Line Code
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275
#!/usr/bin/env python3

"""Usage: dqm-ls [-s SERVER] [-n CONNECTIONS]

Parse ROOT file contents listings on a DQM GUI server.

In order to authenticate to the target server, standard grid certificate
environment must be available. Typically this would be X509_CERT_DIR and
either X509_USER_PROXY or X509_USER_CERT and X509_USER_KEY environment
variables. If these variables are not set, the following defaults are
checked for existence. Note that if the script falls back on using a
key rather than a proxy, it will prompt for the key password.
- $X509_CERT_DIR: /etc/grid-security/certificates
- $X509_USER_KEY: $HOME/.globus/userkey.pem
- $X509_USER_CERT: $HOME/.globus/usercert.pem
"""

from DQMServices.Components.HTTP import RequestManager
from DQMServices.Components.X509 import SSLOptions
import os, sys, re, pycurl, urllib
from optparse import OptionParser
from time import time, strptime, strftime, gmtime
from calendar import timegm
from datetime import datetime
from urllib.parse import urlparse, quote
from tempfile import mkstemp
from traceback import print_exc
from functools import cmp_to_key

# Object types.
DIR = 0
FILE = 1

# HTTP protocol `User-agent` identification string.
ident = "DQMLS/1.0 python/%s.%s.%s" % sys.version_info[:3]

# SSL/X509 options.
ssl_opts = None

# HTTP request manager for content requests.
reqman = None

# Number of HTTP requests made for content.
nfetched = 0

# Found objects.
found = []

def logme(msg, *args):
  """Generate agent log message."""
  procid = "[%s/%d]" % (__file__.rsplit("/", 1)[-1], os.getpid())
  print(datetime.now(), procid, msg % args)

def myumask():
  """Get the current process umask."""
  val = os.umask(0)
  os.umask(val)
  return val

def handle_init(c):
  """Prepare custom properties on download handles."""
  c.temp_file = None
  c.temp_path = None
  c.local_path = None

def request_init(c, options, path, kind):
  """`RequestManager` callback to initialise directory contents request."""

  c.setopt(pycurl.URL, options.server + quote(path) + ((path != "/" and "/") or ""))
  assert c.temp_file == None
  assert c.temp_path == None
  assert c.local_path == None

  # If this is file, prepare temporary destination file in the target
  # directory for possible download. parse_dir_and_files() will finish this off.
  if kind == FILE and options.download:
    try:
      (fd, tmp) = mkstemp()
      fp = os.fdopen(fd, 'wb')
      c.setopt(pycurl.WRITEFUNCTION, fp.write)
      c.temp_file = fp
      c.temp_path = tmp
      c.local_path = path.strip('/')
      c.buffer = None
    except Exception as e:
      logme("ERROR: %s: %s", path, str(e))
      print_exc()

def cleanup(c):
  """Clean up file copy operation, usually after any failures."""
  if c.temp_file:
    try: c.temp_file.close()
    except: pass
  if c.temp_path:
    try: os.remove(c.temp_path)
    except: pass
  if c.local_path:
    try: os.remove(c.local_path)
    except: pass
  c.temp_file = None
  c.temp_path = None
  c.local_path = None
  c.buffer = None

def report_error(c, task, errmsg, errno):
  """`RequestManager` callback to report directory contents request errors."""
  print("FAILED to retrieve %s: %s (%d)" % (task, errmsg, errno), file=sys.stderr)
  global nfetched; nfetched += 1

def parse_dir_and_files(c):
  """`RequestManager` callback to handle directory content response.

This gets called once per every directory which has been successfully
retrieved from the server. It parses the HTML response and turns it
into object listing with all the file meta information.

If verbosity has been requested, also shows simple progress bar on the
search progress, one dot for every ten directories retrieved."""
  options, path, kind = c.task
  root_url = urlparse(options.server).path.rstrip('/')

  if kind == FILE and options.download:
    assert c.local_path, "Expected local path property to be set"
    assert c.temp_file, "Exepected temporary file property to be set"
    try:
      c.setopt(pycurl.WRITEFUNCTION, lambda *args: None)
      c.temp_file.close()
      c.temp_file = None

      os.chmod(c.temp_path, 0o0666 & ~UMASK)
      os.system('mv %s %s'  % (c.temp_path, c.local_path))

      c.local_path = None
      logme("INFO: downloaded %s", path)
    except Exception as e:
      logme("ERROR: downloading %s into %s failed: %s",
            path, c.local_path, str(e))
      print_exc()
    finally:
      cleanup(c)

  elif kind == DIR:
    items = re.findall(r"<tr><td><a href='(.*?)'>(.*?)</a></td><td>(\d+|&nbsp;|-)</td>"
                       r"<td>(&nbsp;|\d\d\d\d-\d\d-\d\d \d\d:\d\d:\d\d UTC)</td>",
                       c.buffer.getvalue().decode('utf-8'))

    for path, name, size, date in items:
      assert path.startswith(root_url)
      path = path[len(root_url):]

      if date == "&nbsp;":
        date = -1
      else:
        date = timegm(strptime(date, "%Y-%m-%d %H:%M:%S %Z"))

      if size == "&nbsp;" or size == "-":
        size = -1
      else:
        size = int(size)

      if path.endswith("/"):
        assert size == -1
        path = path[:-1]
        obj = {'task': c.task,
               'kind': DIR,
               'name': name,
               'size': size,
               'date': date,
               'path': path}
        found.append(obj)
        if options.recursive:
          reqman.put((options, path, DIR))
      else:
        assert size >= 0
        obj = {'task': c.task,
               'kind': FILE,
               'name': name,
               'size': size,
               'date': date,
               'path': path}
        if options.match_file:
          if name == options.match_file:
            found.append(obj)
            if options.download:
              reqman.put((options, path, FILE))
        else:
          found.append(obj)
          if options.download:
            reqman.put((options, path, FILE))

      global nfetched
      nfetched += 1
      if options.verbose and nfetched % 10 == 0:
        sys.stdout.write(".")
        sys.stdout.flush()
        if nfetched % 750 == 0:
          print()

def objcmp(a, b):
  diff = 0
  diff = a['date'] - b['date']
  if diff:
    return diff
  diff = a['size'] - b['size']
  if diff:
    return diff
  return diff

# Parse command line options.
op = OptionParser(usage = __doc__)
op.add_option("-s", "--server", dest = "server",
              type = "string", action = "store", metavar = "SERVER",
              default = "https://cmsweb.cern.ch/dqm/offline/data/browse",
              help = "Pull content from SERVER")
op.add_option("-n", "--connections", dest = "connections",
              type = "int", action = "store", metavar = "NUM",
              default = 10, help = "Use NUM concurrent connections")
op.add_option("-v", "--verbose", dest = "verbose",
              action = "store_true", default = False,
              help = "Show verbose scan information")
op.add_option("-r", "--recursive", dest = "recursive",
              action = "store_true", default = False,
              help = "Perform a recursive scan starting from the root directory.")
op.add_option("-d", "--download", dest = "download",
              action = "store_true", default = False,
              help = "Download all touched ROOT files locally. To be used with extreme care.")
op.add_option("-m", "--match_file", dest = "match_file",
              type= "string", action = "store", default = "",
              help = "Filter results based on exact file name matching.")
options, args = op.parse_args()
if args:
  print("Too many arguments", file=sys.stderr)
  sys.exit(1)
if not options.server:
  print("Server contact string required", sys.stderr)
  sys.exit(1)

UMASK = myumask()

# In case a user specifies a root file as address, remove it from the
# address and return results that match only that name.
gr = re.match("(.*)/(DQM_V.*\.root)$", options.server)
if gr:
  options.server = gr.group(1)
  options.match_file = gr.group(2)

# Get SSL X509 parametres.
ssl_opts = SSLOptions()
if options.verbose:
  print("Using SSL cert dir", ssl_opts.ca_path)
  print("Using SSL private key", ssl_opts.key_file)
  print("Using SSL public key", ssl_opts.cert_file)

# Start a request manager for contents.
reqman = RequestManager(num_connections = options.connections,
                        ssl_opts = ssl_opts,
                        user_agent = ident,
                        request_init = request_init,
                        request_respond = parse_dir_and_files,
                        request_error = report_error,
                        handle_init = handle_init)

# Process from root directory.
start = time()
reqman.put((options, "/", DIR))
reqman.process()
end = time()

for x in sorted(found, key=cmp_to_key(objcmp)):
  print("%20s\t%s\t%s" % (x['size'],
                          strftime("%Y-%m-%d %H:%M:%S %Z", gmtime(x['date'])),
                          ((options.recursive and x['path']) or x['name'])))

if options.verbose:
  print("\nFound %d directories, %d objects in %.3f seconds" % (nfetched, len(found), end - start))