Revised AdultDVDEmpire plugin

I'm not sure how to get this back into github, so I'm posting it here. This updates the AdultDVDEmpire plugin to handle split scenes; if a file has a scene name embedded in it (like "movie movie - Scene 1 - actress" or "1 - scene name"), the plugin uses the parent directory's name as the movie title and attempts to parse the scene name to standardize it.

Known issues:
- Will probably be trouble if you have single files of scenes that aren't in a directory with the name of the movie
- Couldn't get each file's title to reflect Movie : Scene # : Description; apparently setting the title for one file in a directory sets it for all of them

# AdultDVDEmpire
import re

# URLS
ADE_BASEURL = 'http://www.adultdvdempire.com/'
ADE_SEARCH_MOVIES = ADE_BASEURL + 'SearchTitlesPage.aspx?SearchString=%s'
ADE_MOVIE_INFO = ADE_BASEURL + '%s/'

def Start():
  HTTP.CacheTime = CACHE_1DAY
  HTTP.SetHeader('User-agent', 'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.2; Trident/4.0; SLCC2; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0)')

class ADEAgent(Agent.Movies):
  name = 'Adult DVD Empire'
  languages = [Locale.Language.English]
  primary_provider = True

  def ParseScene(self,title,filename):
    Log("ParseScene: checking " + title)
    # Special case: if the filename has a number in brackets, that's the scene number, but Plex eats it
    # so we have to check the filename rather than title
    result = re.search(r"(\[\d{1,2}\])",filename)
    if result:
       sceneNumber = result.group(1)
       return "Scene " + sceneNumber
       
    # If it's an intro, credits, trailer, etc, use that name
    if re.search(r"(^intro$)",title,re.IGNORECASE):
      return "Intro"

    if re.search(r"(opening\scredits|open\scredits)",title,re.IGNORECASE):
      return "Opening credits"

    if re.search(r"(ending\scredits|end\scredits)",title,re.IGNORECASE):
      return "End credits"

    if re.search(r"^end$",title,re.IGNORECASE):
      return "End"

    if re.search(r"^credits$",title,re.IGNORECASE):
      return "Credits"
    
    result = re.match(r"(s?\d{1,2}\s|scene\s?\d{1,2}|sc?\s?\d{1,2}\s|\[\d{1,2}\])(.*)",title,re.IGNORECASE)
    if result:
      Log("Got scene match: " + result.group(1))
      scene = result.group(1)
      extra = result.group(2)
      # We have S01 or Scene1 or Sc1 or something, get the numbers
      result = re.search(r"(\d+)",scene)
      return "Scene " + scene.strip() + " : " + extra.strip()

    # If the title is in the form "Movie name - Scene XX", get scene name
    result = re.search(r"(.+)\s(scene\s?\d{1,2}|s\d{1,2}|sc\d{1,2})(?:\s|$|,)(.*)",title,re.IGNORECASE)
    if result:
        Log("Got embedded scene number" + result.group(2))
        scene = result.group(2)
        extra = result.group(3)
        result = re.search(r"(\d+)",scene)
        return "Scene " + scene.strip() + " : " + extra.strip()

    return False


  def search(self, results, media, lang):
    title = media.name
    Log("Processing " + title)
    if media.primary_metadata is not None:
      title = media.primary_metadata.title

    if title.startswith('The '):
      if title.count(':'):
        title = title.split(':',1)[0].replace('The ','',1) + ', The:' + title.split(':',1)[1]
      else:
        title = title.replace('The ','',1) + ', The'
    
    sceneName = self.ParseScene(title,media.items[0].parts[0].file.decode('utf-8'))
    if (sceneName):
        Log("Got scene name '" + sceneName + "', using parent dir name")
        file = media.items[0].parts[0].file.decode('utf-8')
        pathList = file.split("/")
        pathList.pop()
        parentPath = pathList.pop()
        Log("Parent path: " + parentPath)
        # We don't want to pass in the year if it's there
        title = re.sub(r'\(\d{4}\)','',parentPath)
        # Replace dots and underscores with spaces
        title = re.sub(r'[\s\.]',' ',title)
        Log("Got title " + title)
        
        # If our title includes metadata, remove it
        result = re.search(r"(.*?)(cheggit|bdrip|dvdrip|xxx\sdvdrip|720|split\s?scene|\(split|xvid)",title,re.IGNORECASE)
        if result:
          Log("Stripping scene info: " + result.group(2))
          title = result.group(1)
          Log("New title: " + title)

        # Special case: if our title has " 1" at the end, trim that
        result = re.search(r"(.*)\s1",title)
        if result:
          Log("Removing trailing '1'")
          title = result.group(1)
          Log("New title: " + title)

    query = String.URLEncode(String.StripDiacritics(title.replace('-','')))

    for movie in HTML.ElementFromURL(ADE_SEARCH_MOVIES % query).xpath('//div[contains(@class,"ListItem_ItemTitle")]/a'):
      curName = movie.text_content().strip()
      curID = movie.get('href').split('/',2)[1]
      score = 100 - Util.LevenshteinDistance(title.lower(), curName.lower())
      if score >= 85:
        if curName.count(', The'):
          curName = 'The ' + curName.replace(', The','',1)
        if curName.count(', A'):
          curName = 'A ' + curName.replace(', A','',1)
        if sceneName:
          # ISSUE: If you set one file's title to hold the scene name, all files in that dir are 
          # affected (you end up with 5 files called "movie : scene 5") 
          #curName = curName + " : " + sceneName
          Log("Setting full title to " + curName)

        results.Append(MetadataSearchResult(id = curID, name = curName, score = score, lang = lang))
    
    results.Sort('score', descending=True)

  def update(self, metadata, media, lang):
    html = HTML.ElementFromURL(ADE_MOVIE_INFO % metadata.id)
    Log("Current metadata title: " + metadata.title)
    Log("Current media title: " + media.title)
    metadata.title = media.title
    
    # Get Thumb and Poster
    try:
      img = html.xpath('//div[@id="ctl00_ContentPlaceHolder_ctl00_pnl_Default"]/a/img[contains(@src,"m.jpg")]')[0]
      thumbUrl = img.get('src')
      thumb = HTTP.Request(thumbUrl)
      posterUrl = img.get('src').replace('m.jpg','h.jpg')
      metadata.posters[posterUrl] = Proxy.Preview(thumb)
    except:
      pass

    # Get tagline
    try: metadata.tagline = html.xpath('//span[@class="Item_InfoTagLine"]')[0].text_content().strip()
    except: pass

    # Summary.
    try:
      metadata.summary = html.xpath('//div[@class="Item_InfoContainer"]')[0].text_content().replace('\t','').strip()
      if metadata.summary.find(metadata.tagline) != -1:
        metadata.summary = metadata.summary.replace(metadata.tagline, '').strip()
    except: pass

    # Other data.
    data = {}
    for div in html.xpath('//div[@class="Item_ProductInfoSectionConatiner"]/div'):
      name, value = div.text_content().split(':')
      data[name.strip()] = value.strip()

    if data.has_key('Rating'):
      metadata.content_rating = data['Rating']

    if data.has_key('Studio'):
      metadata.studio = data['Studio']

    if data.has_key('Release Date'):
      try:
        metadata.originally_available_at = Datetime.ParseDate(data['Release Date']).date()
        metadata.year = metadata.originally_available_at.year
      except: pass