Homec4science
Diffusion invenio-infoscience (maint-2.1) Browse

R3600/modules/bibedit/lib/bibedit_utils.py
e255de3235c5maint-2.1

Last Change
Raw File

bibedit_utils.py
View Options

## This file is part of CDS Invenio.
## Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007, 2008 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.
# pylint: disable-msg=C0103
"""BibEdit Utilities.
This module contains support functions (i.e., those that are not called directly
by the web interface), that might be imported by other modules or that is called
by both the web and CLI interfaces.
"""
__revision__ = "$Id$"
import commands
import cPickle
import difflib
import fnmatch
import marshal
import os
import re
import time
import zlib
from invenio.bibedit_config import CFG_BIBEDIT_FILENAME, \
CFG_BIBEDIT_RECORD_TEMPLATES_PATH, CFG_BIBEDIT_TO_MERGE_SUFFIX
from invenio.bibedit_dblayer import get_record_last_modification_date
from invenio.bibrecord import create_record, create_records, \
record_get_field_value, record_has_field, record_xml_output, \
record_strip_empty_fields
from invenio.bibtask import task_low_level_submission
from invenio.config import CFG_BINDIR, CFG_BIBEDIT_LOCKLEVEL, \
CFG_BIBEDIT_TIMEOUT, CFG_BIBUPLOAD_EXTERNAL_OAIID_TAG as OAIID_TAG, \
CFG_BIBUPLOAD_EXTERNAL_SYSNO_TAG as SYSNO_TAG, CFG_TMPDIR
from invenio.dateutils import convert_datetext_to_dategui
from invenio.bibedit_dblayer import get_bibupload_task_opts, \
get_marcxml_of_record_revision, get_record_revisions
from invenio.search_engine import get_fieldvalues, print_record, record_exists
# Precompile regexp:
re_file_option = re.compile(r'^%s' % CFG_TMPDIR)
re_xmlfilename_suffix = re.compile('_(\d+)_\d+\.xml$')
re_revid_split = re.compile('^(\d+)\.(\d{14})$')
re_revdate_split = re.compile('^(\d\d\d\d)(\d\d)(\d\d)(\d\d)(\d\d)(\d\d)')
re_taskid = re.compile('ID="(\d+)"')
re_tmpl_name = re.compile('<!-- BibEdit-Template-Name: (.*) -->')
re_tmpl_description = re.compile('<!-- BibEdit-Template-Description: (.*) -->')
# Operations on the BibEdit cache file
def cache_exists(recid, uid):
"""Check if the BibEdit cache file exists."""
return os.path.isfile('%s.tmp' % _get_file_path(recid, uid))
def get_cache_mtime(recid, uid):
"""Get the last modified time of the BibEdit cache file. Check that the
cache exists before calling this function.
"""
try:
return int(os.path.getmtime('%s.tmp' % _get_file_path(recid, uid)))
except OSError:
pass
def cache_expired(recid, uid):
"""Has it been longer than the number of seconds given by
CFG_BIBEDIT_TIMEOUT since last cache update? Check that the
cache exists before calling this function.
"""
return get_cache_mtime(recid, uid) < int(time.time()) - CFG_BIBEDIT_TIMEOUT
def create_cache_file(recid, uid, record='', cache_dirty=False):
"""Create a BibEdit cache file, and return revision and record. This will
overwrite any existing cache the user has for this record.
"""
if not record:
record = get_bibrecord(recid)
if not record:
return
file_path = '%s.tmp' % _get_file_path(recid, uid)
record_revision = get_record_last_modification_date(recid)
cache_file = open(file_path, 'w')
cPickle.dump([cache_dirty, record_revision, record], cache_file)
cache_file.close()
return record_revision, record
def touch_cache_file(recid, uid):
"""Touch a BibEdit cache file. This should be used to indicate that the
user has again accessed the record, so that locking will work correctly.
"""
if cache_exists(recid, uid):
os.system('touch %s.tmp' % _get_file_path(recid, uid))
def get_bibrecord(recid):
"""Return record in BibRecord wrapping."""
if record_exists(recid):
return create_record(print_record(recid, 'xm'))[0]
def get_cache_file_contents(recid, uid):
"""Return the contents of a BibEdit cache file."""
cache_file = _get_cache_file(recid, uid, 'r')
if cache_file:
cache_dirty, record_revision, record = cPickle.load(cache_file)
cache_file.close()
return cache_dirty, record_revision, record
def update_cache_file_contents(recid, uid, record_revision, record):
"""Save updates to the record in BibEdit cache. Return file modificaton
time.
"""
cache_file = _get_cache_file(recid, uid, 'w')
if cache_file:
cPickle.dump([True, record_revision, record], cache_file)
cache_file.close()
return get_cache_mtime(recid, uid)
def delete_cache_file(recid, uid):
"""Delete a BibEdit cache file."""
os.remove('%s.tmp' % _get_file_path(recid, uid))
def save_xml_record(recid, uid, xml_record='', to_upload=True, to_merge=False):
"""Write XML record to file. Default behaviour is to read the record from
a BibEdit cache file, write it back to an XML file and then pass this
file to BibUpload.
@param xml_record - give XML as string in stead of reading cache file
@param to_upload - pass the XML file to BibUpload
@param to_merge - prepare an XML file for BibMerge to use
"""
if not xml_record:
# Read record from cache file.
cache = get_cache_file_contents(recid, uid)
if cache:
record = cache[2]
record_strip_empty_fields(record)
xml_record = record_xml_output(record)
delete_cache_file(recid, uid)
if xml_record:
# Write XML file.
if not to_merge:
file_path = '%s.xml' % _get_file_path(recid, uid)
else:
file_path = '%s_%s.xml' % (_get_file_path(recid, uid),
CFG_BIBEDIT_TO_MERGE_SUFFIX)
xml_file = open(file_path, 'w')
xml_file.write(xml_record)
xml_file.close()
if to_upload:
# Pass XML file to BibUpload.
task_low_level_submission('bibupload', 'bibedit', '-P', '5', '-r',
file_path)
return True
# Security: Locking and integrity
def latest_record_revision(recid, revision_time):
"""Check if timetuple REVISION_TIME matches latest modification date."""
return revision_time == get_record_last_modification_date(recid)
def record_locked_by_other_user(recid, uid):
"""Return true if any other user than UID has active caches for record
RECID.
"""
active_uids = _uids_with_active_caches(recid)
try:
active_uids.remove(uid)
except ValueError:
pass
return bool(active_uids)
def record_locked_by_queue(recid):
"""Check if record should be locked for editing because of the current state
of the BibUpload queue. The level of checking is based on
CFG_BIBEDIT_LOCKLEVEL.
"""
# Check for *any* scheduled bibupload tasks.
if CFG_BIBEDIT_LOCKLEVEL == 2:
return _get_bibupload_task_ids()
filenames = _get_bibupload_filenames()
# Check for match between name of XML-files and record.
# Assumes that filename ends with _<recid>.xml.
if CFG_BIBEDIT_LOCKLEVEL == 1:
recids = []
for filename in filenames:
filename_suffix = re_xmlfilename_suffix.search(filename)
if filename_suffix:
recids.append(int(filename_suffix.group(1)))
return recid in recids
# Check for match between content of files and record.
if CFG_BIBEDIT_LOCKLEVEL == 3:
while True:
lock = _record_in_files_p(recid, filenames)
# Check if any new files were added while we were searching
if not lock:
filenames_updated = _get_bibupload_filenames()
for filename in filenames_updated:
if not filename in filenames:
break
else:
return lock
else:
return lock
# JSON
def json_unicode_to_utf8(data):
"""Change all strings in a JSON structure to UTF-8."""
if type(data) == unicode:
return data.encode('utf-8')
elif type(data) == dict:
newdict = {}
for key in data:
newdict[json_unicode_to_utf8(key)] = json_unicode_to_utf8(data[key])
return newdict
elif type(data) == list:
return [json_unicode_to_utf8(elem) for elem in data]
else:
return data
# History/revisions
def get_record_revision_ids(recid):
"""Return list of all record revision IDs.
Return revision IDs in chronologically decreasing order (latest first).
"""
res = []
tmp_res = get_record_revisions(recid)
for row in tmp_res:
res.append('%s.%s' % (row[0], row[1]))
return res
def get_marcxml_of_revision_id(revid):
"""Return MARCXML string of revision.
Return empty string if revision does not exist. REVID should be a string.
"""
res = ''
recid, job_date = split_revid(revid, 'datetext')
tmp_res = get_marcxml_of_record_revision(recid, job_date)
if tmp_res:
for row in tmp_res:
res += zlib.decompress(row[0]) + '\n'
return res
def revision_format_valid_p(revid):
"""Test validity of revision ID format (=RECID.REVDATE)."""
if re_revid_split.match(revid):
return True
return False
def split_revid(revid, dateformat=''):
"""Split revid and return tuple (recid, revdate).
Optional dateformat can be datetext or dategui.
"""
recid, revdate = re_revid_split.search(revid).groups()
if dateformat:
datetext = '%s-%s-%s %s:%s:%s' % re_revdate_split.search(
revdate).groups()
if dateformat == 'datetext':
revdate = datetext
elif dateformat == 'dategui':
revdate = convert_datetext_to_dategui(datetext, secs=True)
return recid, revdate
def get_xml_comparison(header1, header2, xml1, xml2):
"""Return diff of two MARCXML records."""
return ''.join(difflib.unified_diff(xml1.splitlines(1),
xml2.splitlines(1), header1, header2))
# Record templates
def get_record_templates():
"""Return list of record template (filename, name, description) tuples."""
template_fnames = fnmatch.filter(os.listdir(
CFG_BIBEDIT_RECORD_TEMPLATES_PATH), '*.xml')
templates = []
for fname in template_fnames:
template_file = open('%s%s%s' % (
CFG_BIBEDIT_RECORD_TEMPLATES_PATH, os.sep, fname),'r')
template = template_file.read()
template_file.close()
fname_stripped = os.path.splitext(fname)[0]
mo_name = re_tmpl_name.search(template)
mo_description = re_tmpl_description.search(template)
if mo_name:
name = mo_name.group(1)
else:
name = fname_stripped
if mo_description:
description = mo_description.group(1)
else:
description = ''
templates.append([fname_stripped, name, description])
return templates
def get_record_template(name):
"""Return an XML record template."""
filepath = '%s%s%s.xml' % (CFG_BIBEDIT_RECORD_TEMPLATES_PATH, os.sep, name)
if os.path.isfile(filepath):
template_file = open(filepath, 'r')
template = template_file.read()
template_file.close()
return template
# Private functions
def _get_cache_file(recid, uid, mode):
"""Return a BibEdit cache file object."""
if cache_exists(recid, uid):
return open('%s.tmp' % _get_file_path(recid, uid), mode)
def _get_file_path(recid, uid, filename=''):
"""Return the file path to a BibEdit file (excluding suffix).
If filename is specified this replaces the config default.
"""
if not filename:
return '%s%s%s_%s_%s' % (CFG_TMPDIR, os.sep, CFG_BIBEDIT_FILENAME,
recid, uid)
else:
return '%s%s%s_%s_%s' % (CFG_TMPDIR, os.sep, filename, recid, uid)
def _uids_with_active_caches(recid):
"""Return list of uids with active caches for record RECID. Active caches
are caches that have been modified a number of seconds ago that is less than
the one given by CFG_BIBEDIT_TIMEOUT.
"""
re_tmpfilename = re.compile('%s_%s_(\d+)\.tmp' % (CFG_BIBEDIT_FILENAME,
recid))
tmpfiles = fnmatch.filter(os.listdir(CFG_TMPDIR), '%s*.tmp' %
CFG_BIBEDIT_FILENAME)
expire_time = int(time.time()) - CFG_BIBEDIT_TIMEOUT
active_uids = []
for tmpfile in tmpfiles:
mo = re_tmpfilename.match(tmpfile)
if mo and int(os.path.getmtime('%s%s%s' % (
CFG_TMPDIR, os.sep, tmpfile))) > expire_time:
active_uids.append(int(mo.group(1)))
return active_uids
def _get_bibupload_task_ids():
"""Return list of all BibUpload task IDs.
Ignore tasks submitted by user bibreformat.
"""
cmd = '%s%sbibsched status -t bibupload' % (CFG_BINDIR, os.sep)
err, out = commands.getstatusoutput(cmd)
if err:
raise StandardError, '%s: %s' % (err, out)
tasks = out.splitlines()[3:-1]
res = []
for task in tasks:
if task.find('USER="bibreformat"') == -1:
matchobj = re_taskid.search(task)
if matchobj:
res.append(matchobj.group(1))
return res
def _get_bibupload_filenames():
"""Return paths to all files scheduled for upload."""
task_ids = _get_bibupload_task_ids()
filenames = []
tasks_opts = get_bibupload_task_opts(task_ids)
for task_opts in tasks_opts:
if task_opts:
record_options = marshal.loads(task_opts[0][0])
for option in record_options[1:]:
if re_file_option.search(option):
filenames.append(option)
return filenames
def _record_in_files_p(recid, filenames):
"""Search XML files for given record."""
# Get id tags of record in question
rec_oaiid = rec_sysno = -1
rec_oaiid_tag = get_fieldvalues(recid, OAIID_TAG)
if rec_oaiid_tag:
rec_oaiid = rec_oaiid_tag[0]
rec_sysno_tag = get_fieldvalues(recid, SYSNO_TAG)
if rec_sysno_tag:
rec_sysno = rec_sysno_tag[0]
# For each record in each file, compare ids and abort if match is found
for filename in filenames:
try:
file_ = open(filename)
records = create_records(file_.read(), 0, 0)
for i in range(0, len(records)):
record, all_good = records[i][:2]
if record and all_good:
if _record_has_id_p(record, recid, rec_oaiid, rec_sysno):
return True
file_.close()
except IOError:
continue
return False
def _record_has_id_p(record, recid, rec_oaiid, rec_sysno):
"""Check if record matches any of the given IDs."""
if record_has_field(record, '001'):
if (record_get_field_value(record, '001', '%', '%')
== str(recid)):
return True
if record_has_field(record, OAIID_TAG[0:3]):
if (record_get_field_value(
record, OAIID_TAG[0:3], OAIID_TAG[3],
OAIID_TAG[4], OAIID_TAG[5]) == rec_oaiid):
return True
if record_has_field(record, SYSNO_TAG[0:3]):
if (record_get_field_value(
record, SYSNO_TAG[0:3], SYSNO_TAG[3],
SYSNO_TAG[4], SYSNO_TAG[5]) == rec_sysno):
return True
return False
c4science · Help