get value of a given option from conf files
--conf-dir path to directory where invenio*.conf files are [optional]
--detect-system-details print system details such as Apache/Python/MySQL versions
"""
__revision__ = "$Id$"
from ConfigParser import ConfigParser
import os
import re
import shutil
import socket
import sys
def print_usage():
"""Print help."""
print __doc__
def print_version():
"""Print version information."""
print __revision__
def convert_conf_option(option_name, option_value):
"""
Convert conf option into Python config.py line, converting
values to ints or strings as appropriate.
"""
## 1) convert option name to uppercase:
option_name = option_name.upper()
## 2) convert option value to int or string:
if option_name in ['CFG_BIBUPLOAD_REFERENCE_TAG',
'CFG_BIBUPLOAD_EXTERNAL_SYSNO_TAG',
'CFG_BIBUPLOAD_EXTERNAL_OAIID_TAG',
'CFG_BIBUPLOAD_EXTERNAL_OAIID_PROVENANCE_TAG',
'CFG_BIBUPLOAD_STRONG_TAGS',
'CFG_BIBFORMAT_HIDDEN_TAGS',]:
# some options are supposed be string even when they look like
# numeric
option_value = '"' + option_value + '"'
else:
try:
option_value = int(option_value)
except ValueError:
option_value = '"' + option_value + '"'
## 3a) special cases: chars regexps
if option_name in ['CFG_BIBINDEX_CHARS_ALPHANUMERIC_SEPARATORS',
'CFG_BIBINDEX_CHARS_PUNCTUATION']:
option_value = 'r"[' + option_value[1:-1] + ']"'
## 3abis) special cases: real regexps
if option_name in ['CFG_BIBINDEX_PERFORM_OCR_ON_DOCNAMES',
'CFG_BATCHUPLOADER_WEB_ROBOT_AGENTS']:
option_value = 'r"' + option_value[1:-1] + '"'
## 3b) special cases: True, False, None
if option_value in ['"True"', '"False"', '"None"']:
option_value = option_value[1:-1]
- ## 3c) special cases: dicts
+ ## 3c) special cases: dicts and real pythonic lists
if option_name in ['CFG_WEBSEARCH_FIELDS_CONVERT',
'CFG_BATCHUPLOADER_WEB_ROBOT_RIGHTS',
'CFG_SITE_EMERGENCY_EMAIL_ADDRESSES',
'CFG_BIBMATCH_FUZZY_WORDLIMITS',
'CFG_BIBMATCH_QUERY_TEMPLATES',
'CFG_WEBSEARCH_SYNONYM_KBRS',
'CFG_BIBINDEX_SYNONYM_KBRS',
'CFG_WEBCOMMENT_EMAIL_REPLIES_TO',
'CFG_WEBCOMMENT_RESTRICTION_DATAFIELD',
'CFG_WEBCOMMENT_ROUND_DATAFIELD',
'CFG_BIBUPLOAD_FFT_ALLOWED_EXTERNAL_URLS',
'CFG_BIBSCHED_NODE_TASKS',
'CFG_BIBEDIT_EXTEND_RECORD_WITH_COLLECTION_TEMPLATE',
'CFG_OAI_METADATA_FORMATS',
'CFG_WEBSUBMIT_DESIRED_CONVERSIONS',
'CFG_BIBDOCFILE_BEST_FORMATS_TO_EXTRACT_TEXT_FROM',
- 'CFG_WEB_API_KEY_ALLOWED_URL']:
+ 'CFG_WEB_API_KEY_ALLOWED_URL',
+ 'CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_MISC',
+ 'CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_DOCTYPES',
+ 'CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_RESTRICTIONS']:
option_value = option_value[1:-1]
## 3cbis) very special cases: dicts with backward compatible string
if option_name in ['CFG_BIBINDEX_SPLASH_PAGES']:
if option_value.startswith('"{') and option_value.endswith('}"'):
option_value = option_value[1:-1]
else:
option_value = """{%s: ".*"}""" % option_value
## 3d) special cases: comma-separated lists
if option_name in ['CFG_SITE_LANGS',
'CFG_WEBSUBMIT_ADDITIONAL_KNOWN_FILE_EXTENSIONS',
'CFG_WEBSEARCH_USE_MATHJAX_FOR_FORMATS',
'CFG_BIBUPLOAD_STRONG_TAGS',
'CFG_BIBFORMAT_HIDDEN_TAGS',
'CFG_BIBSCHED_GC_TASKS_TO_REMOVE',
'CFG_BIBSCHED_GC_TASKS_TO_ARCHIVE',
'CFG_BIBUPLOAD_FFT_ALLOWED_LOCAL_PATHS',
'CFG_BIBUPLOAD_CONTROLLED_PROVENANCE_TAGS',
'CFG_BIBUPLOAD_DELETE_FORMATS',
'CFG_WEBSEARCH_ENABLED_SEARCH_INTERFACES',
'CFG_WEBSTYLE_HTTP_STATUS_ALERT_LIST',
'CFG_WEBSEARCH_RSS_I18N_COLLECTIONS',
'CFG_BATCHUPLOADER_FILENAME_MATCHING_POLICY',
'CFG_BIBAUTHORID_EXTERNAL_CLAIMED_RECORDS_KEY',
'CFG_PLOTEXTRACTOR_DISALLOWED_TEX',
'CFG_OAI_FRIENDS',
'CFG_WEBSTYLE_REVERSE_PROXY_IPS',
'CFG_BIBEDIT_AUTOCOMPLETE_INSTITUTIONS_FIELDS',
'CFG_BIBFORMAT_DISABLE_I18N_FOR_CACHED_FORMATS',
'CFG_BIBFORMAT_HIDDEN_FILE_FORMATS',]:
out = "["
for elem in option_value[1:-1].split(","):
if elem:
if option_name in ['CFG_WEBSEARCH_ENABLED_SEARCH_INTERFACES']:
# 3d1) integer values
out += "%i, " % int(elem)
else:
# 3d2) string values
out += "'%s', " % elem
out += "]"
option_value = out
## 3e) special cases: multiline
if option_name == 'CFG_OAI_IDENTIFY_DESCRIPTION':
# make triple quotes
option_value = '""' + option_value + '""'
## 3f) ignore some options:
if option_name.startswith('CFG_SITE_NAME_INTL'):
# treated elsewhere
return
## 3g) special cases: float
if option_name in ['CFG_BIBDOCFILE_MD5_CHECK_PROBABILITY',
'CFG_BIBMATCH_LOCAL_SLEEPTIME',
'CFG_BIBMATCH_REMOTE_SLEEPTIME',
'CFG_PLOTEXTRACTOR_DOWNLOAD_TIMEOUT',
'CFG_BIBMATCH_FUZZY_MATCH_VALIDATION_LIMIT']:
option_value = float(option_value[1:-1])
## 3h) special cases: bibmatch validation list
if option_name in ['CFG_BIBMATCH_MATCH_VALIDATION_RULESETS']:
option_value = option_value[1:-1]
## 4a) dropped variables
if option_name in ['CFG_BATCHUPLOADER_WEB_ROBOT_AGENT']:
print >> sys.stderr, ("""ERROR: CFG_BATCHUPLOADER_WEB_ROBOT_AGENT has been dropped in favour of
CFG_BATCHUPLOADER_WEB_ROBOT_AGENTS.
Please, update your invenio-local.conf file accordingly.""")
option_value = option_value[1:-1]
## 5) finally, return output line:
return '%s = %s' % (option_name, option_value)
def cli_cmd_update_config_py(conf):
"""
Update new config.py from conf options, keeping previous
config.py in a backup copy.
"""
print ">>> Going to update config.py..."
## location where config.py is:
configpyfile = conf.get("Invenio", "CFG_PYLIBDIR") + \
os.sep + 'invenio' + os.sep + 'config.py'
## backup current config.py file:
if os.path.exists(configpyfile):
shutil.copy(configpyfile, configpyfile + '.OLD')
## here we go:
fdesc = open(configpyfile, 'w')
## generate preamble:
fdesc.write("# -*- coding: utf-8 -*-\n")
fdesc.write("# DO NOT EDIT THIS FILE! IT WAS AUTOMATICALLY GENERATED\n")
fdesc.write("# FROM INVENIO.CONF BY EXECUTING:\n")
fdesc.write("# " + " ".join(sys.argv) + "\n")
## special treatment for CFG_SITE_NAME_INTL options:
fdesc.write("CFG_SITE_NAME_INTL = {}\n")
for lang in conf.get("Invenio", "CFG_SITE_LANGS").split(","):
fdesc.write("CFG_SITE_NAME_INTL['%s'] = \"%s\"\n" % (lang, conf.get("Invenio",
"CFG_SITE_NAME_INTL_" + lang)))
## special treatment for CFG_SITE_SECURE_URL that may be empty, in
## which case it should be put equal to CFG_SITE_URL:
if not conf.get("Invenio", "CFG_SITE_SECURE_URL"):
conf.set("Invenio", "CFG_SITE_SECURE_URL",
conf.get("Invenio", "CFG_SITE_URL"))
## process all the options normally:
sections = conf.sections()
sections.sort()
for section in sections:
options = conf.options(section)
options.sort()
for option in options:
if not option.startswith('CFG_DATABASE_'):
# put all options except for db credentials into config.py
line_out = convert_conf_option(option, conf.get(section, option))
if line_out:
fdesc.write(line_out + "\n")
## FIXME: special treatment for experimental variables
## CFG_WEBSEARCH_ENABLED_SEARCH_INTERFACES and CFG_WEBSEARCH_DEFAULT_SEARCH_INTERFACE
## (not offering them in invenio.conf since they will be refactored)
fdesc.write("CFG_WEBSEARCH_DEFAULT_SEARCH_INTERFACE = 0\n")
fdesc.write("CFG_WEBSEARCH_ENABLED_SEARCH_INTERFACES = [0, 1,]\n")
## generate postamble:
fdesc.write("")
fdesc.write("# END OF GENERATED FILE")
## we are done:
fdesc.close()
print "You may want to restart Apache now."
print ">>> config.py updated successfully."
def cli_cmd_update_dbquery_py(conf):
"""
Update lib/dbquery.py file with DB parameters read from conf file.
Note: this edits dbquery.py in situ, taking a backup first.
Use only when you know what you are doing.
"""
print ">>> Going to update dbquery.py..."
## location where dbquery.py is:
dbquerypyfile = conf.get("Invenio", "CFG_PYLIBDIR") + \
os.sep + 'invenio' + os.sep + 'dbquery.py'
## backup current dbquery.py file:
if os.path.exists(dbquerypyfile):
shutil.copy(dbquerypyfile, dbquerypyfile + '.OLD')
## replace db parameters:
out = ''
for line in open(dbquerypyfile, 'r').readlines():
match = re.search(r'^CFG_DATABASE_(HOST|PORT|NAME|USER|PASS|SLAVE)(\s*=\s*)\'.*\'$', line)
if match:
dbparam = 'CFG_DATABASE_' + match.group(1)
out += "%s%s'%s'\n" % (dbparam, match.group(2),
conf.get('Invenio', dbparam))
else:
out += line
fdesc = open(dbquerypyfile, 'w')
fdesc.write(out)
fdesc.close()
print "You may want to restart Apache now."
print ">>> dbquery.py updated successfully."
def cli_cmd_update_dbexec(conf):
"""
Update bin/dbexec file with DB parameters read from conf file.
Note: this edits dbexec in situ, taking a backup first.
Use only when you know what you are doing.
"""
print ">>> Going to update dbexec..."
## location where dbexec is:
dbexecfile = conf.get("Invenio", "CFG_BINDIR") + \
os.sep + 'dbexec'
## backup current dbexec file:
if os.path.exists(dbexecfile):
shutil.copy(dbexecfile, dbexecfile + '.OLD')
## replace db parameters via sed:
out = ''
for line in open(dbexecfile, 'r').readlines():
match = re.search(r'^CFG_DATABASE_(HOST|PORT|NAME|USER|PASS|SLAVE)(\s*=\s*)\'.*\'$', line)
if match:
dbparam = 'CFG_DATABASE_' + match.group(1)
out += "%s%s'%s'\n" % (dbparam, match.group(2),
conf.get("Invenio", dbparam))
else:
out += line
fdesc = open(dbexecfile, 'w')
fdesc.write(out)
fdesc.close()
print ">>> dbexec updated successfully."
def cli_cmd_update_bibconvert_tpl(conf):
"""
Update bibconvert/config/*.tpl files looking for 856
http://.../CFG_SITE_RECORD lines, replacing URL with CFG_SITE_URL taken
from conf file. Note: this edits tpl files in situ, taking a
backup first. Use only when you know what you are doing.
"""
print ">>> Going to update bibconvert templates..."
## location where bibconvert/config/*.tpl are:
tpldir = conf.get("Invenio", 'CFG_ETCDIR') + \
os.sep + 'bibconvert' + os.sep + 'config'
## find all *.tpl files:
for tplfilename in os.listdir(tpldir):
if tplfilename.endswith(".tpl"):
## change tpl file:
tplfile = tpldir + os.sep + tplfilename
shutil.copy(tplfile, tplfile + '.OLD')
out = ''
for line in open(tplfile, 'r').readlines():
match = re.search(r'^(.*)http://.*?/%s/(.*)$' % conf.get("Invenio", 'CFG_SITE_RECORD'), line)
if match:
out += "%s%s/%s/%s\n" % (match.group(1),
conf.get("Invenio", 'CFG_SITE_URL'),
conf.get("Invenio", 'CFG_SITE_RECORD'),
match.group(2))
else:
out += line
fdesc = open(tplfile, 'w')
fdesc.write(out)
fdesc.close()
print ">>> bibconvert templates updated successfully."
def cli_cmd_update_web_tests(conf):
"""
Update web test cases lib/webtest/test_*.html looking for
| http://.+?[>> Going to update web tests..."
## location where test_*.html files are:
testdir = conf.get("Invenio", 'CFG_PREFIX') + os.sep + \
'lib' + os.sep + 'webtest' + os.sep + 'invenio'
## find all test_*.html files:
for testfilename in os.listdir(testdir):
if testfilename.startswith("test_") and \
testfilename.endswith(".html"):
## change test file:
testfile = testdir + os.sep + testfilename
shutil.copy(testfile, testfile + '.OLD')
out = ''
for line in open(testfile, 'r').readlines():
match = re.search(r'^(.* | )http://.+?([)/opt/invenio(.*)$', line)
if match:
out += "%s%s%s\n" % (match.group(1),
conf.get("Invenio", 'CFG_PREFIX'),
match.group(2))
else:
out += line
fdesc = open(testfile, 'w')
fdesc.write(out)
fdesc.close()
print ">>> web tests updated successfully."
def cli_cmd_reset_sitename(conf):
"""
Reset collection-related tables with new CFG_SITE_NAME and
CFG_SITE_NAME_INTL* read from conf files.
"""
print ">>> Going to reset CFG_SITE_NAME and CFG_SITE_NAME_INTL..."
from invenio.dbquery import run_sql, IntegrityError
# reset CFG_SITE_NAME:
sitename = conf.get("Invenio", "CFG_SITE_NAME")
try:
run_sql("""INSERT INTO collection (id, name, dbquery, reclist) VALUES
(1,%s,NULL,NULL)""", (sitename,))
except IntegrityError:
run_sql("""UPDATE collection SET name=%s WHERE id=1""", (sitename,))
# reset CFG_SITE_NAME_INTL:
for lang in conf.get("Invenio", "CFG_SITE_LANGS").split(","):
sitename_lang = conf.get("Invenio", "CFG_SITE_NAME_INTL_" + lang)
try:
run_sql("""INSERT INTO collectionname (id_collection, ln, type, value) VALUES
(%s,%s,%s,%s)""", (1, lang, 'ln', sitename_lang))
except IntegrityError:
run_sql("""UPDATE collectionname SET value=%s
WHERE ln=%s AND id_collection=1 AND type='ln'""",
(sitename_lang, lang))
print "You may want to restart Apache now."
print ">>> CFG_SITE_NAME and CFG_SITE_NAME_INTL* reset successfully."
def cli_cmd_reset_recstruct_cache(conf):
"""If CFG_BIBUPLOAD_SERIALIZE_RECORD_STRUCTURE is changed, this function
will adapt the database to either store or not store the recstruct
format."""
from invenio.intbitset import intbitset
from invenio.dbquery import run_sql, serialize_via_marshal
from invenio.search_engine import get_record
from invenio.bibsched import server_pid, pidfile
enable_recstruct_cache = conf.get("Invenio", "CFG_BIBUPLOAD_SERIALIZE_RECORD_STRUCTURE")
enable_recstruct_cache = enable_recstruct_cache in ('True', '1')
pid = server_pid(ping_the_process=False)
if pid:
print >> sys.stderr, "ERROR: bibsched seems to run with pid %d, according to %s." % (pid, pidfile)
print >> sys.stderr, " Please stop bibsched before running this procedure."
sys.exit(1)
if enable_recstruct_cache:
print ">>> Searching records which need recstruct cache resetting; this may take a while..."
all_recids = intbitset(run_sql("SELECT id FROM bibrec"))
good_recids = intbitset(run_sql("SELECT bibrec.id FROM bibrec JOIN bibfmt ON bibrec.id = bibfmt.id_bibrec WHERE format='recstruct' AND modification_date < last_updated"))
recids = all_recids - good_recids
print ">>> Generating recstruct cache..."
tot = len(recids)
count = 0
for recid in recids:
value = serialize_via_marshal(get_record(recid))
run_sql("DELETE FROM bibfmt WHERE id_bibrec=%s AND format='recstruct'", (recid, ))
run_sql("INSERT INTO bibfmt(id_bibrec, format, last_updated, value) VALUES(%s, 'recstruct', NOW(), %s)", (recid, value))
count += 1
if count % 1000 == 0:
print " ... done records %s/%s" % (count, tot)
if count % 1000 != 0:
print " ... done records %s/%s" % (count, tot)
print ">>> recstruct cache generated successfully."
else:
print ">>> Cleaning recstruct cache..."
run_sql("DELETE FROM bibfmt WHERE format='recstruct'")
def cli_cmd_reset_siteadminemail(conf):
"""
Reset user-related tables with new CFG_SITE_ADMIN_EMAIL read from conf files.
"""
print ">>> Going to reset CFG_SITE_ADMIN_EMAIL..."
from invenio.dbquery import run_sql
siteadminemail = conf.get("Invenio", "CFG_SITE_ADMIN_EMAIL")
run_sql("DELETE FROM user WHERE id=1")
run_sql("""INSERT INTO user (id, email, password, note, nickname) VALUES
(1, %s, AES_ENCRYPT(email, ''), 1, 'admin')""",
(siteadminemail,))
print "You may want to restart Apache now."
print ">>> CFG_SITE_ADMIN_EMAIL reset successfully."
def cli_cmd_reset_fieldnames(conf):
"""
Reset I18N field names such as author, title, etc and other I18N
ranking method names such as word similarity. Their translations
are taken from the PO files.
"""
print ">>> Going to reset I18N field names..."
from invenio.messages import gettext_set_language, language_list_long
from invenio.dbquery import run_sql, IntegrityError
## get field id and name list:
field_id_name_list = run_sql("SELECT id, name FROM field")
## get rankmethod id and name list:
rankmethod_id_name_list = run_sql("SELECT id, name FROM rnkMETHOD")
## update names for every language:
for lang, dummy in language_list_long():
_ = gettext_set_language(lang)
## this list is put here in order for PO system to pick names
## suitable for translation
field_name_names = {"any field": _("any field"),
"title": _("title"),
"author": _("author"),
"abstract": _("abstract"),
"keyword": _("keyword"),
"report number": _("report number"),
"subject": _("subject"),
"reference": _("reference"),
"fulltext": _("fulltext"),
"collection": _("collection"),
"division": _("division"),
"year": _("year"),
"journal": _("journal"),
"experiment": _("experiment"),
"record ID": _("record ID")}
## update I18N names for every language:
for (field_id, field_name) in field_id_name_list:
if field_name_names.has_key(field_name):
try:
run_sql("""INSERT INTO fieldname (id_field,ln,type,value) VALUES
(%s,%s,%s,%s)""", (field_id, lang, 'ln',
field_name_names[field_name]))
except IntegrityError:
run_sql("""UPDATE fieldname SET value=%s
WHERE id_field=%s AND ln=%s AND type=%s""",
(field_name_names[field_name], field_id, lang, 'ln',))
## ditto for rank methods:
rankmethod_name_names = {"wrd": _("word similarity"),
"demo_jif": _("journal impact factor"),
"citation": _("times cited"),
"citerank_citation_t": _("time-decay cite count"),
"citerank_pagerank_c": _("all-time-best cite rank"),
"citerank_pagerank_t": _("time-decay cite rank"),}
for (rankmethod_id, rankmethod_name) in rankmethod_id_name_list:
if rankmethod_name_names.has_key(rankmethod_name):
try:
run_sql("""INSERT INTO rnkMETHODNAME (id_rnkMETHOD,ln,type,value) VALUES
(%s,%s,%s,%s)""", (rankmethod_id, lang, 'ln',
rankmethod_name_names[rankmethod_name]))
except IntegrityError:
run_sql("""UPDATE rnkMETHODNAME SET value=%s
WHERE id_rnkMETHOD=%s AND ln=%s AND type=%s""",
(rankmethod_name_names[rankmethod_name], rankmethod_id, lang, 'ln',))
print ">>> I18N field names reset successfully."
def cli_check_openoffice(conf):
"""
If OpenOffice.org integration is enabled, checks whether the system is
properly configured.
"""
from invenio.bibtask import check_running_process_user
from invenio.websubmit_file_converter import can_unoconv, get_file_converter_logger
logger = get_file_converter_logger()
for handler in logger.handlers:
logger.removeHandler(handler)
check_running_process_user()
print ">>> Checking if Libre/OpenOffice.org is correctly integrated...",
sys.stdout.flush()
if can_unoconv(True):
print "ok"
else:
sys.exit(1)
def test_db_connection():
"""
Test DB connection, and if fails, advise user how to set it up.
Useful to be called during table creation.
"""
print "Testing DB connection...",
from invenio.textutils import wrap_text_in_a_box
from invenio.dbquery import run_sql, Error
## first, test connection to the DB server:
try:
run_sql("SHOW TABLES")
except Error, err:
from invenio.dbquery import CFG_DATABASE_HOST, CFG_DATABASE_PORT, \
CFG_DATABASE_NAME, CFG_DATABASE_USER, CFG_DATABASE_PASS
print wrap_text_in_a_box("""\
DATABASE CONNECTIVITY ERROR %(errno)d: %(errmsg)s.\n
Perhaps you need to set up database and connection rights?
If yes, then please login as MySQL admin user and run the
following commands now:
$ mysql -h %(dbhost)s -P %(dbport)s -u root -p mysql
mysql> CREATE DATABASE %(dbname)s DEFAULT CHARACTER SET utf8;
mysql> GRANT ALL PRIVILEGES ON %(dbname)s.*
TO %(dbuser)s@%(webhost)s IDENTIFIED BY '%(dbpass)s';
mysql> QUIT
The values printed above were detected from your
configuration. If they are not right, then please edit your
invenio-local.conf file and rerun 'inveniocfg --update-all' first.
If the problem is of different nature, then please inspect
the above error message and fix the problem before continuing.""" % \
{'errno': err.args[0],
'errmsg': err.args[1],
'dbname': CFG_DATABASE_NAME,
'dbhost': CFG_DATABASE_HOST,
'dbport': CFG_DATABASE_PORT,
'dbuser': CFG_DATABASE_USER,
'dbpass': CFG_DATABASE_PASS,
'webhost': CFG_DATABASE_HOST == 'localhost' and 'localhost' or os.popen('hostname -f', 'r').read().strip(),
})
sys.exit(1)
print "ok"
## second, test insert/select of a Unicode string to detect
## possible Python/MySQL/MySQLdb mis-setup:
print "Testing Python/MySQL/MySQLdb UTF-8 chain...",
try:
beta_in_utf8 = "β" # Greek beta in UTF-8 is 0xCEB2
run_sql("CREATE TEMPORARY TABLE test__invenio__utf8 (x char(1), y varbinary(2)) DEFAULT CHARACTER SET utf8")
run_sql("INSERT INTO test__invenio__utf8 (x, y) VALUES (%s, %s)", (beta_in_utf8, beta_in_utf8))
res = run_sql("SELECT x,y,HEX(x),HEX(y),LENGTH(x),LENGTH(y),CHAR_LENGTH(x),CHAR_LENGTH(y) FROM test__invenio__utf8")
assert res[0] == ('\xce\xb2', '\xce\xb2', 'CEB2', 'CEB2', 2L, 2L, 1L, 2L)
run_sql("DROP TEMPORARY TABLE test__invenio__utf8")
except Exception, err:
print wrap_text_in_a_box("""\
DATABASE RELATED ERROR %s\n
A problem was detected with the UTF-8 treatment in the chain
between the Python application, the MySQLdb connector, and
the MySQL database. You may perhaps have installed older
versions of some prerequisite packages?\n
Please check the INSTALL file and please fix this problem
before continuing.""" % err)
sys.exit(1)
print "ok"
def cli_cmd_create_tables(conf):
"""Create and fill Invenio DB tables. Useful for the installation process."""
print ">>> Going to create and fill tables..."
from invenio.config import CFG_PREFIX
test_db_connection()
for cmd in ["%s/bin/dbexec < %s/lib/sql/invenio/tabcreate.sql" % (CFG_PREFIX, CFG_PREFIX),
"%s/bin/dbexec < %s/lib/sql/invenio/tabfill.sql" % (CFG_PREFIX, CFG_PREFIX)]:
if os.system(cmd):
print "ERROR: failed execution of", cmd
sys.exit(1)
cli_cmd_reset_sitename(conf)
cli_cmd_reset_siteadminemail(conf)
cli_cmd_reset_fieldnames(conf)
for cmd in ["%s/bin/webaccessadmin -u admin -c -a" % CFG_PREFIX]:
if os.system(cmd):
print "ERROR: failed execution of", cmd
sys.exit(1)
print ">>> Tables created and filled successfully."
def cli_cmd_load_webstat_conf(conf):
print ">>> Going to load WebStat config..."
from invenio.config import CFG_PREFIX
cmd = "%s/bin/webstatadmin --load-config" % CFG_PREFIX
if os.system(cmd):
print "ERROR: failed execution of", cmd
sys.exit(1)
print ">>> WebStat config load successfully."
def cli_cmd_drop_tables(conf):
"""Drop Invenio DB tables. Useful for the uninstallation process."""
print ">>> Going to drop tables..."
from invenio.config import CFG_PREFIX
from invenio.textutils import wrap_text_in_a_box, wait_for_user
from invenio.webstat import destroy_customevents
wait_for_user(wrap_text_in_a_box("""WARNING: You are going to destroy
your database tables!"""))
msg = destroy_customevents()
if msg:
print msg
cmd = "%s/bin/dbexec < %s/lib/sql/invenio/tabdrop.sql" % (CFG_PREFIX, CFG_PREFIX)
if os.system(cmd):
print "ERROR: failed execution of", cmd
sys.exit(1)
print ">>> Tables dropped successfully."
def cli_cmd_create_demo_site(conf):
"""Create demo site. Useful for testing purposes."""
print ">>> Going to create demo site..."
from invenio.config import CFG_PREFIX
from invenio.dbquery import run_sql
run_sql("TRUNCATE schTASK")
run_sql("TRUNCATE session")
run_sql("DELETE FROM user WHERE email=''")
for cmd in ["%s/bin/dbexec < %s/lib/sql/invenio/democfgdata.sql" % \
(CFG_PREFIX, CFG_PREFIX),]:
if os.system(cmd):
print "ERROR: failed execution of", cmd
sys.exit(1)
cli_cmd_reset_fieldnames(conf) # needed for I18N demo ranking method names
for cmd in ["%s/bin/webaccessadmin -u admin -c -r -D" % CFG_PREFIX,
"%s/bin/webcoll -u admin" % CFG_PREFIX,
"%s/bin/webcoll 1" % CFG_PREFIX,
"%s/bin/bibsort -u admin --load-config" % CFG_PREFIX,
"%s/bin/bibsort 2" % CFG_PREFIX, ]:
if os.system(cmd):
print "ERROR: failed execution of", cmd
sys.exit(1)
print ">>> Demo site created successfully."
def cli_cmd_load_demo_records(conf):
"""Load demo records. Useful for testing purposes."""
from invenio.config import CFG_PREFIX
from invenio.dbquery import run_sql
print ">>> Going to load demo records..."
run_sql("TRUNCATE schTASK")
for cmd in ["%s/bin/bibupload -u admin -i %s/var/tmp/demobibdata.xml" % (CFG_PREFIX, CFG_PREFIX),
"%s/bin/bibupload 1" % CFG_PREFIX,
"%s/bin/bibdocfile --textify --with-ocr --recid 97" % CFG_PREFIX,
"%s/bin/bibdocfile --textify --all" % CFG_PREFIX,
"%s/bin/bibindex -u admin" % CFG_PREFIX,
"%s/bin/bibindex 2" % CFG_PREFIX,
"%s/bin/bibreformat -u admin -o HB" % CFG_PREFIX,
"%s/bin/bibreformat 3" % CFG_PREFIX,
"%s/bin/webcoll -u admin" % CFG_PREFIX,
"%s/bin/webcoll 4" % CFG_PREFIX,
"%s/bin/bibrank -u admin" % CFG_PREFIX,
"%s/bin/bibrank 5" % CFG_PREFIX,
"%s/bin/bibsort -u admin -R" % CFG_PREFIX,
"%s/bin/bibsort 6" % CFG_PREFIX,
"%s/bin/oairepositoryupdater -u admin" % CFG_PREFIX,
"%s/bin/oairepositoryupdater 7" % CFG_PREFIX,
"%s/bin/bibupload 8" % CFG_PREFIX,]:
if os.system(cmd):
print "ERROR: failed execution of", cmd
sys.exit(1)
print ">>> Demo records loaded successfully."
def cli_cmd_remove_demo_records(conf):
"""Remove demo records. Useful when you are finished testing."""
print ">>> Going to remove demo records..."
from invenio.config import CFG_PREFIX
from invenio.dbquery import run_sql
from invenio.textutils import wrap_text_in_a_box, wait_for_user
wait_for_user(wrap_text_in_a_box("""WARNING: You are going to destroy
your records and documents!"""))
if os.path.exists(CFG_PREFIX + os.sep + 'var' + os.sep + 'data'):
shutil.rmtree(CFG_PREFIX + os.sep + 'var' + os.sep + 'data')
run_sql("TRUNCATE schTASK")
for cmd in ["%s/bin/dbexec < %s/lib/sql/invenio/tabbibclean.sql" % (CFG_PREFIX, CFG_PREFIX),
"%s/bin/webcoll -u admin" % CFG_PREFIX,
"%s/bin/webcoll 1" % CFG_PREFIX,]:
if os.system(cmd):
print "ERROR: failed execution of", cmd
sys.exit(1)
print ">>> Demo records removed successfully."
def cli_cmd_drop_demo_site(conf):
"""Drop demo site completely. Useful when you are finished testing."""
print ">>> Going to drop demo site..."
from invenio.textutils import wrap_text_in_a_box, wait_for_user
wait_for_user(wrap_text_in_a_box("""WARNING: You are going to destroy
your site and documents!"""))
cli_cmd_drop_tables(conf)
cli_cmd_create_tables(conf)
cli_cmd_remove_demo_records(conf)
print ">>> Demo site dropped successfully."
def cli_cmd_run_unit_tests(conf):
"""Run unit tests, usually on the working demo site."""
from invenio.testutils import build_and_run_unit_test_suite
if not build_and_run_unit_test_suite():
sys.exit(1)
def cli_cmd_run_regression_tests(conf):
"""Run regression tests, usually on the working demo site."""
from invenio.testutils import build_and_run_regression_test_suite
if not build_and_run_regression_test_suite():
sys.exit(1)
def cli_cmd_run_web_tests(conf):
"""Run web tests in a browser. Requires Firefox with Selenium."""
from invenio.testutils import build_and_run_web_test_suite
if not build_and_run_web_test_suite():
sys.exit(1)
def _detect_ip_address():
"""Detect IP address of this computer. Useful for creating Apache
vhost conf snippet on RHEL like machines.
@return: IP address, or '*' if cannot detect
@rtype: string
@note: creates socket for real in order to detect real IP address,
not the loopback one.
"""
try:
s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
s.connect(('invenio-software.org', 0))
return s.getsockname()[0]
except:
return '*'
def cli_cmd_create_apache_conf(conf):
"""
Create Apache conf files for this site, keeping previous
files in a backup copy.
"""
print ">>> Going to create Apache conf files..."
from invenio.textutils import wrap_text_in_a_box
from invenio.access_control_config import CFG_EXTERNAL_AUTH_USING_SSO
apache_conf_dir = conf.get("Invenio", 'CFG_ETCDIR') + \
os.sep + 'apache'
## Preparation of XSendFile directive
xsendfile_directive_needed = int(conf.get("Invenio", 'CFG_BIBDOCFILE_USE_XSENDFILE')) != 0
if xsendfile_directive_needed:
xsendfile_directive = "XSendFile On\n"
else:
xsendfile_directive = "#XSendFile On\n"
for path in (conf.get('Invenio', 'CFG_WEBSUBMIT_FILEDIR'), # BibDocFile
conf.get('Invenio', 'CFG_WEBDIR'),
conf.get('Invenio', 'CFG_WEBSUBMIT_STORAGEDIR'), # WebSubmit
conf.get('Invenio', 'CFG_TMPDIR'),
os.path.join(conf.get('Invenio', 'CFG_PREFIX'), 'var', 'tmp', 'attachfile'),
os.path.join(conf.get('Invenio', 'CFG_PREFIX'), 'var', 'data', 'comments'),
os.path.join(conf.get('Invenio', 'CFG_PREFIX'), 'var', 'data', 'baskets', 'comments'),
'/tmp'): # BibExport
if xsendfile_directive_needed:
xsendfile_directive += ' XSendFilePath %s\n' % path
else:
xsendfile_directive += ' #XSendFilePath %s\n' % path
xsendfile_directive = xsendfile_directive.strip()
## Preparation of deflate directive
deflate_directive_needed = int(conf.get("Invenio", 'CFG_WEBSTYLE_HTTP_USE_COMPRESSION')) != 0
if deflate_directive_needed:
deflate_directive = r"""
## Configuration snippet taken from:
##
SetOutputFilter DEFLATE
# Netscape 4.x has some problems...
BrowserMatch ^Mozilla/4 gzip-only-text/html
# Netscape 4.06-4.08 have some more problems
BrowserMatch ^Mozilla/4\.0[678] no-gzip
# MSIE masquerades as Netscape, but it is fine
# BrowserMatch \bMSIE !no-gzip !gzip-only-text/html
# NOTE: Due to a bug in mod_setenvif up to Apache 2.0.48
# the above regex won't work. You can use the following
# workaround to get the desired effect:
BrowserMatch \bMSI[E] !no-gzip !gzip-only-text/html
# Don't compress images
SetEnvIfNoCase Request_URI \
\.(?:gif|jpe?g|png)$ no-gzip dont-vary
# Make sure proxies don't deliver the wrong content
Header append Vary User-Agent env=!dont-vary
"""
else:
deflate_directive = ""
if CFG_EXTERNAL_AUTH_USING_SSO:
shibboleth_directive = r"""
SSLRequireSSL # The modules only work using HTTPS
AuthType shibboleth
ShibRequireSession On
ShibRequireAll On
ShibExportAssertion Off
require valid-user
"""
else:
shibboleth_directive = ""
## Apache vhost conf file is distro specific, so analyze needs:
# Gentoo (and generic defaults):
listen_directive_needed = True
ssl_pem_directive_needed = False
ssl_pem_path = '/etc/apache2/ssl/apache.pem'
ssl_crt_path = '/etc/apache2/ssl/server.crt'
ssl_key_path = '/etc/apache2/ssl/server.key'
vhost_ip_address_needed = False
wsgi_socket_directive_needed = False
# Debian:
if os.path.exists(os.path.sep + 'etc' + os.path.sep + 'debian_version'):
listen_directive_needed = False
ssl_pem_directive_needed = True
# RHEL/SLC:
if os.path.exists(os.path.sep + 'etc' + os.path.sep + 'redhat-release'):
listen_directive_needed = False
ssl_crt_path = '/etc/pki/tls/certs/localhost.crt'
ssl_key_path = '/etc/pki/tls/private/localhost.key'
vhost_ip_address_needed = True
wsgi_socket_directive_needed = True
# maybe we are using non-standard ports?
vhost_site_url = conf.get('Invenio', 'CFG_SITE_URL').replace("http://", "")
if vhost_site_url.startswith("https://"):
## The installation is configured to require HTTPS for any connection
vhost_site_url = vhost_site_url.replace("https://", "")
vhost_site_url_port = '80'
vhost_site_secure_url = conf.get('Invenio', 'CFG_SITE_SECURE_URL').replace("https://", "")
vhost_site_secure_url_port = '443'
if ':' in vhost_site_url:
vhost_site_url, vhost_site_url_port = vhost_site_url.split(':', 1)
if ':' in vhost_site_secure_url:
vhost_site_secure_url, vhost_site_secure_url_port = vhost_site_secure_url.split(':', 1)
if vhost_site_url_port != '80' or vhost_site_secure_url_port != '443':
listen_directive_needed = True
## OK, let's create Apache vhost files:
if not os.path.exists(apache_conf_dir):
os.mkdir(apache_conf_dir)
apache_vhost_file = apache_conf_dir + os.sep + \
'invenio-apache-vhost.conf'
apache_vhost_ssl_file = apache_conf_dir + os.sep + \
'invenio-apache-vhost-ssl.conf'
apache_vhost_body = """\
AddDefaultCharset UTF-8
ServerSignature Off
ServerTokens Prod
NameVirtualHost %(vhost_ip_address)s:%(vhost_site_url_port)s
%(listen_directive)s
%(wsgi_socket_directive)s
WSGIRestrictStdout Off
deny from all
deny from all
ServerName %(servername)s
ServerAlias %(serveralias)s
ServerAdmin %(serveradmin)s
DocumentRoot %(webdir)s
Options FollowSymLinks MultiViews
AllowOverride None
Order allow,deny
Allow from all
ErrorLog %(logdir)s/apache.err
LogLevel warn
LogFormat "%%h %%l %%u %%t \\"%%r\\" %%>s %%b \\"%%{Referer}i\\" \\"%%{User-agent}i\\" %%D" combined_with_timing
CustomLog %(logdir)s/apache.log combined_with_timing
DirectoryIndex index.en.html index.html
Alias /img/ %(webdir)s/img/
Alias /js/ %(webdir)s/js/
Alias /flash/ %(webdir)s/flash/
Alias /css/ %(webdir)s/css/
Alias /export/ %(webdir)s/export/
Alias /MathJax/ %(webdir)s/MathJax/
Alias /jsCalendar/ %(webdir)s/jsCalendar/
Alias /ckeditor/ %(webdir)s/ckeditor/
Alias /mediaelement/ %(webdir)s/mediaelement/
AliasMatch /sitemap-(.*) %(webdir)s/sitemap-$1
Alias /robots.txt %(webdir)s/robots.txt
Alias /favicon.ico %(webdir)s/favicon.ico
WSGIDaemonProcess invenio processes=5 threads=1 display-name=%%{GROUP} inactivity-timeout=3600 maximum-requests=10000
WSGIImportScript %(wsgidir)s/invenio.wsgi process-group=invenio application-group=%%{GLOBAL}
WSGIScriptAlias / %(wsgidir)s/invenio.wsgi
WSGIPassAuthorization On
%(xsendfile_directive)s
WSGIProcessGroup invenio
WSGIApplicationGroup %%{GLOBAL}
Options FollowSymLinks MultiViews
AllowOverride None
Order allow,deny
Allow from all
%(deflate_directive)s
""" % {'vhost_site_url_port': vhost_site_url_port,
'servername': vhost_site_url,
'serveralias': vhost_site_url.split('.')[0],
'serveradmin': conf.get('Invenio', 'CFG_SITE_ADMIN_EMAIL'),
'webdir': conf.get('Invenio', 'CFG_WEBDIR'),
'logdir': conf.get('Invenio', 'CFG_LOGDIR'),
'libdir' : conf.get('Invenio', 'CFG_PYLIBDIR'),
'wsgidir': os.path.join(conf.get('Invenio', 'CFG_PREFIX'), 'var', 'www-wsgi'),
'vhost_ip_address': vhost_ip_address_needed and _detect_ip_address() or '*',
'listen_directive': listen_directive_needed and 'Listen ' + vhost_site_url_port or \
'#Listen ' + vhost_site_url_port,
'wsgi_socket_directive': (wsgi_socket_directive_needed and \
'WSGISocketPrefix ' or '#WSGISocketPrefix ') + \
conf.get('Invenio', 'CFG_PREFIX') + os.sep + 'var' + os.sep + 'run',
'xsendfile_directive' : xsendfile_directive,
'deflate_directive': deflate_directive,
}
apache_vhost_ssl_body = """\
ServerSignature Off
ServerTokens Prod
%(listen_directive)s
NameVirtualHost %(vhost_ip_address)s:%(vhost_site_secure_url_port)s
%(ssl_pem_directive)s
%(ssl_crt_directive)s
%(ssl_key_directive)s
WSGIRestrictStdout Off
deny from all
deny from all
ServerName %(servername)s
ServerAlias %(serveralias)s
ServerAdmin %(serveradmin)s
SSLEngine on
DocumentRoot %(webdir)s
Options FollowSymLinks MultiViews
AllowOverride None
Order allow,deny
Allow from all
ErrorLog %(logdir)s/apache-ssl.err
LogLevel warn
LogFormat "%%h %%l %%u %%t \\"%%r\\" %%>s %%b \\"%%{Referer}i\\" \\"%%{User-agent}i\\" %%D" combined_with_timing
CustomLog %(logdir)s/apache-ssl.log combined_with_timing
DirectoryIndex index.en.html index.html
Alias /img/ %(webdir)s/img/
Alias /js/ %(webdir)s/js/
Alias /flash/ %(webdir)s/flash/
Alias /css/ %(webdir)s/css/
Alias /export/ %(webdir)s/export/
Alias /MathJax/ %(webdir)s/MathJax/
Alias /jsCalendar/ %(webdir)s/jsCalendar/
Alias /ckeditor/ %(webdir)s/ckeditor/
Alias /mediaelement/ %(webdir)s/mediaelement/
AliasMatch /sitemap-(.*) %(webdir)s/sitemap-$1
Alias /robots.txt %(webdir)s/robots.txt
Alias /favicon.ico %(webdir)s/favicon.ico
RedirectMatch /sslredirect/(.*) http://$1
WSGIScriptAlias / %(wsgidir)s/invenio.wsgi
WSGIPassAuthorization On
%(xsendfile_directive)s
WSGIProcessGroup invenio
WSGIApplicationGroup %%{GLOBAL}
Options FollowSymLinks MultiViews
AllowOverride None
Order allow,deny
Allow from all
%(deflate_directive)s
%(shibboleth_directive)s
""" % {'vhost_site_secure_url_port': vhost_site_secure_url_port,
'servername': vhost_site_secure_url,
'serveralias': vhost_site_secure_url.split('.')[0],
'serveradmin': conf.get('Invenio', 'CFG_SITE_ADMIN_EMAIL'),
'webdir': conf.get('Invenio', 'CFG_WEBDIR'),
'logdir': conf.get('Invenio', 'CFG_LOGDIR'),
'libdir' : conf.get('Invenio', 'CFG_PYLIBDIR'),
'wsgidir' : os.path.join(conf.get('Invenio', 'CFG_PREFIX'), 'var', 'www-wsgi'),
'vhost_ip_address': vhost_ip_address_needed and _detect_ip_address() or '*',
'listen_directive' : listen_directive_needed and 'Listen ' + vhost_site_secure_url_port or \
'#Listen ' + vhost_site_secure_url_port,
'ssl_pem_directive': ssl_pem_directive_needed and \
'SSLCertificateFile %s' % ssl_pem_path or \
'#SSLCertificateFile %s' % ssl_pem_path,
'ssl_crt_directive': ssl_pem_directive_needed and \
'#SSLCertificateFile %s' % ssl_crt_path or \
'SSLCertificateFile %s' % ssl_crt_path,
'ssl_key_directive': ssl_pem_directive_needed and \
'#SSLCertificateKeyFile %s' % ssl_key_path or \
'SSLCertificateKeyFile %s' % ssl_key_path,
'xsendfile_directive' : xsendfile_directive,
'deflate_directive': deflate_directive,
'shibboleth_directive': shibboleth_directive,
}
# write HTTP vhost snippet:
if os.path.exists(apache_vhost_file):
shutil.copy(apache_vhost_file,
apache_vhost_file + '.OLD')
fdesc = open(apache_vhost_file, 'w')
fdesc.write(apache_vhost_body)
fdesc.close()
print
print "Created file", apache_vhost_file
# write HTTPS vhost snippet:
vhost_ssl_created = False
if conf.get('Invenio', 'CFG_SITE_SECURE_URL').startswith("https://"):
if os.path.exists(apache_vhost_ssl_file):
shutil.copy(apache_vhost_ssl_file,
apache_vhost_ssl_file + '.OLD')
fdesc = open(apache_vhost_ssl_file, 'w')
fdesc.write(apache_vhost_ssl_body)
fdesc.close()
vhost_ssl_created = True
print "Created file", apache_vhost_ssl_file
print wrap_text_in_a_box("""\
Apache virtual host configuration file(s) for your Invenio site
was(were) created. Please check created file(s) and activate virtual
host(s). For example, you can put the following include statements in
your httpd.conf:\n
Include %s
%s
Please see the INSTALL file for more details.
""" % (apache_vhost_file, (vhost_ssl_created and 'Include ' or '#Include ') + apache_vhost_ssl_file))
print ">>> Apache conf files created."
def cli_cmd_get(conf, varname):
"""
Return value of VARNAME read from CONF files. Useful for
third-party programs to access values of conf options such as
CFG_PREFIX. Return None if VARNAME is not found.
"""
# do not pay attention to upper/lower case:
varname = varname.lower()
# do not pay attention to section names yet:
all_options = {}
for section in conf.sections():
for option in conf.options(section):
all_options[option] = conf.get(section, option)
return all_options.get(varname, None)
def cli_cmd_list(conf):
"""
Print a list of all conf options and values from CONF.
"""
sections = conf.sections()
sections.sort()
for section in sections:
options = conf.options(section)
options.sort()
for option in options:
print option.upper(), '=', conf.get(section, option)
def _grep_version_from_executable(path_to_exec, version_regexp):
"""
Try to detect a program version by digging into its binary
PATH_TO_EXEC and looking for VERSION_REGEXP. Return program
version as a string. Return empty string if not succeeded.
"""
from invenio.shellutils import run_shell_command
exec_version = ""
if os.path.exists(path_to_exec):
dummy1, cmd2_out, dummy2 = run_shell_command("strings %s | grep %s",
(path_to_exec, version_regexp))
if cmd2_out:
for cmd2_out_line in cmd2_out.split("\n"):
if len(cmd2_out_line) > len(exec_version):
# the longest the better
exec_version = cmd2_out_line
return exec_version
def detect_apache_version():
"""
Try to detect Apache version by localizing httpd or apache
executables and grepping inside binaries. Return list of all
found Apache versions and paths. (For a given executable, the
returned format is 'apache_version [apache_path]'.) Return empty
list if no success.
"""
from invenio.shellutils import run_shell_command
out = []
dummy1, cmd_out, dummy2 = run_shell_command("locate bin/httpd bin/apache")
for apache in cmd_out.split("\n"):
apache_version = _grep_version_from_executable(apache, '^Apache\/')
if apache_version:
out.append("%s [%s]" % (apache_version, apache))
return out
def cli_cmd_detect_system_details(conf):
"""
Detect and print system details such as Apache/Python/MySQL
versions etc. Useful for debugging problems on various OS.
"""
import MySQLdb
print ">>> Going to detect system details..."
print "* Hostname: " + socket.gethostname()
print "* Invenio version: " + conf.get("Invenio", "CFG_VERSION")
print "* Python version: " + sys.version.replace("\n", " ")
print "* Apache version: " + ";\n ".join(detect_apache_version())
print "* MySQLdb version: " + MySQLdb.__version__
try:
from invenio.dbquery import run_sql
print "* MySQL version:"
for key, val in run_sql("SHOW VARIABLES LIKE 'version%'") + \
run_sql("SHOW VARIABLES LIKE 'charact%'") + \
run_sql("SHOW VARIABLES LIKE 'collat%'"):
if False:
print " - %s: %s" % (key, val)
elif key in ['version',
'character_set_client',
'character_set_connection',
'character_set_database',
'character_set_results',
'character_set_server',
'character_set_system',
'collation_connection',
'collation_database',
'collation_server']:
print " - %s: %s" % (key, val)
except ImportError:
print "* ERROR: cannot import dbquery"
print ">>> System details detected successfully."
def main():
"""Main entry point."""
conf = ConfigParser()
if '--help' in sys.argv or \
'-h' in sys.argv:
print_usage()
elif '--version' in sys.argv or \
'-V' in sys.argv:
print_version()
else:
confdir = None
if '--conf-dir' in sys.argv:
try:
confdir = sys.argv[sys.argv.index('--conf-dir') + 1]
except IndexError:
pass # missing --conf-dir argument value
if not os.path.exists(confdir):
print "ERROR: bad or missing --conf-dir option value."
sys.exit(1)
else:
## try to detect path to conf dir (relative to this bin dir):
confdir = re.sub(r'/bin$', '/etc', sys.path[0])
## read conf files:
for conffile in [confdir + os.sep + 'invenio.conf',
confdir + os.sep + 'invenio-autotools.conf',
confdir + os.sep + 'invenio-local.conf',]:
if os.path.exists(conffile):
conf.read(conffile)
else:
if not conffile.endswith("invenio-local.conf"):
# invenio-local.conf is optional, otherwise stop
print "ERROR: Badly guessed conf file location", conffile
print "(Please use --conf-dir option.)"
sys.exit(1)
## decide what to do:
done = False
for opt_idx in range(0, len(sys.argv)):
opt = sys.argv[opt_idx]
if opt == '--conf-dir':
# already treated before, so skip silently:
pass
elif opt == '--get':
try:
varname = sys.argv[opt_idx + 1]
except IndexError:
print "ERROR: bad or missing --get option value."
sys.exit(1)
if varname.startswith('-'):
print "ERROR: bad or missing --get option value."
sys.exit(1)
varvalue = cli_cmd_get(conf, varname)
if varvalue is not None:
print varvalue
else:
sys.exit(1)
done = True
elif opt == '--list':
cli_cmd_list(conf)
done = True
elif opt == '--detect-system-details':
cli_cmd_detect_system_details(conf)
done = True
elif opt == '--create-tables':
cli_cmd_create_tables(conf)
done = True
elif opt == '--load-webstat-conf':
cli_cmd_load_webstat_conf(conf)
done = True
elif opt == '--drop-tables':
cli_cmd_drop_tables(conf)
done = True
elif opt == '--check-openoffice':
cli_check_openoffice(conf)
done = True
elif opt == '--create-demo-site':
cli_cmd_create_demo_site(conf)
done = True
elif opt == '--load-demo-records':
cli_cmd_load_demo_records(conf)
done = True
elif opt == '--remove-demo-records':
cli_cmd_remove_demo_records(conf)
done = True
elif opt == '--drop-demo-site':
cli_cmd_drop_demo_site(conf)
done = True
elif opt == '--run-unit-tests':
cli_cmd_run_unit_tests(conf)
done = True
elif opt == '--run-regression-tests':
cli_cmd_run_regression_tests(conf)
done = True
elif opt == '--run-web-tests':
cli_cmd_run_web_tests(conf)
done = True
elif opt == '--update-all':
cli_cmd_update_config_py(conf)
cli_cmd_update_dbquery_py(conf)
cli_cmd_update_dbexec(conf)
cli_cmd_update_bibconvert_tpl(conf)
cli_cmd_update_web_tests(conf)
done = True
elif opt == '--update-config-py':
cli_cmd_update_config_py(conf)
done = True
elif opt == '--update-dbquery-py':
cli_cmd_update_dbquery_py(conf)
done = True
elif opt == '--update-dbexec':
cli_cmd_update_dbexec(conf)
done = True
elif opt == '--update-bibconvert-tpl':
cli_cmd_update_bibconvert_tpl(conf)
done = True
elif opt == '--update-web-tests':
cli_cmd_update_web_tests(conf)
done = True
elif opt == '--reset-all':
cli_cmd_reset_sitename(conf)
cli_cmd_reset_siteadminemail(conf)
cli_cmd_reset_fieldnames(conf)
cli_cmd_reset_recstruct_cache(conf)
done = True
elif opt == '--reset-sitename':
cli_cmd_reset_sitename(conf)
done = True
elif opt == '--reset-siteadminemail':
cli_cmd_reset_siteadminemail(conf)
done = True
elif opt == '--reset-fieldnames':
cli_cmd_reset_fieldnames(conf)
done = True
elif opt == '--reset-recstruct-cache':
cli_cmd_reset_recstruct_cache(conf)
done = True
elif opt == '--create-apache-conf':
cli_cmd_create_apache_conf(conf)
done = True
elif opt.startswith("-") and opt != '--yes-i-know':
print "ERROR: unknown option", opt
sys.exit(1)
if not done:
print """ERROR: Please specify a command. Please see '--help'."""
sys.exit(1)
if __name__ == '__main__':
main()
diff --git a/modules/websubmit/lib/websubmit_config.py b/modules/websubmit/lib/websubmit_config.py
index 688629ea5..b888f33fd 100644
--- a/modules/websubmit/lib/websubmit_config.py
+++ b/modules/websubmit/lib/websubmit_config.py
@@ -1,225 +1,264 @@
## This file is part of Invenio.
## Copyright (C) 2004, 2005, 2006, 2007, 2008, 2009, 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.
"""Invenio Submission Web Interface config file."""
__revision__ = "$Id$"
import re
## test:
test = "FALSE"
## CC all action confirmation mails to administrator? (0 == NO; 1 == YES)
CFG_WEBSUBMIT_COPY_MAILS_TO_ADMIN = 0
## During submission, warn user if she is going to leave the
## submission by following some link on the page?
## Does not work with Opera and Konqueror.
## This requires all submission functions to set Javascript variable
## 'user_must_confirm_before_leaving_page' to 'false' before
## programmatically submitting a form , or else users will be asked
## confirmation after each submission step.
## (0 == NO; 1 == YES)
CFG_WEBSUBMIT_CHECK_USER_LEAVES_SUBMISSION = 0
## List of keywords/format parameters that should not write by default
## corresponding files in submission directory (`curdir'). Some other
## filenames not included here are reserved too, such as those
## containing non-alphanumeric chars (excepted underscores '_'), for
## eg all names containing a dot ('bibdocactions.log',
## 'performed_actions.log', etc.)
CFG_RESERVED_SUBMISSION_FILENAMES = ['SuE',
'files',
'lastuploadedfile',
'curdir',
'function_log',
'SN',
'ln']
## CFG_WEBSUBMIT_ICON_SUBFORMAT_RE -- a subformat is an Invenio concept to give
## file formats more semantic. For example "foo.gif;icon" has ".gif;icon"
## 'format', ".gif" 'superformat' and "icon" 'subformat'. That means that this
## particular format/instance of the "foo" document, not only is a ".gif" but
## is in the shape of an "icon", i.e. most probably it will be low-resolution.
## This configuration variable let the administrator to decide which implicit
## convention will be used to know which formats will be meant to be used
## as an icon.
CFG_WEBSUBMIT_ICON_SUBFORMAT_RE = re.compile(r"icon.*")
## CFG_WEBSUBMIT_DEFAULT_ICON_SUBFORMAT -- this is the default subformat used
## when creating new icons.
CFG_WEBSUBMIT_DEFAULT_ICON_SUBFORMAT = "icon"
## Prefix for video uploads, Garbage Collector
CFG_WEBSUBMIT_TMP_VIDEO_PREFIX = "video_upload_"
+try:
+ from invenio.config import CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_MISC
+except ImportError:
+ CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_MISC = {
+ 'can_revise_doctypes': ['*'],
+ 'can_comment_doctypes': ['*'],
+ 'can_describe_doctypes': ['*'],
+ 'can_delete_doctypes': ['*'],
+ 'can_keep_doctypes': ['*'],
+ 'can_rename_doctypes': ['*'],
+ 'can_add_format_to_doctypes': ['*'],
+ 'can_restrict_doctypes': ['*']}
+
+try:
+ from invenio.config import CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_DOCTYPES
+except ImportError:
+ CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_DOCTYPES = [
+ ('Main', 'Main document'),
+ ('LaTeX', 'LaTeX'),
+ ('Source', 'Source'),
+ ('Additional', 'Additional File'),
+ ('Audio', 'Audio file'),
+ ('Video', 'Video file'),
+ ('Script', 'Script'),
+ ('Data', 'Data'),
+ ('Figure', 'Figure'),
+ ('Schema', 'Schema'),
+ ('Graph', 'Graph'),
+ ('Image', 'Image'),
+ ('Drawing', 'Drawing'),
+ ('Slides', 'Slides')]
+
+try:
+ from invenio.config import CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_RESTRICTIONS
+except ImportError:
+ CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_RESTRICTIONS = [
+ ('', 'Public'),
+ ('restricted', 'Restricted')]
+
class InvenioWebSubmitFunctionError(Exception):
"""This exception should only ever be raised by WebSubmit functions.
It will be caught and handled by the WebSubmit core itself.
It is used to signal to WebSubmit core that one of the functions
encountered a FATAL ERROR situation that should all further execution
of the submission.
The exception will carry an error message in its "value" string. This
message will probably be displayed on the user's browser in an Invenio
"error" box, and may be logged for the admin to examine.
Again: If this exception is raised by a WebSubmit function, an error
message will displayed and the submission ends in failure.
Extends: Exception.
"""
def __init__(self, value):
"""Set the internal "value" attribute to that of the passed "value"
parameter.
@param value: (string) - an error string to display to the user.
"""
Exception.__init__(self)
self.value = value
def __str__(self):
"""Return oneself as a string (actually, return the contents of
self.value).
@return: (string)
"""
return str(self.value)
class InvenioWebSubmitFunctionStop(Exception):
"""This exception should only ever be raised by WebSubmit functions.
It will be caught and handled by the WebSubmit core itself.
It is used to signal to WebSubmit core that one of the functions
encountered a situation that should prevent the functions that follow
it from being executed, and that WebSubmit core should display some sort
of message to the user. This message will be stored in the "value"
attribute of the object.
***
NOTE: In the current WebSubmit, this "value" is ususally a JavaScript
string that redirects the user's browser back to the Web form
phase of the submission. The use of JavaScript, however is going
to be removed in the future, so the mechanism may change.
***
Extends: Exception.
"""
def __init__(self, value):
"""Set the internal "value" attribute to that of the passed "value"
parameter.
@param value: (string) - a string to display to the user.
"""
Exception.__init__(self)
self.value = value
def __str__(self):
"""Return oneself as a string (actually, return the contents of
self.value).
@return: (string)
"""
return str(self.value)
class InvenioWebSubmitFunctionWarning(Exception):
"""This exception should be raised by a WebSubmit function
when unexpected behaviour is encountered during the execution
of the function. The unexpected behaviour should not have been
so serious that execution had to be halted, but since the
function was unable to perform its task, the event must be
logged.
Logging of the exception will be performed by WebSubmit.
Extends: Exception.
"""
def __init__(self, value):
"""Set the internal "value" attribute to that of the passed "value"
parameter.
@param value: (string) - a string to write to the log.
"""
Exception.__init__(self)
self.value = value
def __str__(self):
"""Return oneself as a string (actually, return the contents of
self.value).
@return: (string)
"""
return str(self.value)
class InvenioWebSubmitFileStamperError(Exception):
"""This exception should be raised by websubmit_file_stamper when an
error is encoutered that prevents a file from being stamped.
When caught, this exception should be used to stop processing with a
failure signal.
Extends: Exception.
"""
def __init__(self, value):
"""Set the internal "value" attribute to that of the passed "value"
parameter.
@param value: (string) - a string to write to the log.
"""
Exception.__init__(self)
self.value = value
def __str__(self):
"""Return oneself as a string (actually, return the contents of
self.value).
@return: (string)
"""
return str(self.value)
class InvenioWebSubmitIconCreatorError(Exception):
"""This exception should be raised by websubmit_icon_creator when an
error is encoutered that prevents an icon from being created.
When caught, this exception should be used to stop processing with a
failure signal.
Extends: Exception.
"""
def __init__(self, value):
"""Set the internal "value" attribute to that of the passed "value"
parameter.
@param value: (string) - a string to write to the log.
"""
Exception.__init__(self)
self.value = value
def __str__(self):
"""Return oneself as a string (actually, return the contents of
self.value).
@return: (string)
"""
return str(self.value)
class InvenioWebSubmitFileMetadataRuntimeError(Exception):
"""This exception should be raised by websubmit_file_metadata plugins when an
error is encoutered that prevents a extracting/updating a file.
When caught, this exception should be used to stop processing with a
failure signal.
Extends: Exception.
"""
def __init__(self, value):
"""Set the internal "value" attribute to that of the passed "value"
parameter.
@param value: (string) - a string to write to the log.
"""
Exception.__init__(self)
self.value = value
def __str__(self):
"""Return oneself as a string (actually, return the contents of
self.value).
@return: (string)
"""
return str(self.value)
diff --git a/modules/websubmit/lib/websubmit_webinterface.py b/modules/websubmit/lib/websubmit_webinterface.py
index 3412ca3b9..160fc96c9 100644
--- a/modules/websubmit/lib/websubmit_webinterface.py
+++ b/modules/websubmit/lib/websubmit_webinterface.py
@@ -1,1478 +1,1454 @@
## This file is part of Invenio.
## Copyright (C) 2006, 2007, 2008, 2009, 2010, 2011, 2012 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.
__lastupdated__ = """$Date$"""
__revision__ = "$Id$"
import os
import errno
import time
import cgi
import sys
import shutil
from urllib import urlencode
from invenio.config import \
CFG_ACCESS_CONTROL_LEVEL_SITE, \
CFG_SITE_LANG, \
CFG_SITE_NAME, \
CFG_TMPSHAREDDIR, \
CFG_SITE_NAME_INTL, \
CFG_SITE_URL, \
CFG_SITE_SECURE_URL, \
CFG_WEBSUBMIT_STORAGEDIR, \
CFG_PREFIX, \
CFG_CERN_SITE, \
CFG_SITE_RECORD
from invenio import webinterface_handler_config as apache
from invenio.dbquery import run_sql
from invenio.access_control_config import VIEWRESTRCOLL
from invenio.access_control_mailcookie import mail_cookie_create_authorize_action
from invenio.access_control_engine import acc_authorize_action
from invenio.access_control_admin import acc_is_role
from invenio.webpage import page, create_error_box, pageheaderonly, \
pagefooteronly
from invenio.webuser import getUid, page_not_authorized, collect_user_info, isUserSuperAdmin, \
isGuestUser
from invenio.websubmit_config import *
from invenio import webjournal_utils
from invenio.webinterface_handler import wash_urlargd, WebInterfaceDirectory
from invenio.urlutils import make_canonical_urlargd, redirect_to_url
from invenio.messages import gettext_set_language
from invenio.search_engine import \
guess_primary_collection_of_a_record, get_colID, record_exists, \
create_navtrail_links, check_user_can_view_record, record_empty, \
is_user_owner_of_record
from invenio.bibdocfile import BibRecDocs, normalize_format, file_strip_ext, \
stream_restricted_icon, BibDoc, InvenioWebSubmitFileError, stream_file, \
decompose_file, propose_next_docname, get_subformat_from_format
from invenio.errorlib import register_exception
from invenio.htmlutils import is_html_text_editor_installed
from invenio.websubmit_icon_creator import create_icon, InvenioWebSubmitIconCreatorError
from ckeditor_invenio_connector import process_CKEditor_upload, send_response
import invenio.template
websubmit_templates = invenio.template.load('websubmit')
from invenio.websearchadminlib import get_detailed_page_tabs
from invenio.session import get_session
from invenio.jsonutils import json, CFG_JSON_AVAILABLE
import invenio.template
webstyle_templates = invenio.template.load('webstyle')
websearch_templates = invenio.template.load('websearch')
from invenio.websubmit_managedocfiles import \
create_file_upload_interface, \
get_upload_file_interface_javascript, \
get_upload_file_interface_css, \
move_uploaded_files_to_storage
class WebInterfaceFilesPages(WebInterfaceDirectory):
def __init__(self,recid):
self.recid = recid
def _lookup(self, component, path):
# after ///files/ every part is used as the file
# name
filename = component
def getfile(req, form):
args = wash_urlargd(form, websubmit_templates.files_default_urlargd)
ln = args['ln']
_ = gettext_set_language(ln)
uid = getUid(req)
user_info = collect_user_info(req)
verbose = args['verbose']
if verbose >= 1 and not isUserSuperAdmin(user_info):
# Only SuperUser can see all the details!
verbose = 0
if uid == -1 or CFG_ACCESS_CONTROL_LEVEL_SITE > 1:
return page_not_authorized(req, "/%s/%s" % (CFG_SITE_RECORD, self.recid),
navmenuid='submit')
if record_exists(self.recid) < 1:
msg = " %s " % _("Requested record does not seem to exist.")
return warningMsg(msg, req, CFG_SITE_NAME, ln)
if record_empty(self.recid):
msg = "%s " % _("Requested record does not seem to have been integrated.")
return warningMsg(msg, req, CFG_SITE_NAME, ln)
(auth_code, auth_message) = check_user_can_view_record(user_info, self.recid)
if auth_code and user_info['email'] == 'guest':
if webjournal_utils.is_recid_in_released_issue(self.recid):
# We can serve the file
pass
else:
cookie = mail_cookie_create_authorize_action(VIEWRESTRCOLL, {'collection' : guess_primary_collection_of_a_record(self.recid)})
target = '/youraccount/login' + \
make_canonical_urlargd({'action': cookie, 'ln' : ln, 'referer' : \
CFG_SITE_SECURE_URL + user_info['uri']}, {})
return redirect_to_url(req, target, norobot=True)
elif auth_code:
if webjournal_utils.is_recid_in_released_issue(self.recid):
# We can serve the file
pass
else:
return page_not_authorized(req, "../", \
text = auth_message)
readonly = CFG_ACCESS_CONTROL_LEVEL_SITE == 1
# From now on: either the user provided a specific file
# name (and a possible version), or we return a list of
# all the available files. In no case are the docids
# visible.
try:
bibarchive = BibRecDocs(self.recid)
except InvenioWebSubmitFileError, e:
register_exception(req=req, alert_admin=True)
msg = "%s %s " % (
_("The system has encountered an error in retrieving the list of files for this document."),
_("The error has been logged and will be taken in consideration as soon as possible."))
return warningMsg(msg, req, CFG_SITE_NAME, ln)
if bibarchive.deleted_p():
return print_warning(req, _("Requested record does not seem to exist."))
docname = ''
format = ''
version = ''
warn = ''
if filename:
# We know the complete file name, guess which docid it
# refers to
## TODO: Change the extension system according to ext.py from setlink
## and have a uniform extension mechanism...
docname = file_strip_ext(filename)
format = filename[len(docname):]
if format and format[0] != '.':
format = '.' + format
if args['subformat']:
format += ';%s' % args['subformat']
else:
docname = args['docname']
if not format:
format = args['format']
if args['subformat']:
format += ';%s' % args['subformat']
if not version:
version = args['version']
## Download as attachment
is_download = False
if args['download']:
is_download = True
# version could be either empty, or all or an integer
try:
int(version)
except ValueError:
if version != 'all':
version = ''
display_hidden = isUserSuperAdmin(user_info)
if version != 'all':
# search this filename in the complete list of files
for doc in bibarchive.list_bibdocs():
if docname == doc.get_docname():
try:
docfile = doc.get_file(format, version)
(auth_code, auth_message) = docfile.is_restricted(user_info)
if auth_code != 0 and not is_user_owner_of_record(user_info, self.recid):
if CFG_WEBSUBMIT_ICON_SUBFORMAT_RE.match(get_subformat_from_format(format)):
return stream_restricted_icon(req)
if user_info['email'] == 'guest':
cookie = mail_cookie_create_authorize_action('viewrestrdoc', {'status' : docfile.get_status()})
target = '/youraccount/login' + \
make_canonical_urlargd({'action': cookie, 'ln' : ln, 'referer' : \
CFG_SITE_SECURE_URL + user_info['uri']}, {})
redirect_to_url(req, target)
else:
req.status = apache.HTTP_UNAUTHORIZED
warn += print_warning(_("This file is restricted: ") + auth_message)
break
if not docfile.hidden_p():
if not readonly:
ip = str(req.remote_ip)
res = doc.register_download(ip, version, format, uid)
try:
return docfile.stream(req, download=is_download)
except InvenioWebSubmitFileError, msg:
register_exception(req=req, alert_admin=True)
req.status = apache.HTTP_INTERNAL_SERVER_ERROR
return warningMsg(_("An error has happened in trying to stream the request file."), req, CFG_SITE_NAME, ln)
else:
req.status = apache.HTTP_UNAUTHORIZED
warn = print_warning(_("The requested file is hidden and can not be accessed."))
except InvenioWebSubmitFileError, msg:
register_exception(req=req, alert_admin=True)
if docname and format and not warn:
req.status = apache.HTTP_NOT_FOUND
warn += print_warning(_("Requested file does not seem to exist."))
filelist = bibarchive.display("", version, ln=ln, verbose=verbose, display_hidden=display_hidden)
t = warn + websubmit_templates.tmpl_filelist(
ln=ln,
recid=self.recid,
docname=args['docname'],
version=version,
filelist=filelist)
cc = guess_primary_collection_of_a_record(self.recid)
unordered_tabs = get_detailed_page_tabs(get_colID(cc), self.recid, ln)
ordered_tabs_id = [(tab_id, values['order']) for (tab_id, values) in unordered_tabs.iteritems()]
ordered_tabs_id.sort(lambda x,y: cmp(x[1],y[1]))
link_ln = ''
if ln != CFG_SITE_LANG:
link_ln = '?ln=%s' % ln
tabs = [(unordered_tabs[tab_id]['label'], \
'%s/%s/%s/%s%s' % (CFG_SITE_URL, CFG_SITE_RECORD, self.recid, tab_id, link_ln), \
tab_id == 'files',
unordered_tabs[tab_id]['enabled']) \
for (tab_id, order) in ordered_tabs_id
if unordered_tabs[tab_id]['visible'] == True]
top = webstyle_templates.detailed_record_container_top(self.recid,
tabs,
args['ln'])
bottom = webstyle_templates.detailed_record_container_bottom(self.recid,
tabs,
args['ln'])
title, description, keywords = websearch_templates.tmpl_record_page_header_content(req, self.recid, args['ln'])
return pageheaderonly(title=title,
navtrail=create_navtrail_links(cc=cc, aas=0, ln=ln) + \
''' > %s
> %s''' % \
(CFG_SITE_URL, CFG_SITE_RECORD, self.recid, title, _("Access to Fulltext")),
description="",
keywords="keywords",
uid=uid,
language=ln,
req=req,
navmenuid='search',
navtrail_append_title_p=0) + \
websearch_templates.tmpl_search_pagestart(ln) + \
top + t + bottom + \
websearch_templates.tmpl_search_pageend(ln) + \
pagefooteronly(lastupdated=__lastupdated__, language=ln, req=req)
return getfile, []
def __call__(self, req, form):
"""Called in case of URLs like /CFG_SITE_RECORD/123/files without
trailing slash.
"""
args = wash_urlargd(form, websubmit_templates.files_default_urlargd)
ln = args['ln']
link_ln = ''
if ln != CFG_SITE_LANG:
link_ln = '?ln=%s' % ln
return redirect_to_url(req, '%s/%s/%s/files/%s' % (CFG_SITE_URL, CFG_SITE_RECORD, self.recid, link_ln))
def websubmit_legacy_getfile(req, form):
""" Handle legacy /getfile.py URLs """
args = wash_urlargd(form, {
'recid': (int, 0),
'docid': (int, 0),
'version': (str, ''),
'name': (str, ''),
'format': (str, ''),
'ln' : (str, CFG_SITE_LANG)
})
_ = gettext_set_language(args['ln'])
def _getfile_py(req, recid=0, docid=0, version="", name="", format="", ln=CFG_SITE_LANG):
if not recid:
## Let's obtain the recid from the docid
if docid:
try:
bibdoc = BibDoc(docid=docid)
recid = bibdoc.get_recid()
except InvenioWebSubmitFileError, e:
return warningMsg(_("An error has happened in trying to retrieve the requested file."), req, CFG_SITE_NAME, ln)
else:
return warningMsg(_('Not enough information to retrieve the document'), req, CFG_SITE_NAME, ln)
else:
if not name and docid:
## Let's obtain the name from the docid
try:
bibdoc = BibDoc(docid)
name = bibdoc.get_docname()
except InvenioWebSubmitFileError, e:
return warningMsg(_("An error has happened in trying to retrieving the requested file."), req, CFG_SITE_NAME, ln)
format = normalize_format(format)
redirect_to_url(req, '%s/%s/%s/files/%s%s?ln=%s%s' % (CFG_SITE_URL, CFG_SITE_RECORD, recid, name, format, ln, version and 'version=%s' % version or ''), apache.HTTP_MOVED_PERMANENTLY)
return _getfile_py(req, **args)
# --------------------------------------------------
from invenio.websubmit_engine import home, action, interface, endaction, makeCataloguesTable
class WebInterfaceSubmitPages(WebInterfaceDirectory):
_exports = ['summary', 'sub', 'direct', '', 'attachfile', 'uploadfile', \
'getuploadedfile', 'managedocfiles', 'managedocfilesasync', \
'upload_video', ('continue', 'continue_')]
def managedocfiles(self, req, form):
"""
Display admin interface to manage files of a record
"""
argd = wash_urlargd(form, {
'ln': (str, ''),
'access': (str, ''),
'recid': (int, None),
'do': (int, 0),
'cancel': (str, None),
})
_ = gettext_set_language(argd['ln'])
uid = getUid(req)
user_info = collect_user_info(req)
# Check authorization
(auth_code, auth_msg) = acc_authorize_action(req,
'runbibdocfile')
if auth_code and user_info['email'] == 'guest':
# Ask to login
target = '/youraccount/login' + \
make_canonical_urlargd({'ln' : argd['ln'],
'referer' : CFG_SITE_SECURE_URL + user_info['uri']}, {})
return redirect_to_url(req, target)
elif auth_code:
return page_not_authorized(req, referer="/submit/managedocfiles",
uid=uid, text=auth_msg,
ln=argd['ln'],
navmenuid="admin")
# Prepare navtrail
navtrail = '''Admin Area > %(manage_files)s''' \
% {'CFG_SITE_URL': CFG_SITE_URL,
'manage_files': _("Manage Document Files")}
body = ''
if argd['do'] != 0 and not argd['cancel']:
# Apply modifications
working_dir = os.path.join(CFG_TMPSHAREDDIR,
'websubmit_upload_interface_config_' + str(uid),
argd['access'])
move_uploaded_files_to_storage(working_dir=working_dir,
recid=argd['recid'],
icon_sizes=['180>','700>'],
create_icon_doctypes=['*'],
force_file_revision=False)
# Clean temporary directory
shutil.rmtree(working_dir)
# Confirm modifications
body += '%s ' % \
(_('Your modifications to record #%i have been submitted') % argd['recid'])
elif argd['cancel']:
# Clean temporary directory
working_dir = os.path.join(CFG_TMPSHAREDDIR,
'websubmit_upload_interface_config_' + str(uid),
argd['access'])
shutil.rmtree(working_dir)
body += '%s ' % \
(_('Your modifications to record #%i have been cancelled') % argd['recid'])
if not argd['recid'] or argd['do'] != 0:
body += '''
''' % {'edit': _('Edit'),
'edit_record': _('Edit record'),
'CFG_SITE_URL': CFG_SITE_URL}
access = time.strftime('%Y%m%d_%H%M%S')
if argd['recid'] and argd['do'] == 0:
# Displaying interface to manage files
# Prepare navtrail
title, description, keywords = websearch_templates.tmpl_record_page_header_content(req, argd['recid'],
argd['ln'])
navtrail = '''Admin Area >
%(manage_files)s >
%(record)s: %(title)s
''' \
% {'CFG_SITE_URL': CFG_SITE_URL,
'title': title,
'manage_files': _("Document File Manager"),
'record': _("Record #%i") % argd['recid']}
- # FIXME: add parameters to `runbibdocfile' in order to
- # configure the file editor based on role, or at least
- # move configuration below to some config file.
body += create_file_upload_interface(\
recid=argd['recid'],
ln=argd['ln'],
- doctypes_and_desc=[('Main', 'Main document'),
- ('LaTeX', 'LaTeX'),
- ('Source', 'Source'),
- ('Additional', 'Additional File'),
- ('Audio', 'Audio file'),
- ('Video', 'Video file'),
- ('Script', 'Script'),
- ('Data', 'Data'),
- ('Figure', 'Figure'),
- ('Schema', 'Schema'),
- ('Graph', 'Graph'),
- ('Image', 'Image'),
- ('Drawing', 'Drawing'),
- ('Slides', 'Slides')],
- can_revise_doctypes=['*'],
- can_comment_doctypes=['*'],
- can_describe_doctypes=['*'],
- can_delete_doctypes=['*'],
- can_keep_doctypes=['*'],
- can_rename_doctypes=['*'],
- can_add_format_to_doctypes=['*'],
- can_restrict_doctypes=['*'],
- restrictions_and_desc=[('', 'Public'),
- ('restricted', 'Restricted')],
uid=uid,
sbm_access=access,
- display_hidden_files=True)[1]
+ display_hidden_files=True,
+ restrictions_and_desc=CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_RESTRICTIONS,
+ doctypes_and_desc=CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_DOCTYPES,
+ **CFG_WEBSUBMIT_DOCUMENT_FILE_MANAGER_MISC)[1]
body += '''
''' % \
{'apply_changes': _("Apply changes"),
'cancel_changes': _("Cancel all changes"),
'recid': argd['recid'],
'access': access,
'ln': argd['ln'],
'CFG_SITE_URL': CFG_SITE_URL}
body += websubmit_templates.tmpl_page_do_not_leave_submission_js(argd['ln'], enabled=True)
return page(title = _("Document File Manager") + (argd['recid'] and (': ' + _("Record #%i") % argd['recid']) or ''),
navtrail=navtrail,
navtrail_append_title_p=0,
metaheaderadd = get_upload_file_interface_javascript(form_url_params='?access='+access) + \
get_upload_file_interface_css(),
body = body,
uid = uid,
language=argd['ln'],
req=req,
navmenuid='admin')
def managedocfilesasync(self, req, form):
"Upload file and returns upload interface"
argd = wash_urlargd(form, {
'ln': (str, ''),
'recid': (int, 1),
'doctype': (str, ''),
'access': (str, ''),
'indir': (str, ''),
})
user_info = collect_user_info(req)
include_headers = False
# User submitted either through WebSubmit, or admin interface.
if form.has_key('doctype') and form.has_key('indir') \
and form.has_key('access'):
# Submitted through WebSubmit. Check rights
include_headers = True
working_dir = os.path.join(CFG_WEBSUBMIT_STORAGEDIR,
argd['indir'], argd['doctype'],
argd['access'])
try:
assert(working_dir == os.path.abspath(working_dir))
except AssertionError:
raise apache.SERVER_RETURN(apache.HTTP_UNAUTHORIZED)
try:
# Retrieve recid from working_dir, safer.
recid_fd = file(os.path.join(working_dir, 'SN'))
recid = int(recid_fd.read())
recid_fd.close()
except:
recid = ""
try:
act_fd = file(os.path.join(working_dir, 'act'))
action = act_fd.read()
act_fd.close()
except:
action = ""
# Is user authorized to perform this action?
(auth_code, auth_msg) = acc_authorize_action(user_info,
"submit",
authorized_if_no_roles=not isGuestUser(getUid(req)),
doctype=argd['doctype'],
act=action)
if not acc_is_role("submit", doctype=argd['doctype'], act=action):
# There is NO authorization plugged. User should have access
auth_code = 0
else:
# User must be allowed to attach files
(auth_code, auth_msg) = acc_authorize_action(user_info,
'runbibdocfile')
recid = argd['recid']
if auth_code:
raise apache.SERVER_RETURN(apache.HTTP_UNAUTHORIZED)
return create_file_upload_interface(recid=recid,
ln=argd['ln'],
print_outside_form_tag=False,
print_envelope=False,
form=form,
include_headers=include_headers,
sbm_indir=argd['indir'],
sbm_access=argd['access'],
sbm_doctype=argd['doctype'],
uid=user_info['uid'])[1]
def uploadfile(self, req, form):
"""
Similar to /submit, but only consider files. Nice for
asynchronous Javascript uploads. Should be used to upload a
single file.
Also try to create an icon, and return URL to file(s) + icon(s)
Authentication is performed based on session ID passed as
parameter instead of cookie-based authentication, due to the
use of this URL by the Flash plugin (to upload multiple files
at once), which does not route cookies.
FIXME: consider adding /deletefile and /modifyfile functions +
parsing of additional parameters to rename files, add
comments, restrictions, etc.
"""
argd = wash_urlargd(form, {
'doctype': (str, ''),
'access': (str, ''),
'indir': (str, ''),
'session_id': (str, ''),
'rename': (str, ''),
})
curdir = None
if not form.has_key("indir") or \
not form.has_key("doctype") or \
not form.has_key("access"):
raise apache.SERVER_RETURN(apache.HTTP_BAD_REQUEST)
else:
curdir = os.path.join(CFG_WEBSUBMIT_STORAGEDIR,
argd['indir'],
argd['doctype'],
argd['access'])
user_info = collect_user_info(req)
if form.has_key("session_id"):
# Are we uploading using Flash, which does not transmit
# cookie? The expect to receive session_id as a form
# parameter. First check that IP addresses do not
# mismatch. A ValueError will be raises if there is
# something wrong
session = get_session(req=req, sid=argd['session_id'])
try:
session = get_session(req=req, sid=argd['session_id'])
except ValueError, e:
raise apache.SERVER_RETURN(apache.HTTP_BAD_REQUEST)
# Retrieve user information. We cannot rely on the session here.
res = run_sql("SELECT uid FROM session WHERE session_key=%s", (argd['session_id'],))
if len(res):
uid = res[0][0]
user_info = collect_user_info(uid)
try:
act_fd = file(os.path.join(curdir, 'act'))
action = act_fd.read()
act_fd.close()
except:
action = ""
# Is user authorized to perform this action?
(auth_code, auth_message) = acc_authorize_action(uid, "submit",
authorized_if_no_roles=not isGuestUser(uid),
verbose=0,
doctype=argd['doctype'],
act=action)
if acc_is_role("submit", doctype=argd['doctype'], act=action) and auth_code != 0:
# User cannot submit
raise apache.SERVER_RETURN(apache.HTTP_UNAUTHORIZED)
else:
# Process the upload and get the response
added_files = {}
for key, formfields in form.items():
filename = key.replace("[]", "")
file_to_open = os.path.join(curdir, filename)
if hasattr(formfields, "filename") and formfields.filename:
dir_to_open = os.path.abspath(os.path.join(curdir,
'files',
str(user_info['uid']),
key))
try:
assert(dir_to_open.startswith(CFG_WEBSUBMIT_STORAGEDIR))
except AssertionError:
register_exception(req=req, prefix='curdir="%s", key="%s"' % (curdir, key))
raise apache.SERVER_RETURN(apache.HTTP_FORBIDDEN)
if not os.path.exists(dir_to_open):
try:
os.makedirs(dir_to_open)
except OSError, e:
if e.errno != errno.EEXIST:
# If the issue is only that directory
# already exists, then continue, else
# report
register_exception(req=req, alert_admin=True)
raise apache.SERVER_RETURN(apache.HTTP_FORBIDDEN)
filename = formfields.filename
## Before saving the file to disc, wash the filename (in particular
## washing away UNIX and Windows (e.g. DFS) paths):
filename = os.path.basename(filename.split('\\')[-1])
filename = filename.strip()
if filename != "":
# Check that file does not already exist
n = 1
while os.path.exists(os.path.join(dir_to_open, filename)):
#dirname, basename, extension = decompose_file(new_destination_path)
basedir, name, extension = decompose_file(filename)
new_name = propose_next_docname(name)
filename = new_name + extension
# This may be dangerous if the file size is bigger than the available memory
fp = open(os.path.join(dir_to_open, filename), "w")
fp.write(formfields.file.read())
fp.close()
fp = open(os.path.join(curdir, "lastuploadedfile"), "w")
fp.write(filename)
fp.close()
fp = open(file_to_open, "w")
fp.write(filename)
fp.close()
try:
# Create icon
(icon_path, icon_name) = create_icon(
{ 'input-file' : os.path.join(dir_to_open, filename),
'icon-name' : filename, # extension stripped automatically
'icon-file-format' : 'gif',
'multipage-icon' : False,
'multipage-icon-delay' : 100,
'icon-scale' : "300>", # Resize only if width > 300
'verbosity' : 0,
})
icons_dir = os.path.join(os.path.join(curdir,
'icons',
str(user_info['uid']),
key))
if not os.path.exists(icons_dir):
# Create uid/icons dir if needed
try:
os.makedirs(icons_dir)
except OSError, e:
if e.errno != errno.EEXIST:
# If the issue is only that
# directory already exists,
# then continue, else report
register_exception(req=req, alert_admin=True)
raise apache.SERVER_RETURN(apache.HTTP_FORBIDDEN)
os.rename(os.path.join(icon_path, icon_name),
os.path.join(icons_dir, icon_name))
added_files[key] = {'name': filename,
'iconName': icon_name}
except InvenioWebSubmitIconCreatorError, e:
# We could not create the icon
added_files[key] = {'name': filename}
continue
else:
raise apache.SERVER_RETURN(apache.HTTP_BAD_REQUEST)
# Send our response
if CFG_JSON_AVAILABLE:
return json.dumps(added_files)
def upload_video(self, req, form):
"""
A clone of uploadfile but for (large) videos.
Does not copy the uploaded file to the websubmit directory.
Instead, the path to the file is stored inside the submission directory.
"""
def gcd(a,b):
""" the euclidean algorithm """
while a:
a, b = b%a, a
return b
from invenio.bibencode_extract import extract_frames
from invenio.bibencode_config import CFG_BIBENCODE_WEBSUBMIT_ASPECT_SAMPLE_DIR, CFG_BIBENCODE_WEBSUBMIT_ASPECT_SAMPLE_FNAME
from invenio.bibencode_encode import determine_aspect
from invenio.bibencode_utils import probe
from invenio.bibencode_metadata import ffprobe_metadata
from invenio.websubmit_config import CFG_WEBSUBMIT_TMP_VIDEO_PREFIX
argd = wash_urlargd(form, {
'doctype': (str, ''),
'access': (str, ''),
'indir': (str, ''),
'session_id': (str, ''),
'rename': (str, ''),
})
curdir = None
if not form.has_key("indir") or \
not form.has_key("doctype") or \
not form.has_key("access"):
raise apache.SERVER_RETURN(apache.HTTP_BAD_REQUEST)
else:
curdir = os.path.join(CFG_WEBSUBMIT_STORAGEDIR,
argd['indir'],
argd['doctype'],
argd['access'])
user_info = collect_user_info(req)
if form.has_key("session_id"):
# Are we uploading using Flash, which does not transmit
# cookie? The expect to receive session_id as a form
# parameter. First check that IP addresses do not
# mismatch. A ValueError will be raises if there is
# something wrong
session = get_session(req=req, sid=argd['session_id'])
try:
session = get_session(req=req, sid=argd['session_id'])
except ValueError, e:
raise apache.SERVER_RETURN(apache.HTTP_BAD_REQUEST)
# Retrieve user information. We cannot rely on the session here.
res = run_sql("SELECT uid FROM session WHERE session_key=%s", (argd['session_id'],))
if len(res):
uid = res[0][0]
user_info = collect_user_info(uid)
try:
act_fd = file(os.path.join(curdir, 'act'))
action = act_fd.read()
act_fd.close()
except:
act = ""
# Is user authorized to perform this action?
(auth_code, auth_message) = acc_authorize_action(uid, "submit",
authorized_if_no_roles=not isGuestUser(uid),
verbose=0,
doctype=argd['doctype'],
act=action)
if acc_is_role("submit", doctype=argd['doctype'], act=action) and auth_code != 0:
# User cannot submit
raise apache.SERVER_RETURN(apache.HTTP_UNAUTHORIZED)
else:
# Process the upload and get the response
json_response = {}
for key, formfields in form.items():
filename = key.replace("[]", "")
if hasattr(formfields, "filename") and formfields.filename:
dir_to_open = os.path.abspath(os.path.join(curdir,
'files',
str(user_info['uid']),
key))
try:
assert(dir_to_open.startswith(CFG_WEBSUBMIT_STORAGEDIR))
except AssertionError:
register_exception(req=req, prefix='curdir="%s", key="%s"' % (curdir, key))
raise apache.SERVER_RETURN(apache.HTTP_FORBIDDEN)
if not os.path.exists(dir_to_open):
try:
os.makedirs(dir_to_open)
except OSError, e:
if e.errno != errno.EEXIST:
# If the issue is only that directory
# already exists, then continue, else
# report
register_exception(req=req, alert_admin=True)
raise apache.SERVER_RETURN(apache.HTTP_FORBIDDEN)
filename = formfields.filename
## Before saving the file to disc, wash the filename (in particular
## washing away UNIX and Windows (e.g. DFS) paths):
filename = os.path.basename(filename.split('\\')[-1])
filename = filename.strip()
if filename != "":
# Check that file does not already exist
while os.path.exists(os.path.join(dir_to_open, filename)):
#dirname, basename, extension = decompose_file(new_destination_path)
basedir, name, extension = decompose_file(filename)
new_name = propose_next_docname(name)
filename = new_name + extension
#-------------#
# VIDEO STUFF #
#-------------#
## Remove all previous uploads
filelist = os.listdir(os.path.split(formfields.file.name)[0])
for afile in filelist:
if argd['access'] in afile:
os.remove(os.path.join(os.path.split(formfields.file.name)[0], afile))
## Check if the file is a readable video
## We must exclude all image and audio formats that are readable by ffprobe
if (os.path.splitext(filename)[1] in ['jpg', 'jpeg', 'gif', 'tiff', 'bmp', 'png', 'tga',
'jp2', 'j2k', 'jpf', 'jpm', 'mj2', 'biff', 'cgm',
'exif', 'img', 'mng', 'pic', 'pict', 'raw', 'wmf', 'jpe', 'jif',
'jfif', 'jfi', 'tif', 'webp', 'svg', 'ai', 'ps', 'psd',
'wav', 'mp3', 'pcm', 'aiff', 'au', 'flac', 'wma', 'm4a', 'wv', 'oga',
'm4a', 'm4b', 'm4p', 'm4r', 'aac', 'mp4', 'vox', 'amr', 'snd']
or not probe(formfields.file.name)):
formfields.file.close()
raise apache.SERVER_RETURN(apache.HTTP_FORBIDDEN)
## We have no "delete" attribute in Python 2.4
if sys.hexversion < 0x2050000:
## We need to rename first and create a dummy file
## Rename the temporary file for the garbage collector
new_tmp_fullpath = os.path.split(formfields.file.name)[0] + "/" + CFG_WEBSUBMIT_TMP_VIDEO_PREFIX + argd['access'] + "_" + os.path.split(formfields.file.name)[1]
os.rename(formfields.file.name, new_tmp_fullpath)
dummy = open(formfields.file.name, "w")
dummy.close()
formfields.file.close()
else:
# Mark the NamedTemporatyFile as not to be deleted
formfields.file.delete = False
formfields.file.close()
## Rename the temporary file for the garbage collector
new_tmp_fullpath = os.path.split(formfields.file.name)[0] + "/" + CFG_WEBSUBMIT_TMP_VIDEO_PREFIX + argd['access'] + "_" + os.path.split(formfields.file.name)[1]
os.rename(formfields.file.name, new_tmp_fullpath)
# Write the path to the temp file to a file in STORAGEDIR
fp = open(os.path.join(dir_to_open, "filepath"), "w")
fp.write(new_tmp_fullpath)
fp.close()
fp = open(os.path.join(dir_to_open, "filename"), "w")
fp.write(filename)
fp.close()
## We are going to extract some thumbnails for websubmit ##
sample_dir = os.path.join(curdir, 'files', str(user_info['uid']), CFG_BIBENCODE_WEBSUBMIT_ASPECT_SAMPLE_DIR)
try:
## Remove old thumbnails
shutil.rmtree(sample_dir)
except OSError:
register_exception(req=req, alert_admin=False)
try:
os.makedirs(os.path.join(curdir, 'files', str(user_info['uid']), sample_dir))
except OSError:
register_exception(req=req, alert_admin=False)
try:
extract_frames(input_file=new_tmp_fullpath,
output_file=os.path.join(sample_dir, CFG_BIBENCODE_WEBSUBMIT_ASPECT_SAMPLE_FNAME),
size="600x600",
numberof=5)
json_response['frames'] = []
for extracted_frame in os.listdir(sample_dir):
json_response['frames'].append(extracted_frame)
except:
## If the frame extraction fails, something was bad with the video
os.remove(new_tmp_fullpath)
register_exception(req=req, alert_admin=False)
raise apache.SERVER_RETURN(apache.HTTP_FORBIDDEN)
## Try to detect the aspect. if this fails, the video is not readable
## or a wrong file might have been uploaded
try:
(aspect, width, height) = determine_aspect(new_tmp_fullpath)
if aspect:
aspx, aspy = aspect.split(':')
else:
the_gcd = gcd(width, height)
aspx = str(width / the_gcd)
aspy = str(height / the_gcd)
json_response['aspx'] = aspx
json_response['aspy'] = aspy
except TypeError:
## If the aspect detection completely fails
os.remove(new_tmp_fullpath)
register_exception(req=req, alert_admin=False)
raise apache.SERVER_RETURN(apache.HTTP_FORBIDDEN)
## Try to extract some metadata from the video container
metadata = ffprobe_metadata(new_tmp_fullpath)
json_response['meta_title'] = metadata['format'].get('TAG:title')
json_response['meta_description'] = metadata['format'].get('TAG:description')
json_response['meta_year'] = metadata['format'].get('TAG:year')
json_response['meta_author'] = metadata['format'].get('TAG:author')
## Empty file name
else:
raise apache.SERVER_RETURN(apache.HTTP_BAD_REQUEST)
## We found our file, we can break the loop
break;
# Send our response
if CFG_JSON_AVAILABLE:
dumped_response = json.dumps(json_response)
# store the response in the websubmit directory
# this is needed if the submission is not finished and continued later
response_dir = os.path.join(curdir, 'files', str(user_info['uid']), "response")
try:
os.makedirs(response_dir)
except OSError:
# register_exception(req=req, alert_admin=False)
pass
fp = open(os.path.join(response_dir, "response"), "w")
fp.write(dumped_response)
fp.close()
return dumped_response
def getuploadedfile(self, req, form):
"""
Stream uploaded files.
For the moment, restrict to files in ./curdir/files/uid or
./curdir/icons/uid directory, so that we are sure we stream
files only to the user who uploaded them.
"""
argd = wash_urlargd(form, {'indir': (str, None),
'doctype': (str, None),
'access': (str, None),
'icon': (int, 0),
'key': (str, None),
'filename': (str, None),
'nowait': (int, 0)})
if None in argd.values():
raise apache.SERVER_RETURN(apache.HTTP_BAD_REQUEST)
uid = getUid(req)
if argd['icon']:
file_path = os.path.join(CFG_WEBSUBMIT_STORAGEDIR,
argd['indir'],
argd['doctype'],
argd['access'],
'icons',
str(uid),
argd['key'],
argd['filename']
)
else:
file_path = os.path.join(CFG_WEBSUBMIT_STORAGEDIR,
argd['indir'],
argd['doctype'],
argd['access'],
'files',
str(uid),
argd['key'],
argd['filename']
)
abs_file_path = os.path.abspath(file_path)
if abs_file_path.startswith(CFG_WEBSUBMIT_STORAGEDIR):
# Check if file exist. Note that icon might not yet have
# been created.
if not argd['nowait']:
for i in range(5):
if os.path.exists(abs_file_path):
return stream_file(req, abs_file_path)
time.sleep(1)
else:
if os.path.exists(abs_file_path):
return stream_file(req, abs_file_path)
# Send error 404 in all other cases
raise apache.SERVER_RETURN(apache.HTTP_NOT_FOUND)
def attachfile(self, req, form):
"""
Process requests received from CKEditor to upload files.
If the uploaded file is an image, create an icon version
"""
if not is_html_text_editor_installed():
return apache.HTTP_NOT_FOUND
if not form.has_key('type'):
form['type'] = 'File'
if not form.has_key('upload') or \
not form['type'] in \
['File', 'Image', 'Flash', 'Media']:
#return apache.HTTP_NOT_FOUND
pass
filetype = form['type'].lower()
uid = getUid(req)
# URL where the file can be fetched after upload
user_files_path = '%(CFG_SITE_URL)s/submit/getattachedfile/%(uid)s' % \
{'uid': uid,
'CFG_SITE_URL': CFG_SITE_URL,
'filetype': filetype}
# Path to directory where uploaded files are saved
user_files_absolute_path = '%(CFG_PREFIX)s/var/tmp/attachfile/%(uid)s/%(filetype)s' % \
{'uid': uid,
'CFG_PREFIX': CFG_PREFIX,
'filetype': filetype}
try:
os.makedirs(user_files_absolute_path)
except:
pass
user_info = collect_user_info(req)
(auth_code, auth_message) = acc_authorize_action(user_info, 'attachsubmissionfile')
msg = ""
if user_info['email'] == 'guest':
# User is guest: must login prior to upload
msg = 'Please login before uploading file.'
elif auth_code:
# User cannot submit
msg = 'Sorry, you are not allowed to submit files.'
## elif len(form['upload']) != 1:
## msg = 'Sorry, you must upload one single file'
else:
# Process the upload and get the response
(msg, uploaded_file_path, uploaded_file_name, uploaded_file_url, callback_function) = \
process_CKEditor_upload(form, uid, user_files_path, user_files_absolute_path)
if uploaded_file_path:
# Create an icon
if form.get('type','') == 'Image':
try:
(icon_path, icon_name) = create_icon(
{ 'input-file' : uploaded_file_path,
'icon-name' : os.path.splitext(uploaded_file_name)[0],
'icon-file-format' : os.path.splitext(uploaded_file_name)[1][1:] or 'gif',
'multipage-icon' : False,
'multipage-icon-delay' : 100,
'icon-scale' : "300>", # Resize only if width > 300
'verbosity' : 0,
})
# Move original file to /original dir, and replace it with icon file
original_user_files_absolute_path = os.path.join(user_files_absolute_path,
'original')
if not os.path.exists(original_user_files_absolute_path):
# Create /original dir if needed
os.mkdir(original_user_files_absolute_path)
os.rename(uploaded_file_path,
original_user_files_absolute_path + os.sep + uploaded_file_name)
os.rename(icon_path + os.sep + icon_name,
uploaded_file_path)
except InvenioWebSubmitIconCreatorError, e:
pass
user_files_path += '/' + filetype + '/' + uploaded_file_name
else:
user_files_path = ''
if not msg:
msg = 'No valid file found'
# Send our response
send_response(req, msg, user_files_path, callback_function)
def _lookup(self, component, path):
""" This handler is invoked for the dynamic URLs (for getting
and putting attachments) Eg:
/submit/getattachedfile/41336978/image/myfigure.png
/submit/attachfile/41336978/image/myfigure.png
"""
if component == 'getattachedfile' and len(path) > 2:
uid = path[0] # uid of the submitter
file_type = path[1] # file, image, flash or media (as
# defined by CKEditor)
if file_type in ['file', 'image', 'flash', 'media']:
file_name = '/'.join(path[2:]) # the filename
def answer_get(req, form):
"""Accessing files attached to submission."""
form['file'] = file_name
form['type'] = file_type
form['uid'] = uid
return self.getattachedfile(req, form)
return answer_get, []
# All other cases: file not found
return None, []
def getattachedfile(self, req, form):
"""
Returns a file uploaded to the submission 'drop box' by the
CKEditor.
"""
argd = wash_urlargd(form, {'file': (str, None),
'type': (str, None),
'uid': (int, 0)})
# Can user view this record, i.e. can user access its
# attachments?
uid = getUid(req)
user_info = collect_user_info(req)
if not argd['file'] is None:
# Prepare path to file on disk. Normalize the path so that
# ../ and other dangerous components are removed.
path = os.path.abspath(CFG_PREFIX + '/var/tmp/attachfile/' + \
'/' + str(argd['uid']) + \
'/' + argd['type'] + '/' + argd['file'])
# Check that we are really accessing attachements
# directory, for the declared record.
if path.startswith(CFG_PREFIX + '/var/tmp/attachfile/') and os.path.exists(path):
return stream_file(req, path)
# Send error 404 in all other cases
return(apache.HTTP_NOT_FOUND)
def continue_(self, req, form):
"""
Continue an interrupted submission.
"""
args = wash_urlargd(form, {'access': (str, ''), 'doctype': (str, '')})
ln = args['ln']
_ = gettext_set_language(ln)
access = args['access']
doctype = args['doctype']
if not access or not doctype:
return warningMsg(_("Sorry, invalid arguments"), req=req, ln=ln)
user_info = collect_user_info(req)
email = user_info['email']
res = run_sql("SELECT action, status FROM sbmSUBMISSIONS WHERE id=%s AND email=%s and doctype=%s", (access, email, doctype))
if res:
action, status = res[0]
if status == 'finished':
return warningMsg(_("Note: the requested submission has already been completed"), req=req, ln=ln)
redirect_to_url(req, CFG_SITE_SECURE_URL + '/submit/direct?' + urlencode({
'sub': action + doctype,
'access': access}))
return warningMsg(_("Sorry, you don't seem to have initiated a submission with the provided access number"), req=req, ln=ln)
def direct(self, req, form):
"""Directly redirected to an initialized submission."""
args = wash_urlargd(form, {'sub': (str, ''),
'access' : (str, '')})
sub = args['sub']
access = args['access']
ln = args['ln']
_ = gettext_set_language(ln)
uid = getUid(req)
if uid == -1 or CFG_ACCESS_CONTROL_LEVEL_SITE >= 1:
return page_not_authorized(req, "direct",
navmenuid='submit')
myQuery = req.args
if not sub:
return warningMsg(_("Sorry, 'sub' parameter missing..."), req, ln=ln)
res = run_sql("SELECT docname,actname FROM sbmIMPLEMENT WHERE subname=%s", (sub,))
if not res:
return warningMsg(_("Sorry. Cannot analyse parameter"), req, ln=ln)
else:
# get document type
doctype = res[0][0]
# get action name
action = res[0][1]
# retrieve other parameter values
params = dict(form)
# find existing access number
if not access:
# create 'unique' access number
pid = os.getpid()
now = time.time()
access = "%i_%s" % (now,pid)
# retrieve 'dir' value
res = run_sql ("SELECT dir FROM sbmACTION WHERE sactname=%s", (action,))
dir = res[0][0]
mainmenu = req.headers_in.get('referer')
params['access'] = access
params['act'] = action
params['doctype'] = doctype
params['startPg'] = '1'
params['mainmenu'] = mainmenu
params['ln'] = ln
params['indir'] = dir
url = "%s/submit?%s" % (CFG_SITE_SECURE_URL, urlencode(params))
redirect_to_url(req, url)
def sub(self, req, form):
"""DEPRECATED: /submit/sub is deprecated now, so raise email to the admin (but allow submission to continue anyway)"""
args = wash_urlargd(form, {'password': (str, '')})
uid = getUid(req)
if uid == -1 or CFG_ACCESS_CONTROL_LEVEL_SITE >= 1:
return page_not_authorized(req, "../sub/",
navmenuid='submit')
try:
raise DeprecationWarning, 'submit/sub handler has been used. Please use submit/direct. e.g. "submit/sub?RN=123@SBIFOO" -> "submit/direct?RN=123&sub=SBIFOO"'
except DeprecationWarning:
register_exception(req=req, alert_admin=True)
ln = args['ln']
_ = gettext_set_language(ln)
#DEMOBOO_RN=DEMO-BOOK-2008-001&ln=en&password=1223993532.26572%40APPDEMOBOO
params = dict(form)
password = args['password']
if password:
del params['password']
if "@" in password:
params['access'], params['sub'] = password.split('@', 1)
else:
params['sub'] = password
else:
args = str(req.args).split('@')
if len(args) > 1:
params = {'sub' : args[-1]}
args = '@'.join(args[:-1])
params.update(cgi.parse_qs(args))
else:
return warningMsg(_("Sorry, invalid URL..."), req, ln=ln)
url = "%s/submit/direct?%s" % (CFG_SITE_SECURE_URL, urlencode(params, doseq=True))
redirect_to_url(req, url)
def summary(self, req, form):
args = wash_urlargd(form, {
'doctype': (str, ''),
'act': (str, ''),
'access': (str, ''),
'indir': (str, '')})
uid = getUid(req)
if uid == -1 or CFG_ACCESS_CONTROL_LEVEL_SITE >= 1:
return page_not_authorized(req, "../summary",
navmenuid='submit')
t=""
curdir = os.path.join(CFG_WEBSUBMIT_STORAGEDIR, args['indir'], args['doctype'], args['access'])
try:
assert(curdir == os.path.abspath(curdir))
except AssertionError:
register_exception(req=req, alert_admin=True, prefix='Possible cracking tentative: indir="%s", doctype="%s", access="%s"' % (args['indir'], args['doctype'], args['access']))
return warningMsg("Invalid parameters", req)
subname = "%s%s" % (args['act'], args['doctype'])
res = run_sql("select sdesc,fidesc,pagenb,level from sbmFIELD where subname=%s "
"order by pagenb,fieldnb", (subname,))
nbFields = 0
values = []
for arr in res:
if arr[0] != "":
val = {
'mandatory' : (arr[3] == 'M'),
'value' : '',
'page' : arr[2],
'name' : arr[0],
}
if os.path.exists(os.path.join(curdir, curdir,arr[1])):
fd = open(os.path.join(curdir, arr[1]),"r")
value = fd.read()
fd.close()
value = value.replace("\n"," ")
value = value.replace("Select:","")
else:
value = ""
val['value'] = value
values.append(val)
return websubmit_templates.tmpl_submit_summary(
ln = args['ln'],
values = values,
)
def index(self, req, form):
args = wash_urlargd(form, {
'c': (str, CFG_SITE_NAME),
'doctype': (str, ''),
'act': (str, ''),
'startPg': (str, "1"),
'access': (str, ''),
'mainmenu': (str, ''),
'fromdir': (str, ''),
'nextPg': (str, ''),
'nbPg': (str, ''),
'curpage': (str, '1'),
'step': (str, '0'),
'mode': (str, 'U'),
})
## Strip whitespace from beginning and end of doctype and action:
args["doctype"] = args["doctype"].strip()
args["act"] = args["act"].strip()
def _index(req, c, ln, doctype, act, startPg, access,
mainmenu, fromdir, nextPg, nbPg, curpage, step,
mode):
auth_args = {}
if doctype:
auth_args['doctype'] = doctype
if act:
auth_args['act'] = act
uid = getUid(req)
if CFG_CERN_SITE:
## HACK BEGIN: this is a hack for CMS and ATLAS draft
from invenio.webuser import collect_user_info
user_info = collect_user_info(req)
if doctype == 'CMSPUB' and act == "" and 'cds-admin [CERN]' not in user_info['group'] and not user_info['email'].lower() == 'cds.support@cern.ch':
if 'cms-publication-committee-chair [CERN]' not in user_info['group']:
return page_not_authorized(req, "../submit", text="In order to access this submission interface you need to be member of the CMS Publication Committee Chair.",
navmenuid='submit')
elif doctype == 'ATLPUB' and 'cds-admin [CERN]' not in user_info['group'] and not user_info['email'].lower() == 'cds.support@cern.ch':
if 'atlas-gen [CERN]' not in user_info['group']:
return page_not_authorized(req, "../submit", text="In order to access this submission interface you need to be member of ATLAS.",
navmenuid='submit')
## HACK END
if doctype == "":
catalogues_text, at_least_one_submission_authorized, submission_exists= makeCataloguesTable(req, ln=CFG_SITE_LANG)
if not at_least_one_submission_authorized and submission_exists:
if isGuestUser(uid):
return redirect_to_url(req, "%s/youraccount/login%s" % (
CFG_SITE_SECURE_URL,
make_canonical_urlargd({'referer' : CFG_SITE_SECURE_URL + req.unparsed_uri, 'ln' : args['ln']}, {}))
, norobot=True)
else:
return page_not_authorized(req, "../submit",
uid=uid,
navmenuid='submit')
return home(req,catalogues_text, c,ln)
elif act == "":
return action(req,c,ln,doctype)
elif int(step)==0:
return interface(req, c, ln, doctype, act, startPg, access, mainmenu, fromdir, nextPg, nbPg, curpage)
else:
return endaction(req, c, ln, doctype, act, startPg, access,mainmenu, fromdir, nextPg, nbPg, curpage, step, mode)
return _index(req, **args)
# Answer to both /submit/ and /submit
__call__ = index
def errorMsg(title, req, c=None, ln=CFG_SITE_LANG):
# load the right message language
_ = gettext_set_language(ln)
if c is None:
c = CFG_SITE_NAME_INTL.get(ln, CFG_SITE_NAME)
return page(title = _("Error"),
body = create_error_box(req, title=str(title), verbose=0, ln=ln),
description="%s - Internal Error" % c,
keywords="%s, Internal Error" % c,
uid = getUid(req),
language=ln,
req=req,
navmenuid='submit')
def warningMsg(title, req, c=None, ln=CFG_SITE_LANG):
# load the right message language
_ = gettext_set_language(ln)
if c is None:
c = CFG_SITE_NAME_INTL.get(ln, CFG_SITE_NAME)
return page(title = _("Warning"),
body = title,
description="%s - Internal Error" % c,
keywords="%s, Internal Error" % c,
uid = getUid(req),
language=ln,
req=req,
navmenuid='submit')
def print_warning(msg, type='', prologue='
', epilogue='
'):
"""Prints warning message and flushes output."""
if msg:
return websubmit_templates.tmpl_print_warning(
msg = msg,
type = type,
prologue = prologue,
epilogue = epilogue,
)
else:
return ''
## def retrieve_most_recent_attached_file(file_path):
## """
## Retrieve the latest file that has been uploaded with the
## CKEditor. This is the only way to retrieve files that the
## CKEditor has renamed after the upload.
## Eg: 'prefix/image.jpg' was uploaded but did already
## exist. CKEditor silently renamed it to 'prefix/image(1).jpg':
## >>> retrieve_most_recent_attached_file('prefix/image.jpg')
## 'prefix/image(1).jpg'
## """
## (base_path, filename) = os.path.split(file_path)
## base_name = os.path.splitext(filename)[0]
## file_ext = os.path.splitext(filename)[1][1:]
## most_recent_filename = filename
## i = 0
## while True:
## i += 1
## possible_filename = "%s(%d).%s" % \
## (base_name, i, file_ext)
## if os.path.exists(base_path + os.sep + possible_filename):
## most_recent_filename = possible_filename
## else:
## break
## return os.path.join(base_path, most_recent_filename)
| ![[at]](at.gif)
example![[dot]](dot.gif)
com
CFG_WEBSTYLE_EMAIL_ADDRESSES_OBFUSCATION_MODE = 2
## CFG_WEBSTYLE_INSPECT_TEMPLATES -- Do we want to debug all template
## functions so that they would return HTML results wrapped in
## comments indicating which part of HTML page was created by which
## template function? Useful only for debugging Pythonic HTML
## template. See WebStyle Admin Guide for more information.
CFG_WEBSTYLE_INSPECT_TEMPLATES = 0
## (deprecated) CFG_WEBSTYLE_CDSPAGEBOXLEFTTOP -- eventual global HTML
## left top box:
CFG_WEBSTYLE_CDSPAGEBOXLEFTTOP =
## (deprecated) CFG_WEBSTYLE_CDSPAGEBOXLEFTBOTTOM -- eventual global
## HTML left bottom box:
CFG_WEBSTYLE_CDSPAGEBOXLEFTBOTTOM =
## (deprecated) CFG_WEBSTYLE_CDSPAGEBOXRIGHTTOP -- eventual global
## HTML right top box:
CFG_WEBSTYLE_CDSPAGEBOXRIGHTTOP =
## (deprecated) CFG_WEBSTYLE_CDSPAGEBOXRIGHTBOTTOM -- eventual global
## HTML right bottom box:
CFG_WEBSTYLE_CDSPAGEBOXRIGHTBOTTOM =
## CFG_WEBSTYLE_HTTP_STATUS_ALERT_LIST -- when certain HTTP status
## codes are raised to the WSGI handler, the corresponding exceptions
## and error messages can be sent to the system administrator for
## inspecting. This is useful to detect and correct errors. The
## variable represents a comma-separated list of HTTP statuses that
## should alert admin. Wildcards are possible. If the status is
## followed by an "r", it means that a referer is required to exist
## (useful to distinguish broken known links from URL typos when 404
## errors are raised).
CFG_WEBSTYLE_HTTP_STATUS_ALERT_LIST = 404r,400,5*,41*
## CFG_WEBSTYLE_HTTP_USE_COMPRESSION -- whether to enable deflate
## compression of your HTTP/HTTPS connections. This will affect the Apache
## configuration snippets created by inveniocfg --create-apache-conf and
## the OAI-PMH Identify response.
CFG_WEBSTYLE_HTTP_USE_COMPRESSION = 0
## CFG_WEBSTYLE_REVERSE_PROXY_IPS -- if you are setting a multinode
## environment where an HTTP proxy such as mod_proxy is sitting in
## front of the Invenio web application and is forwarding requests to
## worker nodes, set here the the list of IP addresses of the allowed
## HTTP proxies. This is needed in order to avoid IP address spoofing
## when worker nodes are also available on the public Internet and
## might receive forged HTTP requests. Only HTTP requests coming from
## the specified IP addresses will be considered as forwarded from a
## reverse proxy. E.g. set this to '123.123.123.123'.
CFG_WEBSTYLE_REVERSE_PROXY_IPS =
##################################
## Part 3: WebSearch parameters ##
##################################
## This section contains some configuration parameters for WebSearch
## module. Please note that WebSearch is mostly configured on
## run-time via its WebSearch Admin web interface. The parameters
## below are the ones that you do not probably want to modify very
## often during the runtime. (Note that you may modify them
## afterwards too, though.)
## CFG_WEBSEARCH_SEARCH_CACHE_SIZE -- how many queries we want to
## cache in memory per one Apache httpd process? This cache is used
## mainly for "next/previous page" functionality, but it caches also
## "popular" user queries if more than one user happen to search for
## the same thing. Note that large numbers may lead to great memory
## consumption. We recommend a value not greater than 100.
CFG_WEBSEARCH_SEARCH_CACHE_SIZE = 0
## CFG_WEBSEARCH_FIELDS_CONVERT -- if you migrate from an older
## system, you may want to map field codes of your old system (such as
## 'ti') to Invenio/MySQL ("title"). Use Python dictionary syntax
## for the translation table, e.g. {'wau':'author', 'wti':'title'}.
## Usually you don't want to do that, and you would use empty dict {}.
CFG_WEBSEARCH_FIELDS_CONVERT = {}
## CFG_WEBSEARCH_LIGHTSEARCH_PATTERN_BOX_WIDTH -- width of the
## search pattern window in the light search interface, in
## characters. CFG_WEBSEARCH_LIGHTSEARCH_PATTERN_BOX_WIDTH = 60
CFG_WEBSEARCH_LIGHTSEARCH_PATTERN_BOX_WIDTH = 60
## CFG_WEBSEARCH_SIMPLESEARCH_PATTERN_BOX_WIDTH -- width of the search
## pattern window in the simple search interface, in characters.
CFG_WEBSEARCH_SIMPLESEARCH_PATTERN_BOX_WIDTH = 40
## CFG_WEBSEARCH_ADVANCEDSEARCH_PATTERN_BOX_WIDTH -- width of the
## search pattern window in the advanced search interface, in
## characters.
CFG_WEBSEARCH_ADVANCEDSEARCH_PATTERN_BOX_WIDTH = 30
## CFG_WEBSEARCH_NB_RECORDS_TO_SORT -- how many records do we still
## want to sort? For higher numbers we print only a warning and won't
## perform any sorting other than default 'latest records first', as
## sorting would be very time consuming then. We recommend a value of
## not more than a couple of thousands.
CFG_WEBSEARCH_NB_RECORDS_TO_SORT = 1000
## CFG_WEBSEARCH_CALL_BIBFORMAT -- if a record is being displayed but
## it was not preformatted in the "HTML brief" format, do we want to
## call BibFormatting on the fly? Put "1" for "yes" and "0" for "no".
## Note that "1" will display the record exactly as if it were fully
## preformatted, but it may be slow due to on-the-fly processing; "0"
## will display a default format very fast, but it may not have all
## the fields as in the fully preformatted HTML brief format. Note
## also that this option is active only for old (PHP) formats; the new
## (Python) formats are called on the fly by default anyway, since
## they are much faster. When usure, please set "0" here.
CFG_WEBSEARCH_CALL_BIBFORMAT = 0
## CFG_WEBSEARCH_USE_ALEPH_SYSNOS -- do we want to make old SYSNOs
## visible rather than MySQL's record IDs? You may use this if you
## migrate from a different e-doc system, and you store your old
## system numbers into 970__a. Put "1" for "yes" and "0" for
## "no". Usually you don't want to do that, though.
CFG_WEBSEARCH_USE_ALEPH_SYSNOS = 0
## CFG_WEBSEARCH_I18N_LATEST_ADDITIONS -- Put "1" if you want the
## "Latest Additions" in the web collection pages to show
## internationalized records. Useful only if your brief BibFormat
## templates contains internationalized strings. Otherwise put "0" in
## order not to slow down the creation of latest additions by WebColl.
CFG_WEBSEARCH_I18N_LATEST_ADDITIONS = 0
## CFG_WEBSEARCH_INSTANT_BROWSE -- the number of records to display
## under 'Latest Additions' in the web collection pages.
CFG_WEBSEARCH_INSTANT_BROWSE = 10
## CFG_WEBSEARCH_INSTANT_BROWSE_RSS -- the number of records to
## display in the RSS feed.
CFG_WEBSEARCH_INSTANT_BROWSE_RSS = 25
## CFG_WEBSEARCH_RSS_I18N_COLLECTIONS -- comma-separated list of
## collections that feature an internationalized RSS feed on their
## main seach interface page created by webcoll. Other collections
## will have RSS feed using CFG_SITE_LANG.
CFG_WEBSEARCH_RSS_I18N_COLLECTIONS =
## CFG_WEBSEARCH_RSS_TTL -- number of minutes that indicates how long
## a feed cache is valid.
CFG_WEBSEARCH_RSS_TTL = 360
## CFG_WEBSEARCH_RSS_MAX_CACHED_REQUESTS -- maximum number of request kept
## in cache. If the cache is filled, following request are not cached.
CFG_WEBSEARCH_RSS_MAX_CACHED_REQUESTS = 1000
## CFG_WEBSEARCH_AUTHOR_ET_AL_THRESHOLD -- up to how many author names
## to print explicitely; for more print "et al". Note that this is
## used in default formatting that is seldomly used, as usually
## BibFormat defines all the format. The value below is only used
## when BibFormat fails, for example.
CFG_WEBSEARCH_AUTHOR_ET_AL_THRESHOLD = 3
## CFG_WEBSEARCH_NARROW_SEARCH_SHOW_GRANDSONS -- whether to show or
## not collection grandsons in Narrow Search boxes (sons are shown by
## default, grandsons are configurable here). Use 0 for no and 1 for
## yes.
CFG_WEBSEARCH_NARROW_SEARCH_SHOW_GRANDSONS = 1
## CFG_WEBSEARCH_CREATE_SIMILARLY_NAMED_AUTHORS_LINK_BOX -- shall we
## create help links for Ellis, Nick or Ellis, Nicholas and friends
## when Ellis, N was searched for? Useful if you have one author
## stored in the database under several name formats, namely surname
## comma firstname and surname comma initial cataloging policy. Use 0
## for no and 1 for yes.
CFG_WEBSEARCH_CREATE_SIMILARLY_NAMED_AUTHORS_LINK_BOX = 1
## CFG_WEBSEARCH_USE_MATHJAX_FOR_FORMATS -- MathJax is a JavaScript
## library that renders (La)TeX mathematical formulas in the client
## browser. This parameter must contain a comma-separated list of
## output formats for which to apply the MathJax rendering, for example
## "hb,hd". If the list is empty, MathJax is disabled.
CFG_WEBSEARCH_USE_MATHJAX_FOR_FORMATS =
## CFG_WEBSEARCH_EXTERNAL_COLLECTION_SEARCH_TIMEOUT -- when searching
## external collections (e.g. SPIRES, CiteSeer, etc), how many seconds
## do we wait for reply before abandonning?
CFG_WEBSEARCH_EXTERNAL_COLLECTION_SEARCH_TIMEOUT = 5
## CFG_WEBSEARCH_EXTERNAL_COLLECTION_SEARCH_MAXRESULTS -- how many
## results do we fetch?
CFG_WEBSEARCH_EXTERNAL_COLLECTION_SEARCH_MAXRESULTS = 10
## CFG_WEBSEARCH_SPLIT_BY_COLLECTION -- do we want to split the search
## results by collection or not? Use 0 for not, 1 for yes.
CFG_WEBSEARCH_SPLIT_BY_COLLECTION = 1
## CFG_WEBSEARCH_DEF_RECORDS_IN_GROUPS -- the default number of
## records to display per page in the search results pages.
CFG_WEBSEARCH_DEF_RECORDS_IN_GROUPS = 10
## CFG_WEBSEARCH_MAX_RECORDS_IN_GROUPS -- in order to limit denial of
## service attacks the total number of records per group displayed as a
## result of a search query will be limited to this number. Only the superuser
## queries will not be affected by this limit.
CFG_WEBSEARCH_MAX_RECORDS_IN_GROUPS = 200
## CFG_WEBSEARCH_PERMITTED_RESTRICTED_COLLECTIONS_LEVEL -- logged in users
## might have rights to access some restricted collections. This variable
## tweaks the kind of support the system will automatically provide to the
## user with respect to searching into these restricted collections.
## Set this to 0 in order to have the user to explicitly activate restricted
## collections in order to search into them. Set this to 1 in order to
## propose to the user the list of restricted collections to which he/she has
## rights (note: this is not yet implemented). Set this to 2 in order to
## silently add all the restricted collections to which the user has rights to
## to any query.
## Note: the system will discover which restricted collections a user has
## rights to, at login time. The time complexity of this procedure is
## proportional to the number of restricted collections. E.g. for a system
## with ~50 restricted collections, you might expect ~1s of delay in the
## login time, when this variable is set to a value higher than 0.
CFG_WEBSEARCH_PERMITTED_RESTRICTED_COLLECTIONS_LEVEL = 0
## CFG_WEBSEARCH_SHOW_COMMENT_COUNT -- do we want to show the 'N comments'
## links on the search engine pages? (useful only when you have allowed
## commenting)
CFG_WEBSEARCH_SHOW_COMMENT_COUNT = 1
## CFG_WEBSEARCH_SHOW_REVIEW_COUNT -- do we want to show the 'N reviews'
## links on the search engine pages? (useful only when you have allowed
## reviewing)
CFG_WEBSEARCH_SHOW_REVIEW_COUNT = 1
## CFG_WEBSEARCH_FULLTEXT_SNIPPETS -- how many full-text snippets to
## display for full-text searches?
CFG_WEBSEARCH_FULLTEXT_SNIPPETS = 4
## CFG_WEBSEARCH_FULLTEXT_SNIPPETS_WORDS -- how many context words
## to display around the pattern in the snippet?
CFG_WEBSEARCH_FULLTEXT_SNIPPETS_WORDS = 4
## CFG_WEBSEARCH_WILDCARD_LIMIT -- some of the queries, wildcard
## queries in particular (ex: cern*, a*), but also regular expressions
## (ex: [a-z]+), may take a long time to respond due to the high
## number of hits. You can limit the number of terms matched by a
## wildcard by setting this variable. A negative value or zero means
## that none of the queries will be limited (which may be wanted by
## also prone to denial-of-service kind of attacks).
CFG_WEBSEARCH_WILDCARD_LIMIT = 50000
## CFG_WEBSEARCH_SYNONYM_KBRS -- defines which knowledge bases are to
## be used for which index in order to provide runtime synonym lookup
## of user-supplied terms, and what massaging function should be used
## upon search pattern before performing the KB lookup. (Can be one
## of `exact', 'leading_to_comma', `leading_to_number'.)
CFG_WEBSEARCH_SYNONYM_KBRS = {
'journal': ['SEARCH-SYNONYM-JOURNAL', 'leading_to_number'],
}
## CFG_SOLR_URL -- optionally, you may use Solr to serve full-text
## queries. If so, please specify the URL of your Solr instance.
## Example: http://localhost:8983/solr (default solr port)
CFG_SOLR_URL =
## CFG_WEBSEARCH_PREV_NEXT_HIT_LIMIT -- specify the limit when
## the previous/next/back hit links are to be displayed on detailed record pages.
## In order to speeding up list manipulations, if a search returns lots of hits,
## more than this limit, then do not loose time calculating next/previous/back
## hits at all, but display page directly without these.
## Note also that Invenio installations that do not like
## to have the next/previous hit link functionality would be able to set this
## variable to zero and not see anything.
CFG_WEBSEARCH_PREV_NEXT_HIT_LIMIT = 1000
## CFG_WEBSEARCH_VIEWRESTRCOLL_POLICY -- when a record belongs to more than one
## restricted collection, if the viewrestcoll policy is set to "ALL" (default)
## then the user must be authorized to all the restricted collections, in
## order to be granted access to the specific record. If the policy is set to
## "ANY", then the user need to be authorized to only one of the collections
## in order to be granted access to the specific record.
CFG_WEBSEARCH_VIEWRESTRCOLL_POLICY = ALL
## CFG_WEBSEARCH_SPIRES_SYNTAX -- variable to configure the use of the
## SPIRES query syntax in searches. Values: 0 = SPIRES syntax is
## switched off; 1 = leading 'find' is required; 9 = leading 'find' is
## not required (leading SPIRES operator, space-operator-space, etc
## are also accepted).
CFG_WEBSEARCH_SPIRES_SYNTAX = 1
## CFG_WEBSEARCH_DISPLAY_NEAREST_TERMS -- when user search does not
## return any direct result, what do we want to display? Set to 0 in
## order to display a generic message about search returning no hits.
## Set to 1 in order to display list of nearest terms from the indexes
## that may match user query. Note: this functionality may be slow,
## so you may want to disable it on bigger sites.
CFG_WEBSEARCH_DISPLAY_NEAREST_TERMS = 1
## CFG_WEBSEARCH_DETAILED_META_FORMAT -- the output format to use for
## detailed meta tags containing metadata as configured in the tag
## table. Default output format should be 'hdm', included. This
## format will be included in the header of /record/ pages. For
## efficiency this format should be pre-cached with BibReformat. See
## also CFG_WEBSEARCH_ENABLE_GOOGLESCHOLAR and
## CFG_WEBSEARCH_ENABLE_GOOGLESCHOLAR.
CFG_WEBSEARCH_DETAILED_META_FORMAT = hdm
## CFG_WEBSEARCH_ENABLE_GOOGLESCHOLAR -- decides if meta tags for
## Google Scholar shall be included in the detailed record page
## header, when using the standard formatting templates/elements. See
## also CFG_WEBSEARCH_DETAILED_META_FORMAT and
## CFG_WEBSEARCH_ENABLE_OPENGRAPH. When this variable is changed and
## output format defined in CFG_WEBSEARCH_DETAILED_META_FORMAT is
## cached, a bibreformat must be run for the cached records.
CFG_WEBSEARCH_ENABLE_GOOGLESCHOLAR = True
## CFG_WEBSEARCH_ENABLE_OPENGRAPH -- decides if meta tags for the Open
## Graph protocol shall be included in the detailed record page
## header, when using the standard formatting templates/elements. See
## also CFG_WEBSEARCH_DETAILED_META_FORMAT and
## CFG_WEBSEARCH_ENABLE_GOOGLESCHOLAR. When this variable is changed
## and output format defined in CFG_WEBSEARCH_DETAILED_META_FORMAT is
## cached, a bibreformat must be run for the cached records. Note that
## enabling Open Graph produces invalid XHTML/HTML5 markup.
CFG_WEBSEARCH_ENABLE_OPENGRAPH = False
#######################################
## Part 4: BibHarvest OAI parameters ##
#######################################
## This part defines parameters for the Invenio OAI gateway.
## Useful if you are running Invenio as OAI data provider.
## CFG_OAI_ID_FIELD -- OAI identifier MARC field:
CFG_OAI_ID_FIELD = 909COo
## CFG_OAI_SET_FIELD -- OAI set MARC field:
CFG_OAI_SET_FIELD = 909COp
## CFG_OAI_SET_FIELD -- previous OAI set MARC field:
CFG_OAI_PREVIOUS_SET_FIELD = 909COq
## CFG_OAI_DELETED_POLICY -- OAI deletedrecordspolicy
## (no/transient/persistent):
CFG_OAI_DELETED_POLICY = persistent
## CFG_OAI_ID_PREFIX -- OAI identifier prefix:
CFG_OAI_ID_PREFIX = atlantis.cern.ch
## CFG_OAI_SAMPLE_IDENTIFIER -- OAI sample identifier:
CFG_OAI_SAMPLE_IDENTIFIER = oai:atlantis.cern.ch:123
## CFG_OAI_IDENTIFY_DESCRIPTION -- description for the OAI Identify verb:
CFG_OAI_IDENTIFY_DESCRIPTION =