Homec4science
Diffusion invenio-infoscience (maint-1.2) Browse

R3600/modules/miscutil/lib/inveniocfg_dumperloader.py
9036d847ae65maint-1.2

Last Change
Raw File

inveniocfg_dumperloader.py
View Options

# -*- coding: utf-8 -*-
##
## This file is part of Invenio.
## Copyright (C) 2010, 2011,  CERN.
##
## 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.
##
## 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 Invenio; if not, write to the Free Software Foundation, Inc.,
## 59 Temple Place, Suite 330, Boston, MA 02111-1307, USA.
"""
*** HIGHLY EXPERIMENTAL; PLEASE DO NOT USE. ***
Invenio configuration dumper and loader CLI tool.
Usage: python inveniocfg_dumperloader.py [options]
General options:
-h, --help print this help
-V, --version print version number
Dumper options:
-d file dump the collections into a INI file
-col COLLECTION1,COLLECTION2... collection/s to dump
-all dump all the collections
--force-ids also dump the ids of the tables to the file
--output print the file in the screen
Loader options:
-l file load a file into the database
-mode i|c|r select the mode to load(insert, correct, replace)
"""
__revision__ = "$Id$"
import sys
import random
import re
import datetime
import StringIO
from string import Template
from invenio.dbquery import run_sql, wash_table_column_name
from configobj import ConfigObj
IDENT_TYPE = " " #Identation in the *.INI file can be a tab/spaces/etc...
MESSAGES = [] #List of messages to display to the user at the end of the execution
LOAD_DEFAULT_MODE = 'i'
SEPARATOR = '.'
#Dict of blacklisted fields and the message to display
BLACKLIST_TABLE_COLUMNS = {
'collection.reclist': '#INFO Please rerun webcoll.',
'accROLE.firefole_def_ser': '#INFO Please rerun webaccessadmin -c.',
'score':'#INFO Run whatever relevant',
'tag.value':'#INFO Please run inveniocfg --do-something',
'field_tag.score':'#INFO please run inveniocfg --fill-scores'
}
COLLECTIONS = {
'FIELD' : {
'tables':{
'field':'extend(field.id=fieldname.id_field,fieldname.$ln.$type = $value)',
'field_tag':'field_tag.id_field = field.id, field_tag.id_tag = tag.id',
'tag':'normal'},
'relations':'field-field_tag-tag',
},
'COLLECTION' : {
'tables':{
'collection':'normal',
'collection_example':'collection_example.id_example = example.id, collection_example.id_collection = collection.id',
'example':'normal'},
'relations':'collection-collection_example-example',
},
'PORTALBOX' : {
'tables':{
'collection':'normal',
'collection_portalbox':'collection_portalbox.id_portalbox = portalbox.id, collection_portalbox.id_collection = collection.id',
'portalbox':'normal',
},
'relations':'collection-collection_portalbox-portalbox',
},
}
def print_usage():
"""Print help."""
print __doc__
def create_section_id(num, with_date=True):
"""
Generate a unique section id.
Convert the given number in base 18 and append a 5 digit random string
If with_date=True append the date at the beginnig so it can be ordered.
Estructure:
if with_date:
date . base18(id) . 5 random chars e.g. tag.2010-07-30.ddcbz2lf
else:
base18(id) . 5 random chars e.g. field.ddcbz2lf
"""
digits = "abcdefghijklmnopqrstuvwxyz0123456789"
str_id = ""
tail = ''.join([random.choice(digits) for x in range(4)])
while 1:
rest = num % 18
str_id = digits[rest] + str_id
num = num / 18
if num == 0:
break
if with_date == True:
date = str(datetime.date.today())
return date + "." + str_id + tail
return str_id + tail
def dict2db(table_name, dict_data, mode):
"""
Load the dict values into the database
Three modes of operation:
i - insert
r - replace
c - correct
"""
#Escape all the content in dict data to avoid " and '
for data in dict_data:
dict_data[data] = re.escape(dict_data[data])
if mode == 'i': #Insert mode
query_fields = " , " .join(dict_data.keys())
query_values = "' , '" .join(dict_data.values())
query = "INSERT IGNORE INTO %s(%s) VALUES ('%s')" % (wash_table_column_name(table_name),
query_fields,
query_values)
elif mode == 'c': #Correct mode
if '_' in table_name:
query = "SELECT * FROM %s" % table_name#FIXIT Trick to execute something instead of giving error
else:
tbl_id = get_primary_keys(table_name)[0]
del dict_data[tbl_id]
query_update = " , " .join(["%s=\'%s\'" % (field, dict_data[field]) for field in dict_data])
query = "UPDATE %s SET %s" % (wash_table_column_name(table_name),
query_update)
else: #Try in the default mode
dict2db(table_name, dict_data, LOAD_DEFAULT_MODE)
try:
run_sql(query)
except:
print "VALUES: %s ALREADY EXIST IN TABLE %s. SKIPPING" % (query_values, table_name)
pass
def query2list(query, table_name):
"""Given a SQL query return a list of dictionaries with the results"""
results = run_sql(query, with_desc=1)
lst_results = []
dict_results = {}
for section_id, result in enumerate(results[0]):
dict_results = {}
for index, field in enumerate(results[1]):
if not is_blacklisted(table_name, field[0]):
dict_results[field[0]] = result[index]
lst_results.append(dict_results)
return lst_results
def get_primary_keys(table_name):
"""
Get the primary keys from the table with the DESC mysql function
"""
lst_keys = []
query = "DESC %s" % wash_table_column_name(table_name)
results = run_sql(query)
for field in results:
if field[3] == 'PRI':
lst_keys.append(field[0])
return lst_keys
def get_unused_primary_key(table_name):
"""
Returns the first free id from a table
"""
table_id = get_primary_keys(table_name)[0]#FIXIT the table can have more than an id
query = "SELECT %s FROM %s" % (table_id, table_name)
results = query2list(query, table_name)
list_used_ids = [result[table_id] for result in results]
for unused_id in range(1, len(list_used_ids)+2):
if not unused_id in list_used_ids:
return str(unused_id)
def is_blacklisted(table, field):
"""
Check if the current field is blacklisted, if so add the message to the messages list
"""
if (table+ "." + field) in BLACKLIST_TABLE_COLUMNS.keys():
msg = BLACKLIST_TABLE_COLUMNS[(table + "." + field)]
if not msg in MESSAGES:
MESSAGES.append(msg)
return True
return False
def get_relationship(collection, table, field_id):
"""Return the name of the related field"""
tbl_field = table + "." + field_id
dict_relationship = {}
for tbl in collection['tables'].values():
if tbl_field in tbl:
for foo in tbl.split(","):
dict_value, dict_key = foo.split("=")
dict_relationship[dict_key.strip()] = dict_value
return dict_relationship
def delete_keys_from_dict(dict_del, lst_keys):
"""
Delete the keys present in the lst_keys from the dictionary.
Loops recursively over nested dictionaries.
"""
for k in lst_keys:
try:
del dict_del[k]
except KeyError:
pass
for v in dict_del.values():
if isinstance(v, dict):
delete_keys_from_dict(v, lst_keys)
return dict_del
def extract_from_template(template, str_data):
"""
Extract the values from a string given the template
If the template and the string are different, this function may fail
Return a dictionary with the keys from the template and the values from the string
"""
#FIXIT this code can be more elegant
lst_str_data = []
dict_result = {}
pattern = re.compile("\$\w*")
patt_match = pattern.findall(template)
lst_foo = str_data.split("=")
for data in lst_foo:
lst_str_data.extend(data.split("."))
for index, data in enumerate(patt_match):
data = data.replace('$','')
dict_result[data] = lst_str_data[index+1].strip()
return dict_result
def delete_ids(dict_fields, lst_tables):
"""
Remove the ids of the tables from the dictionary
"""
lst_primary = []
for tbl in lst_tables:
lst_primary.extend(get_primary_keys(tbl))
return delete_keys_from_dict(dict_fields, lst_primary)
def add_special_field(collection, tbl_name , dict_data):
"""Add the value for the translation to the dictionary"""
str_template = collection['tables'][tbl_name].split(",")[1][:-1]#FIXIT if the final character is other?
template_key, template_value = str_template.split("=")
template_key = Template(template_key.strip())
template_value = Template(template_value.strip())
id_field = dict_data['id']
query = "SELECT * FROM %s WHERE %s=%s" % ("fieldname", "id_field", id_field)
result = query2list(query, "fieldname")
if result:
for res in result:
dict_data[template_key.safe_substitute(res)] = template_value.safe_substitute(res)
def dump_collection(collection, config, force_ids, print_to_screen=False):
"""
Dump the current collection
Note: there are a special notation, ori(origin) - rel(relation) - fin(final)
For example in the relation field-field_tag-tag:
ori(origin): field table
rel(relation): field_tag
fin(final): tag
"""
tbl_ori, tbl_rel, tbl_fin = collection['relations'].split("-")
query = "SELECT * FROM %s" % (wash_table_column_name(tbl_ori))
lst_ori = query2list(query, tbl_ori)
tbl_ori_id = get_primary_keys(tbl_ori)[0]
for index_ori, result_ori in enumerate(lst_ori):
dict_rels = get_relationship(collection, tbl_ori, tbl_ori_id)
query = "SELECT * FROM %s WHERE %s=%s" % (wash_table_column_name(tbl_rel),
dict_rels[tbl_ori+"."+tbl_ori_id],
result_ori[tbl_ori_id])
if collection['tables'][tbl_ori].startswith('extend'):
add_special_field(collection, tbl_ori, result_ori)
lst_rel = query2list(query, tbl_rel)
for result_rel in lst_rel:
tbl_fin_id = get_primary_keys(tbl_fin)[0]
tbl_rel_id = dict_rels[tbl_fin+"."+tbl_fin_id].split(".")[1].strip()
query = "SELECT * FROM %s WHERE %s=%s" % (wash_table_column_name(tbl_fin),
tbl_fin_id, result_rel[tbl_rel_id])
lst_fin = query2list(query, tbl_fin)
for index_fin, result_fin in enumerate(lst_fin):
result_ori[tbl_fin+"."+create_section_id(index_fin, with_date=False)] = result_fin
section_name = tbl_ori + "." + create_section_id(index_ori)
if force_ids == False:#Remove the ids from the dict
results = delete_ids(result_ori, collection['relations'].split("-"))
config[section_name] = results
else:
config[section_name] = result_ori
if print_to_screen == True:
output = StringIO.StringIO()
config.write(output)#Write to the output string instead of the file
print output.getvalue()
else:
config.write()
def get_collection(table_name):
"""Get the collection asociated with the section"""
for collection in COLLECTIONS.items():
if table_name in collection[1]['relations'].split("-")[0]:
return COLLECTIONS[collection[0]]#this is the collection to load
def load_section(section_name, dict_data, mode):
"""
Load the section back into the database
table_name is the name of the main section
There are some special notation: ori(origin) - rel(related) - fin(final) - ext(extended)
For example for the field-tag collection:
ori: field
ext: fieldname
rel: field_tag
fin:tag
"""
table_ori = section_name.split(".")[0]
collection = get_collection(table_ori)
ori_definition = collection['tables'][table_ori]
if ori_definition.startswith("extend"):
tbl_ext_name = ori_definition.split(",")[1].split(SEPARATOR)[0]
lst_tables = collection['relations'].split("-")
ori_id = get_primary_keys(lst_tables[0])[0]
ori_id_value = get_unused_primary_key(lst_tables[0])
dict_data[ori_id] = ori_id_value#Add the calculated id to the dictionary
#I will separate the dict_data into these 3 dicts corresponding to 3 different tables
dict_ori = {}
dict_rel = {}
dict_ext = {}
for field in dict_data:
if type(dict_data[field]) == str:#the field is a string
if "tbl_ext_name" in locals() and field.startswith(tbl_ext_name):#is extended table
dict2db("fieldname",
extract_from_template("fieldname.$ln.$type = $value",
str(field) + " = " + str(dict_data[field])),
mode)
else:
dict_ori[field] = dict_data[field]
else:#if the field is a dictionary
fin_id = get_primary_keys(lst_tables[2])[0]
fin_id_value = get_unused_primary_key(lst_tables[2])
dict_data[field][fin_id] = fin_id_value
dict2db(lst_tables[2], dict_data[field], mode)#Insert the final into the DB
fieldtag_ids = get_primary_keys(lst_tables[1])
dict_rel[fieldtag_ids[0]] = ori_id_value
dict_rel[fieldtag_ids[1]] = fin_id_value
dict2db(lst_tables[1], dict_rel, mode)#Insert the relation into the DB
dict2db(lst_tables[0], dict_ori, mode)
def cli_cmd_dump_config():
"""Dump the selected collection/s"""
config = ConfigObj(indent_type=IDENT_TYPE)
config.initial_comment = [
str(datetime.datetime.now()),
"This file is automatically generated by Invenio, running:",
" ".join(sys.argv) ,
"" ]
force_ids = False
if "--force-ids" in sys.argv:
force_ids = True
print_to_screen = False
if '--output' in sys.argv:
print_to_screen = True
try:
config.filename = sys.argv[sys.argv.index('-d') + 1]
except:
print_usage()
if '-col' in sys.argv:
try:
collection = COLLECTIONS[sys.argv[sys.argv.index('-col') + 1].upper()]
dump_collection(collection, config, force_ids, print_to_screen)
except:
print "ERROR: you must especify the collection to dump with the -col COLLECTION_NAME option"
elif '-all' in sys.argv:
for collection in COLLECTIONS:
dump_collection(COLLECTIONS[collection], config, force_ids, print_to_screen)
else:
print "Please specify the collection to dump"
def cli_cmd_load_config():
"""Load all the config sections back into the database"""
config = ConfigObj(sys.argv[sys.argv.index('-l') + 1])
mode = "r"
if '-mode' in sys.argv:
try:
mode = sys.argv[sys.argv.index('-mode') + 1]
if mode not in ['i', 'c', 'r']:
print "Not valid mode please select one of the following (i)nsert, (c)orrect or (r)eplace"
sys.exit(1)
except IndexError:
print "You must especify the mode with the -mode option"
sys.exit(1)
for section in config.sections:
load_section(section, config[section], mode)
def main():
"""
Main section, makes the calls to all the functions
"""
if "-d" in sys.argv:
cli_cmd_dump_config()
elif "-l" in sys.argv:
cli_cmd_load_config()
elif "-h" in sys.argv:
print_usage()
else:
print_usage()
for message in MESSAGES:
print message
if __name__ == '__main__':
main()
c4science · Help