diff options
| author | Terry Truong <terry06890@gmail.com> | 2022-06-15 20:48:31 +1000 |
|---|---|---|
| committer | Terry Truong <terry06890@gmail.com> | 2022-06-15 20:48:31 +1000 |
| commit | eabec97d80e5f43e4710dd4c8327d0bc8519ab8c (patch) | |
| tree | 0fd6d93f306b6dc7db2b9233ce63b65ebdc994bc /backend/data/genDbpData.py | |
| parent | 3402f3d6e906afb500b95448d7d0b136b6b5ee86 (diff) | |
Separate node-wiki_id association from 'descs' table
Diffstat (limited to 'backend/data/genDbpData.py')
| -rwxr-xr-x | backend/data/genDbpData.py | 9 |
1 files changed, 5 insertions, 4 deletions
diff --git a/backend/data/genDbpData.py b/backend/data/genDbpData.py index 3755145..e921b6c 100755 --- a/backend/data/genDbpData.py +++ b/backend/data/genDbpData.py @@ -221,8 +221,8 @@ for (name, iri) in nodeToIri.items(): redirectingIriSet.add(name) # Find descriptions, and add to db print("Adding node description data") -dbCur.execute("CREATE TABLE descs (name TEXT PRIMARY KEY, desc TEXT, redirected INT, wiki_id INT, from_dbp INT)") -dbCur.execute("CREATE INDEX descs_id_idx ON descs(wiki_id)") # wiki_id intentionally left non-unique +dbCur.execute("CREATE TABLE wiki_ids (name TEXT PRIMARY KEY, id INT, redirected INT)") +dbCur.execute("CREATE TABLE descs (wiki_id INT PRIMARY KEY, desc TEXT, from_dbp INT)") iterNum = 0 for (name, iri) in nodeToIri.items(): iterNum += 1 @@ -232,8 +232,9 @@ for (name, iri) in nodeToIri.items(): query = "SELECT abstract, id FROM abstracts INNER JOIN ids ON abstracts.iri = ids.iri WHERE ids.iri = ?" row = dbpCur.execute(query, (iri,)).fetchone() if row != None: - dbCur.execute("INSERT INTO descs VALUES (?, ?, ?, ?, ?)", - (name, row[0], 1 if name in redirectingIriSet else 0, row[1], 1)) + (desc, wikiId) = row + dbCur.execute("INSERT INTO wiki_ids VALUES (?, ?, ?)", (name, wikiId, 1 if name in redirectingIriSet else 0)) + dbCur.execute("INSERT OR IGNORE INTO descs VALUES (?, ?, ?)", (wikiId, desc, 1)) # Close dbs dbCon.commit() dbCon.close() |