mp3imager

#!/usr/bin/python3

import subprocess, re, sys, stagger, urllib.request, os, base64
from stagger.id3 import *  

# Amazon Settings

# SearchPrefix      = 'http://www.amazon.co.uk/s/field-keywords='
# ProductImageRE    = re.compile('.*<a href=\"http://www.amazon.co.uk/([^/]*)/.*<img[^>]*src=\"([^\"]*)"[^>]*alt=\"Product Details\".*')
# ProductDescGroup  = 1
# ProductImageGroup = 2
# ImageBaseUrlRE    = re.compile('(.*)\.[^.]*(\.[^.]*)')
# ImageBaseUrlGroup = 1
# ImageRestUrlGroup = 2
# ImgDimension      = '._SL300'
# ImgSuffix         = 'jpg'
# Retries           = 5 

# Last FM Settings

SearchPrefix      = 'http://www.last.fm/search?q='
ProductImageRE    = re.compile('.*<a href=\"([^\"]*)\".*<img\ *width=\"[^\"]*\"\ *src=\"([^\"]*)\"\ *class=\"art\"\ *class=\"albumCover[^\"]*\"\ */>.*')
ProductDescGroup  = 1
ProductImageGroup = 2
AmazonResize      = 0
ImageBaseUrlRE    = re.compile('(.*serve/)[0-9a-z]*(/.*)')
ImageBaseUrlGroup = 1
ImageRestUrlGroup = 2
ImgDimension      = '252'
ImgSuffix         = 'png'
Retries           = 1 

# General Settings

CurlUserAgent     = 'Mozilla/4.0'
Curl              = 'curl'
CurlUserAgentFlag = '-A'
CurlSilent        = '-s'
BufSize           = 1000

FailImgUrl        = 'NotFound'

Find              = 'find'
FindRegExpFlag    = '-regex'
FindMp3RegExp     = '.*\\(mp3\\|MP3\\)'

# To remember what we already tried to download
# to avoid doing it for every song in the album...
tried_and_failed = set()


def GetURL(url):
    print("Getting URL:", url)
    return subprocess.Popen([Curl, 
                             CurlSilent, 
                             CurlUserAgentFlag, 
                             CurlUserAgent, 
                             url], 
                            bufsize=BufSize, 
                            stdout=subprocess.PIPE).stdout


def TakeFirstImage(page_data):
    line = 'blah'
    img_url = FailImgUrl
    while (len(line) > 0) and (img_url == FailImgUrl):
        try:
            line = page_data.readline().decode('utf-8')
            m = ProductImageRE.match(line)
            if m:
                img_url = m.group(ProductImageGroup)
        except UnicodeDecodeError:
            #print('Unicode decode error, ignoring line')
            pass
    return img_url


def GetCovers(page_data):
    line = 'blah'
    covers = [] 
    while (len(line) > 0):
        try:
            line = page_data.readline().decode('utf-8')
            m = ProductImageRE.match(line)
            if m:
                covers.append((m.group(ProductDescGroup),
                               m.group(ProductImageGroup)))
        except UnicodeDecodeError:
            #print('Unicode decode error, ignoring line')
            pass
    return covers

   


def AskImage(page_data):
    covers = GetCovers(page_data)
    for (desc, cover) in covers:
        print(desc)
    return FailImgUrl


def GetSimpleImgUrl(search_term):
#    rawData = urllib.request.urlopen(SearchPrefix+search_term)
    rawData = GetURL(SearchPrefix+search_term)
    if choose_images:
        return AskImage(rawData)
    else:
        return TakeFirstImage(rawData)


def MakeLarge(simple_url):
    m = ImageBaseUrlRE.match(simple_url)
    if m:
        return (m.group(ImageBaseUrlGroup) + 
                ImgDimension + 
                m.group(ImageRestUrlGroup))
    else:
        #print('Error changing image size.')
        return simple_url


def SearchTerm(artist, album):
    both = artist + " " + album
    return urllib.request.pathname2url(both)


def GetImageUrl(artist, album):
    tries = 0
    full_url = FailImgUrl
    while (tries < Retries) and (full_url == FailImgUrl):
        simple_url = GetSimpleImgUrl(SearchTerm(artist, album))
        full_url = MakeLarge(simple_url)
        tries+=1
    return full_url

def SanitiseTag(tag):
    return base64.urlsafe_b64encode(tag.encode("utf-8")).decode("utf-8")

def GetImageFileName(directory, artist, album):
    return (directory + os.sep +
            SanitiseTag(artist) + "-" +
            SanitiseTag(album) +
            "." + ImgSuffix)


def DoMp3(mp3_file):
    try:
        tag = stagger.read_tag(mp3_file)
        if (len(tag.picture) > 0) and clobber_images:
            pass
        elif (len(tag.album) == 0) or (len(tag.artist) == 0):
            print("One not found of " + tag.album + " and " + tag.artist)
        elif ((tag.artist, tag.album) in tried_and_failed):
            print("Already tried and failed",tag.artist,";",tag.album,";",tag.title)
        else:
            print("Doing " + tag.artist + "; " + tag.album + "; " + tag.title)
            image_name = GetImageFileName(os.path.dirname(mp3_file),
                                          tag.artist,
                                          tag.album)
            has_image = 1
            if not os.path.exists(image_name):
                print("Trying to download")
                has_image = 0
                full_url = GetImageUrl(tag.artist, tag.album)
                if full_url != FailImgUrl:
                    rawImage = GetURL(full_url)
                    img_file = open(image_name, mode='wb')
                    img_file.write(rawImage.read())
                    img_file.close()
                    has_image = 1
                else:
                    tried_and_failed.add((tag.artist, tag.album))
                    print("Failed... moving on.")
            if has_image:
                tag.picture = image_name
                tag.write()
    except stagger.NoTagError:
        pass
        
 

def DoMp3s(directory):
    print('Searching for MP3s in ' + directory)
    rawData = subprocess.Popen([Find, 
                                directory, 
                                FindRegExpFlag, 
                                FindMp3RegExp], 
                               bufsize=BufSize, 
                               stdout=subprocess.PIPE).stdout
    line = 'blah'
    while (len(line) > 0): 
        line = rawData.readline().decode('utf-8').strip()
        if len(line) > 0:
            DoMp3(line)
  

mp3dir = ''
clobber_images = 0
choose_images = 0

for arg in sys.argv[1:]:
    if arg == '--clobber':
        clobber_images == 1
    elif arg == '--choose':
        choose_images = 1
    elif len(mp3dir) == 0:
        mp3dir = arg

if len(mp3dir) > 0:
    DoMp3s(mp3dir)
else:
    print("usage: mp3imager <base-dir>")
    print("warning: may trash your mp3s while trying to add images to them.")
    print("    --clobber : override any existing embedded images")
    print("    --choose : offer a menu of choices instead of taking the first one.")