##
##The first rule of VideoScriptClub is : You shouldn't talk about this script.
##I accept no liability if anyone uses this.
##
##Just never, ever use this as it
##certainly contravenes the terms and
##conditions of use your video hosting site
##
##If you do try to use it you need
##python2.4 from www.python.org
##
##Greg MC 1/1/2007

import urllib2,re,string
import sys,os


def getDefaultDir(default = "help?"):
    if os.environ.has_key('HOME'): #nix
        return os.environ['HOME']+'/Desktop/'
    elif os.environ.has_key('USERPROFILE'): #win 
        return os.environ['USERPROFILE']+'\\Bureau\\' #I live in France
    else:
        return default

def openCx(url, data=None):
    req = urllib2.Request(url)
    if not data is None:
            req.add_data(data)
    # Try to mimic Firefox, at least a little bit
    req.add_header('User-Agent', 'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1) Gecko/20061010 Firefox/2.0')
    req.add_header('Accept-Charset', 'ISO-8859-1,utf-8;q=0.7,*;q=0.7')
    req.add_header('Accept', 'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5')
    req.add_header('Accept-Language', 'en-us,en;q=0.5')
    return urllib2.urlopen(req)

class Scraper(object):
    '''wrapper for the download'''
    def __init__(self,url):
        self.baseUrl =  url
        self.fn = None
        self.url = None
        mm = re.search('[\.|/](?P<site>\w+)\.com',url)
        if not mm:
            print "Scraper: not a valid url", url
        else:
            self.site = mm.group('site')
            print url+"\nSITE =", self.site
            cx = openCx(url)
            try:
                cx = openCx(url)
                self.pageData = cx.read()
            except:
                print "Scraper: failed to read page",url
            try:
                self.__getattribute__("scrape_"+self.site+"_page")()
            except:
                self.url = None
                print self.site, "not implemented yet"
        print "scraper done"
        

    def getTitle(self,pageData):
        '''html page contains title in a meta ie.
        Google video, DailyMotion 
        '''
        p = re.compile('<title>(?P<title>.+?)</title>')
        m = p.search(pageData)
        if m.group('title'):
            title = m.group('title').strip()
            title = "".join(title.split('-')[:2])
        return title

    def stripBadChar(self,fn):
        p = re.compile(r'''\w|\s''')
        return  "".join(p.findall(fn))
            
    def scrape_ifilm_page(self):
        title  = re.search('<h2 id="itemtitle[_|\d]*">(?P<title>.*?)</h2>',
                           self.pageData)
        title = title.group('title')
        self.url = "http://download.ifilm.com/flv/%s_300.flv"%self.baseUrl.split('/')[-1]
        self.fn = self.stripBadChar(title)+"_IFILM"


    def scrape_myspace_page(self):  
        title  = re.search('videoid=\d+\&quot;\&gt;(?P<title>.+?)\&lt',
                           self.pageData)
        title = title.group('title')
        vid_id = re.search("\d+",self.baseUrl).group(0) #take the first number in URL
        tail = [x for x in vid_id[-4:]]
        tail.reverse()
        tail = "".join(tail)
        head = vid_id[:len(vid_id)-5]
        head = string.zfill(0,7-len(head))+head #pad to 7 digits
        self.url = "http://content.movies.myspace.com/%s/%s/%s/%s.flv"%(head,tail[:2],tail[2:4],vid_id)
        self.fn = self.stripBadChar(title)+"_myspace"
    
    def scrape_dailymotion_page(self):
        title = self.getTitle(self.pageData)
        fn = title[:title.find('- Dailymotion')] 
        p = '<param name="flashvars" value="url=(?P<vid_url>[^&]+)'
        m = re.search(p,self.pageData)
        self.url = urllib2.unquote(m.group('vid_url'))
        self.fn = self.stripBadChar(fn)

    def scrape_youtube_page(self):
        m = re.search('watch_fullscreen\\?(?P<vid_query>.*?)&fs=1&title=" \\+ "(?P<title>[^"]*)"',
                      self.pageData)
        self.fn = self.stripBadChar(m.group('title'))
        self.url = "http://youtube.com/get_video.php?%s" % m.group('vid_query')

    def scrape_google_page(self):
        p = re.compile(r'%2Fvideodownload([^&]+)')
        m = p.search(self.pageData)
        if m:
            url = "http%3A%2F%2Fvp.video.google.com"+m.group(0)
            self.url = urllib2.unquote(url)
            self.fn = self.stripBadChar(self.getTitle(self.pageData)) 
        else:
            self.url = None      


class Downloader(object):
    def __init__(self):
        self.reset()


    def reset(self):
        self.block_size = 1024**2 #ask for  a gig at a time
        self.size = self.totalSize = 0 #download stats
        self.url = self.fn = None 
        self.cx = self.fp = None    #filelikes
        self.busy = self.ok = False #boolean states
        freeMessage()
        if self.fp:
            self.fp.close() #housekeeping

    def setData(self,vidData):
        if not self.busy and vidData.url:
            self.url = vidData.url
            self.cx = None
            self.fn = "%s%s.flv"%(vidDir.get(),vidData.fn)
            self.busy = False
            self.ok = True
            busyMessage()
            STDOUT.clear()
        else:
            self.ok = False

    def getChunk(self):
        #this is a callback for a progress bar
        #returns fraction left
        if self.ok and not self.cx:
            self.openCx()        
        if self.cx and not self.busy and self.size >0:
            try:
                self.fp.write(self.cx.read(self.block_size))
                self.size -= self.block_size
                if self.size<0:
                    STDOUT.write("DONE")
                    freeMessage()
            except:
                STDOUT.write('FAILED: download interrupted')
                self.reset()
                return 1.0
            return 1.0*self.size/self.totalSize
        return 1.0

    def openCx(self):
        STDOUT.write("opening cx")
        try:
            self.cx = openCx(self.url)
        except:
            STDOUT.write("FAILED: can't open cx")
            self.cx = None
            return
        
        self.totalSize = self.size = int(self.cx.info()['Content-Length'])
        STDOUT.write("video : %s \nsaving to %s.flv"%(self.url,self.fn) )
        STDOUT.write("SIZE = %dK"%(self.totalSize/1000))
        self.fp = open(self.fn,"wb") #this shouldn't be here
        

##gui stuff
##Tkinter sucks should use swt
import Tkinter

class textPad(object):
    width = 400
    height = 100
    curLine = 0 #line count

    def __init__(self,parent):
        self.canvas = Tkinter.Canvas(parent,width=self.width,height=self.height, bg="white")
        self.canvas.pack()
        #self.canvas.pack(side=Tkinter.BOTTOM, fill=Tkinter.BOTH,expand=1)

    def write(self,txt):
        self.canvas.create_text(10,10+20*self.curLine,text=txt,fill='red',
                                anchor=Tkinter.NW, tag ="TXT")
        self.curLine += (1 + len(re.findall("\n",txt)) )#take a new line

    def clear(self):
        self.canvas.delete("TXT")
        
class progressBar(object):
    width = 300
    height = 40
    callback = None

    def __init__(self,parent,callback = None):
        box = Tkinter.Frame(parent,height=self.height,width=90,bd=1, relief=Tkinter.SUNKEN)
        box.pack(fill=Tkinter.X, padx=1, pady=1)
        #populate frame
        downLoad2 = Tkinter.Label(box, text="DL progress")
        downLoad2.pack(side = Tkinter.LEFT)   
        self.bar = Tkinter.Canvas(box,width = self.width, height = self.height)
        self.bar.pack(side = Tkinter.RIGHT)
        self.bar.create_rectangle(0,self.height,self.width,0,
                                  fill = "blue")
        self.callback = callback
        #behavior
        self.bar.after(500,self.update)

    def update(self):      
        fraction = 1.0-self.callback()
        self.bar.create_rectangle(0,self.height,self.width,0,
                                  fill = "blue")
        self.bar.create_rectangle(0,self.height,int(self.width*fraction),0,
                                  fill = "red")
        self.bar.after(500, self.update) #this is the right order

def doVid():
    '''callback talks to Scraper and Download'''
    url = sourceURL.get()#take copy
    vidData = Scraper(url)
    download.setData(vidData)

def busyMessage():
    sourceURL.set("I'm busy.....")#overwrite

def freeMessage():
    sourceURL.set('done, your url?')#overwrite
              
def main():
    global root,  sourceURL, vidDir, download, STDOUT
    #create root first else ERROR
    root = Tkinter.Tk()
    root.title('Greg\'s GenericVideoDownloader: paste URL below')
    sourceURL = Tkinter.StringVar()#this should belong to download
    download = Downloader()
    
    vidDir = Tkinter.StringVar()
    vidDir.set(getDefaultDir("paste download path here?")) #default download
    box1 = Tkinter.Frame(root,height=18,width=90,bd=1, relief=Tkinter.SUNKEN)
    box1.pack(fill=Tkinter.X, padx=1, pady=1)
    urlFrom = Tkinter.Label(box1, text="Source URL")
    urlFrom.pack(side = Tkinter.LEFT)
    Tkinter.Button(root, text="OK",
                   command=doVid ).pack()

    progBar = progressBar(root,callback = download.getChunk)
    
    Tkinter.Entry(box1, width=60,textvariable= sourceURL).pack(side=Tkinter.RIGHT)
    box2 = Tkinter.Frame(root,height=18,width=90,bd=1, relief=Tkinter.SUNKEN)
    box2.pack(fill=Tkinter.X, padx=1, pady=1)
    downLoad2 = Tkinter.Label(box2, text="Download directory")
    downLoad2.pack(side = Tkinter.LEFT)    
    Tkinter.Entry(box2, width=50,
                  textvariable= vidDir).pack(side=Tkinter.RIGHT)
    

    STDOUT = textPad(root)#text window
  
    root.mainloop()
    
main()





##
##You shouldn't use this.
##I accept no liability if anyone uses this.
##
##Just never, ever use this as it
##probably contravenes the terms and
##conditions of use of video hosting sites
##
##If you do try to use it you need
##python2.4 from www.python.org
##
##Greg MC 1/1/2007

import urllib2,re
import sys,os


def stripBadChar(fn):
    p = re.compile(r'''\w|\s''')
    return  "".join(p.findall(fn))

def url2file(cx,fp):
    '''downloads to a file
    this has a text progress bar'''
    headers = cx.info()
    size = int(headers['Content-Length'])
    block_size= 1024**2 #ask for  a gig at a time
    print ".-"*int(size/block_size)
    for i in range(int(size/block_size)+1):
        fp.write (cx.read(block_size))
        print '*',
        sys.stdout.flush()
    print "\n done"
    
def openCx(url, data=None):
    req = urllib2.Request(url)
    if not data is None:
            req.add_data(data)
    # Try to mimic Firefox, at least a little bit
    req.add_header('User-Agent', 'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1) Gecko/20061010 Firefox/2.0')
    req.add_header('Accept-Charset', 'ISO-8859-1,utf-8;q=0.7,*;q=0.7')
    req.add_header('Accept', 'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5')
    req.add_header('Accept-Language', 'en-us,en;q=0.5')
    return urllib2.urlopen(req)
  
def parseIFILMPage(url,pageData):
    title  = re.search('<h2 id="itemtitle[_|\d]*">(?P<title>.*?)</h2>',pageData)
    title = title.group('title')
    url = "http://download.ifilm.com/flv/%s_300.flv"%url.split('/')[-1]
    return url, stripBadChar(fn)+"_IFILM"

def parseDailyMotionPage(url,pageData):
    title = re.search('<title>(?P<title>.*?)</title>',pageData)
    title = title.group('title')
    fn = title[:title.find('- Dailymotion')] 
    #print urllib2.unquote(soup.findAll('meta')[3]['content'])
    #ss = soup.findAll('param')[1]#this [1] is prety lame
    p = '<param name="flashvars" value="url=(?P<vid_url>[^&]+)'
    m = re.search(p,pageData)
    url = urllib2.unquote(m.group('vid_url'))
    return url, stripBadChar(fn)

def parseYoutubePage(url,data):
    m = re.search('watch_fullscreen\\?(?P<vid_query>.*?)&fs=1&title=" \\+ "(?P<title>[^"]*)"', data)   
    fn = m.group('title')
    url = "http://youtube.com/get_video.php?%s" % m.group('vid_query')
    return url, stripBadChar(fn)


def getVid(url,fn,vidDir):
    print "video :",fn
    print "saving to %s.FLV"%fn
    try:
        cx = openCx(url)
    except:
        print 'FAILED'
        return None
    fp = open("%s%s.FLV"%(vidDir,fn),"wb")
    try:
        url2file(cx,fp)
    except:
        print 'FAILED',
        fp.close()
        return None
    fp.close()
    return 1
    

def doVid():
    url = sourceURL.get()#take copy
    if url ==_prompt1:
        return 
    sourceURL.set("I'm busy.....")#overwrite
    site = url.split('.')[1].lower()
    print site
    cx = openCx(url)
    pageData = cx.read()
    if site == "ifilm":
        url,fn = parseIFILMPage(url,pageData)
    elif site =="dailymotion":
        url,fn = parseDailyMotionPage(url,pageData)
    elif url.lower().find("youtube")>-1:
        url,fn = parseYoutubePage(url,pageData)
    print url

    
    if getVid(url,fn,vidDir.get()):
        sourceURL.set('...OK next vid')
    else:
        sourceURL.set('...FAILED another vid?')
         
import Tkinter
_prompt1 = 'your url?'

def main():
    global root,  sourceURL, vidDir
    #create root first else ERROR
    root = Tkinter.Tk()
    root.title('Greg\'s GenericVideoDownloader: paste URL below')
    sourceURL = Tkinter.StringVar()
    sourceURL.set(_prompt1)
    vidDir = Tkinter.StringVar()
    vidDir.set(os.environ['HOME']+'/Desktop/') #default download
    Tkinter.Entry(root, width=70,
                  textvariable= sourceURL).pack()
    box = Tkinter.Frame(root,height=25,width=90)
    box.pack(fill=Tkinter.X, padx=5, pady=5)

    downLoad2 = Tkinter.Label(box, text="Dowload directory")
    downLoad2.pack(side = Tkinter.LEFT)
    
    Tkinter.Entry(root, width=70,
                  textvariable= vidDir).pack()
    Tkinter.Button(root, text="OK",
                   command=doVid ).pack()
    root.mainloop()
    
main()