Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-07 10:14:46 -04:00
Code Issues Packages Projects Releases Wiki Activity

Allow FTS searches to not return text

Browse Source 
Allows efficiently querying for which books match without th eoverhead
of converting the book text to a python string
This commit is contained in:
Kovid Goyal 2022-05-05 12:39:18 +05:30
parent 0a6dc87fcd
commit cac2a91df9
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 17 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
src/calibre/db/fts/connect.py
View File

@ -137,31 +137,36 @@ class FTS:
def search(self,
fts_engine_query, use_stemming, highlight_start, highlight_end, snippet_size, restrict_to_book_ids,
return_text=True,
):
fts_engine_query = unicode_normalize(fts_engine_query)
fts_table = 'books_fts' + ('_stemmed' if use_stemming else '')
text = 'books_text.searchable_text'
if highlight_start is not None and highlight_end is not None:
if snippet_size is not None:
text = f'snippet("{fts_table}", 0, "{highlight_start}", "{highlight_end}", "…", {max(1, min(snippet_size, 64))})'
else:
text = f'highlight("{fts_table}", 0, "{highlight_start}", "{highlight_end}")'
query = 'SELECT {0}.id, {0}.book, {0}.format, {1} FROM {0} '
query = query.format('books_text', text)
if return_text:
text = 'books_text.searchable_text'
if highlight_start is not None and highlight_end is not None:
if snippet_size is not None:
text = f'snippet("{fts_table}", 0, "{highlight_start}", "{highlight_end}", "…", {max(1, min(snippet_size, 64))})'
else:
text = f'highlight("{fts_table}", 0, "{highlight_start}", "{highlight_end}")'
text = ', ' + text
else:
text = ''
query = 'SELECT {0}.id, {0}.book, {0}.format {1} FROM {0} '.format('books_text', text)
query += f' JOIN {fts_table} ON fts_db.books_text.id = {fts_table}.rowid'
query += f' WHERE "{fts_table}" MATCH ?'
data = [fts_engine_query]
query += f' ORDER BY {fts_table}.rank '
conn = self.get_connection()
try:
for (rowid, book_id, fmt, text) in conn.execute(query, tuple(data)):
for record in conn.execute(query, tuple(data)):
book_id = record[1]
if restrict_to_book_ids is not None and book_id not in restrict_to_book_ids:
continue
yield {
'id': rowid,
'id': record[0],
'book_id': book_id,
'format': fmt,
'text': text,
'format': record[2],
'text': record[3] if return_text else '',
}
except apsw.SQLError as e:
raise FTSQueryError(fts_engine_query, query, e) from e