| 1 |
#!/usr/bin/env python
|
| 2 |
import sys, os, string, re
|
| 3 |
import urllib, urllister
|
| 4 |
import urllib2
|
| 5 |
|
| 6 |
# ####################################
|
| 7 |
# Exception with use of refDB
|
| 8 |
class RefDBError:
|
| 9 |
def __init__(self, owner, dataset):
|
| 10 |
print '\nERROR accessing RefDB for Owner/Dataset: '+owner+'/'+dataset+'\n'
|
| 11 |
pass
|
| 12 |
|
| 13 |
# ####################################
|
| 14 |
class PubDBError:
|
| 15 |
def __init__(self, url):
|
| 16 |
print '\nERROR accessing PubDB at '+url+'\n'
|
| 17 |
pass
|
| 18 |
|
| 19 |
# ####################################
|
| 20 |
class NoPHPError:
|
| 21 |
def __init__(self, url):
|
| 22 |
#print '\nERROR accessing PHP at '+url+' \n'
|
| 23 |
print 'ERROR accessing PHP: ',url,'isn\'t updated version \n'
|
| 24 |
pass
|
| 25 |
|
| 26 |
# ####################################
|
| 27 |
class pubDBResult:
|
| 28 |
def __init__(self,
|
| 29 |
contents):
|
| 30 |
self.contents=contents
|
| 31 |
|
| 32 |
|
| 33 |
def dump(self):
|
| 34 |
print 'Contents : ',self.contents
|
| 35 |
|
| 36 |
# ####################################
|
| 37 |
# class to access PubDBs
|
| 38 |
class PubDB:
|
| 39 |
def __init__(self, owner, dataset, dataTiers):
|
| 40 |
self.owner = owner
|
| 41 |
self.dataset = dataset
|
| 42 |
self.dataTiers = dataTiers
|
| 43 |
|
| 44 |
self.RefDBurl_ = 'http://cmsdoc.cern.ch/cms/production/www/'
|
| 45 |
self.RefDBphp_ = 'PubDB/GetIdCollection.php'
|
| 46 |
self.RefDBMotherphp_ = 'cgi/SQL/CollectionTree.php'
|
| 47 |
|
| 48 |
self.PubDBCentralUrl_ = 'http://cmsdoc.cern.ch/cms/production/www/PubDB/'
|
| 49 |
self.PubDBCentralPhp_ = 'GetPublishedCollectionInfoFromRefDB.php'
|
| 50 |
|
| 51 |
self.PubDBAnalysisPhp_ = 'get-pubdb-analysisinfo.php'
|
| 52 |
self.PubDBAnalysisPhpOld_ = 'get-pubdb-analysisinfo.php'
|
| 53 |
|
| 54 |
try:
|
| 55 |
self.collid=self.findAllCollId()
|
| 56 |
except PubDBError:
|
| 57 |
raise RefDBError(self.owner, self.dataset)
|
| 58 |
|
| 59 |
########################################################################
|
| 60 |
def findAllCollId(self):
|
| 61 |
collId=self.findCollId()
|
| 62 |
|
| 63 |
|
| 64 |
NeededCollID = []
|
| 65 |
NeededCollID.append(collId)
|
| 66 |
#dataTypeReq = ['Digi' ]#, 'Digi', 'Hit', 'PU']
|
| 67 |
|
| 68 |
if len(self.dataTiers)>0:
|
| 69 |
dataTypeReq = self.dataTiers
|
| 70 |
CollInfos=self.findMotherCollId(collId)
|
| 71 |
while (CollInfos[1][2]!='PU'):
|
| 72 |
|
| 73 |
for TypeReq in dataTypeReq:
|
| 74 |
for CollInfo in CollInfos[1:]:
|
| 75 |
if TypeReq==CollInfo[2]:
|
| 76 |
NeededCollID.append(CollInfo[0])
|
| 77 |
break
|
| 78 |
pass
|
| 79 |
CollInfos=self.findMotherCollId(CollInfo[0])
|
| 80 |
### no more parents
|
| 81 |
if len(CollInfos)==1:
|
| 82 |
break
|
| 83 |
|
| 84 |
|
| 85 |
print NeededCollID
|
| 86 |
return NeededCollID
|
| 87 |
|
| 88 |
|
| 89 |
|
| 90 |
########################################################################
|
| 91 |
def findCollId(self):
|
| 92 |
"""
|
| 93 |
Contact RefDB and get CollId given Dataset and Owner
|
| 94 |
"""
|
| 95 |
|
| 96 |
#anche questa info viene dal cfg. E' PubDB centrale
|
| 97 |
url = self.RefDBurl_+self.RefDBphp_+'?Owner=' + self.owner + '&Dataset=' + self.dataset
|
| 98 |
|
| 99 |
try:
|
| 100 |
f = urllib.urlopen(url)
|
| 101 |
except IOError:
|
| 102 |
# print 'Cannot access URL: '+url
|
| 103 |
raise PubDBError(url)
|
| 104 |
|
| 105 |
line = f.read()
|
| 106 |
try:
|
| 107 |
collid = string.split(line,": ")
|
| 108 |
#part = string.strip(part[1])
|
| 109 |
collid = string.split(collid[1],"<")
|
| 110 |
collid = string.strip(collid[0])
|
| 111 |
except IndexError:
|
| 112 |
raise PubDBError(url)
|
| 113 |
|
| 114 |
print 'CollectionId: '+collid+' \n'
|
| 115 |
return collid
|
| 116 |
|
| 117 |
########################################################################
|
| 118 |
def findMotherCollId(self, collid):
|
| 119 |
"""
|
| 120 |
Contact RefDB and get CollId of mother of current Dataset Owner (eg. Digi if DST, Hit if Digi)
|
| 121 |
"""
|
| 122 |
|
| 123 |
url = self.RefDBurl_+self.RefDBMotherphp_+'?cid=' + collid
|
| 124 |
|
| 125 |
try:
|
| 126 |
f = urllib.urlopen(url)
|
| 127 |
except IOError:
|
| 128 |
# print 'Cannot access URL: '+url
|
| 129 |
raise PubDBError(url)
|
| 130 |
|
| 131 |
reEmptyLine = re.compile( r'^$' )
|
| 132 |
|
| 133 |
collInfos = []
|
| 134 |
for line in f.readlines():
|
| 135 |
#print '#',line,'#'
|
| 136 |
line = string.strip(line)
|
| 137 |
if reEmptyLine.match(line):
|
| 138 |
pass
|
| 139 |
else:
|
| 140 |
#print '#',line,'#'
|
| 141 |
keys = string.split(line,',')
|
| 142 |
#print '#',keys,'#'
|
| 143 |
collInfo = []
|
| 144 |
for key in keys:
|
| 145 |
collInfo.append(string.split(key, '=')[1])
|
| 146 |
collInfos.append(collInfo)
|
| 147 |
|
| 148 |
return collInfos
|
| 149 |
|
| 150 |
########################################################################
|
| 151 |
def getPubDBInfo(self, url):
|
| 152 |
"""
|
| 153 |
Contact a local PubDB to collect all the relevant information
|
| 154 |
"""
|
| 155 |
|
| 156 |
result = []
|
| 157 |
end=string.rfind(url,'/')
|
| 158 |
lastEq=string.rfind(url,'=')
|
| 159 |
urlphp=url[:end+1]+self.PubDBAnalysisPhp_+'?CollID='+url[lastEq+1:]
|
| 160 |
# print 'PHP URL: '+urlphp+' \n'
|
| 161 |
|
| 162 |
reOld=re.compile( r'V24' )
|
| 163 |
#print urlphp,'Old PubDB ',reOld.search(urlphp)
|
| 164 |
if reOld.search(urlphp):
|
| 165 |
raise NoPHPError(urlphp)
|
| 166 |
# try:
|
| 167 |
# urldev=string.replace(urlphp,'V24','V3_1')
|
| 168 |
# print "urldev URL: ",urldev
|
| 169 |
# f = urllib2.urlopen(urldev)
|
| 170 |
# except urllib2.HTTPError, msg:
|
| 171 |
# raise NoPHPError(urldev)
|
| 172 |
else:
|
| 173 |
try:
|
| 174 |
f = urllib2.urlopen(urlphp)
|
| 175 |
except urllib2.HTTPError, msg:
|
| 176 |
raise NoPHPError(urlphp)
|
| 177 |
|
| 178 |
content = f.read()
|
| 179 |
return pubDBResult(content)
|
| 180 |
|
| 181 |
########################################################################
|
| 182 |
def findPubDBsUrls(self):
|
| 183 |
"""
|
| 184 |
Find the URL of the PubDB of all the sites which publish the collid
|
| 185 |
"""
|
| 186 |
|
| 187 |
completeResult = []
|
| 188 |
|
| 189 |
### first collId is the primary one, (Dataset/Owner asker by user)
|
| 190 |
### The other CollIDs are the parents
|
| 191 |
|
| 192 |
### Get all the pubDb's URL containig the primary collection _and_ the requested parents
|
| 193 |
primaryCollId=self.collid[0]
|
| 194 |
|
| 195 |
primaryUrl = self.PubDBCentralUrl_+self.PubDBCentralPhp_+'?CollID=' + primaryCollId
|
| 196 |
#print "primaryUrl=", primaryUrl
|
| 197 |
|
| 198 |
try:
|
| 199 |
sock = urllib.urlopen(primaryUrl)
|
| 200 |
except IOError:
|
| 201 |
raise PubDBError(primaryUrl)
|
| 202 |
|
| 203 |
parser = urllister.URLLister()
|
| 204 |
parser.feed(sock.read())
|
| 205 |
sock.close()
|
| 206 |
parser.close()
|
| 207 |
|
| 208 |
# this are all the href links found in the page
|
| 209 |
for url in parser.linksList:
|
| 210 |
# return only those which contains "CollID"
|
| 211 |
result=[]
|
| 212 |
if string.find(url, primaryCollId) != -1 :
|
| 213 |
#print 'URL ',url
|
| 214 |
result.append(url)
|
| 215 |
|
| 216 |
try:
|
| 217 |
for tmp in self.checkPubDBs(url):
|
| 218 |
result.append(tmp)
|
| 219 |
except PubDBError:
|
| 220 |
continue
|
| 221 |
|
| 222 |
completeResult.append(result)
|
| 223 |
|
| 224 |
return completeResult
|
| 225 |
|
| 226 |
#########################################################
|
| 227 |
def checkPubDBs(self, url):
|
| 228 |
"""
|
| 229 |
Check if the given PubDB contains also the CollId's collections
|
| 230 |
"""
|
| 231 |
|
| 232 |
result = []
|
| 233 |
primaryCollId=self.collid[0]
|
| 234 |
|
| 235 |
reNotCollId=re.compile( r'no such collection' )
|
| 236 |
reDBError=re.compile( r'DB Error: connect failed' )
|
| 237 |
for collid in self.collid[1:]:
|
| 238 |
newurl = string.replace(url,primaryCollId,collid)
|
| 239 |
try:
|
| 240 |
sock = urllib.urlopen(newurl)
|
| 241 |
for line in sock.readlines():
|
| 242 |
line = string.strip(line)
|
| 243 |
if reNotCollId.search(line):
|
| 244 |
raise PubDBError(newurl)
|
| 245 |
if reDBError.search(line):
|
| 246 |
raise PubDBError(newurl)
|
| 247 |
except IOError:
|
| 248 |
raise PubDBError(newurl)
|
| 249 |
sock.close()
|
| 250 |
result.append(newurl)
|
| 251 |
|
| 252 |
return result
|
| 253 |
|
| 254 |
#########################################################
|
| 255 |
def getAllPubDBsInfo(self):
|
| 256 |
"""
|
| 257 |
Prepare the file to send in InputSandbox, with the info retrieved by local PubDBs
|
| 258 |
"""
|
| 259 |
|
| 260 |
# get all the URLs of PubDBs which publish CollId
|
| 261 |
pubDBUrlsSet = self.findPubDBsUrls()
|
| 262 |
# print 'pubDBUrls ',pubDBUrlsSet
|
| 263 |
|
| 264 |
# get the contents of each PubDBs
|
| 265 |
completeResult = []
|
| 266 |
for pubDBUrls in pubDBUrlsSet:
|
| 267 |
result = []
|
| 268 |
for url in pubDBUrls:
|
| 269 |
# print 'URL ',url
|
| 270 |
try:
|
| 271 |
result.append(self.getPubDBInfo(url))
|
| 272 |
except NoPHPError:
|
| 273 |
continue
|
| 274 |
# for r in result:
|
| 275 |
# r.dump()
|
| 276 |
if len(result)>0: completeResult.append(result)
|
| 277 |
|
| 278 |
# print 'getAllPubDBsInfo ',completeResult
|
| 279 |
# for result in completeResult:
|
| 280 |
# print '.....'
|
| 281 |
# for r in result:
|
| 282 |
# r.dump()
|
| 283 |
# print '.....'
|
| 284 |
return completeResult
|
