1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
|
# This file is part of Senator Sperma
# Licensed under the GPL-3.0 License
# Copyright (C) 2026 Salmonik @ salmonik@tuta.io
# https://codeberg.org/salmonik/senator-sperma
import json
import logging
import re
from pathlib import Path
import aiofiles
from rapidfuzz import fuzz
logger = logging.getLogger(__name__)
class SearchUtils:
"""Utilities for searching and scoring audio files."""
@staticmethod
def normalize_name(p: Path, keep_ext=False):
"""Preprocess filenames: lower, replace separators, strip extensions optionally."""
name = str(p.name if keep_ext else p.stem)
return name.lower().replace('_', ' ').replace('-', ' ')
@staticmethod
def length_penalty(query, candidate):
"""Penalize tiny filenames when query is long; returns multiplier [0.0,1.0]."""
ql = len(query)
cl = len(candidate)
if ql == 0:
return 1.0
ratio = cl / ql
# Prefer candidate lengths near query length; gentle falloff otherwise
return math.exp(-0.5 * ((math.log(max(ratio, 1e-6)))**2))
@staticmethod
def combined_score(query, candidate):
"""Calculate a combined fuzzy matching score using multiple strategies.
Args:
query: The search query
candidate: The candidate string to score against
Returns:
float: Score between 0-100, where higher means better match
"""
if not query or not candidate:
return 0.0
query = query.lower().strip()
candidate = candidate.lower()
# Check for direct matches first
direct_match_score = SearchUtils._check_direct_matches(query, candidate)
if direct_match_score is not None:
return direct_match_score
# Handle short queries specially
short_query_score = SearchUtils._handle_short_queries(query, candidate)
if short_query_score is not None:
return short_query_score
# For longer queries, use complex scoring
return SearchUtils._calculate_complex_score(query, candidate)
@staticmethod
def _check_direct_matches(query: str, candidate: str) -> float:
"""Check for direct substring matches."""
if query in candidate:
base_score = 100
if re.search(rf'\b{re.escape(query)}\b', candidate):
base_score = 110
return min(100, base_score)
return 0
@staticmethod
def _handle_short_queries(query: str, candidate: str) -> float:
"""Handle scoring for short queries (2-3 chars)."""
if len(query) <= 3:
if re.search(rf'\b{re.escape(query)}', candidate):
return 90
if query in candidate:
return 80
return 0
@staticmethod
def _calculate_complex_score(query: str, candidate: str) -> float:
"""Calculate complex score for longer queries."""
query_words = {word for word in re.split(r'\W+', query) if len(word) > 1}
if not query_words:
return 0.0
candidate_words = set(re.split(r'\W+', candidate))
word_ratio = SearchUtils._calculate_word_ratio(query_words, candidate_words)
if word_ratio == 0:
return 0.0
# Calculate various similarity metrics
similarity_scores = SearchUtils._get_similarity_scores(query, candidate)
word_scores = SearchUtils._get_word_scores(query, candidate, query_words, candidate_words)
# Calculate final weighted score
final_score = SearchUtils._calculate_weighted_score(similarity_scores, word_scores, word_ratio)
# Apply penalties and boosts
final_score = SearchUtils._apply_adjustments(final_score, query, candidate, word_ratio)
return min(100, max(0, int(final_score)))
@staticmethod
def _calculate_word_ratio(query_words: set, candidate_words: set) -> float:
"""Calculate the ratio of matching words."""
matching_words = sum(1 for word in query_words if word in candidate_words)
return matching_words / len(query_words)
@staticmethod
def _get_similarity_scores(query: str, candidate: str) -> dict:
"""Get various fuzzy similarity scores."""
return {
'ratio': fuzz.ratio(query, candidate),
'partial': fuzz.partial_ratio(query, candidate),
'token_set': fuzz.token_set_ratio(query, candidate)
}
@staticmethod
def _get_word_scores(query: str, candidate: str, query_words: set, candidate_words: set) -> dict:
"""Calculate word-based scores."""
query_tokens = set(query.split())
candidate_tokens = set(candidate.split())
word_intersection = len(query_tokens & candidate_tokens)
word_union = len(query_tokens | candidate_tokens)
jaccard_sim = word_intersection / word_union if word_union > 0 else 0
prefix_bonus = SearchUtils._calculate_prefix_bonus(query_words, candidate_words)
return {
'jaccard': jaccard_sim,
'prefix': min(prefix_bonus, 100)
}
@staticmethod
def _calculate_prefix_bonus(query_words: set, candidate_words: set) -> float:
"""Calculate prefix matching bonus."""
prefix_bonus = 0
for q_word in query_words:
for c_word in candidate_words:
if c_word.startswith(q_word):
prefix_bonus += (len(q_word) / max(len(c_word), 1)) * 20
return prefix_bonus
@staticmethod
def _calculate_weighted_score(similarity_scores: dict, word_scores: dict, word_ratio: float) -> float:
"""Calculate the final weighted score."""
return (
0.40 * similarity_scores['token_set'] +
0.25 * similarity_scores['ratio'] +
0.15 * similarity_scores['partial'] +
0.15 * (word_scores['jaccard'] * 100) +
0.05 * word_scores['prefix']
)
@staticmethod
def _apply_adjustments(score: float, query: str, candidate: str, word_ratio: float) -> float:
"""Apply length penalty and word ratio boost."""
import math
# Apply length penalty
score *= min(1.0, 0.9 + (0.1 * SearchUtils.length_penalty(query, candidate)))
# Apply word ratio boost
score = score * (0.7 + (0.3 * word_ratio))
return score
class AudioSearcher:
"""Handles searching for audio files in configured paths."""
@staticmethod
def _is_valid_audio_file(file_path: Path) -> bool:
"""Check if file is a valid audio file."""
valid_extensions = {'.mp4', '.mp3', '.wav', '.opus', '.m4a'}
return file_path.is_file() and file_path.suffix.lower() in valid_extensions
@staticmethod
def _calculate_match_score(query: str, file_name: str, autocomplete: bool) -> tuple[int, str]:
"""Calculate match score for a file."""
if not query and autocomplete:
return 0, file_name
if query and len(query) >= 2:
if query in file_name:
logger.debug(f"Found direct match: {file_name} (contains '{query}')")
score = SearchUtils.combined_score(query, file_name)
logger.debug(f"File: {file_name} - Score: {score}")
if score >= 30:
return score, file_name
return -1, file_name
@staticmethod
async def _search_single_path(path: Path, query: str, max_results: int, autocomplete: bool) -> list[tuple[int, str]]:
"""Search for audio files in a single path."""
matches = []
logger.debug(f"Searching in directory: {path}")
for file_path in path.rglob('*'):
if not AudioSearcher._is_valid_audio_file(file_path):
continue
file_name = file_path.name.lower()
score, _ = AudioSearcher._calculate_match_score(query, file_name, autocomplete)
if score >= 0:
matches.append((score, str(file_path.absolute())))
if len(matches) >= max_results * 5:
break
return matches
@classmethod
async def search_local(cls, query: str = None, max_results: int = 25, autocomplete: bool = False):
"""Search for audio files with fuzzy matching and optional autocomplete support.
Args:
query: Search query string (can be None for autocomplete)
max_results: Maximum number of results to return
autocomplete: If True, returns results even with empty query
Returns:
List of matching file paths with scores
"""
if not query and not autocomplete:
return []
query = (query or '').lower().strip()
logger.info(f"Starting local search for query: '{query}'")
try:
async with aiofiles.open('settings.json') as f:
data = json.load(f)
search_paths = data.get('audio_paths', [])
logger.info(f"Searching in paths: {search_paths}")
matches = []
file_count = 0
for path_str in search_paths:
path = Path(path_str)
if not path.exists():
logger.warning(f"Path does not exist: {path}")
continue
path_matches = await cls._search_single_path(path, query, max_results, autocomplete)
matches.extend(path_matches)
file_count += len(path_matches)
if len(matches) >= max_results * 5:
break
logger.info(f"Scanned {file_count} files total")
matches.sort(reverse=True, key=lambda x: x[0])
result = [path for score, path in matches[:max_results]]
logger.info(f"Search complete. Found {len(result)} matches out of {file_count} files.")
if not result and query:
logger.warning(f"No matches found for query: '{query}'. Try a different search term.")
return result
except Exception as e:
logger.error(f"Error during local search: {str(e)}", exc_info=True)
return []
@classmethod
async def search_local_autocomplete(cls, query: str = None, max_results: int = 10):
"""Wrapper for autocomplete searches that returns more results and allows empty query."""
# For autocomplete, we want to show more options and allow empty queries
if not query:
return []
return await cls.search_local(query, max_results=max_results, autocomplete=True)
|
