详解crifan的Python库：crifanLib.py

2.4. 网络方面的函数

#------------------------------------------------------------------------------
# check file validation:
# open file url to check return info is match or not
# with exception support
# note: should handle while the file url is redirect
# eg :
# http://publish.it168.com/2007/0627/images/500754.jpg ->
# http://img.publish.it168.com/2007/0627/images/500754.jpg
# other special one:
# sina pic url: 
# http://s14.sinaimg.cn/middle/3d55a9b7g9522d474a84d&690
# http://s14.sinaimg.cn/orignal/3d55a9b7g9522d474a84d
# the real url is same with above url
def isFileValid(fileUrl) :
    fileIsValid = False;
    errReason = "Unknown error";

    try :
        #print "original fileUrl=",fileUrl;
        origFileName = fileUrl.split('/')[-1];
        #print "origFileName=",origFileName;
        
        #old: https://ie2zeq.bay.livefilestore.com/y1mo7UWr-TrmqbBhkw52I0ii__WE6l2UtMRSTZHSky66-uDxnCdKPr3bdqVrpUcQHcoJLedlFXa43bvCp_O0zEGF3JdG_yZ4wRT-c2AQmJ_TNcWvVZIXfBDgGerouWyx19WpA4I0XQR1syRJXjDNpwAbQ/IMG_5214_thumb[1].jpg
        #new: https://kxoqva.bay.livefilestore.com/y1mQlGjwNAYiHKoH5Aw6TMNhsCmX2YDR3vPKnP86snuqQEtnZgy3dHkwUvZ61Ah8zU3AGiS4whmm_ADrvxdufEAfMGo56KjLdhIbosn9F34olQ/IMG_5214_thumb%5b1%5d.jpg
        unquotedOrigFilenname = urllib.unquote(origFileName);
        #print "unquotedOrigFilenname=",unquotedOrigFilenname
        lowUnquotedOrigFilename = unquotedOrigFilenname.lower();
        #print "lowUnquotedOrigFilename=",lowUnquotedOrigFilename;
        
        resp = urllib2.urlopen(fileUrl, timeout=gConst['defaultTimeout']); # note: Python 2.6 has added timeout support.
        #print "resp=",resp;
        realUrl = resp.geturl();
        #print "realUrl=",realUrl;
        newFilename = realUrl.split('/')[-1];
        #print "newFilename=",newFilename;
        
        #http://blog.sina.com.cn/s/blog_696e50390100ntxs.html
        unquotedNewFilename = urllib.unquote(newFilename);
        #print "unquotedNewFilename=",unquotedNewFilename;
        unquotedLowNewFilename = unquotedNewFilename.lower();
        #print "unquotedLowNewFilename=",unquotedLowNewFilename;
        
        respInfo = resp.info();
        #print "respInfo=",respInfo;
        respCode = resp.getcode();
        #print "respCode=",respCode;

        # special:
        # http://116.img.pp.sohu.com/images/blog/2007/5/24/17/24/11355bf42a9.jpg
        # return no content-length
        #contentLen = respInfo['Content-Length'];
        
        # for redirect, if returned size>0 and filename is same, also should be considered valid
        #if (origFileName == newFilename) and (contentLen > 0):
        # for redirect, if returned response code is 200(OK) and filename is same, also should be considered valid
        #if (origFileName == newFilename) and (respCode == 200):
        if (lowUnquotedOrigFilename == unquotedLowNewFilename) and (respCode == 200):
            fileIsValid = True;
        else :
            fileIsValid = False;
            
            # eg: Content-Type= image/gif, ContentTypes : audio/mpeg
            # more ContentTypes can refer: http://kenya.bokee.com/3200033.html
            contentType = respInfo['Content-Type'];
        
            errReason = "file url returned info: type=%s, len=%d, realUrl=%s"%(contentType, contentLen, realUrl);
    except urllib2.URLError,reason :
        fileIsValid = False;
        errReason = reason;
    except urllib2.HTTPError,code :
        fileIsValid = False;
        errReason = code;
    except :
        fileIsValid = False;
        errReason = "Unknown error";

    # here type(errReason)= <class 'urllib2.HTTPError'>, so just convert it to str
    errReason = str(errReason);
    return (fileIsValid, errReason);
        

# indeed is pic, process it
(picIsValid, errReason) = isFileValid(curUrl);