def do_upgrade():
# first step: change tables
stmt = run_sql("SHOW CREATE TABLE idxINDEX")[0][1]
if "`synonym_kbrs` varchar(255)" not in stmt:
run_sql("ALTER TABLE idxINDEX ADD COLUMN synonym_kbrs varchar(255) NOT NULL default '' AFTER indexer")
# second step: fill tables
run_sql("UPDATE idxINDEX SET synonym_kbrs='INDEX-SYNONYM-TITLE,exact' WHERE name IN ('global','title')")
# third step: check invenio.conf
from invenio.config import CFG_BIBINDEX_SYNONYM_KBRS
from invenio.bibindex_engine_utils import get_index_id_from_index_name
if CFG_BIBINDEX_SYNONYM_KBRS:
for index in CFG_BIBINDEX_SYNONYM_KBRS:
index_id = get_index_id_from_index_name(index)
synonym = ",".join(CFG_BIBINDEX_SYNONYM_KBRS[index])
query = "UPDATE idxINDEX SET synonym_kbrs='%s' WHERE id=%s" % (synonym, index_id)
run_sql(query)
def initialise_term_collector(
index_name,
collector_type=TermCollector,
table_type=CFG_BIBINDEX_INDEX_TABLE_TYPE["Words"]):
"""Initialise any term collector for given index and WORD index table"""
index_id = get_index_id_from_index_name(index_name)
remove_stopwords = get_index_remove_stopwords(index_id)
remove_html_markup = get_index_remove_html_markup(index_id)
remove_latex_markup = get_index_remove_latex_markup(index_id)
stemming = get_index_stemming_language(index_id)
tokenizer = get_index_tokenizer(index_id)(stemming, remove_stopwords,
remove_html_markup,
remove_latex_markup)
args = [tokenizer, detect_tokenizer_type(tokenizer), table_type]
tagtype = collector_type == TermCollector and "marc" or "nonmarc"
args.append(get_index_tags(index_name, tagtype=tagtype))
args.append([1, get_max_recid()])
return collector_type(*args)
def initialise_term_collector(index_name,
collector_type=TermCollector,
table_type=CFG_BIBINDEX_INDEX_TABLE_TYPE["Words"]):
"""Initialise any term collector for given index and WORD index table"""
index_id = get_index_id_from_index_name(index_name)
remove_stopwords = get_index_remove_stopwords(index_id)
remove_html_markup = get_index_remove_html_markup(index_id)
remove_latex_markup = get_index_remove_latex_markup(index_id)
stemming = get_index_stemming_language(index_id)
tokenizer = get_index_tokenizer(index_id)(stemming,
remove_stopwords,
remove_html_markup,
remove_latex_markup)
args = [tokenizer,
detect_tokenizer_type(tokenizer),
table_type]
tagtype = collector_type == TermCollector and "marc" or "nonmarc"
args.append(get_index_tags(index_name, tagtype=tagtype))
args.append([1, get_max_recid()])
return collector_type(*args)
def do_upgrade():
#first step: change tables
stmt = run_sql('SHOW CREATE TABLE idxINDEX')[0][1]
if '`synonym_kbrs` varchar(255)' not in stmt:
run_sql(
"ALTER TABLE idxINDEX ADD COLUMN synonym_kbrs varchar(255) NOT NULL default '' AFTER indexer"
)
#second step: fill tables
run_sql(
"UPDATE idxINDEX SET synonym_kbrs='INDEX-SYNONYM-TITLE,exact' WHERE name IN ('global','title')"
)
#third step: check invenio.conf
from invenio.config import CFG_BIBINDEX_SYNONYM_KBRS
from invenio.bibindex_engine_utils import get_index_id_from_index_name
if CFG_BIBINDEX_SYNONYM_KBRS:
for index in CFG_BIBINDEX_SYNONYM_KBRS:
index_id = get_index_id_from_index_name(index)
synonym = ",".join(CFG_BIBINDEX_SYNONYM_KBRS[index])
query = "UPDATE idxINDEX SET synonym_kbrs='%s' WHERE id=%s" % (
synonym, index_id)
run_sql(query)
