CMS 3D CMS Logo

/data/refman/pasoursint/CMSSW_4_1_8_patch13/src/Calibration/Tools/scripts/alcaDQMUpload.py

Go to the documentation of this file.
00001 #!/usr/bin/env python
00002 
00003 import sys, os, os.path, re, string, httplib, mimetypes, urllib, urllib2, httplib, gzip, md5
00004 from cStringIO import StringIO
00005 from stat import *
00006 import optparse
00007 
00008 try:
00009         import hashlib
00010 except:
00011         pass
00012 
00013 HTTPS = httplib.HTTPS
00014 if sys.version_info[:3] >= (2, 4, 0):
00015   HTTPS = httplib.HTTPSConnection
00016 
00017 ssl_key_file = None
00018 ssl_cert_file = None
00019 
00020 class HTTPSCertAuth(HTTPS):
00021   def __init__(self, host, *args, **kwargs):
00022     HTTPS.__init__(self, host, key_file = ssl_key_file, cert_file = ssl_cert_file, **kwargs)
00023 
00024 class HTTPSCertAuthenticate(urllib2.AbstractHTTPHandler):
00025   def default_open(self, req):
00026     return self.do_open(HTTPSCertAuth, req)
00027 
00028 def filetype(filename):
00029   return mimetypes.guess_type(filename)[0] or 'application/octet-stream'
00030 
00031 def encode(args, files):
00032   """
00033     Encode form (name, value) and (name, filename, type) elements into
00034     multi-part/form-data. We don't actually need to know what we are
00035     uploading here, so just claim it's all text/plain.
00036   """
00037   boundary = '----------=_DQM_FILE_BOUNDARY_=-----------'
00038   (body, crlf) = ('', '\r\n')
00039   for (key, value) in args.items():
00040     body += '--' + boundary + crlf
00041     body += ('Content-disposition: form-data; name="%s"' % key) + crlf
00042     body += crlf + str(value) + crlf
00043   for (key, filename) in files.items():
00044     body += '--' + boundary + crlf
00045     body += ('Content-Disposition: form-data; name="%s"; filename="%s"'
00046              % (key, os.path.basename(filename))) + crlf
00047     body += ('Content-Type: %s' % filetype(filename)) + crlf
00048     body += crlf + open(filename, "r").read() + crlf
00049   body += '--' + boundary + '--' + crlf + crlf
00050   return ('multipart/form-data; boundary=' + boundary, body)
00051 
00052 def marshall(args, files, request):
00053   """
00054     Marshalls the arguments to the CGI script as multi-part/form-data,
00055     not the default application/x-www-form-url-encoded.  This improves
00056     the transfer of the large inputs and eases command line invocation
00057     of the CGI script.
00058   """
00059   (type, body) = encode(args, files)
00060   request.add_header('Content-type', type)
00061   request.add_header('Content-length', str(len(body)))
00062   request.add_data(body)
00063 
00064 def upload(url, args, files):
00065   ident = "visDQMUpload DQMGUI/%s CMSSW/%s python/%s" % \
00066     (os.getenv('DQMGUI_VERSION', '?'),
00067      os.getenv('DQM_CMSSW_VERSION', os.getenv('CMSSW_VERSION', '?')),
00068      "%d.%d.%d" % sys.version_info[:3])
00069   cookie = None
00070   if url.startswith("https:"):
00071     authreq = urllib2.Request(url + '/authenticate')
00072     authreq.add_header('User-agent', ident)
00073     result = urllib2.build_opener(HTTPSCertAuthenticate()).open(authreq)
00074     cookie = result.headers.get('Set-Cookie')
00075     if not cookie:
00076       raise RuntimeError("Did not receive authentication cookie")
00077     cookie = cookie.split(";")[0]
00078 
00079   datareq = urllib2.Request(url + '/data/put')
00080   datareq.add_header('Accept-encoding', 'gzip')
00081   datareq.add_header('User-agent', ident)
00082   if cookie:
00083     datareq.add_header('Cookie', cookie)
00084   marshall(args, files, datareq)
00085   result = urllib2.build_opener().open(datareq)
00086   data = result.read()
00087   if result.headers.get ('Content-encoding', '') == 'gzip':
00088     data = gzip.GzipFile (fileobj=StringIO(data)).read ()
00089   return (result.headers, data)
00090 
00091 def print_help(*args):
00092     sys.stdout.write("\n")
00093     sys.stdout.write("This scripts intends to do the upload of files to the DQM server. It\n")
00094     sys.stdout.write("runs some basic checks on the file name as it is crucial to follow\n")
00095     sys.stdout.write("the naming convention.\n")
00096     sys.stdout.write("\n")
00097     sys.stdout.write("Mandatory Option\n")
00098     sys.stdout.write("    -d, --destination parameter to specify the DQM server\n")
00099     sys.stdout.write("\n")
00100     sys.stdout.write("  Proxy Options\n")
00101     sys.stdout.write("  The script will try to find your grid proxy automatically. To\n")
00102     sys.stdout.write("  do so, it checks $X509_* environment variables and your globus\n")
00103     sys.stdout.write("  directory (~/.globus). To override this automatism, you can use\n")
00104     sys.stdout.write("  the following two options:\n")
00105     sys.stdout.write("    --ssl-key-file          location of your private key\n")
00106     sys.stdout.write("    --ssl-cert-file         location of your public key\n")
00107     sys.stdout.write("\n")
00108     sys.stdout.write("Other Options\n")
00109     sys.stdout.write("    -h, --help                show this help message\n")
00110     sys.stdout.write("    -s, --no-submission       suppress the submission\n")
00111     sys.stdout.write("    -r, --no-registration     suppress the submission\n")
00112     sys.stdout.write("    --no-filename-check       omit the file name check\n")
00113     sys.stdout.write("\n")
00114     sys.exit(0)
00115 
00116 def checkSSL(opts):
00117     global ssl_key_file
00118     global ssl_cert_file
00119     
00120     if opts.ssl_key_file and os.path.exists(opts.ssl_key_file):
00121         ssl_key_file = opts.ssl_key_file
00122     if opts.ssl_cert_file and os.path.exists(opts.ssl_cert_file):
00123         ssl_cert_file = opts.ssl_cert_file
00124 
00125     if not ssl_key_file:
00126         x509_path = os.getenv("X509_USER_PROXY", None)
00127         if x509_path and os.path.exists(x509_path):
00128             ssl_key_file = ssl_cert_file = x509_path
00129 
00130     if not ssl_key_file:
00131         x509_path = os.getenv("X509_USER_KEY", None)
00132         if x509_path and os.path.exists(x509_path):
00133             ssl_key_file = x509_path
00134 
00135     if not ssl_cert_file:
00136         x509_path = os.getenv("X509_USER_CERT", None)
00137         if x509_path and os.path.exists(x509_path):
00138             ssl_cert_file = x509_path
00139     
00140     if not ssl_key_file:
00141         x509_path = os.getenv("HOME") + "/.globus/userkey.pem"
00142         if os.path.exists(x509_path):
00143             ssl_key_file = x509_path
00144 
00145     if not ssl_cert_file:
00146         x509_path = os.getenv("HOME") + "/.globus/usercert.pem"
00147         if os.path.exists(x509_path):
00148             ssl_cert_file = x509_path
00149     
00150     if not ssl_key_file or not os.path.exists(ssl_key_file):
00151         sys.stderr.write("no certificate private key file found, please specify one via $X509_USER_PROXY, $X509_USER_KEY or --ssl-key-file\n")
00152         sys.exit(2)
00153                   
00154     if not ssl_cert_file or not os.path.exists(ssl_cert_file):
00155         sys.stderr.write("no certificate public key file found, please specify one via $X509_USER_CERT or --ssl-cert-file\n")
00156         sys.exit(3)
00157 
00158     print "Using SSL private key", ssl_key_file
00159     print "Using SSL public key", ssl_cert_file
00160 
00161                                                                                   
00162 def checkFileName(fileName):
00163     regWhitespace = re.compile('.*\s.*')
00164     if regWhitespace.match(fileName):
00165         sys.stderr.write("whitespace detected!\n")
00166         return False
00167             
00168     regRelval=re.compile('.*relval.*')
00169     regCMSSW=re.compile('.*CMSSW_[0-9]+_[0-9]+_[0-9]+_.*')
00170     regCMSSWpre=re.compile('.*CMSSW_[0-9]+_[0-9]+_[0-9]+_pre[0-9]+_.*')
00171     if regRelval.match(fileName):
00172         # TODO check for pre-versions
00173         if not regCMSSW.match(fileName):
00174             print "no CMSSW"
00175     return True
00176     # DQM stuff
00177 
00178 def startUpload(url, filename):
00179     global ssl_key_file
00180     global ssl_cert_file
00181 
00182     print url, filename
00183     try:
00184         (headers, data) = \
00185                   upload(url,
00186                      { 'size': os.stat(filename).st_size,
00187                        'checksum': "md5:%s" % md5.new(file(filename).read()).hexdigest() },
00188                      { 'file': filename })
00189         print 'Status code: ', headers.get("Dqm-Status-Code", "None")
00190         print 'Message:     ', headers.get("Dqm-Status-Message", "None")
00191         print 'Detail:      ', headers.get("Dqm-Status-Detail", "None")
00192         print data
00193     except urllib2.HTTPError, e:
00194         print "ERROR", e
00195         print 'Status code: ', e.hdrs.get("Dqm-Status-Code", "None")
00196         print 'Message:     ', e.hdrs.get("Dqm-Status-Message", "None")
00197         print 'Detail:      ', e.hdrs.get("Dqm-Status-Detail", "None")
00198         sys.exit(1)
00199 
00200 def getURL(filename, destination):
00201         filename = filename.split("/")[-1]
00202         regMC=re.compile('.*_R([0-9]*)__*')
00203         if regMC.match(filename):
00204                 m = re.search('.*_R([0-9]*)(__.*).root',filename)
00205                 runNr = m.group(1)
00206                 dataset = m.group(2).replace("__","/")
00207         else:
00208                 m = re.search('.*_R([0-9]*)(_?.*).root',filename)
00209                 runNr = m.group(1)
00210                 dataset = m.group(2).replace("__","/")
00211                 if dataset=="":
00212                         dataset="/Global/Online/ALL"
00213         if not runNr:
00214                 runNr="1"
00215         if (int(runNr)==1):
00216                 return destination+"start?workspace=summary;dataset="+dataset+";sampletype=offline_data"
00217         else:
00218                 return destination+"start?workspace=summary;runnr="+runNr+";dataset="+dataset+";sampletype=online_data"
00219         
00220 def registerFileAtLogServer(filename, destination, tags):
00221         filename = filename.split("/")[-1]
00222         regMC=re.compile('.*_R([0-9]*)__*')
00223         if regMC.match(filename):
00224                 m = re.search('.*_R([0-9]*)(__.*).root',filename)
00225                 runNr = m.group(1)
00226                 dataset = m.group(2).replace("__","/")
00227         else:
00228                 m = re.search('.*_R([0-9]*)(_?.*).root',filename)
00229                 runNr = m.group(1)
00230                 dataset = m.group(2).replace("__","/")
00231                 if dataset=="":
00232                         dataset="/Global/Online/ALL"
00233         tempurl = "https://www-ekp.physik.uni-karlsruhe.de/~zeise/cgi-bin/register.py?run="+runNr+"&dataset="+dataset+"&filename="+filename+"&tags="+tags+"&instance="+destination
00234         print "Link that is used to register: ", tempurl
00235         urllib.urlopen(tempurl)
00236 
00237 def main(args):
00238     global opts
00239     parser = optparse.OptionParser(add_help_option=False)
00240     parser.add_option("-h", "--help",          action="callback", callback=print_help),
00241     parser.add_option("",   "--no-filename-check", dest="no_filename_check",  default=False, action="store_true")
00242     parser.add_option("-d", "--destination", dest="destination", default="",  action="store")
00243     parser.add_option("-t", "--tags", dest="tags", default="",  action="store")
00244     parser.add_option("-s", "--no-submission", dest="submission", default=True,  action="store_false")
00245     parser.add_option("-r", "--no-registration", dest="registration", default=True,  action="store_false")
00246     parser.add_option("","--ssl-key-file", dest="ssl_key_file", default="", action="store")
00247     parser.add_option("","--ssl-cert-file", dest="ssl_cert_file", default="", action="store")
00248     (opts, args) = parser.parse_args()
00249     opts.abort = False
00250 
00251     if not opts.destination:
00252         sys.stderr.write("no destination specified\n")
00253         sys.exit(4)
00254     checkSSL(opts)
00255 
00256     if len(args)==0:
00257         sys.stderr.write("no input files specified\n")
00258         sys.exit(1)
00259     for fileName in args:
00260         fileName=fileName.strip()
00261         if not os.path.exists(fileName):
00262             sys.stderr.write("file '%s' doesn't exist!\n" % fileName)
00263             continue
00264         if not opts.no_filename_check and not checkFileName(fileName):
00265             continue
00266         sys.stderr.write("file '%s' passed name check, upload will follow!\n" % fileName)
00267         if opts.submission:
00268           startUpload(opts.destination, fileName)
00269         else:
00270           sys.stdout.write("file '%s' would be uploaded to '%s'\n" % (fileName, opts.destination))
00271         if opts.registration:
00272           registerFileAtLogServer(fileName, opts.destination, opts.tags)       
00273         print "You should see the plots here: "+getURL(fileName, opts.destination)
00274 
00275 if __name__ == '__main__':
00276     sys.exit(main(sys.argv[1:]))
00277     
00278