """
@@ -175,4 +177,4 @@ class AZlyrics(Requester):
         
         # Store songs for later usage
         self.songs = parseSongs(albums_page)
-        return self.songs
+        return self.songs

requester.py

@@ -0,0 +1,52 @@
+import requests, random
+
+userAgents = '''Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2228.0 Safari/537.36
+Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2227.1 Safari/537.36
+Mozilla/5.0 (Macintosh; Intel Mac OS X 10_6_8) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.11 Safari/535.19
+Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.11 (KHTML, like Gecko) Chrome/17.0.963.66 Safari/535.11
+Mozilla/5.0 (X11; U; Linux x86_64; en-US) AppleWebKit/532.2 (KHTML, like Gecko) Chrome/4.0.221.3 Safari/532.2
+Mozilla/5.0 (X11; U; Linux i686; en-US) AppleWebKit/532.2 (KHTML, like Gecko) Chrome/4.0.221.0 Safari/532.2
+Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US) AppleWebKit/532.1 (KHTML, like Gecko) Chrome/4.0.220.1 Safari/532.1
+Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/532.1 (KHTML, like Gecko) Chrome/4.0.219.6 Safari/532.1
+Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US) AppleWebKit/532.1 (KHTML, like Gecko) Chrome/4.0.219.5 Safari/532.1
+Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/532.1 (KHTML, like Gecko) Chrome/4.0.219.5 Safari/532.1
+Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/532.1 (KHTML, like Gecko) Chrome/4.0.219.4 Safari/532.1
+Mozilla/5.0 (X11; U; Linux x86_64; en-US) AppleWebKit/532.1 (KHTML, like Gecko) Chrome/4.0.219.3 Safari/532.1
+Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US) AppleWebKit/532.1 (KHTML, like Gecko) Chrome/4.0.219.3 Safari/532.1
+Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/532.1 (KHTML, like Gecko) Chrome/4.0.219.3 Safari/532.1
+Mozilla/5.0 (X11; U; Linux i686; en-US) AppleWebKit/532.0 (KHTML, like Gecko) Chrome/3.0.197.0 Safari/532.0
+Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US) AppleWebKit/532.0 (KHTML, like Gecko) Chrome/3.0.197.0 Safari/532.0
+Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US) AppleWebKit/530.5 (KHTML, like Gecko) Chrome/2.0.172.23 Safari/530.5
+Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US) AppleWebKit/530.5 (KHTML, like Gecko) Chrome/2.0.172.2 Safari/530.5
+Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/530.5 (KHTML, like Gecko) Chrome/2.0.172.2 Safari/530.5
+Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US) AppleWebKit/530.4 (KHTML, like Gecko) Chrome/2.0.172.0 Safari/530.4
+Mozilla/5.0 (Windows; U; Windows NT 5.2; eu) AppleWebKit/530.4 (KHTML, like Gecko) Chrome/2.0.172.0 Safari/530.4
+Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US) AppleWebKit/530.4 (KHTML, like Gecko) Chrome/2.0.172.0 Safari/530.4
+Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/530.5 (KHTML, like Gecko) Chrome/2.0.172.0 Safari/530.5
+Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US) AppleWebKit/530.4 (KHTML, like Gecko) Chrome/2.0.171.0 Safari/530.4
+Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/530.1 (KHTML, like Gecko) Chrome/2.0.170.0 Safari/530.1
+Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/530.1 (KHTML, like Gecko) Chrome/2.0.169.0 Safari/530.1
+Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US) AppleWebKit/530.1 (KHTML, like Gecko) Chrome/2.0.168.0 Safari/530.1
+Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US) AppleWebKit/530.1 (KHTML, like Gecko) Chrome/2.0.164.0 Safari/530.1
+Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US) AppleWebKit/530.0 (KHTML, like Gecko) Chrome/2.0.162.0 Safari/530.0
+Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US) AppleWebKit/530.0 (KHTML, like Gecko) Chrome/2.0.160.0 Safari/530.0
+Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US) AppleWebKit/528.10 (KHTML, like Gecko) Chrome/2.0.157.2 Safari/528.10
+Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/528.10 (KHTML, like Gecko) Chrome/2.0.157.2 Safari/528.10
+Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US) AppleWebKit/528.11 (KHTML, like Gecko) Chrome/2.0.157.0 Safari/528.11
+Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/528.9 (KHTML, like Gecko) Chrome/2.0.157.0 Safari/528.9
+Mozilla/5.0 (Linux; U; en-US) AppleWebKit/525.13 (KHTML, like Gecko) Chrome/0.2.149.27 Safari/525.13
+Mozilla/5.0 (Macintosh; U; Mac OS X 10_6_1; en-US) AppleWebKit/530.5 (KHTML, like Gecko) Chrome/ Safari/530.5
+Mozilla/5.0 (Macintosh; U; Mac OS X 10_5_7; en-US) AppleWebKit/530.5 (KHTML, like Gecko) Chrome/ Safari/530.5
+Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10_5_6; en-US) AppleWebKit/530.9 (KHTML, like Gecko) Chrome/ Safari/530.9
+Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10_5_6; en-US) AppleWebKit/530.6 (KHTML, like Gecko) Chrome/ Safari/530.6
+Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10_5_6; en-US) AppleWebKit/530.5 (KHTML, like Gecko) Chrome/ Safari/530.5'''
+
+class Requester():
+    USER_AGENTS = userAgents.split('\n')
+    
+    # Inspired from: https://github.com/brianchesley/Lyrics/blob/master/lyrics_data_scrape.py
+    def get(self, url, _proxies={}):
+        return requests.get(url, headers={'User-Agent': random.choice(self.USER_AGENTS)}, proxies=_proxies)
+    
+    def head(self, url, _proxies={}):
+        return requests.head(url, headers={'User-Agent': random.choice(self.USER_AGENTS)}, proxies=_proxies)

tools.py

@@ -0,0 +1,149 @@
+import bs4, re, time, os
+from urllib.parse import quote
+# from .jaro import jaro_distance
+from songscope import jaro_distance
+
+letters = 'abcdefghijklmnopqrstuvwxyz0123456789'
+
+def htmlFind(page):
+    # v3.0
+    # Changed page.text -> page.content.decode() to support variant unicodes
+    soup = bs4.BeautifulSoup(
+                        page.content.decode(),
+                        "html.parser"
+                        )
+    return soup.find
+
+def htmlFindAll(page):
+    # v3.0
+    # Changed page.text -> page.content.decode() to support variant unicodes
+    soup = bs4.BeautifulSoup(
+                        page.content.decode(),
+                        "html.parser"
+                        )
+    return soup.findAll
+
+def filtr(inpt, isFile=False):
+    if isFile:
+        return ''.join(i for i in inpt if i not in r'<>:"/\|?*')
+    return ''.join(i.lower() for i in inpt if i.lower() in letters)
+
+def normalGet(artist='', title='', _type=0):
+    art, tit = filtr(artist), filtr(title)
+    if _type:
+        print('https://www.azlyrics.com/{}/{}.html'.format(art[0], art))
+        return 'https://www.azlyrics.com/{}/{}.html'.format(art[0], art)
+    return 'https://www.azlyrics.com/lyrics/{}/{}.html'.format(art, tit)
+
+def googleGet(srch_eng, acc, get_func, artist='', title='', _type=0, proxies={}):
+    # Encode artist and title to avoid url encoding errors
+    data = artist + ' ' * (title != '' and artist != '') + title
+    encoded_data = quote(data.replace(' ', '+'))
+
+    # Perform a search (for accuracy) [Custom search engine]
+    search_engines = {
+        'google': 'https://www.google.com/search?q=',
+        'duckduckgo': 'https://duckduckgo.com/html/?q='
+    }
+
+    slctd_srch_engn = 'google'
+    if srch_eng in search_engines:
+        slctd_srch_engn = srch_eng
+
+    google_page = get_func('{}{}+site%3Aazlyrics.com'.format(
+                                    search_engines[slctd_srch_engn],
+                                    encoded_data
+                                    ),
+                            proxies
+                            )
+    
+    # Choose between lyrics or song according to function used
+    regex = [
+        r'(azlyrics\.com\/lyrics\/(\w+)\/(\w+).html)',
+        r'(azlyrics\.com\/[a-z0-9]+\/(\w+).html)'
+    ]
+    
+    # ex result: [('azlyrics.com/t/taylorswift.html', 'taylorswift')]
+    # result[0][0] = 'azlyrics.com/t/taylorswift.html'
+    results = re.findall(
+                        regex[_type],
+                        google_page.text
+                        )
+
+    if len(results):
+        # calculate jaro similarity for artist and title
+        jaro_artist = 1.0
+        jaro_title = 1.0
+        
+        if artist:
+            jaro_artist = jaro_distance(
+                                        artist.replace(' ', ''),
+                                        results[0][1]
+                                        )
+        if title:
+            jaro_title = jaro_distance(
+                                        title.replace(' ', ''),
+                                        results[0][2]
+                                        )
+        
+        if jaro_artist >= acc and jaro_title >= acc:
+            return 'https://www.' + results[0][0]
+        else:
+            print('Similarity <', acc)
+    else:
+        print(srch_eng.title(), 'found nothing!')
+    
+    return 0
+
+# v3.0.5: Re-coded ParseLyrics to be more efficient
+def parseLyric(page):
+    divs = [i.text for i in htmlFindAll(page)('div', {'class': None})]
+    return max(divs, key=len)
+
+def parseSongs(page):
+    songs = {}
+    Parent = htmlFind(page)('div', {'id':'listAlbum'})
+    if Parent:
+        Raw_Data = Parent.findChildren()
+
+        curType, curName, curYear = '', '', ''
+
+        for elmnt in Raw_Data:
+
+            # v3.0.3: Removed break after script due to google ads inside listAlbum
+            # is using script tag, which results in not all songs retrieved
+            #if elmnt.name == 'script':
+            #    break
+            
+            # album info are inside divs
+            if elmnt.name == 'div':
+                if elmnt.text == 'other songs:':
+                    curType, curName, curYear = 'Others', '', ''
+                else:
+                    # Separating to (album, name, year)
+                    rgx = re.findall(r'(.*):\s"(.*)"\s\(([0-9]+)\)', elmnt.text)
+                    if rgx:
+                        curType, curName, curYear = rgx[0]
+            if elmnt.name == 'a':
+                songs[elmnt.text] = {
+                    'year': curYear,
+                    'album': curName,
+                    'type': curType,
+                    # Azlyrics puts hrefs with/without base url
+                    'url': 'http://www.azlyrics.com' + elmnt['href'].strip('.') \
+                            if elmnt['href'].startswith('/lyrics/') else elmnt['href']
+                }
+    # v 3.0
+    # Some artists have no albums, so we cover this
+    else:
+        for div in htmlFindAll(page)('div', {'class':'listalbum-item'}):
+            a = div.find('a')
+            songs[a.text] = {
+                'year': '',
+                'album': '',
+                'type': '',
+                # v3.0.1: fix relative urls -> absolute url
+                'url': 'http://www.azlyrics.com' + a['href'][2:] \
+                        if a['href'][:2] == '..' else a['href']
+                }
+    return songs