Homec4science
Diffusion invenio-infoscience (maint-2.1) Browse

R3600/modules/bibupload/lib/bibupload.py
803e1afe5d8dmaint-2.1

Last Change
Raw File

bibupload.py
View Options

# -*- coding: utf-8 -*-
##
## $Id$
##
## This file is part of CDS Invenio.
## Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007 CERN.
##
## CDS Invenio is free software; you can redistribute it and/or
## modify it under the terms of the GNU General Public License as
## published by the Free Software Foundation; either version 2 of the
## License, or (at your option) any later version.
##
## CDS Invenio is distributed in the hope that it will be useful, but
## WITHOUT ANY WARRANTY; without even the implied warranty of
## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
## General Public License for more details.
##
## You should have received a copy of the GNU General Public License
## along with CDS Invenio; if not, write to the Free Software Foundation, Inc.,
## 59 Temple Place, Suite 330, Boston, MA 02111-1307, USA.
"""
BibUpload: Receive MARC XML file and update the appropriate database
tables according to options.
Usage: bibupload [options] input.xml
Examples:
$ bibupload -i input.xml
Options:
-a, --append new fields are appended to the existing record
-c, --correct fields are replaced by the new ones in the
existing record
-f, --format takes only the FMT fields into account.
Does not update
-i, --insert insert the new record in the database
-r, --replace the existing record is entirely replaced
by the new one
-z, --reference update references (update only 999 fields)
-s, --stage=STAGE stage to start from in the algorithm
(0: always done; 1: FMT tags;
2: FFT tags; 3: BibFmt; 4: Metadata update; 5: time update)
-n, --notimechange do not change record last modification date
when updating
Scheduling options:
-u, --user=USER user name to store task, password needed
General options:
-h, --help print this help and exit
-v, --verbose=LEVEL verbose level (from 0 to 9, default 1)
-V --version print the script version
"""
__revision__ = "$Id$"
import os
import sys
import time
from zlib import compress
import MySQLdb
import re
import urllib2
import tempfile
from invenio.config import CFG_OAI_ID_FIELD, weburl
from invenio.bibupload_config import *
from invenio.dbquery import run_sql, \
Error
from invenio.bibrecord import create_records, \
create_record, \
record_add_field, \
record_delete_field, \
record_xml_output, \
record_get_field_instances, \
record_get_field_values, \
field_get_subfield_values
from invenio.dateutils import convert_datestruct_to_datetext
from invenio.bibformat import format_record
from invenio.config import filedir, \
filedirsize, \
htdocsurl, \
tmpdir, \
CFG_PREFIX
from invenio.bibtask import task_init, write_message, get_datetime, \
task_set_option, task_get_option, task_get_task_param, task_update_status, \
task_update_progress
from invenio.file import BibRecDocs, file_strip_ext
#Statistic variables
stat = {}
stat['nb_records_to_upload'] = 0
stat['nb_records_updated'] = 0
stat['nb_records_inserted'] = 0
stat['nb_errors'] = 0
stat['exectime'] = time.localtime()
### bibupload engine functions:
def bibupload(record, opt_tag=None, opt_mode=None,
opt_stage_to_start_from=1, opt_notimechange=0):
"""Main function: process a record and fit it in the tables
bibfmt, bibrec, bibrec_bibxxx, bibxxx with proper record
metadata.
Return (error_code, recID) of the processed record.
"""
assert(opt_mode in ('insert', 'replace', 'replace_or_insert', 'reference',
'correct', 'append', 'format'))
error = None
# If there are special tags to proceed check if it exists in the record
if opt_tag is not None and not(record.has_key(opt_tag)):
write_message(" Failed: Tag not found, enter a valid tag to update.",
verbose=1, stream=sys.stderr)
return (1, -1)
# Extraction of the Record Id from 001, SYSNO or OAIID tags:
rec_id = retrieve_rec_id(record, opt_mode)
if rec_id == -1:
return (1, -1)
elif rec_id > 0:
write_message(" -Retrieve record ID (found %s): DONE." % rec_id, verbose=2)
if not record.has_key('001'):
# Found record ID by means of SYSNO or OAIID, and the
# input MARCXML buffer does not have this 001 tag, so we
# should add it now:
error = record_add_field(record, '001', '', '', rec_id, [], 0)
if error is None:
write_message(" Failed: " \
"Error during adding the 001 controlfield " \
"to the record", verbose=1, stream=sys.stderr)
return (1, int(rec_id))
else:
error = None
write_message(" -Added tag 001: DONE.", verbose=2)
write_message(" -Check if the xml marc file is already in the database: DONE" , verbose=2)
# Reference mode check if there are reference tag
if opt_mode == 'reference':
error = extract_tag_from_record(record, CFG_BIBUPLOAD_REFERENCE_TAG)
if error is None:
write_message(" Failed: No reference tags has been found...",
verbose=1, stream=sys.stderr)
return (1, -1)
else:
error = None
write_message(" -Check if reference tags exist: DONE", verbose=2)
if opt_mode == 'insert' or \
(opt_mode == 'replace_or_insert' and rec_id is None):
insert_mode_p = True
# Insert the record into the bibrec databases to have a recordId
rec_id = create_new_record()
write_message(" -Creation of a new record id (%d): DONE" % rec_id, verbose=2)
# we add the record Id control field to the record
error = record_add_field(record, '001', '', '', rec_id, [], 0)
if error is None:
write_message(" Failed: " \
"Error during adding the 001 controlfield " \
"to the record", verbose=1, stream=sys.stderr)
return (1, int(rec_id))
else:
error = None
elif opt_mode != 'insert' and opt_mode != 'format' and \
opt_stage_to_start_from != 5:
insert_mode_p = False
# Update Mode
# Retrieve the old record to update
rec_old = create_record(format_record(int(rec_id), 'xm'), 2)[0]
if rec_old is None:
write_message(" Failed during the creation of the old record!",
verbose=1, stream=sys.stderr)
return (1, int(rec_id))
else:
write_message(" -Retrieve the old record to update: DONE", verbose=2)
# In Replace mode, take over old strong tags if applicable:
if opt_mode == 'replace' or \
opt_mode == 'replace_or_insert':
copy_strong_tags_from_old_record(record, rec_old)
# Delete tags to correct in the record
if opt_mode == 'correct' or opt_mode == 'reference':
delete_tags_to_correct(record, rec_old, opt_tag)
write_message(" -Delete the old tags to correct in the old record: DONE",
verbose=2)
# Append new tag to the old record and update the new record with the old_record modified
if opt_mode == 'append' or opt_mode == 'correct' or \
opt_mode == 'reference':
record = append_new_tag_to_old_record(record, rec_old,
opt_tag, opt_mode)
write_message(" -Append new tags to the old record: DONE", verbose=2)
# now we clear all the rows from bibrec_bibxxx from the old
# record (they will be populated later (if needed) during
# stage 4 below):
delete_bibrec_bibxxx(rec_old, rec_id)
write_message(" -Clean bibrec_bibxxx: DONE", verbose=2)
write_message(" -Stage COMPLETED", verbose=2)
# Have a look if we have FMT tags
write_message("Stage 1: Start (Insert of FMT tags if exist).", verbose=2)
if opt_stage_to_start_from <= 1 and \
extract_tag_from_record(record, 'FMT') is not None:
record = insert_fmt_tags(record, rec_id, opt_mode)
if record is None:
write_message(" Stage 1 failed: Error while inserting FMT tags",
verbose=1, stream=sys.stderr)
return (1, int(rec_id))
elif record == 0:
# Mode format finished
stat['nb_records_updated'] += 1
return (0, int(rec_id))
write_message(" -Stage COMPLETED", verbose=2)
else:
write_message(" -Stage NOT NEEDED", verbose=2)
# Have a look if we have FFT tags
write_message("Stage 2: Start (Process FFT tags if exist).", verbose=2)
if opt_stage_to_start_from <= 2 and \
extract_tag_from_record(record, 'FFT') is not None:
record = elaborate_fft_tags(record, rec_id, opt_mode)
write_message(" -Stage COMPLETED", verbose=2)
else:
write_message(" -Stage NOT NEEDED", verbose=2)
# Update of the BibFmt
write_message("Stage 3: Start (Update bibfmt).", verbose=2)
if opt_stage_to_start_from <= 3:
# format the single record as xml
rec_xml_new = record_xml_output(record)
# Update bibfmt with the format xm of this record
if opt_mode != 'format':
error = update_bibfmt_format(rec_id, rec_xml_new, 'xm')
if error == 1:
write_message(" Failed: error during update_bibfmt_format",
verbose=1, stream=sys.stderr)
return (1, int(rec_id))
write_message(" -Stage COMPLETED", verbose=2)
# Update the database MetaData
write_message("Stage 4: Start (Update the database with the metadata).",
verbose=2)
if opt_stage_to_start_from <= 4:
if opt_mode == 'insert' or \
opt_mode == 'replace' or \
opt_mode == 'replace_or_insert' or \
opt_mode == 'append' or \
opt_mode == 'correct' or \
opt_mode == 'reference':
update_database_with_metadata(record, rec_id)
else:
write_message(" -Stage NOT NEEDED in mode %s" % opt_mode,
verbose=2)
write_message(" -Stage COMPLETED", verbose=2)
else:
write_message(" -Stage NOT NEEDED", verbose=2)
# Finally we update the bibrec table with the current date
write_message("Stage 5: Start (Update bibrec table with current date).",
verbose=2)
if opt_stage_to_start_from <= 5 and \
opt_notimechange == 0 and \
not insert_mode_p:
now = convert_datestruct_to_datetext(time.localtime())
write_message(" -Retrieved current localtime: DONE", verbose=2)
update_bibrec_modif_date(now, rec_id)
write_message(" -Stage COMPLETED", verbose=2)
else:
write_message(" -Stage NOT NEEDED", verbose=2)
# Increase statistics
if insert_mode_p:
stat['nb_records_inserted'] += 1
else:
stat['nb_records_updated'] += 1
# Upload of this record finish
write_message("Record "+str(rec_id)+" DONE", verbose=1)
return (0, int(rec_id))
def print_out_bibupload_statistics():
"""Print the statistics of the process"""
out = "Task stats: %(nb_input)d input records, %(nb_updated)d updated, " \
"%(nb_inserted)d inserted, %(nb_errors)d errors. " \
"Time %(nb_sec).2f sec." % { \
'nb_input': stat['nb_records_to_upload'],
'nb_updated': stat['nb_records_updated'],
'nb_inserted': stat['nb_records_inserted'],
'nb_errors': stat['nb_errors'],
'nb_sec': time.time() - time.mktime(stat['exectime']) }
write_message(out)
def open_marc_file(path):
"""Open a file and return the data"""
try:
# open the file containing the marc document
marc_file = open(path,'r')
marc = marc_file.read()
marc_file.close()
except IOError, erro:
write_message("Error: %s" % erro, verbose=1, stream=sys.stderr)
write_message("Exiting.", sys.stderr)
task_update_status("ERROR")
sys.exit(1)
return marc
def xml_marc_to_records(xml_marc):
"""create the records"""
# Creation of the records from the xml Marc in argument
recs = create_records(xml_marc, 1, 1)
if recs == []:
write_message("Error: Cannot parse MARCXML file.", verbose=1, stream=sys.stderr)
write_message("Exiting.", sys.stderr)
task_update_status("ERROR")
sys.exit(1)
elif recs[0][0] is None:
write_message("Error: MARCXML file has wrong format: %s" % recs,
verbose=1, stream=sys.stderr)
write_message("Exiting.", sys.stderr)
task_update_status("ERROR")
sys.exit(1)
else:
recs = map((lambda x:x[0]), recs)
return recs
def find_record_format(rec_id, format):
"""Look whether record REC_ID is formatted in FORMAT,
i.e. whether FORMAT exists in the bibfmt table for this record.
Return the number of times it is formatted: 0 if not, 1 if yes,
2 if found more than once (should never occur).
"""
out = 0
query = """SELECT COUNT(id) FROM bibfmt WHERE id_bibrec=%s AND format=%s"""
params = (rec_id, format)
res = []
try:
res = run_sql(query, params)
out = res[0][0]
except Error, error:
write_message(" Error during find_record_format() : %s " % error, verbose=1, stream=sys.stderr)
return out
def find_record_bibfmt(marc):
""" receives the xmlmarc containing a record and returns the id in bibrec
if the record exists in bibfmt"""
# compress the marc value
pickled_marc = MySQLdb.escape_string(compress(marc))
query = """SELECT id_bibrec FROM bibfmt WHERE value = %s"""
# format for marc xml is xm
params = (pickled_marc,)
try:
res = run_sql(query, params)
except Error, error:
write_message(" Error during find_record_bibfmt function : %s "
% error, verbose=1, stream=sys.stderr)
if len(res):
return res
else:
return None
def find_record_from_recid(rec_id):
"""
Try to find record in the database from the REC_ID number.
Return record ID if found, None otherwise.
"""
try:
res = run_sql("SELECT id FROM bibrec WHERE id=%s",
(rec_id,))
except Error, error:
write_message(" Error during find_record_bibrec() : %s "
% error, verbose=1, stream=sys.stderr)
if res:
return res[0][0]
else:
return None
def find_record_from_sysno(sysno):
"""
Try to find record in the database from the external SYSNO number.
Return record ID if found, None otherwise.
"""
bibxxx = 'bib'+CFG_BIBUPLOAD_EXTERNAL_SYSNO_TAG[0:2]+'x'
bibrec_bibxxx = 'bibrec_' + bibxxx
try:
res = run_sql("""SELECT bb.id_bibrec FROM %(bibrec_bibxxx)s AS bb,
%(bibxxx)s AS b WHERE b.tag=%%s AND b.value=%%s
AND bb.id_bibxxx=b.id""" % \
{'bibxxx': bibxxx,
'bibrec_bibxxx': bibrec_bibxxx},
(CFG_BIBUPLOAD_EXTERNAL_SYSNO_TAG, sysno,))
except Error, error:
write_message(" Error during find_record_from_sysno(): %s " % error,
verbose=1, stream=sys.stderr)
if res:
return res[0][0]
else:
return None
def find_record_from_extoaiid(extoaiid):
"""
Try to find record in the database from the external EXTOAIID number.
Return record ID if found, None otherwise.
"""
bibxxx = 'bib'+CFG_BIBUPLOAD_EXTERNAL_OAIID_TAG[0:2]+'x'
bibrec_bibxxx = 'bibrec_' + bibxxx
try:
res = run_sql("""SELECT bb.id_bibrec FROM %(bibrec_bibxxx)s AS bb,
%(bibxxx)s AS b WHERE b.tag=%%s AND b.value=%%s
AND bb.id_bibxxx=b.id""" % \
{'bibxxx': bibxxx,
'bibrec_bibxxx': bibrec_bibxxx},
(CFG_BIBUPLOAD_EXTERNAL_OAIID_TAG, extoaiid,))
except Error, error:
write_message(" Error during find_record_from_extoaiid(): %s "
% error, verbose=1, stream=sys.stderr)
if res:
return res[0][0]
else:
return None
def find_record_from_oaiid(oaiid):
"""
Try to find record in the database from the OAI ID number.
Return record ID if found, None otherwise.
"""
bibxxx = 'bib'+CFG_OAI_ID_FIELD[0:2]+'x'
bibrec_bibxxx = 'bibrec_' + bibxxx
try:
res = run_sql("""SELECT bb.id_bibrec FROM %(bibrec_bibxxx)s AS bb,
%(bibxxx)s AS b WHERE b.tag=%%s AND b.value=%%s
AND bb.id_bibxxx=b.id""" % \
{'bibxxx': bibxxx,
'bibrec_bibxxx': bibrec_bibxxx},
(CFG_OAI_ID_FIELD, oaiid,))
except Error, error:
write_message(" Error during find_record_from_oaiid(): %s " % error,
verbose=1, stream=sys.stderr)
if res:
return res[0][0]
else:
return None
def extract_tag_from_record(record, tag_number):
""" Extract the tag_number for record."""
# first step verify if the record is not already in the database
if record:
return record.get(tag_number, None)
return None
def retrieve_rec_id(record, opt_mode):
"""Retrieve the record Id from a record by using tag 001 or SYSNO or OAI ID
tag. opt_mod is the desired mode."""
rec_id = None
# 1st step: we look for the tag 001
tag_001 = extract_tag_from_record(record, '001')
if tag_001 is not None:
# We extract the record ID from the tag
rec_id = tag_001[0][3]
# if we are in insert mode => error
if opt_mode == 'insert':
write_message(" Failed : Error tag 001 found in the xml" \
" submitted, you should use the option replace," \
" correct or append to replace an existing" \
" record. (-h for help)",
verbose=1, stream=sys.stderr)
return -1
else:
# we found the rec id and we are not in insert mode => continue
# we try to match rec_id against the database:
if find_record_from_recid(rec_id) is not None:
# okay, 001 corresponds to some known record
return rec_id
else:
# The record doesn't exist yet. We shall have try to check
# the SYSNO or OAI id later.
write_message(" -Tag 001 value not found in database.",
verbose=9)
rec_id = None
else:
write_message(" -Tag 001 not found in the xml marc file.", verbose=9)
if rec_id is None:
# 2nd step we look for the SYSNO
sysnos = record_get_field_values(record,
CFG_BIBUPLOAD_EXTERNAL_SYSNO_TAG[0:3],
CFG_BIBUPLOAD_EXTERNAL_SYSNO_TAG[3:4] != "_" and \
CFG_BIBUPLOAD_EXTERNAL_SYSNO_TAG[3:4] or "",
CFG_BIBUPLOAD_EXTERNAL_SYSNO_TAG[4:5] != "_" and \
CFG_BIBUPLOAD_EXTERNAL_SYSNO_TAG[4:5] or "",
CFG_BIBUPLOAD_EXTERNAL_SYSNO_TAG[5:6])
if sysnos:
sysno = sysnos[0] # there should be only one external SYSNO
write_message(" -Checking if SYSNO " + sysno + \
" exists in the database", verbose=9)
# try to find the corresponding rec id from the database
rec_id = find_record_from_sysno(sysno)
if rec_id is not None:
# rec_id found
pass
else:
# The record doesn't exist yet. We will try to check
# external and internal OAI ids later.
write_message(" -Tag SYSNO value not found in database.",
verbose=9)
rec_id = None
else:
write_message(" -Tag SYSNO not found in the xml marc file.",
verbose=9)
if rec_id is None:
# 2nd step we look for the external OAIID
extoaiids = record_get_field_values(record,
CFG_BIBUPLOAD_EXTERNAL_OAIID_TAG[0:3],
CFG_BIBUPLOAD_EXTERNAL_OAIID_TAG[3:4] != "_" and \
CFG_BIBUPLOAD_EXTERNAL_OAIID_TAG[3:4] or "",
CFG_BIBUPLOAD_EXTERNAL_OAIID_TAG[4:5] != "_" and \
CFG_BIBUPLOAD_EXTERNAL_OAIID_TAG[4:5] or "",
CFG_BIBUPLOAD_EXTERNAL_OAIID_TAG[5:6])
if extoaiids:
extoaiid = extoaiids[0] # there should be only one external OAI ID
write_message(" -Checking if EXTOAIID " + extoaiid + \
" exists in the database", verbose=9)
# try to find the corresponding rec id from the database
rec_id = find_record_from_extoaiid(extoaiid)
if rec_id is not None:
# rec_id found
pass
else:
# The record doesn't exist yet. We will try to check
# OAI id later.
write_message(" -Tag EXTOAIID value not found in database.",
verbose=9)
rec_id = None
else:
write_message(" -Tag EXTOAIID not found in the xml marc file.", verbose=9)
if rec_id is None:
# 4th step we look for the OAI ID
oaiidvalues = record_get_field_values(record,
CFG_OAI_ID_FIELD[0:3],
CFG_OAI_ID_FIELD[3:4] != "_" and \
CFG_OAI_ID_FIELD[3:4] or "",
CFG_OAI_ID_FIELD[4:5] != "_" and \
CFG_OAI_ID_FIELD[4:5] or "",
CFG_OAI_ID_FIELD[5:6])
if oaiidvalues:
oaiid = oaiidvalues[0] # there should be only one OAI ID
write_message(" -Check if local OAI ID " + oaiid + \
" exist in the database", verbose=9)
# try to find the corresponding rec id from the database
rec_id = find_record_from_oaiid(oaiid)
if rec_id is not None:
# rec_id found
pass
else:
write_message(" -Tag OAI ID value not found in database.",
verbose=9)
rec_id = None
else:
write_message(" -Tag SYSNO not found in the xml marc file.",
verbose=9)
# Now we should have detected rec_id from SYSNO or OAIID
# tags. (None otherwise.)
if rec_id:
if opt_mode == 'insert':
write_message(" Failed : Record found in the database," \
" you should use the option replace," \
" correct or append to replace an existing" \
" record. (-h for help)",
verbose=1, stream=sys.stderr)
return -1
else:
if opt_mode != 'insert' and \
opt_mode != 'replace_or_insert':
write_message(" Failed : Record not found in the database."\
" Please insert the file before updating it."\
" (-h for help)", verbose=1, stream=sys.stderr)
return -1
return rec_id
### Insert functions
def create_new_record():
"""Create new record in the database"""
now = convert_datestruct_to_datetext(time.localtime())
query = """INSERT INTO bibrec (creation_date, modification_date)
VALUES (%s, %s)"""
params = (now, now)
try:
rec_id = run_sql(query, params)
return rec_id
except Error, error:
write_message(" Error during the creation_new_record function : %s "
% error, verbose=1, stream=sys.stderr)
return None
def insert_bibfmt(id_bibrec, marc, format):
"""Insert the format in the table bibfmt"""
# compress the marc value
#pickled_marc = MySQLdb.escape_string(compress(marc))
pickled_marc = compress(marc)
# get the current time
now = convert_datestruct_to_datetext(time.localtime())
query = """INSERT INTO bibfmt (id_bibrec, format, last_updated, value)
VALUES (%s, %s, %s, %s)"""
try:
row_id = run_sql(query, (id_bibrec, format, now, pickled_marc))
return row_id
except Error, error:
write_message(" Error during the insert_bibfmt function : %s "
% error, verbose=1, stream=sys.stderr)
return None
def insert_record_bibxxx(tag, value):
"""Insert the record into bibxxx"""
# determine into which table one should insert the record
table_name = 'bib'+tag[0:2]+'x'
# check if the tag, value combination exists in the table
query = """SELECT id,value FROM %s """ % table_name
query += """ WHERE tag=%s AND value=%s"""
params = (tag, value)
try:
res = run_sql(query, params)
except Error, error:
write_message(" Error during the insert_record_bibxxx function : %s "
% error, verbose=1, stream=sys.stderr)
# Note: compare now the found values one by one and look for
# string binary equality (e.g. to respect lowercase/uppercase
# match), regardless of the charset etc settings. Ideally we
# could use a BINARY operator in the above SELECT statement, but
# we would have to check compatibility on various MySQLdb versions
# etc; this approach checks all matched values in Python, not in
# MySQL, which is less cool, but more conservative, so it should
# work better on most setups.
for row in res:
row_id = row[0]
row_value = row[1]
if row_value == value:
return (table_name, row_id)
# We got here only when the tag,value combination was not found,
# so it is now necessary to insert the tag,value combination into
# bibxxx table as new.
query = """INSERT INTO %s """ % table_name
query += """ (tag, value) values (%s , %s)"""
params = (tag, value)
try:
row_id = run_sql(query, params)
except Error, error:
write_message(" Error during the insert_record_bibxxx function : %s "
% error, verbose=1, stream=sys.stderr)
return (table_name, row_id)
def insert_record_bibrec_bibxxx(table_name, id_bibxxx,
field_number, id_bibrec):
"""Insert the record into bibrec_bibxxx"""
# determine into which table one should insert the record
full_table_name = 'bibrec_'+ table_name
# insert the proper row into the table
query = """INSERT INTO %s """ % full_table_name
query += """(id_bibrec,id_bibxxx, field_number) values (%s , %s, %s)"""
params = (id_bibrec, id_bibxxx, field_number)
try:
res = run_sql(query, params)
except Error, error:
write_message(" Error during the insert_record_bibrec_bibxxx"
" function 2nd query : %s " % error, verbose=1, stream=sys.stderr)
return res
def _download_url(url, format):
"""Download a url (if it corresponds to a remote file) and return a local url
to it."""
protocol = urllib2.urlparse.urlsplit(url)[0]
(tmp, tmppath) = tempfile.mkstemp(suffix=format, dir=tmpdir)
if protocol in ('', 'file'):
path = urllib2.urlparse.urlsplit(url)[2]
ok = False
for allowed_path in CFG_BIBUPLOAD_FFT_ALLOWED_LOCAL_PATHS:
if path.startswith(allowed_path):
tmp.write(open(path).read())
ok = True
if not ok:
raise StandardError, "%s is not in one of the allowed paths." % path
tmp.write(urllib2.urlopen(url).read())
return tmppath
def make_mark_url(recid, docname, format):
"""Return a url valid to be used for MARC."""
return '%s/record/%s/files/%s%s' % (weburl, recid, docname, format)
def _get_docname_from_local_url(local_url):
path = urllib2.urlparse.urlsplit(local_url)[2]
filename = os.path.split(path)[-1]
return file_strip_ext(filename)
def _synchronize_8564(rec_id, record, descriptions, comments):
"""Sinchronize the 8564 tags for record with actual files. descriptions
should be a dictionary docname:description for the new description to be
inserted."""
tags8564s = record_get_field_instances(record, '856', '4', ' ')
filtered_tags8564s = []
# Let's discover all the previous internal urls, in order to rebuild them!
for field in tags8564s:
to_be_removed = False
for value in field_get_subfield_values(field, 'u'):
if value.startswith('%s/record/%s/files/' % (weburl, rec_id)):
to_be_removed = True
old_description = field_get_subfield_values(field, 'y')
if old_description:
docname = _get_docname_from_local_url(value)
if not descriptions.has_key(docname):
descriptions[docname] = old_description[0]
if not to_be_removed:
filtered_tags8564s.append(field)
# Let's keep in the record only external 8564
record_delete_field(record, '856', '4', ' ') # First we delete 8564
for field in filtered_tags8564s: # Then we readd external ones
record_add_field(record, '856', '4', ' ', '', field[0])
# Now we refresh with existing internal 8564
bibrecdocs = BibRecDocs(rec_id)
latest_files = bibrecdocs.listLatestFiles()
for file in latest_files:
new_url = '%s/record/%s/files/%s' % (weburl, rec_id, file.getFullName())
new_subfield = [('u', new_url)]
if descriptions.get(file.getName(), ''):
new_subfield.append(('y', descriptions[file.getName()]))
if comments.get(new_url, ''):
new_subfield.append(('z', comments[new_url]))
record_add_field(record, '856', '4', ' ', '', new_subfield)
return record
def elaborate_fft_tags(record, rec_id, mode, cfg_bibupload_wget_sleep_time=1, cfg_bibupload_max_download_retires=10):
"""
Process FFT tags that should contain $a with file pathes or URLs
to get the fulltext from. This function enriches record with
proper 8564 URL tags, downloads fulltext files and stores them
into var/data structure where appropriate.
CFG_BIBUPLOAD_WGET_SLEEP_TIME defines time to sleep in seconds in
between URL downloads.
Note: if an FFT tag contains multiple $a subfields, we upload them
into different 856 URL tags in the metadata. See regression test
case test_multiple_fft_insert_via_http().
"""
tuple_list = extract_tag_from_record(record, 'FFT')
if tuple_list:
write_message("insert_fft_tags, all FFTs: "+str(tuple_list), verbose=9)
docs = {}
comments = {}
for fft in record_get_field_instances(record, 'FFT', ' ', ' '):
doctype = field_get_subfield_values(fft, 't')
if doctype:
doctype = doctype[0]
else:
doctype = 'Main'
url = field_get_subfield_values(fft, 'a')
if url:
url = url[0]
elif doctype not in ('PURGE', 'DELETE'):
write_message('url not specified for fft %s' % str(fft), stream=sys.stderr)
description = field_get_subfield_values(fft, 'd')
if description:
description = description[0]
else:
description = ''
name = field_get_subfield_values(fft, 'n')
if name:
name = file_strip_ext(name[0])
else:
name = _get_docname_from_local_url(url)
newname = field_get_subfield_values(fft, 'm')
if newname:
newname = file_strip_ext(newname[0])
else:
newname = name
format = field_get_subfield_values(fft, 'f')
if format:
format = format[0]
else:
path = urllib2.urlparse.urlsplit(url)[2]
filename = os.path.split(path)[-1]
format = filename[len(file_strip_ext(filename)):].lower()
comment = field_get_subfield_values(fft, 'z')
if comment:
comment = comment[0]
else:
comment = ''
if docs.has_key(name):
(description2, doctype2, urls) = docs[name]
if description2 != description:
write_message("fft '%s' specifies a different description from previous fft with docname '%s'" % (str(fft), name), stream=sys.stderr)
break
if doctype2 != doctype:
write_message("fft '%s' specifies a different doctype from previous fft with docname '%s'" % (str(fft), name), stream=sys.stderr)
break
for (url2, format2) in urls:
if format == format2:
write_message("fft '%s' specifies a second file '%s' with the same format '%s' from previous fft with docname '%s'" % (str(fft), url, format, name), stream=sys.stderr)
docs[name] = (description, doctype, newname, urls.append((url, format)))
else:
docs[name] = (description, doctype, newname, [(url, format)])
if comment:
comments['%s/record/%s/files/%s%s' % (weburl, rec_id, newname, format)] = comment
bibrecdocs = BibRecDocs(rec_id)
descriptions = {}
for docname, (description, doctype, newname, urls) in docs.iteritems():
descriptions[newname] = description
if mode == 'insert': # new bibdocs, new docnames, new marc
if newname in bibrecdocs.getBibDocNames(doctype):
write_message("('%s', '%s', '%s', '%s') not inserted because docname already exists." % (description, doctype, newname, urls), stream=sys.stderr)
break
bibdoc = bibrecdocs.addBibDoc(doctype, newname)
for (url, format) in urls:
try:
tmpurl = _download_url(url, format)
try:
bibdoc.addFilesNewFormat([tmpurl])
except StandardError:
write_message("('%s', '%s', '%s', '%s') not inserted because format already exists." % (description, doctype, newname, urls), stream=sys.stderr)
os.remove(tmpurl)
except Exception, e:
write_message("Error in downloading '%s' because of: %s" % (url, e))
elif mode == 'replace': # only add new revisions, no new docnames, no marc
for bibdoc in bibrecdocs.listBibDocs():
bibdoc.delete()
bibrecdocs.buildBibDocList()
bibdoc = bibrecdocs.addBibDoc(doctype, newname)
for (url, format) in urls:
try:
tmpurl = _download_url(url, format)
try:
bibdoc.addFilesNewFormat([tmpurl])
except StandardError, e:
write_message("('%s', '%s', '%s', '%s') not added because '%s'." % (description, doctype, newname, urls, e))
os.remove(tmpurl)
except Exception, e:
write_message("Error in downloading '%s' because of: %s" % (url, e))
elif mode == 'replace_or_insert': # e
found_bibdoc = False
for bibdoc in bibrecdocs.listBibDocs():
if bibdoc.getDocName() == docname:
found_bibdoc = True
if newname != docname:
try:
bibdoc.changeName(newname)
except StandardError, e:
write_message(e, stream=sys.stderr)
break
(first_url, first_format) = urls[0]
other_urls = urls[1:]
try:
tmpurl = _download_url(first_url, first_format)
try:
bibdoc.addFilesNewVersion(tmpurl)
except StandardError, e:
write_message("('%s', '%s', '%s', '%s') not added because '%s'." % (description, doctype, newname, urls, e), stream=sys.stderr)
os.remove(tmpurl)
except Exception, e:
write_message("Error in downloading '%s' because of: %s" % (first_url, e))
continue
for (url, format) in other_urls:
try:
tmpurl = _download_url(url, format)
try:
bibdoc.addFilesNewFormat(tmpurl)
except StandardError, e:
write_message("('%s', '%s', '%s', '%s') not added because '%s'." % (description, doctype, newname, urls, e), stream=sys.stderr)
os.remove(tmpurl)
except Exception, e:
write_message("Error in downloading '%s' because of: %s" % (url, e))
continue
break
if not found_bibdoc:
bibdoc = bibrecdocs.addBibDoc(doctype, newname)
for (url, format) in urls:
try:
tmpurl = _download_url(url, format)
try:
bibdoc.addFilesNewFormat([tmpurl])
except StandardError:
write_message("('%s', '%s', '%s', '%s') not inserted because format already exists." % (description, doctype, newname, urls), stream=sys.stderr)
pass
os.remove(tmpurl)
except Exception, e:
write_message("Error in downloading '%s' because of: %s" % (url, e))
continue
elif mode == 'correct':
found_bibdoc = False
for bibdoc in bibrecdocs.listBibDocs():
if bibdoc.getDocName() == docname:
found_bibdoc = True
if doctype == 'PURGE':
bibdoc.purge()
elif doctype == 'DELETE':
bibdoc.delete()
else:
if newname != docname:
try:
bibdoc.changeName(newname)
except StandardError, e:
write_message(e, stream=sys.stderr)
break
(first_url, first_format) = urls[0]
other_urls = urls[1:]
try:
tmpurl = _download_url(first_url, first_format)
except Exception, e:
write_message("Error in downloading '%s' because of: %s" % (first_url, e))
continue
try:
bibdoc.addFilesNewVersion(tmpurl)
except StandardError, e:
write_message("('%s', '%s', '%s', '%s') not added because '%s'." % (description, doctype, newname, urls, e), stream=sys.stderr)
os.remove(tmpurl)
for (url, format) in other_urls:
try:
tmpurl = _download_url(url, format)
try:
bibdoc.addFilesNewFormat(tmpurl)
except StandardError, e:
write_message("('%s', '%s', '%s', '%s') not added because '%s'." % (description, doctype, newname, urls, e), stream=sys.stderr)
os.remove(tmpurl)
except Exception, e:
write_message("Error in downloading '%s' because of: %s" % (url, e))
continue
break
if not found_bibdoc:
write_message("('%s', '%s', '%s', '%s') not added because '%s' docname didn't existed." % (description, doctype, newname, urls, docname), stream=sys.stderr)
elif mode == 'append':
found_bibdoc = False
for bibdoc in bibrecdocs.listBibDocs():
if bibdoc.getDocName() == docname:
found_bibdoc = True
for (url, format) in urls:
try:
tmpurl = _download_url(url, format)
try:
bibdoc.addFilesNewFormat([tmpurl])
except StandardError:
write_message("('%s', '%s', '%s', '%s') not appended because format already exists." % (description, doctype, newname, urls), stream=sys.stderr)
pass
os.remove(tmpurl)
except Exception, e:
write_message("Error in downloading '%s' because of: %s" % (url, e))
continue
if not found_bibdoc:
try:
bibdoc = bibrecdocs.addBibDoc(doctype, docname)
for (url, format) in urls:
try:
tmpurl = _download_url(url, format)
try:
bibdoc.addFilesNewFormat([tmpurl])
except StandardError:
write_message("('%s', '%s', '%s', '%s') not appended because format already exists." % (description, doctype, newname, urls), stream=sys.stderr)
pass
os.remove(tmpurl)
except Exception, e:
write_message("Error in downloading '%s' because of: %s" % (url, e))
continue
except Exception, e:
write_message("('%s', '%s', '%s', '%s') not appended." % (description, doctype, newname, urls), stream=sys.stderr)
return _synchronize_8564(rec_id, record, descriptions, comments)
else:
return record
def insert_fmt_tags(record, rec_id, opt_mode):
"""Process and insert FMT tags"""
fmt_fields = record_get_field_instances(record, 'FMT')
if fmt_fields:
for fmt_field in fmt_fields:
# Get the f, g subfields of the FMT tag
try:
f_value = field_get_subfield_values(fmt_field, "f")[0]
except IndexError:
f_value = ""
try:
g_value = field_get_subfield_values(fmt_field, "g")[0]
except IndexError:
g_value = ""
# Update the format
res = update_bibfmt_format(rec_id, g_value, f_value)
if res == 1:
write_message(" Failed: Error during update_bibfmt", verbose=1, stream=sys.stderr)
# If we are in format mode, we only care about the FMT tag
if opt_mode == 'format':
return 0
# We delete the FMT Tag of the record
record_delete_field(record, 'FMT')
write_message(" -Delete field FMT from record : DONE", verbose=2)
return record
elif opt_mode == 'format':
write_message(" Failed: Format updated failed : No tag FMT found", verbose=1, stream=sys.stderr)
return None
else:
return record
### Update functions
def update_bibrec_modif_date(now, bibrec_id):
"""Update the date of the record in bibrec table """
query = """UPDATE bibrec SET modification_date=%s WHERE id=%s"""
params = (now, bibrec_id)
try:
run_sql(query, params)
write_message(" -Update record modification date : DONE" , verbose=2)
except Error, error:
write_message(" Error during update_bibrec_modif_date function : %s" % error,
verbose=1, stream=sys.stderr)
def update_bibfmt_format(id_bibrec, format_value, format_name):
"""Update the format in the table bibfmt"""
# We check if the format is already in bibFmt
nb_found = find_record_format(id_bibrec, format_name)
if nb_found == 1:
# Update the format
# get the current time
now = convert_datestruct_to_datetext(time.localtime())
# compress the format_value value
pickled_format_value = compress(format_value)
query = """UPDATE bibfmt SET last_updated=%s, value=%s WHERE id_bibrec=%s AND format=%s"""
params = (now, pickled_format_value, id_bibrec, format_name)
try:
row_id = run_sql(query, params)
if row_id is None:
write_message(" Failed: Error during update_bibfmt_format function", verbose=1, stream=sys.stderr)
return 1
else:
write_message(" -Update the format %s in bibfmt : DONE" % format_name , verbose=2)
return 0
except Error, error:
write_message(" Error during the update_bibfmt_format function : %s " % error, verbose=1, stream=sys.stderr)
elif nb_found > 1:
write_message(" Failed: Same format %s found several time in bibfmt for the same record." % format_name, verbose=1, stream=sys.stderr)
return 1
else:
# Insert the format information in BibFMT
res = insert_bibfmt(id_bibrec, format_value, format_name)
if res is None:
write_message(" Failed: Error during insert_bibfmt", verbose=1, stream=sys.stderr)
return 1
else:
write_message(" -Insert the format %s in bibfmt : DONE" % format_name , verbose=2)
return 0
def update_database_with_metadata(record, rec_id):
"""Update the database tables with the record and the record id given in parameter"""
for tag in record.keys():
# check if tag is not a special one:
if tag not in CFG_BIBUPLOAD_SPECIAL_TAGS:
# for each tag there is a list of tuples representing datafields
tuple_list = record[tag]
# this list should contain the elements of a full tag [tag, ind1, ind2, subfield_code]
tag_list = []
tag_list.append(tag)
for single_tuple in tuple_list:
# these are the contents of a single tuple
subfield_list = single_tuple[0]
ind1 = single_tuple[1]
ind2 = single_tuple[2]
# append the ind's to the full tag
if ind1 == '' or ind1 == ' ':
tag_list.append('_')
else:
tag_list.append(ind1)
if ind2 == '' or ind2 == ' ':
tag_list.append('_')
else:
tag_list.append(ind2)
datafield_number = single_tuple[4]
if tag in CFG_BIBUPLOAD_SPECIAL_TAGS:
# nothing to do for special tags (FFT, FMT)
pass
elif tag in CFG_BIBUPLOAD_CONTROLFIELD_TAGS and tag != "001":
value = single_tuple[3]
# get the full tag
full_tag = ''.join(tag_list)
# update the tables
write_message(" insertion of the tag "+full_tag+" with the value "+value, verbose=9)
# insert the tag and value into into bibxxx
(table_name, bibxxx_row_id) = insert_record_bibxxx(full_tag, value)
#print 'tname, bibrow', table_name, bibxxx_row_id;
if table_name is None or bibxxx_row_id is None:
write_message(" Failed : during insert_record_bibxxx", verbose=1, stream=sys.stderr)
# connect bibxxx and bibrec with the table bibrec_bibxxx
res = insert_record_bibrec_bibxxx(table_name, bibxxx_row_id, datafield_number, rec_id)
if res is None:
write_message(" Failed : during insert_record_bibrec_bibxxx", verbose=1, stream=sys.stderr)
else:
# get the tag and value from the content of each subfield
for subfield in subfield_list:
subtag = subfield[0]
value = subfield[1]
tag_list.append(subtag)
# get the full tag
full_tag = ''.join(tag_list)
# update the tables
write_message(" insertion of the tag "+full_tag+" with the value "+value, verbose=9)
# insert the tag and value into into bibxxx
(table_name, bibxxx_row_id) = insert_record_bibxxx(full_tag, value)
if table_name is None or bibxxx_row_id is None:
write_message(" Failed : during insert_record_bibxxx", verbose=1, stream=sys.stderr)
# connect bibxxx and bibrec with the table bibrec_bibxxx
res = insert_record_bibrec_bibxxx(table_name, bibxxx_row_id, datafield_number, rec_id)
if res is None:
write_message(" Failed : during insert_record_bibrec_bibxxx", verbose=1, stream=sys.stderr)
# remove the subtag from the list
tag_list.pop()
tag_list.pop()
tag_list.pop()
tag_list.pop()
write_message(" -Update the database with metadata : DONE", verbose=2)
def append_new_tag_to_old_record(record, rec_old, opt_tag, opt_mode):
"""Append new tags to a old record"""
if opt_tag is not None:
tag = opt_tag
if tag in CFG_BIBUPLOAD_CONTROLFIELD_TAGS:
if tag == '001':
pass
else:
# if it is a controlfield,just access the value
for single_tuple in record[tag]:
controlfield_value = single_tuple[3]
# add the field to the old record
newfield_number = record_add_field(rec_old, tag, "", "", controlfield_value)
if newfield_number is None:
write_message(" Error when adding the field"+tag, verbose=1, stream=sys.stderr)
else:
# For each tag there is a list of tuples representing datafields
for single_tuple in record[tag]:
# We retrieve the information of the tag
subfield_list = single_tuple[0]
ind1 = single_tuple[1]
ind2 = single_tuple[2]
# We add the datafield to the old record
write_message(" Adding tag: %s ind1=%s ind2=%s code=%s" % (tag, ind1, ind2, subfield_list), verbose=9)
newfield_number = record_add_field(rec_old, tag, ind1, ind2, "", subfield_list)
if newfield_number is None:
write_message("Error when adding the field"+tag, verbose=1, stream=sys.stderr)
else:
# Go through each tag in the appended record
for tag in record.keys():
# Reference mode append only reference tag
if opt_mode == 'reference':
if tag == CFG_BIBUPLOAD_REFERENCE_TAG:
for single_tuple in record[tag]:
# We retrieve the information of the tag
subfield_list = single_tuple[0]
ind1 = single_tuple[1]
ind2 = single_tuple[2]
# We add the datafield to the old record
write_message(" Adding tag: %s ind1=%s ind2=%s code=%s" % (tag, ind1, ind2, subfield_list), verbose=9)
newfield_number = record_add_field(rec_old, tag, ind1, ind2, "", subfield_list)
if newfield_number is None:
write_message(" Error when adding the field"+tag, verbose=1, stream=sys.stderr)
else:
if tag in CFG_BIBUPLOAD_CONTROLFIELD_TAGS:
if tag == '001':
pass
else:
# if it is a controlfield,just access the value
for single_tuple in record[tag]:
controlfield_value = single_tuple[3]
# add the field to the old record
newfield_number = record_add_field(rec_old, tag, "", "", controlfield_value)
if newfield_number is None:
write_message(" Error when adding the field"+tag, verbose=1, stream=sys.stderr)
else:
# For each tag there is a list of tuples representing datafields
for single_tuple in record[tag]:
# We retrieve the information of the tag
subfield_list = single_tuple[0]
ind1 = single_tuple[1]
ind2 = single_tuple[2]
# We add the datafield to the old record
write_message(" Adding tag: %s ind1=%s ind2=%s code=%s" % (tag, ind1, ind2, subfield_list), verbose=9)
newfield_number = record_add_field(rec_old, tag, ind1, ind2, "", subfield_list)
if newfield_number is None:
write_message(" Error when adding the field"+tag, verbose=1, stream=sys.stderr)
return rec_old
def copy_strong_tags_from_old_record(record, rec_old):
"""
Look for strong tags in RECORD and REC_OLD. If no strong tags are
found in RECORD, then copy them over from REC_OLD. This function
modifies RECORD structure on the spot.
"""
for strong_tag in CFG_BIBUPLOAD_STRONG_TAGS:
if not record_get_field_instances(record, strong_tag):
strong_tag_old_field_instances = record_get_field_instances(rec_old, strong_tag)
if strong_tag_old_field_instances:
for strong_tag_old_field_instance in strong_tag_old_field_instances:
sf_vals, fi_ind1, fi_ind2, controlfield, dummy = strong_tag_old_field_instance
record_add_field(record, strong_tag, fi_ind1, fi_ind2, controlfield, sf_vals)
return
### Delete functions
def delete_tags_to_correct(record, rec_old, opt_tag):
"""
Delete tags from REC_OLD which are also existing in RECORD. When
deleting, pay attention not only to tags, but also to indicators,
so that fields with the same tags but different indicators are not
deleted.
"""
# browse through all the tags from the MARCXML file:
for tag in record.keys():
# do we have to delete only a special tag or any tag?
if opt_tag is None or opt_tag == tag:
# check if the tag exists in the old record too:
if rec_old.has_key(tag) and tag != '001':
# the tag does exist, so delete all record's tag+ind1+ind2 combinations from rec_old
for dummy_sf_vals, ind1, ind2, dummy_cf, dummy_field_number in record[tag]:
write_message(" Delete tag: " + tag + " ind1=" + ind1 + " ind2=" + ind2, verbose=9)
record_delete_field(rec_old, tag, ind1, ind2)
def delete_bibrec_bibxxx(record, id_bibrec):
"""Delete the database record from the table bibxxx given in parameters"""
# we clear all the rows from bibrec_bibxxx from the old record
for tag in record.keys():
if tag not in CFG_BIBUPLOAD_SPECIAL_TAGS:
# for each name construct the bibrec_bibxxx table name
table_name = 'bibrec_bib'+tag[0:2]+'x'
# delete all the records with proper id_bibrec
query = """DELETE FROM `%s` where id_bibrec = %s"""
params = (table_name, id_bibrec)
try:
run_sql(query % params)
except Error, error:
write_message(" Error during the delete_bibrec_bibxxx function : %s " % error, verbose=1, stream=sys.stderr)
def wipe_out_record_from_all_tables(recid):
"""
Wipe out completely the record and all its traces of RECID from
the database (bibrec, bibrec_bibxxx, bibxxx, bibfmt). Useful for
the time being for test cases.
"""
# delete from bibrec:
run_sql("DELETE FROM bibrec WHERE id=%s", (recid,))
# delete from bibrec_bibxxx:
for i in range(0, 10):
for j in range(0, 10):
run_sql("DELETE FROM %(bibrec_bibxxx)s WHERE id_bibrec=%%s" % \
{'bibrec_bibxxx': "bibrec_bib%i%ix" % (i, j)},
(recid,))
# delete all unused bibxxx values:
for i in range(0, 10):
for j in range(0, 10):
run_sql("DELETE %(bibxxx)s FROM %(bibxxx)s " \
" LEFT JOIN %(bibrec_bibxxx)s " \
" ON %(bibxxx)s.id=%(bibrec_bibxxx)s.id_bibxxx " \
" WHERE %(bibrec_bibxxx)s.id_bibrec IS NULL" % \
{'bibxxx': "bib%i%ix" % (i, j),
'bibrec_bibxxx': "bibrec_bib%i%ix" % (i, j)})
# delete from bibfmt:
run_sql("DELETE FROM bibfmt WHERE id_bibrec=%s", (recid,))
# delete from bibrec_bibdoc:
run_sql("DELETE FROM bibrec_bibdoc WHERE id_bibrec=%s", (recid,))
return
def delete_bibdoc(id_bibrec):
"""Delete document from bibdoc which correspond to the bibrec id given in parameter"""
query = """UPDATE bibdoc SET status='deleted'
WHERE id IN (SELECT id_bibdoc FROM bibrec_bibdoc
WHERE id_bibrec=%s)"""
params = (id_bibrec,)
try:
run_sql(query, params)
except Error, error:
write_message(" Error during the delete_bibdoc function : %s " % error,
verbose=1, stream=sys.stderr)
def delete_bibrec_bibdoc(id_bibrec):
"""Delete the bibrec record from the table bibrec_bibdoc given in parameter"""
# delete all the records with proper id_bibrec
query = """DELETE FROM bibrec_bibdoc WHERE id_bibrec=%s"""
params = (id_bibrec,)
try:
run_sql(query, params)
except Error, error:
write_message(" Error during the delete_bibrec_bibdoc function : %s " % error,
verbose=1, stream=sys.stderr)
def main():
"""Main that construct all the bibtask."""
task_set_option('mode', None)
task_set_option('verbose', 1)
task_set_option("tag", None)
task_set_option("file_path", None)
task_set_option("notimechange", 0)
task_set_option("stage_to_start_from", 1)
task_init(authorization_action='runbibupload',
authorization_msg="BibUpload Task Submission",
description="""Receive MARC XML file and update appropriate database
tables according to options.
Examples:
$ bibupload -i input.xml
""",
help_specific_usage=""" -a, --append\t\tnew fields are appended to the existing record
-c, --correct\t\tfields are replaced by the new ones in the existing record
-f, --format\t\ttakes only the FMT fields into account. Does not update
-i, --insert\t\tinsert the new record in the database
-r, --replace\t\tthe existing record is entirely replaced by the new one
-z, --reference\tupdate references (update only 999 fields)
-s, --stage=STAGE\tstage to start from in the algorithm (0: always done; 1: FMT tags;
\t\t\t2: FFT tags; 3: BibFmt; 4: Metadata update; 5: time update)
-n, --notimechange\tdo not change record last modification date when updating
""",
version=__revision__,
specific_params=("ircazs:fn",
[
"insert",
"replace",
"correct",
"append",
"reference",
"stage=",
"format",
"notimechange",
]),
task_submit_elaborate_specific_parameter_fnc=task_submit_elaborate_specific_parameter,
task_run_fnc=task_run_core)
def task_submit_elaborate_specific_parameter(key, value, opts, args):
""" Given the string key it checks it's meaning, eventually using the
value. Usually it fills some key in the options dict.
It must return True if it has elaborated the key, False, if it doesn't
know that key.
eg:
if key in ['-n', '--number']:
task_get_option(\1) = value
return True
return False
"""
# No time change option
if key in ("-n", "--notimechange"):
task_set_option('notimechange', 1)
# Insert mode option
elif key in ("-i", "--insert"):
if task_get_option('mode') == 'replace':
# if also replace found, then set to replace_or_insert
task_set_option('mode', 'replace_or_insert')
else:
task_set_option('mode', 'insert')
task_set_option('file_path', os.path.abspath(args[0]))
# Replace mode option
elif key in ("-r", "--replace"):
if task_get_option('mode') == 'insert':
# if also insert found, then set to replace_or_insert
task_set_option('mode', 'replace_or_insert')
else:
task_set_option('mode', 'replace')
task_set_option('file_path', os.path.abspath(args[0]))
# Correct mode option
elif key in ("-c", "--correct"):
task_set_option('mode', 'correct')
task_set_option('file_path', os.path.abspath(args[0]))
# Append mode option
elif key in ("-a", "--append"):
task_set_option('mode', 'append')
task_set_option('file_path', os.path.abspath(args[0]))
# Reference mode option
elif key in ("-z", "--reference"):
task_set_option('mode', 'reference')
task_set_option('file_path', os.path.abspath(args[0]))
# Format mode option
elif key in ("-f", "--format"):
task_set_option('mode', 'format')
task_set_option('file_path', os.path.abspath(args[0]))
else:
return False
return True
def task_submit_check_options():
""" Reimplement this method for having the possibility to check options
before submitting the task, in order for example to provide default
values. It must return False if there are errors in the options.
"""
if task_get_option('mode') is None:
write_message("Please specify at least one update/insert mode!")
return False
if task_get_option('file_path') is None:
write_message("Missing filename! -h for help.")
return False
return True
def task_run_core():
""" Reimplement to add the body of the task."""
error = 0
write_message("Input file '%s', input mode '%s'." %
(task_get_option('file_path'), task_get_option('mode')))
write_message("STAGE 0:", verbose=2)
if task_get_option('file_path') is not None:
recs = xml_marc_to_records(open_marc_file(task_get_option('file_path')))
stat['nb_records_to_upload'] = len(recs)
write_message(" -Open XML marc: DONE", verbose=2)
if recs is not None:
# We proceed each record by record
for record in recs:
error = bibupload(
record,
opt_tag=task_get_option('tag'),
opt_mode=task_get_option('mode'),
opt_stage_to_start_from=task_get_option('stage_to_start_from'),
opt_notimechange=task_get_option('notimechange'))
if error[0] == 1:
if record:
write_message(record_xml_output(record),
stream=sys.stderr)
else:
write_message("Record could not have been parsed",
stream=sys.stderr)
stat['nb_errors'] += 1
task_update_progress("Done %d out of %d." % \
(stat['nb_records_inserted'] + \
stat['nb_records_updated'],
stat['nb_records_to_upload']))
else:
write_message(" Error bibupload failed: No record found",
verbose=1, stream=sys.stderr)
if task_get_option('verbose') >= 1:
# Print out the statistics
print_out_bibupload_statistics()
# Check if they were errors
return not stat['nb_errors'] >= 1
if __name__ == "__main__":
main()
c4science · Help