mirror of
https://github.com/paperless-ngx/paperless-ngx.git
synced 2026-11-21 21:12:40 +00:00
Enhancement: add highlighting to title + content searches (#12593)
This commit is contained in:
1 parent
814fdf5892
commit
f784a74eba
3 files changed
+52
-7
No files matched your search
@@ -21,6 +21,7 @@ from guardian.shortcuts import get_users_with_perms
|
||||
|
||||
from documents.search._normalize import ascii_fold
|
||||
from documents.search._query import build_permission_filter
|
||||
from documents.search._query import parse_simple_text_highlight_query
|
||||
from documents.search._query import parse_simple_text_query
|
||||
from documents.search._query import parse_simple_title_query
|
||||
from documents.search._query import parse_user_query
|
||||
@@ -549,6 +550,9 @@ class TantivyBackend:
|
||||
|
||||
self._ensure_open()
|
||||
user_query = self._parse_query(query, search_mode)
|
||||
highlight_query = user_query
|
||||
if search_mode is SearchMode.TEXT:
|
||||
highlight_query = parse_simple_text_highlight_query(self._index, query)
|
||||
|
||||
# For notes_text snippet generation, we need a query that targets the
|
||||
# notes_text field directly. user_query may contain JSON-field terms
|
||||
@@ -601,7 +605,7 @@ class TantivyBackend:
|
||||
if snippet_generator is None:
|
||||
snippet_generator = tantivy.SnippetGenerator.create(
|
||||
searcher,
|
||||
user_query,
|
||||
highlight_query,
|
||||
self._schema,
|
||||
"content",
|
||||
)
|
||||
@@ -610,7 +614,7 @@ class TantivyBackend:
|
||||
if content_html:
|
||||
highlights["content"] = content_html
|
||||
|
||||
if "notes_text" in doc_dict:
|
||||
if search_mode is SearchMode.QUERY and "notes_text" in doc_dict:
|
||||
# Use notes_text (plain text) for snippet generation — tantivy's
|
||||
# SnippetGenerator does not support JSON fields.
|
||||
if notes_snippet_generator is None:
|
||||
|
||||
@@ -490,6 +490,14 @@ _FIELD_BOOSTS = {"title": 2.0}
|
||||
_SIMPLE_FIELD_BOOSTS = {"simple_title": 2.0}
|
||||
|
||||
|
||||
def _simple_query_tokens(raw_query: str) -> list[str]:
|
||||
tokens = [
|
||||
ascii_fold(token.lower())
|
||||
for token in _SIMPLE_QUERY_TOKEN_RE.findall(raw_query, timeout=_REGEX_TIMEOUT)
|
||||
]
|
||||
return [token for token in tokens if token]
|
||||
|
||||
|
||||
def _build_simple_field_query(
|
||||
index: tantivy.Index,
|
||||
field: str,
|
||||
@@ -585,11 +593,7 @@ def parse_simple_query(
|
||||
|
||||
Query string is escaped and normalized to be treated as "simple" text query.
|
||||
"""
|
||||
tokens = [
|
||||
ascii_fold(token.lower())
|
||||
for token in _SIMPLE_QUERY_TOKEN_RE.findall(raw_query, timeout=_REGEX_TIMEOUT)
|
||||
]
|
||||
tokens = [token for token in tokens if token]
|
||||
tokens = _simple_query_tokens(raw_query)
|
||||
if not tokens:
|
||||
return tantivy.Query.empty_query()
|
||||
|
||||
@@ -602,6 +606,23 @@ def parse_simple_query(
|
||||
return tantivy.Query.boolean_query(field_queries)
|
||||
|
||||
|
||||
def parse_simple_text_highlight_query(
|
||||
index: tantivy.Index,
|
||||
raw_query: str,
|
||||
) -> tantivy.Query:
|
||||
"""Build a snippet-friendly query for simple text searches.
|
||||
|
||||
Simple search matching uses regex queries but for compatibility with Tantivy
|
||||
SnippetGenerator we build a plain term query over the content field instead.
|
||||
"""
|
||||
|
||||
tokens = _simple_query_tokens(raw_query)
|
||||
if not tokens:
|
||||
return tantivy.Query.empty_query()
|
||||
|
||||
return index.parse_query(" ".join(tokens), ["content"])
|
||||
|
||||
|
||||
def parse_simple_text_query(
|
||||
index: tantivy.Index,
|
||||
raw_query: str,
|
||||
|
||||
Reference in new issue
Block a user