Annotation of ZSQLExtend/importFMPXML.py, revision 1.26

1.1       casties     1: #!/usr/local/bin/python
                      2: #
                      3: 
                      4: import string
                      5: import logging
                      6: import sys
1.7       dwinter     7: import types
1.8       casties     8: import time
1.1       casties     9: 
1.5       casties    10: from xml import sax
                     11: from amara import saxtools
                     12: 
1.2       casties    13: try:
                     14:     import psycopg2 as psycopg
1.23      casties    15:     import psycopg2.extensions
                     16:     # switch to unicode
                     17:     psycopg2.extensions.register_type(psycopg2.extensions.UNICODE)
1.2       casties    18:     psyco = 2
                     19: except:
                     20:     import psycopg
                     21:     psyco = 1
                     22: 
1.5       casties    23: fm_ns = 'http://www.filemaker.com/fmpxmlresult'
1.1       casties    24: 
1.26    ! casties    25: version_string = "V0.6.1 ROC 2.7.2008"
1.20      casties    26: 
1.21      casties    27: def unicodify(text, withNone=False):
1.20      casties    28:     """decode str (utf-8 or latin-1 representation) into unicode object"""
1.21      casties    29:     if withNone and text is None:
1.20      casties    30:         return None
1.21      casties    31:     if not text:
1.20      casties    32:         return u""
1.21      casties    33:     if isinstance(text, str):
1.20      casties    34:         try:
1.21      casties    35:             return text.decode('utf-8')
1.20      casties    36:         except:
1.21      casties    37:             return text.decode('latin-1')
1.20      casties    38:     else:
1.21      casties    39:         return text
1.20      casties    40: 
1.21      casties    41: def utf8ify(text, withNone=False):
1.20      casties    42:     """encode unicode object or string into byte string in utf-8 representation"""
1.21      casties    43:     if withNone and text is None:
1.20      casties    44:         return None
1.21      casties    45:     if not text:
1.20      casties    46:         return ""
1.22      casties    47:     if isinstance(text, unicode):
                     48:         return text.encode('utf-8')
                     49:     else:
1.21      casties    50:         return text
1.8       casties    51: 
1.1       casties    52: def getTextFromNode(nodename):
                     53:     """get the cdata content of a node"""
                     54:     if nodename is None:
                     55:         return ""
                     56:     nodelist=nodename.childNodes
                     57:     rc = ""
                     58:     for node in nodelist:
                     59:         if node.nodeType == node.TEXT_NODE:
                     60:            rc = rc + node.data
                     61:     return rc
                     62: 
                     63: def sql_quote(v):
                     64:     # quote dictionary
                     65:     quote_dict = {"\'": "''", "\\": "\\\\"}
                     66:     for dkey in quote_dict.keys():
                     67:         if string.find(v, dkey) >= 0:
                     68:             v=string.join(string.split(v,dkey),quote_dict[dkey])
1.5       casties    69:     return "'%s'"%v
1.1       casties    70: 
1.8       casties    71: def SimpleSearch(curs,query, args=None, ascii=False):
1.1       casties    72:     """execute sql query and return data"""
1.15      casties    73:     #logger.debug("executing: "+query)
1.8       casties    74:     if ascii:
                     75:         # encode all in UTF-8
1.20      casties    76:         query = utf8ify(query)
1.8       casties    77:         if args is not None:
                     78:             encargs = []
                     79:             for a in args:
1.20      casties    80:                 encargs.append(utf8ify(a, withNone=True))
1.8       casties    81:             
                     82:             args = encargs
1.7       dwinter    83: 
1.1       casties    84:     curs.execute(query, args)
1.15      casties    85:     #logger.debug("sql done")
1.4       casties    86:     try:
                     87:         return curs.fetchall()
                     88:     except:
                     89:         return None
1.1       casties    90: 
                     91: 
1.8       casties    92: class TableColumn:
                     93:     """simple type for storing sql column name and type"""
                     94:     
                     95:     def __init__(self, name, type=None):
                     96:         #print "new tablecolumn(%s,%s)"%(name, type)
                     97:         self.name = name
                     98:         self.type = type
                     99:         
                    100:     def getName(self):
                    101:         return self.name
                    102:     
                    103:     def getType(self):
                    104:         if self.type is not None:
                    105:             return self.type
                    106:         else:
                    107:             return "text"
1.5       casties   108: 
1.8       casties   109:     def __str__(self):
                    110:         return self.name
                    111:     
                    112:     
1.5       casties   113: class xml_handler:
1.8       casties   114:     def __init__(self,options):
1.9       dwinter   115:         """SAX handler to import FileMaker XML file (FMPXMLRESULT format) into the table.
1.8       casties   116:         @param options: dict of options
                    117:         @param options.dsn: database connection string
                    118:         @param options.table: name of the table the xml shall be imported into
                    119:         @param options.filename: xmlfile filename
                    120:         @param options.update_fields: (optional) list of fields to update; default is to create all fields
                    121:         @param options.id_field: (optional) field which uniquely identifies an entry for updating purposes.
                    122:         @param options.sync_mode: (optional) really synchronise, i.e. delete entries not in XML file
                    123:         @param options.lc_names: (optional) lower case and clean up field names from XML
                    124:         @param options.keep_fields: (optional) don't add fields to SQL database
                    125:         @param options.ascii_db: (optional) assume ascii encoding in db
                    126:         @param options.replace_table: (optional) delete and re-insert data
1.14      casties   127:         @param options.backup_table: (optional) create backup of old table (breaks indices)
1.15      casties   128:         @param options.use_logger_instance: (optional) use this instance of a logger
1.9       dwinter   129:         """
                    130:         
1.15      casties   131:         # set up logger
                    132:         if hasattr(options, 'use_logger_instance'):
                    133:             self.logger = options.use_logger_instance
                    134:         else:
                    135:             self.logger = logging.getLogger('db.import.fmpxml')
                    136: 
                    137:         
1.5       casties   138:         # set up parser
                    139:         self.event = None
                    140:         self.top_dispatcher = { 
                    141:             (saxtools.START_ELEMENT, fm_ns, u'METADATA'): 
                    142:             self.handle_meta_fields,
                    143:             (saxtools.START_ELEMENT, fm_ns, u'RESULTSET'): 
1.8       casties   144:             self.handle_data_fields,
1.5       casties   145:             }
                    146:         
                    147:         # connect database
1.8       casties   148:         self.dbCon = psycopg.connect(options.dsn)
1.25      casties   149:         logging.debug("DB encoding: %s"%getattr(self.dbCon, 'encoding', 'UNKNOWN'))
1.5       casties   150:         self.db = self.dbCon.cursor()
1.8       casties   151:         assert self.db, "AIIEE no db cursor for %s!!"%options.dsn
1.1       casties   152:     
1.9       dwinter   153:         self.table = getattr(options,"table",None)
                    154:         self.update_fields = getattr(options,"update_fields",None)
                    155:         self.id_field = getattr(options,"id_field",None)
                    156:         self.sync_mode = getattr(options,"sync_mode",None)
                    157:         self.lc_names = getattr(options,"lc_names",None)
                    158:         self.keep_fields = getattr(options,"keep_fields",None)
                    159:         self.ascii_db = getattr(options,"ascii_db",None)
                    160:         self.replace_table = getattr(options,"replace_table",None)
                    161:         self.backup_table = getattr(options,"backup_table",None)
1.23      casties   162:         self.read_before_update = getattr(options,"read_before_update",None)
1.8       casties   163: 
1.15      casties   164:         self.logger.debug("dsn: "+repr(getattr(options,"dsn",None)))
                    165:         self.logger.debug("table: "+repr(self.table))
                    166:         self.logger.debug("update_fields: "+repr(self.update_fields))
                    167:         self.logger.debug("id_field: "+repr(self.id_field))
                    168:         self.logger.debug("sync_mode: "+repr(self.sync_mode))
                    169:         self.logger.debug("lc_names: "+repr(self.lc_names))
                    170:         self.logger.debug("keep_fields: "+repr(self.keep_fields))
                    171:         self.logger.debug("ascii_db: "+repr(self.ascii_db))
                    172:         self.logger.debug("replace_table: "+repr(self.replace_table))
                    173:         self.logger.debug("backup_table: "+repr(self.backup_table))
1.23      casties   174:         self.logger.debug("read_before_update: "+repr(self.read_before_update))
1.5       casties   175:         
                    176:         self.dbIDs = {}
                    177:         self.rowcnt = 0
1.8       casties   178:         
                    179:         if self.id_field is not None:
1.5       casties   180:             # prepare a list of ids for sync mode
1.8       casties   181:             qstr="select %s from %s"%(self.id_field,self.table)
1.5       casties   182:             for id in SimpleSearch(self.db, qstr):
                    183:                 # value 0: not updated
                    184:                 self.dbIDs[id[0]] = 0;
                    185:                 self.rowcnt += 1
                    186:                 
1.15      casties   187:             self.logger.info("%d entries in DB to sync"%self.rowcnt)
1.5       casties   188:         
1.8       casties   189:         # names of fields in XML file
                    190:         self.xml_field_names = []
                    191:         # map XML field names to SQL field names
                    192:         self.xml_field_map = {}
                    193:         # and vice versa
                    194:         self.sql_field_map = {}
1.5       casties   195:         
                    196:         return
                    197: 
                    198:     def handle_meta_fields(self, end_condition):
                    199:         dispatcher = {
                    200:             (saxtools.START_ELEMENT, fm_ns, u'FIELD'):
                    201:             self.handle_meta_field,
                    202:             }
                    203:         #First round through the generator corresponds to the
                    204:         #start element event
1.15      casties   205:         self.logger.info("reading metadata...")
                    206:         self.logger.debug("START METADATA")
1.5       casties   207:         yield None
1.1       casties   208:     
1.5       casties   209:         #delegate is a generator that handles all the events "within"
                    210:         #this element
                    211:         delegate = None
                    212:         while not self.event == end_condition:
                    213:             delegate = saxtools.tenorsax.event_loop_body(
                    214:                 dispatcher, delegate, self.event)
                    215:             yield None
                    216:         
                    217:         #Element closed. Wrap up
1.15      casties   218:         self.logger.debug("END METADATA")
1.8       casties   219:         
                    220:         # rename table for backup
                    221:         if self.backup_table:
                    222:             self.orig_table = self.table
1.16      casties   223:             self.tmp_table = self.table + "_tmp"
                    224:             backup_name = "%s_%s"%(self.table,time.strftime('%Y_%m_%d_%H_%M_%S'))
                    225:             
1.8       casties   226:             # remove old temp table
1.16      casties   227:             qstr = "DROP TABLE %s"%(self.tmp_table)
1.8       casties   228:             try:
                    229:                 self.db.execute(qstr)
                    230:             except:
                    231:                 pass
                    232:             
                    233:             self.dbCon.commit()
                    234:            
                    235:             if self.id_field:
1.16      casties   236:                 # sync mode -- copy backup table, update current table 
                    237:                 self.logger.info("copy table %s to %s"%(self.table,backup_name))
                    238:                 qstr = "CREATE TABLE %s AS (SELECT * FROM %s)"%(backup_name,self.table)
1.8       casties   239: 
                    240:             else:
1.16      casties   241:                 # replace mode -- create empty tmp table, insert into tmp table
                    242:                 self.table = self.tmp_table
1.15      casties   243:                 self.logger.info("create empty table %s"%(self.table))
1.8       casties   244:                 qstr = "CREATE TABLE %s AS (SELECT * FROM %s WHERE 1=0)"%(self.table,self.orig_table)
                    245:             
                    246:             self.db.execute(qstr)
                    247:             self.dbCon.commit()
                    248:         
                    249:         # delete data from table for replace
                    250:         if self.replace_table:
1.15      casties   251:             self.logger.info("delete data from table %s"%(self.table))
1.8       casties   252:             qstr = "TRUNCATE TABLE %s"%(self.table)
                    253:             self.db.execute(qstr)
                    254:             self.dbCon.commit()
                    255:            
                    256:         # try to match date style with XML
                    257:         self.db.execute("set datestyle to 'german'")
                    258:         
1.15      casties   259:         #self.logger.debug("xml-fieldnames:"+repr(self.xml_field_names))
1.8       casties   260:         # get list of fields and types of db table
                    261:         qstr="select attname, format_type(pg_attribute.atttypid, pg_attribute.atttypmod) from pg_attribute, pg_class where attrelid = pg_class.oid and pg_attribute.attnum > 0 and relname = '%s'"
                    262:         self.sql_fields={}
                    263:         for f in SimpleSearch(self.db, qstr%self.table):
1.24      casties   264:             fn = f[0]
                    265:             ft = f[1]
1.8       casties   266:             #print "SQL fields: %s (%s)"%(n,t)
1.24      casties   267:             self.sql_fields[fn] = TableColumn(fn,ft)
1.8       casties   268:         
1.17      casties   269:         # translate id_field (SQL-name) to XML-name
                    270:         self.xml_id = self.sql_field_map.get(self.id_field, None)
                    271:         # get type of id_field
                    272:         if self.id_field:
                    273:             self.id_type = self.sql_fields[self.id_field].getType()
                    274:         else:
                    275:             self.id_type = None
                    276:         
1.8       casties   277:         # check fields to update
1.5       casties   278:         if self.update_fields is None:
1.8       casties   279:             if self.keep_fields:
1.13      casties   280:                 # update all existing fields from sql (when they are in the xml file)
                    281:                 self.update_fields = {}
                    282:                 for f in self.sql_fields.keys():
                    283:                     if self.sql_field_map.has_key(f):
                    284:                         xf = self.sql_field_map[f]
                    285:                         self.update_fields[f] = self.xml_field_map[xf]
                    286: 
1.8       casties   287:             else:
                    288:                 # update all fields
                    289:                 if self.lc_names:
                    290:                     # create dict with sql names
                    291:                     self.update_fields = {}
                    292:                     for f in self.xml_field_map.values():
                    293:                         self.update_fields[f.getName()] = f
                    294:                         
                    295:                 else:
                    296:                     self.update_fields = self.xml_field_map
1.17      casties   297:                                 
1.8       casties   298:         # and translate to list of xml fields
                    299:         if self.lc_names:
                    300:             self.xml_update_list = [self.sql_field_map[x] for x in self.update_fields]
                    301:         else:
                    302:             self.xml_update_list = self.update_fields.keys()
1.17      casties   303: 
1.8       casties   304:         if not self.keep_fields:
                    305:             # adjust db table to fields in XML and update_fields
                    306:             for f in self.xml_field_map.values():
1.15      casties   307:                 self.logger.debug("sync-fieldname: %s"%f.getName())
1.8       casties   308:                 sf = self.sql_fields.get(f.getName(), None)
                    309:                 uf = self.update_fields.get(f.getName(), None)
                    310:                 if sf is not None:
                    311:                     # name in db -- check type
                    312:                     if f.getType() != sf.getType():
1.15      casties   313:                         self.logger.debug("field %s has different type (%s vs %s)"%(f,f.getType(),sf.getType()))
1.8       casties   314:                 elif uf is not None:
                    315:                     # add field to table
1.24      casties   316:                     fn = uf.getName()
                    317:                     ft = uf.getType()
                    318:                     qstr="alter table %s add \"%s\" %s"%(self.table,fn,ft)
1.15      casties   319:                     self.logger.info("db add field:"+qstr)
1.8       casties   320:                     
                    321:                     if self.ascii_db and type(qstr)==types.UnicodeType:
                    322:                         qstr=qstr.encode('utf-8')
                    323:                         
                    324:                     self.db.execute(qstr)
                    325:                     self.dbCon.commit()
1.24      casties   326:                     # add field to field list
                    327:                     self.sql_fields[fn] = TableColumn(fn, ft)
1.7       dwinter   328:                 
1.17      casties   329:         # prepare sql statements for update (do not update id_field)
1.24      casties   330:         setStr=string.join(["\"%s\" = %%s"%self.xml_field_map[f] for f in self.xml_update_list if f != self.xml_id], ', ')
                    331:         self.updQuery="UPDATE %s SET %s WHERE \"%s\" = %%s"%(self.table,setStr,self.id_field)
1.23      casties   332:         # and select (for update check)
                    333:         selStr=string.join([self.xml_field_map[f].getName() for f in self.xml_update_list if f != self.xml_id], ', ')
1.24      casties   334:         self.selQuery="SELECT %s FROM %s WHERE \"%s\" = %%s"%(selStr,self.table,self.id_field)
1.5       casties   335:         # and insert
1.24      casties   336:         fields=string.join(["\"%s\""%self.xml_field_map[x].getName() for x in self.xml_update_list], ',')
1.8       casties   337:         values=string.join(['%s' for f in self.xml_update_list], ',')
1.5       casties   338:         self.addQuery="INSERT INTO %s (%s) VALUES (%s)"%(self.table,fields,values)
1.15      casties   339:         self.logger.debug("update-query: "+self.updQuery)
1.23      casties   340:         self.logger.debug("sel-query: "+self.selQuery)
1.15      casties   341:         self.logger.debug("add-query: "+self.addQuery)
1.5       casties   342:         return
                    343: 
                    344:     def handle_meta_field(self, end_condition):
                    345:         name = self.params.get((None, u'NAME'))
                    346:         yield None
                    347:         #Element closed.  Wrap up
1.8       casties   348:         if self.lc_names:
                    349:             # clean name
                    350:             sqlname = name.replace(" ","_").lower() 
                    351:         else:
                    352:             sqlname = name
                    353:         self.xml_field_names.append(name)
                    354:         # map to sql name and default text type
                    355:         self.xml_field_map[name] = TableColumn(sqlname, 'text')
                    356:         self.sql_field_map[sqlname] = name
1.15      casties   357:         self.logger.debug("FIELD name: "+name)
1.5       casties   358:         return
                    359: 
1.8       casties   360:     def handle_data_fields(self, end_condition):
1.5       casties   361:         dispatcher = {
                    362:             (saxtools.START_ELEMENT, fm_ns, u'ROW'):
                    363:             self.handle_row,
                    364:             }
                    365:         #First round through the generator corresponds to the
                    366:         #start element event
1.15      casties   367:         self.logger.info("reading data...")
                    368:         self.logger.debug("START RESULTSET")
1.5       casties   369:         self.rowcnt = 0
                    370:         yield None
1.1       casties   371:     
1.5       casties   372:         #delegate is a generator that handles all the events "within"
                    373:         #this element
                    374:         delegate = None
                    375:         while not self.event == end_condition:
                    376:             delegate = saxtools.tenorsax.event_loop_body(
                    377:                 dispatcher, delegate, self.event)
                    378:             yield None
                    379:         
                    380:         #Element closed.  Wrap up
1.15      casties   381:         self.logger.debug("END RESULTSET")
1.5       casties   382:         self.dbCon.commit()
1.1       casties   383:         
1.5       casties   384:         if self.sync_mode:
                    385:             # delete unmatched entries in db
1.15      casties   386:             self.logger.info("deleting unmatched rows from db")
1.24      casties   387:             delQuery = "DELETE FROM %s WHERE \"%s\" = %%s"%(self.table,self.id_field)
1.5       casties   388:             for id in self.dbIDs.keys():
                    389:                 # find all not-updated fields
                    390:                 if self.dbIDs[id] == 0:
1.15      casties   391:                     self.logger.info(" delete:"+id)
1.8       casties   392:                     SimpleSearch(self.db, delQuery, [id], ascii=self.ascii_db)
1.1       casties   393:                     
1.5       casties   394:                 elif self.dbIDs[id] > 1:
1.15      casties   395:                     self.logger.info(" sync: ID %s used more than once?"%id)
1.8       casties   396:             
                    397:             self.dbCon.commit()
1.1       casties   398:             
1.8       casties   399:         # reinstate backup tables
1.16      casties   400:         if self.backup_table and not self.id_field:
1.8       casties   401:             backup_name = "%s_%s"%(self.orig_table,time.strftime('%Y_%m_%d_%H_%M_%S'))
1.15      casties   402:             self.logger.info("rename backup table %s to %s"%(self.orig_table,backup_name))
1.8       casties   403:             qstr = "ALTER TABLE %s RENAME TO %s"%(self.orig_table,backup_name)
                    404:             self.db.execute(qstr)
1.15      casties   405:             self.logger.info("rename working table %s to %s"%(self.table,self.orig_table))
1.8       casties   406:             qstr = "ALTER TABLE %s RENAME TO %s"%(self.table,self.orig_table)
                    407:             self.db.execute(qstr)
1.5       casties   408:             self.dbCon.commit()
1.1       casties   409:         
1.5       casties   410:         return
                    411: 
                    412:     def handle_row(self, end_condition):
                    413:         dispatcher = {
                    414:             (saxtools.START_ELEMENT, fm_ns, u'COL'):
                    415:             self.handle_col,
                    416:             }
1.15      casties   417:         self.logger.debug("START ROW")
1.8       casties   418:         self.xml_data = {}
1.5       casties   419:         self.colIdx = 0
                    420:         yield None
1.1       casties   421:     
1.5       casties   422:         #delegate is a generator that handles all the events "within"
                    423:         #this element
                    424:         delegate = None
                    425:         while not self.event == end_condition:
                    426:             delegate = saxtools.tenorsax.event_loop_body(
                    427:                 dispatcher, delegate, self.event)
                    428:             yield None
                    429:         
                    430:         #Element closed.  Wrap up
1.15      casties   431:         self.logger.debug("END ROW")
1.5       casties   432:         self.rowcnt += 1
                    433:         # process collected row data
                    434:         update=False
                    435:         id_val=''
                    436:         # synchronize by id_field
                    437:         if self.id_field:
1.17      casties   438:             if self.id_type == 'integer':
                    439:                 id_val = int(self.xml_data[self.xml_id])
                    440:             else:
                    441:                 id_val = self.xml_data[self.xml_id]
                    442:                 
1.5       casties   443:             if id_val in self.dbIDs:
                    444:                 self.dbIDs[id_val] += 1
                    445:                 update=True
1.8       casties   446: 
                    447:         # collect all values
                    448:         args = []
                    449:         for fn in self.xml_update_list:
1.17      casties   450:             # do not update id_field
1.19      casties   451:             if update and fn == self.xml_id:
1.17      casties   452:                 continue
                    453:             
1.8       casties   454:             f = self.xml_field_map[fn]
                    455:             val = self.xml_data[fn]
                    456:             type = self.sql_fields[f.getName()].getType()
                    457:             if type == "date" and len(val) == 0: 
                    458:                 # empty date field
                    459:                 val = None
                    460:                 
                    461:             elif type == "integer" and len(val) == 0: 
                    462:                 # empty int field
                    463:                 val = None
                    464:                 
                    465:             args.append(val)
                    466:                     
1.5       casties   467:         if update:
                    468:             # update existing row (by id_field)
1.23      casties   469:             if self.read_before_update:
                    470:                 # read data
                    471:                 self.logger.debug("update check: %s = %s"%(id_val, args))
                    472:                 oldrow = SimpleSearch(self.db, self.selQuery, [id_val], ascii=self.ascii_db)
                    473:                 #i = 0
                    474:                 #for v in oldrow[0]:
                    475:                 #    logging.debug("v: %s = %s (%s)"%(v,args[i],v==args[i]))
                    476:                 #    i += 1
                    477:                 if tuple(oldrow[0]) != tuple(args):
                    478:                     # data has changed -- update
                    479:                     self.logger.debug("really update: %s = %s"%(id_val, args))
                    480:                     args.append(id_val) # last arg is id
                    481:                     SimpleSearch(self.db, self.updQuery, args, ascii=self.ascii_db)
                    482:                     
                    483:             else:
                    484:                 # always update
                    485:                 self.logger.debug("update: %s = %s"%(id_val, args))
                    486:                 args.append(id_val) # last arg is id
                    487:                 SimpleSearch(self.db, self.updQuery, args, ascii=self.ascii_db)
1.8       casties   488: 
1.5       casties   489:         else:
                    490:             # create new row
1.15      casties   491:             self.logger.debug("insert: %s"%args)
1.8       casties   492:             SimpleSearch(self.db, self.addQuery, args, ascii=self.ascii_db)
1.5       casties   493: 
1.15      casties   494:         #self.logger.info(" row:"+"%d (%s)"%(self.rowcnt,id_val))
                    495:         if (self.rowcnt % 100) == 0:
                    496:             self.logger.info(" row:"+"%d (id:%s)"%(self.rowcnt,id_val))
1.5       casties   497:             self.dbCon.commit()
                    498:             
                    499:         return
                    500: 
                    501:     def handle_col(self, end_condition):
                    502:         dispatcher = {
                    503:             (saxtools.START_ELEMENT, fm_ns, u'DATA'):
                    504:             self.handle_data_tag,
                    505:             }
                    506:         #print "START COL"
                    507:         yield None
                    508:         #delegate is a generator that handles all the events "within"
                    509:         #this element
                    510:         delegate = None
                    511:         while not self.event == end_condition:
                    512:             delegate = saxtools.tenorsax.event_loop_body(
                    513:                 dispatcher, delegate, self.event)
                    514:             yield None
                    515:         #Element closed.  Wrap up
                    516:         #print "END COL"
                    517:         self.colIdx += 1
                    518:         return
                    519: 
                    520:     def handle_data_tag(self, end_condition):
                    521:         #print "START DATA"
                    522:         content = u''
                    523:         yield None
                    524:         # gather child elements
                    525:         while not self.event == end_condition:
                    526:             if self.event[0] == saxtools.CHARACTER_DATA:
                    527:                 content += self.params
                    528:             yield None
                    529:         #Element closed.  Wrap up
1.8       casties   530:         fn = self.xml_field_names[self.colIdx]
                    531:         self.xml_data[fn] = content
1.5       casties   532:         return
                    533: 
                    534: 
1.11      dwinter   535: def importFMPXML(options):
1.14      casties   536:     """import FileMaker XML file (FMPXMLRESULT format) into the table.     
1.11      dwinter   537:         @param options: dict of options
                    538:         @param options.dsn: database connection string
                    539:         @param options.table: name of the table the xml shall be imported into
                    540:         @param options.filename: xmlfile filename
                    541:         @param options.update_fields: (optional) list of fields to update; default is to create all fields
                    542:         @param options.id_field: (optional) field which uniquely identifies an entry for updating purposes.
                    543:         @param options.sync_mode: (optional) really synchronise, i.e. delete entries not in XML file
                    544:         @param options.lc_names: (optional) lower case and clean up field names from XML
                    545:         @param options.keep_fields: (optional) don't add fields to SQL database
                    546:         @param options.ascii_db: (optional) assume ascii encoding in db
                    547:         @param options.replace_table: (optional) delete and re-insert data
1.16      casties   548:         @param options.backup_table: (optional) create backup of old table
1.11      dwinter   549:         """
1.15      casties   550:         
1.11      dwinter   551:     if getattr(options,'update_fields',None):
                    552:         uf = {}
                    553:         for f in options.update_fields.split(','):
1.12      casties   554:             if f.find(':') > 0:
                    555:                 (n,t) = f.split(':')
                    556:             else:
                    557:                 n = f
                    558:                 t = None
1.11      dwinter   559:             uf[n] = TableColumn(n,t)
                    560:             
                    561:         options.update_fields = uf
                    562:     
                    563:     if getattr(options,'id_field',None) and getattr(options,'replace_table',None):
                    564:         logging.error("ABORT: sorry, you can't do both sync (id_field) and replace")
1.15      casties   565:         return
1.11      dwinter   566:         
                    567:     parser = sax.make_parser()
                    568:     #The "consumer" is our own handler
                    569:     consumer = xml_handler(options)
                    570:     #Initialize Tenorsax with handler
                    571:     handler = saxtools.tenorsax(consumer)
                    572:     #Resulting tenorsax instance is the SAX handler 
                    573:     parser.setContentHandler(handler)
                    574:     parser.setFeature(sax.handler.feature_namespaces, 1)
                    575:     parser.parse(options.filename)  
                    576:     
1.1       casties   577: 
1.9       dwinter   578: if __name__ == "__main__":
                    579:     from optparse import OptionParser
1.5       casties   580: 
1.9       dwinter   581:     opars = OptionParser()
                    582:     opars.add_option("-f", "--file", 
                    583:                      dest="filename",
                    584:                      help="FMPXML file name", metavar="FILE")
                    585:     opars.add_option("-c", "--dsn", 
                    586:                      dest="dsn", 
                    587:                      help="database connection string")
                    588:     opars.add_option("-t", "--table", 
                    589:                      dest="table", 
                    590:                      help="database table name")
                    591:     opars.add_option("--fields", default=None, 
                    592:                      dest="update_fields", 
                    593:                      help="list of fields to update (comma separated, sql-names)", metavar="LIST")
                    594:     opars.add_option("--id-field", default=None, 
                    595:                      dest="id_field", 
                    596:                      help="name of id field for synchronisation (only appends data otherwise, sql-name)", metavar="NAME")
                    597:     opars.add_option("--sync", "--sync-mode", default=False, action="store_true", 
                    598:                      dest="sync_mode", 
                    599:                      help="do full sync based on id field (remove unmatched fields from db)")
                    600:     opars.add_option("--lc-names", default=False, action="store_true", 
                    601:                      dest="lc_names", 
                    602:                      help="clean and lower case field names from XML")
                    603:     opars.add_option("--keep-fields", default=False, action="store_true", 
                    604:                      dest="keep_fields", 
                    605:                      help="don't add fields from XML to SQL table")
                    606:     opars.add_option("--ascii-db", default=False, action="store_true", 
                    607:                      dest="ascii_db", 
                    608:                      help="the SQL database stores ASCII instead of unicode")
                    609:     opars.add_option("--replace", default=False, action="store_true", 
                    610:                      dest="replace_table", 
                    611:                      help="replace table i.e. delete and re-insert data")
                    612:     opars.add_option("--backup", default=False, action="store_true", 
                    613:                      dest="backup_table", 
1.16      casties   614:                      help="create backup of old table")
1.23      casties   615:     opars.add_option("--read-before-update", default=False, action="store_true", 
                    616:                      dest="read_before_update", 
                    617:                      help="read all data to check if it really changed")
1.9       dwinter   618:     opars.add_option("-d", "--debug", default=False, action="store_true", 
                    619:                      dest="debug", 
                    620:                      help="debug mode (more output)")
                    621:     
                    622:     (options, args) = opars.parse_args()
                    623:     
                    624:     if len(sys.argv) < 2 or options.filename is None or options.dsn is None:
                    625:         print "importFMPXML "+version_string
                    626:         opars.print_help()
                    627:         sys.exit(1)
                    628:     
                    629:     if options.debug:
                    630:         loglevel = logging.DEBUG
                    631:     else:
                    632:         loglevel = logging.INFO
                    633:     
                    634:     logging.basicConfig(level=loglevel, 
                    635:                         format='%(asctime)s %(levelname)s %(message)s',
                    636:                         datefmt='%H:%M:%S')
                    637: 
1.11      dwinter   638:     importFMPXML(options)
                    639: 
1.9       dwinter   640: 
                    641:     
1.1       casties   642: 

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>