def task_submit_check_options():
"""Last checks and updating on the options..."""
if not (task_has_option('all') or task_has_option('collection')
or task_has_option('field') or task_has_option('pattern')
or task_has_option('matching') or task_has_option('recids')):
task_set_option('last', 1)
return True
def task_submit_check_options():
"""Last checks and updating on the options..."""
if not (task_has_option('all') or task_has_option('collection')
or task_has_option('field') or task_has_option('pattern')
or task_has_option('matching') or task_has_option('recids')):
task_set_option('last', 1)
return True
def task_run_core():
"""Run the task by fetching arguments from the BibSched task queue.
This is what BibSched will be invoking via daemon call.
"""
fmts = task_get_option('format', 'HB,RECJSON')
for fmt in fmts.split(','):
last_updated = fetch_last_updated(fmt)
write_message("last stored run date is %s" % last_updated)
recids = intbitset()
if task_has_option("all"):
recids += all_records()
if task_has_option("last"):
recids += outdated_caches(fmt, last_updated)
if task_has_option('ignore_without'):
without_fmt = intbitset()
else:
without_fmt = missing_caches(fmt)
recids += without_fmt
cli_recids = split_cli_ids_arg(task_get_option('recids', ''))
recids += cli_recids
query_params = {
'collection': task_get_option('collection', ''),
'field': task_get_option('field', ''),
'pattern': task_get_option('pattern', ''),
'matching': task_get_option('matching', '')
}
recids += query_records(query_params)
bibreformat_task(fmt, recids, without_fmt,
not task_has_option('noprocess'))
return True
def task_run_core():
"""Run the task by fetching arguments from the BibSched task queue.
This is what BibSched will be invoking via daemon call.
"""
fmts = task_get_option('format', 'HB,RECJSON')
for fmt in fmts.split(','):
last_updated = fetch_last_updated(fmt)
write_message("last stored run date is %s" % last_updated)
recids = intbitset()
if task_has_option("all"):
recids += all_records()
if task_has_option("last"):
recids += outdated_caches(fmt, last_updated)
if task_has_option('ignore_without'):
without_fmt = intbitset()
else:
without_fmt = missing_caches(fmt)
recids += without_fmt
cli_recids = split_cli_ids_arg(task_get_option('recids', ''))
recids += cli_recids
query_params = {'collection': task_get_option('collection', ''),
'field': task_get_option('field', ''),
'pattern': task_get_option('pattern', ''),
'matching': task_get_option('matching', '')}
recids += query_records(query_params)
bibreformat_task(fmt,
recids,
without_fmt,
not task_has_option('noprocess'))
return True
def task_submit_check_options():
"""Check that options are valid."""
if task_has_option('wjob'):
jobnames = task_get_option('wjob')
if jobnames:
jobnames = jobnames.split(',')
for jobname in jobnames:
res = run_sql("SELECT COUNT(*) FROM expJOB WHERE jobname=%s", (jobname,))
if res and res[0][0]:
# okay, jobname exists
pass
else:
write_message("Sorry, job name %s is not known. Exiting." % jobname)
return False
return True
def update_rule_last_run(rule_name):
"""
Set the last time a rule was run to now. This function should be called
after a rule has been ran.
"""
if task_has_option('record_ids') or task_get_option('no_upload', False) \
or task_get_option('no_tickets', False):
return # We don't want to update the database in this case
updated = run_sql("UPDATE bibcheck_rules SET last_run=%s WHERE name=%s;",
(task_get_task_param('task_starting_time'), rule_name,))
if not updated: # rule not in the database, insert it
run_sql("INSERT INTO bibcheck_rules(name, last_run) VALUES (%s, %s)",
(rule_name, task_get_task_param('task_starting_time')))
def task_submit_check_options():
"""Check that options are valid."""
if task_has_option('wjob'):
jobnames = task_get_option('wjob')
if jobnames:
jobnames = jobnames.split(',')
for jobname in jobnames:
res = run_sql("SELECT COUNT(*) FROM expJOB WHERE jobname=%s",
(jobname, ))
if res and res[0][0]:
# okay, jobname exists
pass
else:
write_message("Sorry, job name %s is not known. Exiting." %
jobname)
return False
return True
def task_submit_check_options():
"""
NOTE: Depending on the parameters, either "BibSched mode" or plain
straigh-forward execution mode is entered.
"""
if task_has_option("create_event_with_id"):
print(webstat.create_customevent(task_get_option("create_event_with_id"),
task_get_option("event_name", None),
task_get_option("column_headers", [])))
sys.exit(0)
elif task_has_option("destroy_event_with_id"):
print(webstat.destroy_customevent(task_get_option("destroy_event_with_id")))
sys.exit(0)
elif task_has_option("list_events"):
events = webstat._get_customevents()
if len(events) == 0:
print("There are no custom events available.")
else:
print("Available custom events are:\n")
print('\n'.join([x[0] + ": " + ((x[1] == None) and "No descriptive name" or str(x[1])) for x in events]))
sys.exit(0)
elif task_has_option("cache_events"):
events = task_get_option("cache_events")
write_message(str(events), verbose=9)
if events[0] == 'ALL':
keyevents_to_cache = webstat.KEYEVENT_REPOSITORY.keys()
customevents_to_cache = [x[0] for x in webstat._get_customevents()]
elif events[0] == 'KEYEVENTS':
keyevents_to_cache = webstat.KEYEVENT_REPOSITORY.keys()
customevents_to_cache = []
elif events[0] == 'CUSTOMEVENTS':
keyevents_to_cache = []
customevents_to_cache = [x[0] for x in webstat._get_customevents()]
elif events[0] != '':
keyevents_to_cache = [x for x in webstat.KEYEVENT_REPOSITORY.keys() if x in events]
customevents_to_cache = [x[0] for x in webstat._get_customevents() if x in events]
# Control so that we have valid event names
if len(keyevents_to_cache + customevents_to_cache) == 0:
# Oops, no events. Abort and display help.
return False
else:
task_set_option("keyevents", keyevents_to_cache)
task_set_option("customevents", customevents_to_cache)
return True
elif task_has_option("dump_config"):
print("""\
[general]
visitors_box = True
search_box = True
record_box = True
bibsched_box = True
basket_box = True
apache_box = True
uptime_box = True
[webstat_custom_event_1]
name = baskets
param1 = action
param2 = basket
param3 = user
[apache_log_analyzer]
profile = nil
nb-histogram-items-to-print = 20
exclude-ip-list = ("137.138.249.162")
home-collection = "Atlantis Institute of Fictive Science"
search-interface-url = "/?"
detailed-record-url = "/%s/"
search-engine-url = "/search?"
search-engine-url-old-style = "/search.py?"
basket-url = "/yourbaskets/"
add-to-basket-url = "/yourbaskets/add"
display-basket-url = "/yourbaskets/display"
display-public-basket-url = "/yourbaskets/display_public"
alert-url = "/youralerts/"
display-your-alerts-url = "/youralerts/list"
display-your-searches-url = "/youralerts/display"
""" % CFG_SITE_RECORD)
sys.exit(0)
elif task_has_option("load_config"):
from ConfigParser import ConfigParser
conf = ConfigParser()
conf.read(CFG_WEBSTAT_CONFIG_PATH)
for section in conf.sections():
if section[:21] == "webstat_custom_event_":
cols = []
name = ""
for option, value in conf.items(section):
if option == "name":
name = value
if option[:5] == "param":
# add the column name in it's position
index = int(option[-1]) - 1
while len(cols) <= index:
cols.append("")
cols[index] = value
if name:
res = run_sql("SELECT COUNT(id) FROM staEVENT WHERE id = %s", (name, ))
if res[0][0] == 0:
# name does not exist, create customevent
webstat.create_customevent(name, name, cols)
else:
# name already exists, update customevent
webstat.modify_customevent(name, cols=cols)
sys.exit(0)
else:
# False means that the --help should be displayed
return False
def bibreformat_task(fmt, recids, without_fmt, process):
"""BibReformat main task.
@param fmt: output format to use
@param process:
@param recids: a list of record IDs to reformat
@return: None
"""
write_message("Processing format %s" % fmt)
t1 = os.times()[4]
start_date = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
latest_bibrank_run = get_bibrankmethod_lastupdate('citation')
def related_records(recids, recids_processed):
if fmt == "HDREF" and recids:
# HDREF represents the references tab
# the tab needs to be recomputed not only when the record changes
# but also when one of the citations changes
sql = """SELECT id, modification_date FROM bibrec
WHERE id in (%s)""" % ','.join(str(r) for r in recids)
def check_date(mod_date):
return mod_date.strftime(
"%Y-%m-%d %H:%M:%S") < latest_bibrank_run
rel_recids = intbitset([
recid for recid, mod_date in run_sql(sql)
if check_date(mod_date)
])
for r in rel_recids:
recids |= intbitset(get_cited_by(r))
# To not process recids twice
recids -= recids_processed
# Adds to the set of processed recids
recids_processed += recids
return recids
def recid_chunker(recids):
recids_processed = intbitset()
chunk = intbitset()
for recid in recids:
if len(chunk) == 5000:
for r in related_records(chunk, recids_processed):
yield r
recids_processed += chunk
chunk = intbitset()
if recid not in recids_processed:
chunk.add(recid)
if chunk:
for r in related_records(chunk, recids_processed):
yield r
recIDs = list(recid_chunker(recids))
### list of corresponding record IDs was retrieved
### now format the selected records
if without_fmt:
write_message("Records to be processed: %d" % len(recIDs))
write_message("Out of it records without existing cache: %d" %
len(without_fmt))
else:
write_message("Records to be processed: %d" % len(recIDs))
### Initialize main loop
total_rec = 0 # Total number of records
tbibformat = 0 # time taken up by external call
tbibupload = 0 # time taken up by external call
### Iterate over all records prepared in lists I (option)
if process:
total_rec_1, tbibformat_1, tbibupload_1 = iterate_over_new(recIDs, fmt)
total_rec += total_rec_1
tbibformat += tbibformat_1
tbibupload += tbibupload_1
### Store last run time
if task_has_option("last"):
write_message("storing run date to %s" % start_date)
store_last_updated(fmt, start_date)
### Final statistics
t2 = os.times()[4]
elapsed = t2 - t1
message = "total records processed: %d" % total_rec
write_message(message)
message = "total processing time: %2f sec" % elapsed
write_message(message)
message = "Time spent on external call (os.system):"
write_message(message)
message = " bibformat: %2f sec" % tbibformat
write_message(message)
message = " bibupload: %2f sec" % tbibupload
write_message(message)
def bibreformat_task(fmt, recids, without_fmt, process):
"""BibReformat main task.
@param fmt: output format to use
@param process:
@param recids: a list of record IDs to reformat
@return: None
"""
write_message("Processing format %s" % fmt)
t1 = os.times()[4]
start_date = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
latest_bibrank_run = get_bibrankmethod_lastupdate('citation')
def related_records(recids, recids_processed):
if fmt == "HDREF" and recids:
# HDREF represents the references tab
# the tab needs to be recomputed not only when the record changes
# but also when one of the citations changes
sql = """SELECT id, modification_date FROM bibrec
WHERE id in (%s)""" % ','.join(str(r) for r in recids)
def check_date(mod_date):
return mod_date.strftime(
"%Y-%m-%d %H:%M:%S") < latest_bibrank_run
rel_recids = intbitset([recid for recid, mod_date in run_sql(sql)
if check_date(mod_date)])
for r in rel_recids:
recids |= intbitset(get_cited_by(r))
# To not process recids twice
recids -= recids_processed
# Adds to the set of processed recids
recids_processed += recids
return recids
def recid_chunker(recids):
recids_processed = intbitset()
chunk = intbitset()
for recid in recids:
if len(chunk) == 5000:
for r in related_records(chunk, recids_processed):
yield r
recids_processed += chunk
chunk = intbitset()
if recid not in recids_processed:
chunk.add(recid)
if chunk:
for r in related_records(chunk, recids_processed):
yield r
recIDs = list(recid_chunker(recids))
### list of corresponding record IDs was retrieved
### now format the selected records
if without_fmt:
write_message("Records to be processed: %d" % len(recIDs))
write_message("Out of it records without existing cache: %d" %
len(without_fmt))
else:
write_message("Records to be processed: %d" % len(recIDs))
### Initialize main loop
total_rec = 0 # Total number of records
tbibformat = 0 # time taken up by external call
tbibupload = 0 # time taken up by external call
### Iterate over all records prepared in lists I (option)
if process:
total_rec_1, tbibformat_1, tbibupload_1 = iterate_over_new(recIDs, fmt)
total_rec += total_rec_1
tbibformat += tbibformat_1
tbibupload += tbibupload_1
### Store last run time
if task_has_option("last"):
write_message("storing run date to %s" % start_date)
store_last_updated(fmt, start_date)
### Final statistics
t2 = os.times()[4]
elapsed = t2 - t1
message = "total records processed: %d" % total_rec
write_message(message)
message = "total processing time: %2f sec" % elapsed
write_message(message)
message = "Time spent on external call (os.system):"
write_message(message)
message = " bibformat: %2f sec" % tbibformat
write_message(message)
message = " bibupload: %2f sec" % tbibupload
write_message(message)
def task_submit_check_options():
"""
NOTE: Depending on the parameters, either "BibSched mode" or plain
straigh-forward execution mode is entered.
"""
if task_has_option("create_event_with_id"):
print(
webstat.create_customevent(task_get_option("create_event_with_id"),
task_get_option("event_name", None),
task_get_option("column_headers", [])))
sys.exit(0)
elif task_has_option("destroy_event_with_id"):
print(
webstat.destroy_customevent(
task_get_option("destroy_event_with_id")))
sys.exit(0)
elif task_has_option("list_events"):
events = webstat._get_customevents()
if len(events) == 0:
print("There are no custom events available.")
else:
print("Available custom events are:\n")
print('\n'.join([
x[0] + ": " +
((x[1] == None) and "No descriptive name" or str(x[1]))
for x in events
]))
sys.exit(0)
elif task_has_option("cache_events"):
events = task_get_option("cache_events")
write_message(str(events), verbose=9)
if events[0] == 'ALL':
keyevents_to_cache = webstat.KEYEVENT_REPOSITORY.keys()
customevents_to_cache = [x[0] for x in webstat._get_customevents()]
elif events[0] == 'KEYEVENTS':
keyevents_to_cache = webstat.KEYEVENT_REPOSITORY.keys()
customevents_to_cache = []
elif events[0] == 'CUSTOMEVENTS':
keyevents_to_cache = []
customevents_to_cache = [x[0] for x in webstat._get_customevents()]
elif events[0] != '':
keyevents_to_cache = [
x for x in webstat.KEYEVENT_REPOSITORY.keys() if x in events
]
customevents_to_cache = [
x[0] for x in webstat._get_customevents() if x in events
]
# Control so that we have valid event names
if len(keyevents_to_cache + customevents_to_cache) == 0:
# Oops, no events. Abort and display help.
return False
else:
task_set_option("keyevents", keyevents_to_cache)
task_set_option("customevents", customevents_to_cache)
return True
elif task_has_option("dump_config"):
print("""\
[general]
visitors_box = True
search_box = True
record_box = True
bibsched_box = True
basket_box = True
apache_box = True
uptime_box = True
[webstat_custom_event_1]
name = baskets
param1 = action
param2 = basket
param3 = user
[apache_log_analyzer]
profile = nil
nb-histogram-items-to-print = 20
exclude-ip-list = ("137.138.249.162")
home-collection = "Atlantis Institute of Fictive Science"
search-interface-url = "/?"
detailed-record-url = "/%s/"
search-engine-url = "/search?"
search-engine-url-old-style = "/search.py?"
basket-url = "/yourbaskets/"
add-to-basket-url = "/yourbaskets/add"
display-basket-url = "/yourbaskets/display"
display-public-basket-url = "/yourbaskets/display_public"
alert-url = "/youralerts/"
display-your-alerts-url = "/youralerts/list"
display-your-searches-url = "/youralerts/display"
""" % CFG_SITE_RECORD)
sys.exit(0)
elif task_has_option("load_config"):
from ConfigParser import ConfigParser
conf = ConfigParser()
conf.read(CFG_WEBSTAT_CONFIG_PATH)
for section in conf.sections():
if section[:21] == "webstat_custom_event_":
cols = []
name = ""
for option, value in conf.items(section):
if option == "name":
name = value
if option[:5] == "param":
# add the column name in it's position
index = int(option[-1]) - 1
while len(cols) <= index:
cols.append("")
cols[index] = value
if name:
res = run_sql(
"SELECT COUNT(id) FROM staEVENT WHERE id = %s",
(name, ))
if res[0][0] == 0:
# name does not exist, create customevent
webstat.create_customevent(name, name, cols)
else:
# name already exists, update customevent
webstat.modify_customevent(name, cols=cols)
sys.exit(0)
else:
# False means that the --help should be displayed
return False
def bibreformat_task(fmt, sql, sql_queries, cds_query, process_format, process, recids):
"""
BibReformat main task
@param fmt: output format to use
@param sql: dictionary with pre-created sql queries for various cases (for selecting records). Some of these queries will be picked depending on the case
@param sql_queries: a list of sql queries to be executed to select records to reformat.
@param cds_query: a search query to be executed to select records to reformat
@param process_format:
@param process:
@param recids: a list of record IDs to reformat
@return: None
"""
write_message("Processing format %s" % fmt)
t1 = os.times()[4]
start_date = datetime.now()
### Query the database
###
task_update_progress('Fetching records to process')
if process_format: # '-without' parameter
write_message("Querying database for records without cache...")
without_format = without_fmt(sql)
recIDs = intbitset(recids)
if cds_query['field'] != "" or \
cds_query['collection'] != "" or \
cds_query['pattern'] != "":
write_message("Querying database (CDS query)...")
if cds_query['collection'] == "":
# use search_pattern() whenever possible, as it can search
# even in private collections
res = search_pattern(p=cds_query['pattern'],
f=cds_query['field'],
m=cds_query['matching'])
else:
# use perform_request_search when '-c' argument has been
# defined, as it is not supported by search_pattern()
res = intbitset(perform_request_search(req=None, of='id',
c=cds_query['collection'],
p=cds_query['pattern'],
f=cds_query['field']))
recIDs |= res
for sql_query in sql_queries:
write_message("Querying database (%s) ..." % sql_query, verbose=2)
recIDs |= intbitset(run_sql(sql_query))
if fmt == "HDREF" and recIDs:
# HDREF represents the references tab
# the tab needs to be recomputed not only when the record changes
# but also when one of the citations changes
latest_bibrank_run = get_bibrankmethod_lastupdate('citation')
start_date = latest_bibrank_run
sql = """SELECT id, modification_date FROM bibrec
WHERE id in (%s)""" % ','.join(str(r) for r in recIDs)
def check_date(mod_date):
return mod_date < latest_bibrank_run
recIDs = intbitset([recid for recid, mod_date in run_sql(sql) \
if check_date(mod_date)])
for r in recIDs:
recIDs |= intbitset(get_cited_by(r))
### list of corresponding record IDs was retrieved
### now format the selected records
if process_format:
write_message("Records to be processed: %d" % (len(recIDs) \
+ len(without_format)))
write_message("Out of it records without existing cache: %d" % len(without_format))
else:
write_message("Records to be processed: %d" % (len(recIDs)))
### Initialize main loop
total_rec = 0 # Total number of records
tbibformat = 0 # time taken up by external call
tbibupload = 0 # time taken up by external call
### Iterate over all records prepared in lists I (option)
if process:
if CFG_BIBFORMAT_USE_OLD_BIBFORMAT: # FIXME: remove this
# when migration from php to
# python bibformat is done
(total_rec_1, tbibformat_1, tbibupload_1) = iterate_over_old(recIDs,
fmt)
else:
(total_rec_1, tbibformat_1, tbibupload_1) = iterate_over_new(recIDs,
fmt)
total_rec += total_rec_1
tbibformat += tbibformat_1
tbibupload += tbibupload_1
### Iterate over all records prepared in list II (no_format)
if process_format and process:
if CFG_BIBFORMAT_USE_OLD_BIBFORMAT: # FIXME: remove this
# when migration from php to
# python bibformat is done
(total_rec_2, tbibformat_2, tbibupload_2) = iterate_over_old(without_format,
fmt)
else:
(total_rec_2, tbibformat_2, tbibupload_2) = iterate_over_new(without_format,
fmt)
total_rec += total_rec_2
tbibformat += tbibformat_2
tbibupload += tbibupload_2
### Store last run time
if task_has_option("last"):
write_message("storing run date to %s" % start_date)
store_last_updated(fmt, start_date)
### Final statistics
t2 = os.times()[4]
elapsed = t2 - t1
message = "total records processed: %d" % total_rec
write_message(message)
message = "total processing time: %2f sec" % elapsed
write_message(message)
message = "Time spent on external call (os.system):"
write_message(message)
message = " bibformat: %2f sec" % tbibformat
write_message(message)
message = " bibupload: %2f sec" % tbibupload
write_message(message)
def task_run_core():
"""Runs the task by fetching arguments from the BibSched task queue. This is what BibSched will be invoking via daemon call."""
## initialize parameters
if task_get_option('format'):
fmts = task_get_option('format')
else:
fmts = 'HB' # default value if no format option given
for fmt in fmts.split(','):
last_updated = fetch_last_updated(fmt)
write_message("last stored run date is %s" % last_updated)
sql = {
"all" : """SELECT br.id FROM bibrec AS br, bibfmt AS bf
WHERE bf.id_bibrec = br.id AND bf.format = '%s'""" % fmt,
"last": """SELECT br.id FROM bibrec AS br
INNER JOIN bibfmt AS bf ON bf.id_bibrec = br.id
WHERE br.modification_date >= '%(last_updated)s'
AND bf.format='%(format)s'
AND bf.last_updated < br.modification_date""" \
% {'format': fmt,
'last_updated': last_updated.strftime('%Y-%m-%d %H:%M:%S')},
"missing" : """SELECT br.id
FROM bibrec as br
LEFT JOIN bibfmt as bf
ON bf.id_bibrec = br.id AND bf.format ='%s'
WHERE bf.id_bibrec IS NULL
AND br.id BETWEEN %%s AND %%s
""" % fmt,
}
sql_queries = []
cds_query = {}
if task_has_option("all"):
sql_queries.append(sql['all'])
if task_has_option("last"):
sql_queries.append(sql['last'])
if task_has_option("collection"):
cds_query['collection'] = task_get_option('collection')
else:
cds_query['collection'] = ""
if task_has_option("field"):
cds_query['field'] = task_get_option('field')
else:
cds_query['field'] = ""
if task_has_option("pattern"):
cds_query['pattern'] = task_get_option('pattern')
else:
cds_query['pattern'] = ""
if task_has_option("matching"):
cds_query['matching'] = task_get_option('matching')
else:
cds_query['matching'] = ""
if task_has_option("recids"):
recids = list(split_cli_ids_arg(task_get_option('recids')))
else:
recids = []
### sql commands to be executed during the script run
###
bibreformat_task(fmt, sql, sql_queries, cds_query, task_has_option('without'), not task_has_option('noprocess'), recids)
return True
