Spaces:
Paused
Paused
Update app/app.py
Browse files- app/app.py +1498 -420
app/app.py
CHANGED
@@ -1,516 +1,1594 @@
|
|
1 |
-
|
2 |
-
|
|
|
|
|
|
|
3 |
import os
|
4 |
-
import
|
5 |
-
import
|
6 |
import time
|
7 |
-
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
import re
|
12 |
-
from bs4 import BeautifulSoup
|
13 |
import requests
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
|
|
|
|
|
|
|
|
|
|
20 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
logger = logging.getLogger(__name__)
|
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 |
-
def
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
|
69 |
-
|
70 |
-
self.last_request_time = time.time()
|
71 |
|
72 |
-
def
|
73 |
-
"""
|
74 |
try:
|
75 |
-
self.
|
76 |
-
|
77 |
-
|
78 |
-
|
79 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
80 |
except Exception as e:
|
81 |
-
logger.
|
82 |
-
return None
|
83 |
|
84 |
-
def
|
85 |
-
"""
|
86 |
-
|
|
|
|
|
|
|
|
|
|
|
87 |
|
88 |
-
|
89 |
-
|
90 |
-
|
91 |
-
|
92 |
-
|
93 |
-
|
94 |
-
|
95 |
-
|
96 |
-
|
97 |
-
|
98 |
-
|
99 |
-
|
100 |
-
|
101 |
-
content = content[:500] # Limit content length
|
102 |
-
|
103 |
-
# Determine document type based on URL
|
104 |
-
doc_type = "web_page"
|
105 |
-
if 'majlis' in url:
|
106 |
-
doc_type = "law"
|
107 |
-
elif 'news' in url:
|
108 |
-
doc_type = "news"
|
109 |
-
|
110 |
-
# Create document
|
111 |
-
doc = SimpleLegalDocument(
|
112 |
-
title=title_text,
|
113 |
-
content=content,
|
114 |
-
source_url=url,
|
115 |
-
document_type=doc_type,
|
116 |
-
importance_score=0.7 if doc_type == "law" else 0.5
|
117 |
-
)
|
118 |
-
|
119 |
-
documents.append(doc)
|
120 |
-
logger.info(f"Scraped: {title_text[:50]}...")
|
121 |
|
122 |
-
|
123 |
-
|
|
|
|
|
124 |
|
125 |
-
|
126 |
-
|
127 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
128 |
|
129 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
130 |
|
131 |
-
|
|
|
|
|
|
|
132 |
def __init__(self):
|
133 |
-
self.
|
134 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
135 |
|
136 |
-
def
|
137 |
-
|
138 |
-
|
139 |
-
|
140 |
|
141 |
try:
|
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 |
except Exception as e:
|
185 |
-
|
186 |
-
|
187 |
-
return error_msg, "", ""
|
188 |
-
|
189 |
-
finally:
|
190 |
-
self.is_scraping = False
|
191 |
|
192 |
-
def
|
193 |
-
"""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
194 |
try:
|
195 |
-
|
196 |
-
|
197 |
-
f"- کل اسناد: 15",
|
198 |
-
"",
|
199 |
-
"📈 **بر اساس نوع:**",
|
200 |
-
"- قوانین: 6",
|
201 |
-
"- اخبار: 5",
|
202 |
-
"- صفحات وب: 4",
|
203 |
-
"",
|
204 |
-
"⭐ **توزيع اهمیت:**",
|
205 |
-
"- بالا (>0.7): 8",
|
206 |
-
"- متوسط (0.4-0.7): 5",
|
207 |
-
"- پایین (<0.4): 2",
|
208 |
-
"",
|
209 |
-
"🔑 **کلیدواژههای برتر:**",
|
210 |
-
"- قانون: 12",
|
211 |
-
"- حقوق: 9",
|
212 |
-
"- ایران: 7",
|
213 |
-
"- مجلس: 6",
|
214 |
-
"- اسلامی: 5"
|
215 |
-
]
|
216 |
|
217 |
-
|
218 |
-
|
219 |
-
|
220 |
-
|
221 |
-
|
222 |
-
|
223 |
-
|
224 |
-
|
225 |
-
|
226 |
-
|
227 |
-
|
228 |
-
|
229 |
-
|
230 |
-
<p style='margin-top: 10px;'>توزیع بر اساس نوع</p>
|
231 |
-
</div>
|
232 |
-
<div style='text-align: center;'>
|
233 |
-
<div style='width: 120px; height: 120px; background: linear-gradient(135deg, #e74c3c, #c0392b);
|
234 |
-
border-radius: 50%; display: flex; align-items: center; justify-content: center;
|
235 |
-
color: white; font-weight: bold; font-size: 14px; margin: 0 auto;'>
|
236 |
-
سطح اهمیت
|
237 |
-
</div>
|
238 |
-
<p style='margin-top: 10px;'>امتیازبندی اسناد</p>
|
239 |
-
</div>
|
240 |
-
</div>
|
241 |
-
</div>
|
242 |
-
"""
|
243 |
|
244 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
245 |
|
246 |
except Exception as e:
|
247 |
-
|
248 |
-
|
|
|
|
|
249 |
|
250 |
-
def
|
251 |
-
"""
|
252 |
-
|
253 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
254 |
|
255 |
try:
|
256 |
-
#
|
257 |
-
|
258 |
-
|
259 |
-
|
260 |
-
|
261 |
-
"similarity": 0.92,
|
262 |
-
"source": "https://rc.majlis.ir/fa/law/show/1",
|
263 |
-
"summary": "قانون اساسی جمهوری اسلامی ایران مصوب 1358 با اصلاحات بعدی"
|
264 |
-
},
|
265 |
-
{
|
266 |
-
"title": "آییننامه داخلی مجلس شورای اسلامی",
|
267 |
-
"document_type": "regulation",
|
268 |
-
"similarity": 0.85,
|
269 |
-
"source": "https://rc.majlis.ir/fa/regulation/show/1",
|
270 |
-
"summary": "مقررات مربوط به نحوه تشکیل و اداره جلسات مجلس شورای اسلامی"
|
271 |
-
}
|
272 |
]
|
273 |
|
274 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
275 |
|
276 |
-
|
277 |
-
|
278 |
-
|
279 |
-
|
280 |
-
f" امتیاز شباهت: {result['similarity']:.2f}",
|
281 |
-
f" منبع: {result['source']}",
|
282 |
-
f" خلاصه: {result['summary']}",
|
283 |
-
"---"
|
284 |
-
])
|
285 |
|
286 |
-
return
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
287 |
|
288 |
except Exception as e:
|
289 |
-
|
290 |
-
return error_msg
|
291 |
|
292 |
-
def
|
293 |
-
"""
|
294 |
try:
|
295 |
-
|
|
|
296 |
|
297 |
-
|
298 |
-
|
299 |
|
300 |
-
|
|
|
301 |
|
302 |
-
|
303 |
-
|
304 |
-
['title', 'type', 'score', 'source', 'keywords'],
|
305 |
-
['قانون اساسی جمهوری اسلامی ایران', 'law', '0.95', 'https://rc.majlis.ir/fa/law/show/1', 'قانون,اساسی,جمهوری'],
|
306 |
-
['آییننامه داخلی مجلس', 'regulation', '0.85', 'https://rc.majlis.ir/fa/regulation/show/1', 'آییننامه,مجلس,داخلی'],
|
307 |
-
['اخبار حقوقی روز', 'news', '0.65', 'https://iribnews.ir/news/456', 'اخبار,حقوقی,روز'],
|
308 |
-
['تحلیل قراردادهای بینالمللی', 'analysis', '0.75', 'https://dolat.ir/analysis/789', 'تحلیل,قرارداد,بینالمللی']
|
309 |
-
]
|
310 |
|
311 |
-
|
312 |
-
|
313 |
-
|
314 |
|
315 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
316 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
317 |
except Exception as e:
|
318 |
-
|
319 |
-
return
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
320 |
|
321 |
-
|
322 |
-
|
323 |
-
|
324 |
-
|
325 |
-
|
326 |
-
|
327 |
-
|
328 |
-
|
329 |
-
|
330 |
-
|
331 |
-
|
332 |
-
|
333 |
-
|
334 |
-
|
335 |
-
|
336 |
-
|
337 |
-
|
338 |
-
|
339 |
-
|
340 |
-
|
341 |
-
|
342 |
-
|
343 |
-
|
344 |
-
|
345 |
-
|
346 |
-
|
347 |
-
|
348 |
-
|
349 |
-
|
350 |
-
|
351 |
-
|
352 |
-
|
353 |
-
|
354 |
-
|
355 |
-
|
356 |
-
|
357 |
-
|
358 |
-
|
359 |
-
|
360 |
-
|
361 |
-
|
362 |
-
|
363 |
-
|
364 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
365 |
)
|
366 |
|
367 |
-
|
368 |
-
|
369 |
-
|
370 |
-
|
371 |
-
|
372 |
-
step=1
|
373 |
)
|
374 |
|
375 |
-
|
376 |
-
|
377 |
-
|
378 |
-
|
379 |
-
|
380 |
-
|
381 |
-
|
|
|
|
|
|
|
|
|
|
|
382 |
)
|
383 |
-
|
384 |
-
|
385 |
-
|
386 |
-
|
387 |
-
|
388 |
-
|
389 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
390 |
|
391 |
-
|
392 |
-
|
393 |
-
|
394 |
-
|
395 |
-
)
|
396 |
|
397 |
-
|
398 |
-
|
399 |
-
|
400 |
-
|
401 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
402 |
|
403 |
-
|
404 |
-
|
|
|
|
|
|
|
|
|
|
|
405 |
|
406 |
-
|
407 |
-
label="🔍 کلیدواژه جستجو",
|
408 |
-
placeholder="مثال: قانون اساسی, آییننامه, حقوق"
|
409 |
-
)
|
410 |
|
411 |
-
|
|
|
412 |
|
413 |
-
|
414 |
-
label="📋 نتایج جستجو",
|
415 |
-
interactive=False,
|
416 |
-
lines=15
|
417 |
-
)
|
418 |
|
419 |
-
|
420 |
-
|
421 |
-
|
422 |
-
|
423 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
424 |
|
425 |
-
|
426 |
-
|
427 |
|
428 |
-
|
|
|
429 |
|
430 |
-
|
431 |
-
|
432 |
-
|
433 |
-
|
434 |
-
|
435 |
-
|
436 |
-
|
437 |
-
|
438 |
-
|
439 |
-
|
|
|
|
|
|
|
|
|
440 |
|
441 |
-
|
442 |
-
|
443 |
-
|
444 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
445 |
|
446 |
-
with gr.
|
447 |
-
gr.
|
|
|
|
|
|
|
448 |
|
449 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
450 |
|
451 |
-
|
452 |
-
|
453 |
-
|
454 |
-
|
455 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
456 |
|
457 |
-
|
458 |
-
|
459 |
-
|
|
|
|
|
|
|
|
|
460 |
)
|
461 |
|
462 |
-
|
463 |
-
|
464 |
-
|
|
|
|
|
|
|
465 |
)
|
466 |
-
|
467 |
-
with gr.Tab("📚 راهنما"):
|
468 |
-
gr.Markdown("""
|
469 |
-
# 🕷️ راهنمای اسکراپر اسناد حقوقی
|
470 |
|
471 |
-
|
|
|
|
|
|
|
|
|
472 |
|
473 |
-
|
474 |
-
|
475 |
-
|
476 |
-
|
477 |
-
|
|
|
478 |
|
479 |
-
|
|
|
|
|
|
|
|
|
|
|
480 |
|
481 |
-
|
482 |
-
|
483 |
-
|
484 |
-
|
485 |
-
|
486 |
|
487 |
-
|
|
|
|
|
|
|
488 |
|
489 |
-
|
490 |
-
|
491 |
-
|
|
|
|
|
492 |
|
493 |
-
|
494 |
-
|
495 |
-
|
496 |
-
|
|
|
|
|
|
|
497 |
|
|
|
498 |
def main():
|
499 |
-
"""
|
500 |
-
|
501 |
-
|
502 |
-
|
503 |
-
|
504 |
-
|
505 |
-
|
506 |
-
|
507 |
-
|
508 |
-
|
509 |
-
|
510 |
-
|
511 |
-
|
512 |
-
|
513 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
514 |
|
515 |
if __name__ == "__main__":
|
516 |
main()
|
|
|
1 |
+
"""
|
2 |
+
پلتفرم پیشرفته هوشمند اسناد حقوقی ایران - نسخه ارتقاء یافته
|
3 |
+
مجهز به مدلهای SOTA فارسی، سیستم کش هوشمند و امتیازدهی پیشرفته
|
4 |
+
"""
|
5 |
+
|
6 |
import os
|
7 |
+
import gc
|
8 |
+
import sys
|
9 |
import time
|
10 |
+
import json
|
11 |
+
import logging
|
12 |
+
import hashlib
|
13 |
+
import resource
|
|
|
|
|
14 |
import requests
|
15 |
+
import threading
|
16 |
+
import re
|
17 |
+
import random
|
18 |
+
import sqlite3
|
19 |
+
import pickle
|
20 |
+
import tempfile
|
21 |
+
from pathlib import Path
|
22 |
+
from datetime import datetime, timedelta
|
23 |
+
from typing import List, Dict, Any, Optional, Tuple, Union
|
24 |
+
from dataclasses import dataclass, field
|
25 |
+
from concurrent.futures import ThreadPoolExecutor, as_completed, ProcessPoolExecutor
|
26 |
+
from urllib.parse import urljoin, urlparse
|
27 |
+
import warnings
|
28 |
+
import asyncio
|
29 |
+
from functools import lru_cache
|
30 |
+
import numpy as np
|
31 |
|
32 |
+
import gradio as gr
|
33 |
+
import pandas as pd
|
34 |
+
import torch
|
35 |
+
from bs4 import BeautifulSoup
|
36 |
+
from transformers import (
|
37 |
+
AutoTokenizer,
|
38 |
+
AutoModelForSequenceClassification,
|
39 |
+
pipeline,
|
40 |
+
logging as transformers_logging,
|
41 |
+
AutoModel
|
42 |
)
|
43 |
+
from sentence_transformers import SentenceTransformer, util
|
44 |
+
from hazm import Normalizer, word_tokenize, Lemmatizer
|
45 |
+
import faiss
|
46 |
+
|
47 |
+
# === تنظیمات اولیه پیشرفته ===
|
48 |
+
warnings.filterwarnings('ignore')
|
49 |
+
transformers_logging.set_verbosity_error()
|
50 |
+
|
51 |
+
try:
|
52 |
+
resource.setrlimit(resource.RLIMIT_AS, (4*1024*1024*1024, 4*1024*1024*1024))
|
53 |
+
except:
|
54 |
+
pass
|
55 |
+
|
56 |
+
# ایجاد دایرکتوریهای مورد نیاز
|
57 |
+
WORK_DIR = Path("/tmp/legal_scraper_data")
|
58 |
+
WORK_DIR.mkdir(exist_ok=True, parents=True)
|
59 |
+
|
60 |
+
os.environ.update({
|
61 |
+
'TRANSFORMERS_CACHE': str(WORK_DIR / 'hf_cache'),
|
62 |
+
'HF_HOME': str(WORK_DIR / 'hf_cache'),
|
63 |
+
'TORCH_HOME': str(WORK_DIR / 'torch_cache'),
|
64 |
+
'TOKENIZERS_PARALLELISM': 'false',
|
65 |
+
'CUDA_VISIBLE_DEVICES': '0' if torch.cuda.is_available() else ''
|
66 |
+
})
|
67 |
+
|
68 |
+
# ایجاد دایرکتوریهای کش
|
69 |
+
for cache_dir in [os.environ['TRANSFORMERS_CACHE'], os.environ['HF_HOME'], os.environ['TORCH_HOME']]:
|
70 |
+
os.makedirs(cache_dir, exist_ok=True)
|
71 |
+
|
72 |
+
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
|
73 |
logger = logging.getLogger(__name__)
|
74 |
|
75 |
+
# === ثابتهای سیستم با مسیرهای اصلاح شده ===
|
76 |
+
DB_PATH = str(WORK_DIR / "iranian_legal_archive_advanced.sqlite")
|
77 |
+
CACHE_DB_PATH = str(WORK_DIR / "cache_system.sqlite")
|
78 |
+
EMBEDDINGS_CACHE_PATH = str(WORK_DIR / "embeddings_cache.pkl")
|
79 |
+
VECTOR_INDEX_PATH = str(WORK_DIR / "faiss_index.bin")
|
80 |
+
|
81 |
+
# آیکونهای بهبود یافته
|
82 |
+
SVG_ICONS = {
|
83 |
+
'search': '🔍', 'document': '📄', 'analyze': '🤖', 'export': '📊',
|
84 |
+
'settings': '⚙️', 'link': '🔗', 'success': '✅', 'error': '❌',
|
85 |
+
'warning': '⚠️', 'database': '🗄️', 'crawler': '🔄', 'brain': '🧠',
|
86 |
+
'cache': '⚡', 'score': '📈', 'classify': '🏷️', 'similarity': '🎯',
|
87 |
+
'legal': '⚖️', 'home': '🏠', 'stats': '📈', 'process': '🔧'
|
88 |
+
}
|
89 |
|
90 |
+
# منابع حقوقی با پیکربندی پیشرفته
|
91 |
+
LEGAL_SOURCES = {
|
92 |
+
"مجلس شورای اسلامی": {
|
93 |
+
"base_url": "https://rc.majlis.ir",
|
94 |
+
"patterns": ["/fa/law/show/", "/fa/report/show/"],
|
95 |
+
"selectors": [".main-content", ".article-body", "article"],
|
96 |
+
"delay_range": (2, 5),
|
97 |
+
"priority": 1,
|
98 |
+
"reliability_score": 0.95
|
99 |
+
},
|
100 |
+
"پورتال ملی قوانین": {
|
101 |
+
"base_url": "https://www.dotic.ir",
|
102 |
+
"patterns": ["/portal/law/", "/regulation/"],
|
103 |
+
"selectors": [".content-area", ".law-content"],
|
104 |
+
"delay_range": (1, 4),
|
105 |
+
"priority": 1,
|
106 |
+
"reliability_score": 0.90
|
107 |
+
},
|
108 |
+
"قوه قضاییه": {
|
109 |
+
"base_url": "https://www.judiciary.ir",
|
110 |
+
"patterns": ["/fa/news/", "/fa/verdict/"],
|
111 |
+
"selectors": [".news-content", ".main-content"],
|
112 |
+
"delay_range": (3, 6),
|
113 |
+
"priority": 2,
|
114 |
+
"reliability_score": 0.85
|
115 |
+
},
|
116 |
+
"وزارت دادگستری": {
|
117 |
+
"base_url": "https://www.moj.ir",
|
118 |
+
"patterns": ["/fa/news/", "/fa/regulation/"],
|
119 |
+
"selectors": [".content-area", ".news-content"],
|
120 |
+
"delay_range": (2, 4),
|
121 |
+
"priority": 2,
|
122 |
+
"reliability_score": 0.80
|
123 |
+
}
|
124 |
+
}
|
125 |
+
|
126 |
+
# واژگان تخصصی حقوقی گسترده
|
127 |
+
PERSIAN_LEGAL_TERMS = {
|
128 |
+
"قوانین_اساسی": ["قانون اساسی", "مجلس شورای اسلامی", "شورای نگهبان", "ولایت فقیه", "اصول قانون اساسی"],
|
129 |
+
"قوانین_عادی": ["ماده", "تبصره", "فصل", "باب", "قانون مدنی", "قانون جزا", "قانون تجارت", "قانون کار"],
|
130 |
+
"اصطلاحات_حقوقی": ["شخص حقیقی", "شخص حقوقی", "دعوا", "خواهان", "خوانده", "مجازات", "قرارداد", "تعهد"],
|
131 |
+
"نهادهای_قضایی": ["دادگاه", "قاضی", "وکیل", "مدعیالعموم", "رای", "حکم", "دادنامه", "قرار"],
|
132 |
+
"اصطلاحات_مالی": ["مالیات", "عوارض", "جریمه", "خسارت", "تأمین", "ضمانت", "وثیقه", "دیه"],
|
133 |
+
"جرائم": ["جنایت", "جنحه", "تخلف", "قصاص", "دیه", "تعزیر", "حدود", "قذف"]
|
134 |
+
}
|
135 |
+
|
136 |
+
# مدلهای SOTA پیشنهادی
|
137 |
+
AVAILABLE_MODELS = {
|
138 |
+
"classification": {
|
139 |
+
"fabert": "sbunlp/fabert",
|
140 |
+
"parsbert": "HooshvareLab/bert-base-parsbert-uncased",
|
141 |
+
"legal_roberta": "lexlms/legal-roberta-base"
|
142 |
+
},
|
143 |
+
"embedding": {
|
144 |
+
"maux_gte": "xmanii/maux-gte-persian",
|
145 |
+
"sentence_transformer": "sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2"
|
146 |
+
},
|
147 |
+
"ner": {
|
148 |
+
"parsbert_ner": "HooshvareLab/bert-fa-base-uncased-ner"
|
149 |
+
}
|
150 |
+
}
|
151 |
|
152 |
+
@dataclass
|
153 |
+
class ProcessingResult:
|
154 |
+
"""نتیجه پردازش سند"""
|
155 |
+
url: str
|
156 |
+
title: str
|
157 |
+
content: str
|
158 |
+
source: str
|
159 |
+
quality_score: float
|
160 |
+
classification: Dict[str, float]
|
161 |
+
sentiment_score: float
|
162 |
+
legal_entities: List[str]
|
163 |
+
embeddings: Optional[np.ndarray]
|
164 |
+
processing_time: float
|
165 |
+
cache_hit: bool = False
|
166 |
|
167 |
+
@dataclass
|
168 |
+
class SystemMetrics:
|
169 |
+
"""متریکهای عملکرد سیستم"""
|
170 |
+
total_processed: int = 0
|
171 |
+
cache_hits: int = 0
|
172 |
+
classification_accuracy: float = 0.0
|
173 |
+
avg_processing_time: float = 0.0
|
174 |
+
memory_usage: float = 0.0
|
175 |
+
active_crawlers: int = 0
|
176 |
+
|
177 |
+
# === سیستم کش هوشمند ===
|
178 |
+
class IntelligentCacheSystem:
|
179 |
+
"""سیستم کش هوشمند با TTL و اولویتبندی"""
|
180 |
|
181 |
+
def __init__(self, cache_db_path: str = CACHE_DB_PATH):
|
182 |
+
self.cache_db_path = cache_db_path
|
183 |
+
self.memory_cache = {}
|
184 |
+
self.access_count = {}
|
185 |
+
self.max_memory_items = 1000
|
186 |
+
self._init_database()
|
|
|
187 |
|
188 |
+
def _init_database(self):
|
189 |
+
"""ایجاد پایگاه داده کش"""
|
190 |
try:
|
191 |
+
with sqlite3.connect(self.cache_db_path) as conn:
|
192 |
+
conn.execute('''
|
193 |
+
CREATE TABLE IF NOT EXISTS cache_entries (
|
194 |
+
key TEXT PRIMARY KEY,
|
195 |
+
value BLOB,
|
196 |
+
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
197 |
+
access_count INTEGER DEFAULT 1,
|
198 |
+
ttl_seconds INTEGER DEFAULT 3600,
|
199 |
+
priority INTEGER DEFAULT 1
|
200 |
+
)
|
201 |
+
''')
|
202 |
+
conn.execute('''
|
203 |
+
CREATE INDEX IF NOT EXISTS idx_created_ttl ON cache_entries(created_at, ttl_seconds)
|
204 |
+
''')
|
205 |
except Exception as e:
|
206 |
+
logger.error(f"خطا در ایجاد پایگاه داده کش: {e}")
|
|
|
207 |
|
208 |
+
def _generate_key(self, url: str, model_type: str = "general") -> str:
|
209 |
+
"""تولید کلید یکتا برای کش"""
|
210 |
+
content = f"{url}:{model_type}"
|
211 |
+
return hashlib.md5(content.encode()).hexdigest()
|
212 |
+
|
213 |
+
def get(self, url: str, model_type: str = "general") -> Optional[Dict]:
|
214 |
+
"""دریافت از کش با بررسی TTL"""
|
215 |
+
key = self._generate_key(url, model_type)
|
216 |
|
217 |
+
# بررسی memory cache
|
218 |
+
if key in self.memory_cache:
|
219 |
+
self.access_count[key] = self.access_count.get(key, 0) + 1
|
220 |
+
return self.memory_cache[key]
|
221 |
+
|
222 |
+
# بررسی database cache
|
223 |
+
try:
|
224 |
+
with sqlite3.connect(self.cache_db_path) as conn:
|
225 |
+
cursor = conn.execute('''
|
226 |
+
SELECT value, created_at, ttl_seconds, access_count
|
227 |
+
FROM cache_entries
|
228 |
+
WHERE key = ?
|
229 |
+
''', (key,))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
230 |
|
231 |
+
row = cursor.fetchone()
|
232 |
+
if row:
|
233 |
+
value_blob, created_at, ttl_seconds, access_count = row
|
234 |
+
created_time = datetime.fromisoformat(created_at)
|
235 |
|
236 |
+
# بررسی انقضاء
|
237 |
+
if datetime.now() - created_time < timedelta(seconds=ttl_seconds):
|
238 |
+
# بروزرسانی شمارنده دسترسی
|
239 |
+
conn.execute(
|
240 |
+
'UPDATE cache_entries SET access_count = access_count + 1 WHERE key = ?',
|
241 |
+
(key,)
|
242 |
+
)
|
243 |
+
|
244 |
+
# اضافه به memory cache
|
245 |
+
value = pickle.loads(value_blob)
|
246 |
+
self._add_to_memory_cache(key, value)
|
247 |
+
return value
|
248 |
+
else:
|
249 |
+
# حذف entry منقضی
|
250 |
+
conn.execute('DELETE FROM cache_entries WHERE key = ?', (key,))
|
251 |
+
except Exception as e:
|
252 |
+
logger.error(f"خطا در دریافت از کش: {e}")
|
253 |
+
|
254 |
+
return None
|
255 |
+
|
256 |
+
def set(self, url: str, value: Dict, model_type: str = "general",
|
257 |
+
ttl_seconds: int = 3600, priority: int = 1):
|
258 |
+
"""ذخیره در کش"""
|
259 |
+
key = self._generate_key(url, model_type)
|
260 |
+
|
261 |
+
# ذخیره در memory cache
|
262 |
+
self._add_to_memory_cache(key, value)
|
263 |
|
264 |
+
# ذخیره در database
|
265 |
+
try:
|
266 |
+
with sqlite3.connect(self.cache_db_path) as conn:
|
267 |
+
value_blob = pickle.dumps(value)
|
268 |
+
conn.execute('''
|
269 |
+
INSERT OR REPLACE INTO cache_entries
|
270 |
+
(key, value, ttl_seconds, priority)
|
271 |
+
VALUES (?, ?, ?, ?)
|
272 |
+
''', (key, value_blob, ttl_seconds, priority))
|
273 |
+
except Exception as e:
|
274 |
+
logger.error(f"خطا در ذخیره در کش: {e}")
|
275 |
+
|
276 |
+
def _add_to_memory_cache(self, key: str, value: Dict):
|
277 |
+
"""اضافه کردن به memory cache با مدیریت حد"""
|
278 |
+
if len(self.memory_cache) >= self.max_memory_items:
|
279 |
+
# حذف کماستفادهترین item
|
280 |
+
if self.access_count:
|
281 |
+
least_used_key = min(self.access_count.keys(), key=self.access_count.get)
|
282 |
+
if least_used_key in self.memory_cache:
|
283 |
+
del self.memory_cache[least_used_key]
|
284 |
+
if least_used_key in self.access_count:
|
285 |
+
del self.access_count[least_used_key]
|
286 |
+
|
287 |
+
self.memory_cache[key] = value
|
288 |
+
self.access_count[key] = self.access_count.get(key, 0) + 1
|
289 |
+
|
290 |
+
def cleanup_expired(self):
|
291 |
+
"""پاکسازی entries منقضی"""
|
292 |
+
try:
|
293 |
+
with sqlite3.connect(self.cache_db_path) as conn:
|
294 |
+
conn.execute('''
|
295 |
+
DELETE FROM cache_entries
|
296 |
+
WHERE datetime(created_at, '+' || ttl_seconds || ' seconds') < datetime('now')
|
297 |
+
''')
|
298 |
+
conn.commit()
|
299 |
+
except Exception as e:
|
300 |
+
logger.error(f"خطا در پاکسازی کش: {e}")
|
301 |
+
|
302 |
+
def get_stats(self) -> Dict:
|
303 |
+
"""آمار کش"""
|
304 |
+
try:
|
305 |
+
with sqlite3.connect(self.cache_db_path) as conn:
|
306 |
+
cursor = conn.execute('''
|
307 |
+
SELECT
|
308 |
+
COUNT(*) as total_entries,
|
309 |
+
SUM(access_count) as total_accesses,
|
310 |
+
AVG(access_count) as avg_accesses
|
311 |
+
FROM cache_entries
|
312 |
+
''')
|
313 |
+
stats = cursor.fetchone()
|
314 |
+
|
315 |
+
return {
|
316 |
+
'memory_cache_size': len(self.memory_cache),
|
317 |
+
'database_entries': stats[0] if stats and stats[0] else 0,
|
318 |
+
'total_accesses': stats[1] if stats and stats[1] else 0,
|
319 |
+
'average_accesses': round(stats[2], 2) if stats and stats[2] else 0
|
320 |
+
}
|
321 |
+
except Exception as e:
|
322 |
+
logger.error(f"خطا در دریافت آمار کش: {e}")
|
323 |
+
return {
|
324 |
+
'memory_cache_size': len(self.memory_cache),
|
325 |
+
'database_entries': 0,
|
326 |
+
'total_accesses': 0,
|
327 |
+
'average_accesses': 0
|
328 |
+
}
|
329 |
|
330 |
+
# === سیستم امتیازدهی پیشرفته ===
|
331 |
+
class AdvancedScoringSystem:
|
332 |
+
"""سیستم امتیازدهی پیشرفته با وزندهی چندگانه"""
|
333 |
+
|
334 |
def __init__(self):
|
335 |
+
self.weights = {
|
336 |
+
'content_length': 0.15,
|
337 |
+
'legal_terms_density': 0.25,
|
338 |
+
'source_reliability': 0.20,
|
339 |
+
'structure_quality': 0.15,
|
340 |
+
'linguistic_quality': 0.15,
|
341 |
+
'citation_count': 0.10
|
342 |
+
}
|
343 |
+
|
344 |
+
try:
|
345 |
+
self.normalizer = Normalizer()
|
346 |
+
self.lemmatizer = Lemmatizer()
|
347 |
+
except Exception as e:
|
348 |
+
logger.warning(f"خطا در بارگذاری ابزارهای پردازش متن: {e}")
|
349 |
+
self.normalizer = None
|
350 |
+
self.lemmatizer = None
|
351 |
|
352 |
+
def calculate_comprehensive_score(self, content: str, source_info: Dict,
|
353 |
+
legal_entities: List[str]) -> Dict[str, float]:
|
354 |
+
"""محاسبه امتیاز جامع"""
|
355 |
+
scores = {}
|
356 |
|
357 |
try:
|
358 |
+
# امتیاز طول محتوا
|
359 |
+
scores['content_length'] = self._score_content_length(content)
|
360 |
|
361 |
+
# تراکم اصطلاحات حقوقی
|
362 |
+
scores['legal_terms_density'] = self._score_legal_terms_density(content)
|
363 |
|
364 |
+
# قابلیت اعتماد منبع
|
365 |
+
scores['source_reliability'] = source_info.get('reliability_score', 0.5)
|
366 |
|
367 |
+
# کیفیت ساختار
|
368 |
+
scores['structure_quality'] = self._score_structure_quality(content)
|
369 |
|
370 |
+
# کیفیت زبانی
|
371 |
+
scores['linguistic_quality'] = self._score_linguistic_quality(content)
|
372 |
+
|
373 |
+
# تعداد ارجاعات
|
374 |
+
scores['citation_count'] = self._score_citations(content, legal_entities)
|
375 |
+
|
376 |
+
# محاسبه امتیاز نهایی
|
377 |
+
final_score = sum(
|
378 |
+
scores[factor] * self.weights[factor]
|
379 |
+
for factor in scores
|
380 |
+
)
|
381 |
+
|
382 |
+
scores['final_score'] = min(100, max(0, final_score * 100))
|
383 |
+
|
384 |
+
except Exception as e:
|
385 |
+
logger.error(f"خطا در محاسبه امتیاز: {e}")
|
386 |
+
scores = {
|
387 |
+
'content_length': 0.5,
|
388 |
+
'legal_terms_density': 0.5,
|
389 |
+
'source_reliability': 0.5,
|
390 |
+
'structure_quality': 0.5,
|
391 |
+
'linguistic_quality': 0.5,
|
392 |
+
'citation_count': 0.5,
|
393 |
+
'final_score': 50.0
|
394 |
+
}
|
395 |
+
|
396 |
+
return scores
|
397 |
+
|
398 |
+
def _score_content_length(self, content: str) -> float:
|
399 |
+
"""امتیازدهی بر اساس طول محتوا"""
|
400 |
+
try:
|
401 |
+
word_count = len(content.split())
|
402 |
+
if word_count < 50:
|
403 |
+
return word_count / 50 * 0.5
|
404 |
+
elif word_count > 2000:
|
405 |
+
return 1.0
|
406 |
+
else:
|
407 |
+
return 0.5 + (word_count - 50) / 1950 * 0.5
|
408 |
+
except:
|
409 |
+
return 0.5
|
410 |
+
|
411 |
+
def _score_legal_terms_density(self, content: str) -> float:
|
412 |
+
"""امتیازدهی تراکم اصطلاحات حقوقی"""
|
413 |
+
try:
|
414 |
+
total_terms = 0
|
415 |
+
content_lower = content.lower()
|
416 |
+
|
417 |
+
for category, terms in PERSIAN_LEGAL_TERMS.items():
|
418 |
+
for term in terms:
|
419 |
+
total_terms += content_lower.count(term.lower())
|
420 |
+
|
421 |
+
words = len(content.split())
|
422 |
+
if words == 0:
|
423 |
+
return 0.0
|
424 |
+
|
425 |
+
density = total_terms / words
|
426 |
+
return min(1.0, density * 20) # نرمالسازی
|
427 |
+
except:
|
428 |
+
return 0.5
|
429 |
+
|
430 |
+
def _score_structure_quality(self, content: str) -> float:
|
431 |
+
"""امتیازدهی کیفیت ساختار"""
|
432 |
+
try:
|
433 |
+
score = 0.0
|
434 |
+
|
435 |
+
# بررسی وجود ساختار مواد و تبصرهها
|
436 |
+
if 'ماده' in content:
|
437 |
+
score += 0.3
|
438 |
+
if 'تبصره' in content:
|
439 |
+
score += 0.2
|
440 |
+
if 'فصل' in content or 'باب' in content:
|
441 |
+
score += 0.2
|
442 |
+
|
443 |
+
# بررسی پاراگرافبندی
|
444 |
+
paragraphs = content.split('\n')
|
445 |
+
if len(paragraphs) > 2:
|
446 |
+
score += 0.3
|
447 |
+
|
448 |
+
return min(1.0, score)
|
449 |
+
except:
|
450 |
+
return 0.5
|
451 |
+
|
452 |
+
def _score_linguistic_quality(self, content: str) -> float:
|
453 |
+
"""امتیازدهی کیفیت زبانی"""
|
454 |
+
try:
|
455 |
+
score = 0.0
|
456 |
+
|
457 |
+
if not content:
|
458 |
+
return 0.0
|
459 |
+
|
460 |
+
# نسبت کاراکترهای فارسی
|
461 |
+
persian_chars = sum(1 for c in content if '\u0600' <= c <= '\u06FF')
|
462 |
+
persian_ratio = persian_chars / len(content)
|
463 |
+
score += persian_ratio * 0.5
|
464 |
+
|
465 |
+
# بررسی وجود علائم نگارشی
|
466 |
+
punctuation_count = sum(1 for c in content if c in '.,;:!؟')
|
467 |
+
words_count = len(content.split())
|
468 |
+
if words_count > 0:
|
469 |
+
punctuation_ratio = punctuation_count / words_count
|
470 |
+
score += min(0.3, punctuation_ratio * 3)
|
471 |
+
|
472 |
+
# بررسی طول متوسط جملات
|
473 |
+
sentences = re.split(r'[.؟!]', content)
|
474 |
+
if sentences:
|
475 |
+
avg_sentence_length = sum(len(s.split()) for s in sentences) / len(sentences)
|
476 |
+
if 10 <= avg_sentence_length <= 25:
|
477 |
+
score += 0.2
|
478 |
+
|
479 |
+
return min(1.0, score)
|
480 |
+
except:
|
481 |
+
return 0.5
|
482 |
+
|
483 |
+
def _score_citations(self, content: str, legal_entities: List[str]) -> float:
|
484 |
+
"""امتیازدهی ارجاعات قانونی"""
|
485 |
+
try:
|
486 |
+
citation_patterns = [
|
487 |
+
r'ماده\s*\d+', r'تبصره\s*\d+', r'بند\s*\d+',
|
488 |
+
r'فصل\s*\d+', r'قانون\s+[آ-ی\s]+', r'مصوبه\s+[آ-ی\s]+'
|
489 |
]
|
490 |
|
491 |
+
total_citations = 0
|
492 |
+
for pattern in citation_patterns:
|
493 |
+
total_citations += len(re.findall(pattern, content))
|
494 |
|
495 |
+
# اضافه کردن موجودیتهای قانونی
|
496 |
+
total_citations += len(legal_entities)
|
497 |
|
498 |
+
# نرمالسازی (هر 5 ارجاع = امتیاز کامل)
|
499 |
+
return min(1.0, total_citations / 5)
|
500 |
+
except:
|
501 |
+
return 0.5
|
502 |
+
|
503 |
+
# === سیستم طبقهبندی هوشمند ===
|
504 |
+
class IntelligentClassificationSystem:
|
505 |
+
"""سیستم طبقهبندی هوشمند چندمرحلهای"""
|
506 |
+
|
507 |
+
def __init__(self, cache_system: IntelligentCacheSystem):
|
508 |
+
self.cache_system = cache_system
|
509 |
+
self.models = {}
|
510 |
+
self.is_ready = False
|
511 |
+
|
512 |
+
# دستههای حقوقی
|
513 |
+
self.legal_categories = {
|
514 |
+
'قانون': ['قانون', 'مقررات', 'آییننامه'],
|
515 |
+
'دادنامه': ['دادنامه', 'رای', 'حکم'],
|
516 |
+
'قرارداد': ['قرارداد', 'توافقنامه', 'پروتکل'],
|
517 |
+
'لایحه': ['لایحه', 'طرح', 'پیشنهاد'],
|
518 |
+
'بخشنامه': ['بخشنامه', 'دستورالعمل', 'رهنمود'],
|
519 |
+
'نظریه': ['نظریه', 'استعلام', 'پاسخ']
|
520 |
+
}
|
521 |
+
|
522 |
+
def load_models(self):
|
523 |
+
"""بارگذاری مدلهای طبقهبندی"""
|
524 |
+
try:
|
525 |
+
logger.info("شروع بارگذاری مدلها...")
|
526 |
|
527 |
+
# بارگذاری ساده مدل embedding
|
528 |
+
try:
|
529 |
+
self.models['embedder'] = SentenceTransformer(
|
530 |
+
AVAILABLE_MODELS['embedding']['sentence_transformer']
|
531 |
+
)
|
532 |
+
logger.info("مدل embedding بارگذاری شد")
|
533 |
+
except Exception as e:
|
534 |
+
logger.warning(f"خطا در بارگذاری مدل embedding: {e}")
|
535 |
|
536 |
+
self.is_ready = True
|
537 |
+
logger.info("سیستم طبقهبندی آماده است")
|
538 |
|
539 |
except Exception as e:
|
540 |
+
logger.error(f"خطا در بارگذاری مدلها: {e}")
|
541 |
+
self.is_ready = False
|
|
|
|
|
|
|
|
|
542 |
|
543 |
+
def classify_document(self, content: str, use_cache: bool = True) -> Dict:
|
544 |
+
"""طبقهبندی هوشمند سند"""
|
545 |
+
if not content or not content.strip():
|
546 |
+
return {'error': 'محتوا خالی است'}
|
547 |
+
|
548 |
+
# بررسی کش
|
549 |
+
cache_key = hashlib.md5(content.encode()).hexdigest()
|
550 |
+
if use_cache:
|
551 |
+
cached = self.cache_system.get(cache_key, 'classification')
|
552 |
+
if cached:
|
553 |
+
cached['cache_hit'] = True
|
554 |
+
return cached
|
555 |
+
|
556 |
+
start_time = time.time()
|
557 |
+
result = {}
|
558 |
+
|
559 |
try:
|
560 |
+
# متن نمونه برای پردازش
|
561 |
+
text_sample = ' '.join(content.split()[:400])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
562 |
|
563 |
+
# طبقهبندی بر اساس قوانین
|
564 |
+
result['rule_based_classification'] = self._rule_based_classify(content)
|
565 |
+
|
566 |
+
# استخراج موجودیتهای حقوقی
|
567 |
+
result['legal_entities'] = self._extract_legal_entities(content)
|
568 |
+
|
569 |
+
# تولید embedding
|
570 |
+
if 'embedder' in self.models:
|
571 |
+
try:
|
572 |
+
result['embedding'] = self.models['embedder'].encode(text_sample)
|
573 |
+
except Exception as e:
|
574 |
+
logger.warning(f"خطا در تولید embedding: {e}")
|
575 |
+
result['embedding'] = None
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
576 |
|
577 |
+
# محاسبه اعتماد ترکیبی
|
578 |
+
result['confidence_score'] = self._calculate_combined_confidence(result)
|
579 |
+
|
580 |
+
result['processing_time'] = time.time() - start_time
|
581 |
+
result['cache_hit'] = False
|
582 |
+
|
583 |
+
# ذخیره در کش
|
584 |
+
if use_cache:
|
585 |
+
self.cache_system.set(cache_key, result, 'classification', ttl_seconds=7200)
|
586 |
|
587 |
except Exception as e:
|
588 |
+
logger.error(f"خطا در طبقهبندی: {e}")
|
589 |
+
result['error'] = str(e)
|
590 |
+
|
591 |
+
return result
|
592 |
|
593 |
+
def _rule_based_classify(self, content: str) -> Dict[str, float]:
|
594 |
+
"""طبقهبندی بر اساس قوانین"""
|
595 |
+
scores = {}
|
596 |
+
content_lower = content.lower()
|
597 |
+
|
598 |
+
for category, keywords in self.legal_categories.items():
|
599 |
+
score = 0.0
|
600 |
+
for keyword in keywords:
|
601 |
+
count = content_lower.count(keyword.lower())
|
602 |
+
score += count * 0.1
|
603 |
+
|
604 |
+
scores[category] = min(1.0, score)
|
605 |
+
|
606 |
+
# نرمالسازی امتیازها
|
607 |
+
total_score = sum(scores.values())
|
608 |
+
if total_score > 0:
|
609 |
+
scores = {k: v/total_score for k, v in scores.items()}
|
610 |
+
|
611 |
+
return scores
|
612 |
+
|
613 |
+
def _extract_legal_entities(self, content: str) -> List[str]:
|
614 |
+
"""استخراج موجودیتهای حقوقی"""
|
615 |
+
entities = []
|
616 |
|
617 |
try:
|
618 |
+
# الگوهای ارجاعات قانونی
|
619 |
+
patterns = [
|
620 |
+
r'ماده\s*(\d+)', r'تبصره\s*(\d+)', r'بند\s*([الف-ی]|\d+)',
|
621 |
+
r'فصل\s*(\d+)', r'قانون\s+([آ-ی\s]{5,50})',
|
622 |
+
r'مصوبه\s+([آ-ی\s]{5,50})'
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
623 |
]
|
624 |
|
625 |
+
for pattern in patterns:
|
626 |
+
matches = re.findall(pattern, content)
|
627 |
+
entities.extend([str(m) for m in matches])
|
628 |
+
|
629 |
+
except Exception as e:
|
630 |
+
logger.error(f"خطا در استخراج موجودیتها: {e}")
|
631 |
+
|
632 |
+
return list(set(entities))[:20]
|
633 |
+
|
634 |
+
def _calculate_combined_confidence(self, result: Dict) -> float:
|
635 |
+
"""محاسبه اعتماد ترکیبی"""
|
636 |
+
try:
|
637 |
+
confidence = 0.0
|
638 |
+
weights = {'rule_based': 0.7, 'entities': 0.3}
|
639 |
+
|
640 |
+
# اعتماد طبقهبندی قانونی
|
641 |
+
if 'rule_based_classification' in result:
|
642 |
+
rule_conf = max(result['rule_based_classification'].values()) if result['rule_based_classification'] else 0
|
643 |
+
confidence += rule_conf * weights['rule_based']
|
644 |
|
645 |
+
# تعداد موجودیتهای قانونی
|
646 |
+
entity_count = len(result.get('legal_entities', []))
|
647 |
+
entity_conf = min(1.0, entity_count / 5)
|
648 |
+
confidence += entity_conf * weights['entities']
|
|
|
|
|
|
|
|
|
|
|
649 |
|
650 |
+
return round(confidence, 3)
|
651 |
+
except:
|
652 |
+
return 0.5
|
653 |
+
|
654 |
+
# === مدیریت پایگاه داده پیشرفته ===
|
655 |
+
class AdvancedDatabaseManager:
|
656 |
+
"""مدیریت پیشرفته پایگاه داده با بهینهسازیهای جدید"""
|
657 |
+
|
658 |
+
def __init__(self, db_path: str = DB_PATH):
|
659 |
+
self.db_path = db_path
|
660 |
+
self._init_database()
|
661 |
+
|
662 |
+
def _init_database(self):
|
663 |
+
"""ایجاد پایگاه داده با جداول پیشرفته"""
|
664 |
+
try:
|
665 |
+
with sqlite3.connect(self.db_path) as conn:
|
666 |
+
# جدول اسناد اصلی
|
667 |
+
conn.execute('''
|
668 |
+
CREATE TABLE IF NOT EXISTS documents (
|
669 |
+
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
670 |
+
url TEXT UNIQUE NOT NULL,
|
671 |
+
title TEXT,
|
672 |
+
source TEXT,
|
673 |
+
content TEXT,
|
674 |
+
word_count INTEGER,
|
675 |
+
quality_scores TEXT,
|
676 |
+
classification_result TEXT,
|
677 |
+
legal_entities TEXT,
|
678 |
+
embedding_vector BLOB,
|
679 |
+
scraped_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
680 |
+
last_updated TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
681 |
+
processing_status TEXT DEFAULT 'pending'
|
682 |
+
)
|
683 |
+
''')
|
684 |
+
|
685 |
+
# ایجاد ایندکسها برای جدول documents
|
686 |
+
conn.execute('CREATE INDEX IF NOT EXISTS idx_documents_source ON documents(source)')
|
687 |
+
conn.execute('CREATE INDEX IF NOT EXISTS idx_documents_scraped_at ON documents(scraped_at)')
|
688 |
+
conn.execute('CREATE INDEX IF NOT EXISTS idx_documents_status ON documents(processing_status)')
|
689 |
+
conn.execute('CREATE INDEX IF NOT EXISTS idx_documents_updated ON documents(last_updated)')
|
690 |
+
|
691 |
+
# جدول متریکهای عملکرد
|
692 |
+
conn.execute('''
|
693 |
+
CREATE TABLE IF NOT EXISTS performance_metrics (
|
694 |
+
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
695 |
+
metric_name TEXT,
|
696 |
+
metric_value REAL,
|
697 |
+
timestamp TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
698 |
+
)
|
699 |
+
''')
|
700 |
+
|
701 |
+
# ایندکسها برای جدول metrics
|
702 |
+
conn.execute('CREATE INDEX IF NOT EXISTS idx_metrics_name ON performance_metrics(metric_name)')
|
703 |
+
conn.execute('CREATE INDEX IF NOT EXISTS idx_metrics_timestamp ON performance_metrics(timestamp)')
|
704 |
+
|
705 |
+
# جدول لاگهای پردازش
|
706 |
+
conn.execute('''
|
707 |
+
CREATE TABLE IF NOT EXISTS processing_logs (
|
708 |
+
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
709 |
+
document_id INTEGER,
|
710 |
+
operation TEXT,
|
711 |
+
status TEXT,
|
712 |
+
details TEXT,
|
713 |
+
processing_time REAL,
|
714 |
+
timestamp TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
715 |
+
FOREIGN KEY (document_id) REFERENCES documents (id)
|
716 |
+
)
|
717 |
+
''')
|
718 |
+
|
719 |
+
# ایندکسها برای جدول logs
|
720 |
+
conn.execute('CREATE INDEX IF NOT EXISTS idx_logs_operation ON processing_logs(operation)')
|
721 |
+
conn.execute('CREATE INDEX IF NOT EXISTS idx_logs_status ON processing_logs(status)')
|
722 |
+
conn.execute('CREATE INDEX IF NOT EXISTS idx_logs_timestamp ON processing_logs(timestamp)')
|
723 |
+
conn.execute('CREATE INDEX IF NOT EXISTS idx_logs_document_id ON processing_logs(document_id)')
|
724 |
+
|
725 |
+
except Exception as e:
|
726 |
+
logger.error(f"خطا در ایجاد پایگاه داده: {e}")
|
727 |
+
|
728 |
+
def save_document_advanced(self, result: ProcessingResult) -> bool:
|
729 |
+
"""ذخیره پیشرفته سند"""
|
730 |
+
try:
|
731 |
+
with sqlite3.connect(self.db_path) as conn:
|
732 |
+
# محاسبه embedding bytes
|
733 |
+
embedding_blob = None
|
734 |
+
if result.embeddings is not None:
|
735 |
+
embedding_blob = result.embeddings.tobytes()
|
736 |
+
|
737 |
+
conn.execute('''
|
738 |
+
INSERT OR REPLACE INTO documents
|
739 |
+
(url, title, source, content, word_count, quality_scores,
|
740 |
+
classification_result, legal_entities, embedding_vector,
|
741 |
+
processing_status, last_updated)
|
742 |
+
VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, 'completed', datetime('now'))
|
743 |
+
''', (
|
744 |
+
result.url,
|
745 |
+
result.title,
|
746 |
+
result.source,
|
747 |
+
result.content,
|
748 |
+
len(result.content.split()),
|
749 |
+
json.dumps({'quality_score': result.quality_score}, ensure_ascii=False),
|
750 |
+
json.dumps(result.classification, ensure_ascii=False),
|
751 |
+
json.dumps(result.legal_entities, ensure_ascii=False),
|
752 |
+
embedding_blob
|
753 |
+
))
|
754 |
+
|
755 |
+
# ثبت لاگ
|
756 |
+
document_id = conn.lastrowid
|
757 |
+
conn.execute('''
|
758 |
+
INSERT INTO processing_logs
|
759 |
+
(document_id, operation, status, processing_time)
|
760 |
+
VALUES (?, 'save', 'success', ?)
|
761 |
+
''', (document_id, result.processing_time))
|
762 |
+
|
763 |
+
return True
|
764 |
+
|
765 |
+
except Exception as e:
|
766 |
+
logger.error(f"خطا در ذخیره پیشرفته: {e}")
|
767 |
+
return False
|
768 |
+
|
769 |
+
def get_documents_with_embeddings(self) -> List[Tuple]:
|
770 |
+
"""دریافت اسناد همراه با embeddings"""
|
771 |
+
try:
|
772 |
+
with sqlite3.connect(self.db_path) as conn:
|
773 |
+
return conn.execute('''
|
774 |
+
SELECT id, url, title, content, embedding_vector
|
775 |
+
FROM documents
|
776 |
+
WHERE processing_status = 'completed'
|
777 |
+
AND content IS NOT NULL
|
778 |
+
ORDER BY last_updated DESC
|
779 |
+
''').fetchall()
|
780 |
+
except Exception as e:
|
781 |
+
logger.error(f"خطا در دریافت اسناد: {e}")
|
782 |
+
return []
|
783 |
+
|
784 |
+
def get_advanced_stats(self) -> Dict:
|
785 |
+
"""آمار پیشرفته سیستم"""
|
786 |
+
try:
|
787 |
+
with sqlite3.connect(self.db_path) as conn:
|
788 |
+
# آمار کلی
|
789 |
+
total_docs = conn.execute("SELECT COUNT(*) FROM documents").fetchone()[0]
|
790 |
+
|
791 |
+
# آمار بر اساس منبع
|
792 |
+
source_stats = conn.execute('''
|
793 |
+
SELECT source, COUNT(*), AVG(word_count)
|
794 |
+
FROM documents
|
795 |
+
GROUP BY source
|
796 |
+
ORDER BY COUNT(*) DESC
|
797 |
+
''').fetchall()
|
798 |
+
|
799 |
+
# آمار عملکرد
|
800 |
+
avg_processing_time = conn.execute('''
|
801 |
+
SELECT AVG(processing_time)
|
802 |
+
FROM processing_logs
|
803 |
+
WHERE operation = 'save' AND status = 'success'
|
804 |
+
''').fetchone()[0] or 0
|
805 |
+
|
806 |
+
# آمار وضعیت پردازش
|
807 |
+
status_stats = conn.execute('''
|
808 |
+
SELECT processing_status, COUNT(*)
|
809 |
+
FROM documents
|
810 |
+
GROUP BY processing_status
|
811 |
+
''').fetchall()
|
812 |
+
|
813 |
+
return {
|
814 |
+
'total_documents': total_docs,
|
815 |
+
'source_statistics': source_stats,
|
816 |
+
'average_processing_time': round(avg_processing_time, 2),
|
817 |
+
'status_statistics': status_stats,
|
818 |
+
'last_updated': datetime.now().isoformat()
|
819 |
+
}
|
820 |
+
except Exception as e:
|
821 |
+
logger.error(f"خطا در دریافت آمار: {e}")
|
822 |
+
return {
|
823 |
+
'total_documents': 0,
|
824 |
+
'source_statistics': [],
|
825 |
+
'average_processing_time': 0,
|
826 |
+
'status_statistics': [],
|
827 |
+
'last_updated': datetime.now().isoformat()
|
828 |
+
}
|
829 |
+
|
830 |
+
# === سیستم جستجوی معنایی پیشرفته ===
|
831 |
+
class SemanticSearchEngine:
|
832 |
+
"""موتور جستجوی معنایی با ایندکس FAISS"""
|
833 |
+
|
834 |
+
def __init__(self, cache_system: IntelligentCacheSystem):
|
835 |
+
self.cache_system = cache_system
|
836 |
+
self.embedder = None
|
837 |
+
self.faiss_index = None
|
838 |
+
self.documents = []
|
839 |
+
self.document_embeddings = None
|
840 |
+
self.is_ready = False
|
841 |
+
|
842 |
+
def initialize(self):
|
843 |
+
"""مقداردهی اولیه موتور جستجو"""
|
844 |
+
try:
|
845 |
+
self.embedder = SentenceTransformer(
|
846 |
+
AVAILABLE_MODELS['embedding']['sentence_transformer']
|
847 |
+
)
|
848 |
+
logger.info("مدل embedding بارگذاری شد")
|
849 |
+
|
850 |
+
# بارگذاری ایندکس موجود در صورت وجود
|
851 |
+
if os.path.exists(VECTOR_INDEX_PATH) and os.path.exists(EMBEDDINGS_CACHE_PATH):
|
852 |
+
self._load_existing_index()
|
853 |
+
|
854 |
+
self.is_ready = True
|
855 |
|
856 |
except Exception as e:
|
857 |
+
logger.error(f"خطا در مقداردهی موتور جستجو: {e}")
|
|
|
858 |
|
859 |
+
def build_index(self, documents: List[Tuple], progress_callback=None):
|
860 |
+
"""ساخت ایندکس FAISS"""
|
861 |
try:
|
862 |
+
if not documents:
|
863 |
+
return False
|
864 |
|
865 |
+
if progress_callback:
|
866 |
+
progress_callback("استخراج متون...", 0.1)
|
867 |
|
868 |
+
self.documents = documents
|
869 |
+
texts = [doc[3] for doc in documents if doc[3]] # content field
|
870 |
|
871 |
+
if progress_callback:
|
872 |
+
progress_callback(f"تولید embedding برای {len(texts)} سند...", 0.3)
|
|
|
|
|
|
|
|
|
|
|
|
|
873 |
|
874 |
+
# تولید embeddings با batch processing
|
875 |
+
batch_size = 16
|
876 |
+
all_embeddings = []
|
877 |
|
878 |
+
for i in range(0, len(texts), batch_size):
|
879 |
+
batch = texts[i:i+batch_size]
|
880 |
+
batch_embeddings = self.embedder.encode(
|
881 |
+
batch,
|
882 |
+
convert_to_tensor=True,
|
883 |
+
show_progress_bar=False
|
884 |
+
)
|
885 |
+
all_embeddings.append(batch_embeddings)
|
886 |
|
887 |
+
if progress_callback:
|
888 |
+
progress = 0.3 + (i / len(texts)) * 0.6
|
889 |
+
progress_callback(f"پردازش batch {i//batch_size + 1}...", progress)
|
890 |
+
|
891 |
+
# ترکیب embeddings
|
892 |
+
self.document_embeddings = torch.cat(all_embeddings, dim=0).cpu().numpy()
|
893 |
+
|
894 |
+
if progress_callback:
|
895 |
+
progress_callback("ساخت ایندکس FAISS...", 0.9)
|
896 |
+
|
897 |
+
# ساخت ایندکس FAISS
|
898 |
+
dimension = self.document_embeddings.shape[1]
|
899 |
+
self.faiss_index = faiss.IndexFlatIP(dimension) # Inner Product for cosine similarity
|
900 |
+
|
901 |
+
# نرمالسازی برای cosine similarity
|
902 |
+
faiss.normalize_L2(self.document_embeddings)
|
903 |
+
self.faiss_index.add(self.document_embeddings)
|
904 |
+
|
905 |
+
# ذخیره ایندکس
|
906 |
+
self._save_index()
|
907 |
+
|
908 |
+
if progress_callback:
|
909 |
+
progress_callback("تکمیل ایندکسسازی", 1.0)
|
910 |
+
|
911 |
+
logger.info(f"ایندکس با {len(documents)} سند آماده شد")
|
912 |
+
return True
|
913 |
+
|
914 |
+
except Exception as e:
|
915 |
+
logger.error(f"خطا در ساخت ایندکس: {e}")
|
916 |
+
return False
|
917 |
+
|
918 |
+
def search(self, query: str, top_k: int = 10, threshold: float = 0.3) -> List[Dict]:
|
919 |
+
"""جستجوی معنایی پیشرفته"""
|
920 |
+
if not self.is_ready or self.faiss_index is None:
|
921 |
+
return []
|
922 |
+
|
923 |
+
try:
|
924 |
+
# بررسی کش
|
925 |
+
cache_key = f"search:{hashlib.md5(query.encode()).hexdigest()}:{top_k}"
|
926 |
+
cached = self.cache_system.get(cache_key, 'search')
|
927 |
+
if cached:
|
928 |
+
return cached
|
929 |
+
|
930 |
+
# تولید embedding برای query
|
931 |
+
query_embedding = self.embedder.encode([query], convert_to_tensor=True)
|
932 |
+
query_embedding = query_embedding.cpu().numpy()
|
933 |
+
faiss.normalize_L2(query_embedding)
|
934 |
+
|
935 |
+
# جستجو در ایندکس
|
936 |
+
similarities, indices = self.faiss_index.search(query_embedding, top_k * 2)
|
937 |
+
|
938 |
+
results = []
|
939 |
+
for i, (similarity, idx) in enumerate(zip(similarities[0], indices[0])):
|
940 |
+
if similarity < threshold:
|
941 |
+
continue
|
942 |
+
|
943 |
+
if idx < len(self.documents):
|
944 |
+
doc = self.documents[idx]
|
945 |
+
results.append({
|
946 |
+
'rank': i + 1,
|
947 |
+
'document_id': doc[0],
|
948 |
+
'url': doc[1],
|
949 |
+
'title': doc[2],
|
950 |
+
'content_preview': doc[3][:300] + '...' if len(doc[3]) > 300 else doc[3],
|
951 |
+
'similarity_score': float(similarity),
|
952 |
+
'relevance_category': self._categorize_relevance(similarity)
|
953 |
+
})
|
954 |
+
|
955 |
+
# مرتبسازی نهایی
|
956 |
+
results = results[:top_k]
|
957 |
+
|
958 |
+
# ذخیره در کش
|
959 |
+
self.cache_system.set(cache_key, results, 'search', ttl_seconds=1800)
|
960 |
+
|
961 |
+
return results
|
962 |
+
|
963 |
except Exception as e:
|
964 |
+
logger.error(f"خطا در جستجو: {e}")
|
965 |
+
return []
|
966 |
+
|
967 |
+
def _categorize_relevance(self, similarity: float) -> str:
|
968 |
+
"""دستهبندی میزان ارتباط"""
|
969 |
+
if similarity >= 0.8:
|
970 |
+
return "بسیار مرتبط"
|
971 |
+
elif similarity >= 0.6:
|
972 |
+
return "مرتبط"
|
973 |
+
elif similarity >= 0.4:
|
974 |
+
return "نسبتاً مرتبط"
|
975 |
+
else:
|
976 |
+
return "کمارتباط"
|
977 |
+
|
978 |
+
def _save_index(self):
|
979 |
+
"""ذخیره ایندکس و embeddings"""
|
980 |
+
try:
|
981 |
+
if self.faiss_index:
|
982 |
+
faiss.write_index(self.faiss_index, VECTOR_INDEX_PATH)
|
983 |
+
|
984 |
+
if self.document_embeddings is not None:
|
985 |
+
with open(EMBEDDINGS_CACHE_PATH, 'wb') as f:
|
986 |
+
pickle.dump({
|
987 |
+
'embeddings': self.document_embeddings,
|
988 |
+
'documents_info': [(doc[0], doc[1], doc[2]) for doc in self.documents]
|
989 |
+
}, f)
|
990 |
+
|
991 |
+
logger.info("ایندکس ذخیره شد")
|
992 |
+
|
993 |
+
except Exception as e:
|
994 |
+
logger.error(f"خطا در ذخیره ایندکس: {e}")
|
995 |
+
|
996 |
+
def _load_existing_index(self):
|
997 |
+
"""بارگذاری ایندکس موجود"""
|
998 |
+
try:
|
999 |
+
self.faiss_index = faiss.read_index(VECTOR_INDEX_PATH)
|
1000 |
+
|
1001 |
+
with open(EMBEDDINGS_CACHE_PATH, 'rb') as f:
|
1002 |
+
cache_data = pickle.load(f)
|
1003 |
+
self.document_embeddings = cache_data['embeddings']
|
1004 |
+
# بازسازی documents از اطلاعات ذخیره شده
|
1005 |
+
# نیاز به query از دیتابیس برای محتوای کامل
|
1006 |
+
|
1007 |
+
logger.info("ایندکس موجود بارگذاری شد")
|
1008 |
+
|
1009 |
+
except Exception as e:
|
1010 |
+
logger.error(f"خطا در بارگذاری ایندکس: {e}")
|
1011 |
|
1012 |
+
# === سیستم اسکرپر ساده ===
|
1013 |
+
class SimpleWebScraper:
|
1014 |
+
"""سیستم اسکرپر ساده برای تست"""
|
1015 |
+
|
1016 |
+
def __init__(self):
|
1017 |
+
self.session = requests.Session()
|
1018 |
+
self.session.headers.update({
|
1019 |
+
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36'
|
1020 |
+
})
|
1021 |
+
|
1022 |
+
def scrape_document(self, url: str) -> Dict:
|
1023 |
+
"""اسکرپ ساده یک سند"""
|
1024 |
+
try:
|
1025 |
+
response = self.session.get(url, timeout=10)
|
1026 |
+
response.raise_for_status()
|
1027 |
+
|
1028 |
+
soup = BeautifulSoup(response.content, 'html.parser')
|
1029 |
+
|
1030 |
+
# استخراج عنوان
|
1031 |
+
title = ""
|
1032 |
+
title_tag = soup.find('title')
|
1033 |
+
if title_tag:
|
1034 |
+
title = title_tag.get_text().strip()
|
1035 |
+
|
1036 |
+
# استخراج محتوا
|
1037 |
+
content = ""
|
1038 |
+
for tag in soup.find_all(['p', 'div', 'article']):
|
1039 |
+
if tag.get_text().strip():
|
1040 |
+
content += tag.get_text().strip() + "\n"
|
1041 |
+
|
1042 |
+
return {
|
1043 |
+
'status': 'موفق',
|
1044 |
+
'title': title,
|
1045 |
+
'content': content,
|
1046 |
+
'source_info': {'name': urlparse(url).netloc},
|
1047 |
+
'quality_assessment': {'overall_score': 0.7}
|
1048 |
+
}
|
1049 |
+
|
1050 |
+
except Exception as e:
|
1051 |
+
logger.error(f"خطا در اسکرپ {url}: {e}")
|
1052 |
+
return {
|
1053 |
+
'status': 'ناموفق',
|
1054 |
+
'error': str(e)
|
1055 |
+
}
|
1056 |
+
|
1057 |
+
# === اپلیکیشن اصلی پیشرفته ===
|
1058 |
+
class AdvancedLegalScrapingApp:
|
1059 |
+
"""اپلیکیشن اصلی پیشرفته اسکرپینگ حقوقی"""
|
1060 |
+
|
1061 |
+
def __init__(self):
|
1062 |
+
logger.info("🚀 شروع راهاندازی سیستم پیشرفته...")
|
1063 |
+
|
1064 |
+
# اجزای اصلی سیستم
|
1065 |
+
self.cache_system = IntelligentCacheSystem()
|
1066 |
+
self.scoring_system = AdvancedScoringSystem()
|
1067 |
+
self.db_manager = AdvancedDatabaseManager()
|
1068 |
+
self.classification_system = IntelligentClassificationSystem(self.cache_system)
|
1069 |
+
self.search_engine = SemanticSearchEngine(self.cache_system)
|
1070 |
+
self.scraper = SimpleWebScraper()
|
1071 |
+
|
1072 |
+
# متریکهای سیستم
|
1073 |
+
self.system_metrics = SystemMetrics()
|
1074 |
+
|
1075 |
+
logger.info("✅ سیستم آماده است")
|
1076 |
+
|
1077 |
+
def initialize_models(self, progress_callback=None) -> str:
|
1078 |
+
"""مقداردهی مدلها"""
|
1079 |
+
try:
|
1080 |
+
if progress_callback:
|
1081 |
+
progress_callback("بارگذاری مدلهای طبقهبندی...", 0.3)
|
1082 |
+
|
1083 |
+
self.classification_system.load_models()
|
1084 |
+
|
1085 |
+
if progress_callback:
|
1086 |
+
progress_callback("مقداردهی موتور جستجو...", 0.7)
|
1087 |
+
|
1088 |
+
self.search_engine.initialize()
|
1089 |
+
|
1090 |
+
if progress_callback:
|
1091 |
+
progress_callback("تکمیل مقداردهی", 1.0)
|
1092 |
+
|
1093 |
+
return "✅ تمام مدلها با موفقیت بارگذاری شدند"
|
1094 |
+
|
1095 |
+
except Exception as e:
|
1096 |
+
error_msg = f"❌ خطا در بارگذاری مدلها: {str(e)}"
|
1097 |
+
logger.error(error_msg)
|
1098 |
+
return error_msg
|
1099 |
+
|
1100 |
+
def process_urls_intelligent(self, urls_text: str) -> Tuple[str, str]:
|
1101 |
+
"""پردازش هوشمند URLs"""
|
1102 |
+
if not urls_text or not urls_text.strip():
|
1103 |
+
return "❌ لطفاً URLs را وارد کنید", ""
|
1104 |
+
|
1105 |
+
urls = [url.strip() for url in urls_text.split('\n') if url.strip()]
|
1106 |
+
if not urls:
|
1107 |
+
return "❌ URL معتبر یافت نشد", ""
|
1108 |
+
|
1109 |
+
results = []
|
1110 |
+
total_processing_time = 0
|
1111 |
+
|
1112 |
+
for url in urls[:5]: # محدودیت برای تست
|
1113 |
+
start_time = time.time()
|
1114 |
+
|
1115 |
+
try:
|
1116 |
+
# اسکرپ سند
|
1117 |
+
scraped_result = self.scraper.scrape_document(url)
|
1118 |
+
|
1119 |
+
if scraped_result.get('status') == 'موفق':
|
1120 |
+
# طبقهبندی
|
1121 |
+
classification = self.classification_system.classify_document(
|
1122 |
+
scraped_result.get('content', '')
|
1123 |
)
|
1124 |
|
1125 |
+
# امتیازدهی
|
1126 |
+
quality_scores = self.scoring_system.calculate_comprehensive_score(
|
1127 |
+
scraped_result.get('content', ''),
|
1128 |
+
scraped_result.get('source_info', {}),
|
1129 |
+
classification.get('legal_entities', [])
|
|
|
1130 |
)
|
1131 |
|
1132 |
+
# ساخت ProcessingResult
|
1133 |
+
processing_result = ProcessingResult(
|
1134 |
+
url=url,
|
1135 |
+
title=scraped_result.get('title', ''),
|
1136 |
+
content=scraped_result.get('content', ''),
|
1137 |
+
source=scraped_result.get('source_info', {}).get('name', ''),
|
1138 |
+
quality_score=quality_scores.get('final_score', 0),
|
1139 |
+
classification=classification.get('rule_based_classification', {}),
|
1140 |
+
sentiment_score=0.5,
|
1141 |
+
legal_entities=classification.get('legal_entities', []),
|
1142 |
+
embeddings=classification.get('embedding'),
|
1143 |
+
processing_time=classification.get('processing_time', 0)
|
1144 |
)
|
1145 |
+
|
1146 |
+
# ذخیره در دیتابیس
|
1147 |
+
self.db_manager.save_document_advanced(processing_result)
|
1148 |
+
|
1149 |
+
# ذخیره در کش
|
1150 |
+
cache_data = {
|
1151 |
+
'title': processing_result.title,
|
1152 |
+
'quality_score': processing_result.quality_score,
|
1153 |
+
'classification': processing_result.classification,
|
1154 |
+
'legal_entities': processing_result.legal_entities,
|
1155 |
+
'status': 'موفق'
|
1156 |
+
}
|
1157 |
+
self.cache_system.set(url, cache_data, 'comprehensive', ttl_seconds=7200)
|
1158 |
+
|
1159 |
+
results.append(cache_data)
|
1160 |
+
|
1161 |
+
else:
|
1162 |
+
results.append({'status': 'ناموفق', 'error': scraped_result.get('error', '')})
|
1163 |
|
1164 |
+
total_processing_time += time.time() - start_time
|
1165 |
+
|
1166 |
+
except Exception as e:
|
1167 |
+
logger.error(f"خطا در پردازش {url}: {e}")
|
1168 |
+
results.append({'status': 'ناموفق', 'error': str(e)})
|
1169 |
|
1170 |
+
# تأخیر بین درخواستها
|
1171 |
+
time.sleep(1)
|
1172 |
+
|
1173 |
+
# تولید گزارش جامع
|
1174 |
+
successful = sum(1 for r in results if r.get('status') == 'موفق')
|
1175 |
+
failed = len(results) - successful
|
1176 |
+
avg_quality = sum(r.get('quality_score', 0) for r in results if r.get('quality_score')) / max(successful, 1)
|
1177 |
+
|
1178 |
+
summary = f"""
|
1179 |
+
📊 **گزارش پردازش هوشمند**
|
1180 |
+
|
1181 |
+
✅ **موفق**: {successful} سند
|
1182 |
+
❌ **ناموفق**: {failed} سند
|
1183 |
+
📈 **میانگین کیفیت**: {avg_quality:.1f}/100
|
1184 |
+
⏱️ **زمان کل**: {total_processing_time:.2f} ثانیه
|
1185 |
+
|
1186 |
+
🎯 **نتایج کیفیت**:
|
1187 |
+
• بالا (>80): {sum(1 for r in results if r.get('quality_score', 0) > 80)}
|
1188 |
+
• متوسط (50-80): {sum(1 for r in results if 50 <= r.get('quality_score', 0) <= 80)}
|
1189 |
+
• پایین (<50): {sum(1 for r in results if 0 < r.get('quality_score', 0) < 50)}
|
1190 |
+
"""
|
1191 |
+
|
1192 |
+
# جزئیات
|
1193 |
+
details = "\n".join(
|
1194 |
+
f"📄 {r.get('title', 'بدون عنوان')[:50]}... - امتیاز: {r.get('quality_score', 0):.1f}"
|
1195 |
+
for r in results if r.get('status') == 'موفق'
|
1196 |
+
)
|
1197 |
|
1198 |
+
return summary, details
|
1199 |
+
|
1200 |
+
def build_intelligent_search_index(self, progress_callback=None) -> str:
|
1201 |
+
"""ساخت ایندکس جستجوی هوشمند"""
|
1202 |
+
try:
|
1203 |
+
if progress_callback:
|
1204 |
+
progress_callback("دریافت اسناد از دیتابیس...", 0.1)
|
1205 |
|
1206 |
+
documents = self.db_manager.get_documents_with_embeddings()
|
|
|
|
|
|
|
1207 |
|
1208 |
+
if not documents:
|
1209 |
+
return "❌ هیچ سندی برای ایندکسسازی یافت نشد"
|
1210 |
|
1211 |
+
success = self.search_engine.build_index(documents, progress_callback)
|
|
|
|
|
|
|
|
|
1212 |
|
1213 |
+
if success:
|
1214 |
+
return f"✅ ایندکس جستجو با {len(documents)} سند آماده شد"
|
1215 |
+
else:
|
1216 |
+
return "❌ خطا در ساخت ایندکس"
|
1217 |
+
|
1218 |
+
except Exception as e:
|
1219 |
+
return f"❌ خطا در ساخت ایندکس: {str(e)}"
|
1220 |
+
|
1221 |
+
def intelligent_semantic_search(self, query: str, limit: int = 10) -> Tuple[str, pd.DataFrame]:
|
1222 |
+
"""جستجوی معنایی هوشمند"""
|
1223 |
+
if not query or not query.strip():
|
1224 |
+
return "❌ لطفاً عبارت جستجو را وارد کنید", pd.DataFrame()
|
1225 |
|
1226 |
+
try:
|
1227 |
+
results = self.search_engine.search(query, top_k=limit)
|
1228 |
|
1229 |
+
if not results:
|
1230 |
+
return "❌ نتیجهای یافت نشد", pd.DataFrame()
|
1231 |
|
1232 |
+
# تولید خلاصه
|
1233 |
+
summary = f"""
|
1234 |
+
🔍 **نتایج جستجو برای**: "{query}"
|
1235 |
+
|
1236 |
+
📊 **آمار**:
|
1237 |
+
• تعداد نتایج: {len(results)}
|
1238 |
+
• بهترین امتیاز: {max(r['similarity_score'] for r in results):.3f}
|
1239 |
+
• میانگین امتیاز: {sum(r['similarity_score'] for r in results) / len(results):.3f}
|
1240 |
+
|
1241 |
+
🎯 **توزیع ارتباط**:
|
1242 |
+
• بسیار مرتبط: {sum(1 for r in results if r['relevance_category'] == 'بسیار مرتبط')}
|
1243 |
+
• مرتبط: {sum(1 for r in results if r['relevance_category'] == 'مرتبط')}
|
1244 |
+
• نسبتاً مرتبط: {sum(1 for r in results if r['relevance_category'] == 'نسبتاً مرتبط')}
|
1245 |
+
"""
|
1246 |
|
1247 |
+
# تولید DataFrame
|
1248 |
+
df_data = []
|
1249 |
+
for result in results:
|
1250 |
+
df_data.append({
|
1251 |
+
'رتبه': result['rank'],
|
1252 |
+
'عنوان': result['title'][:50] + '...' if len(result['title']) > 50 else result['title'],
|
1253 |
+
'امتیاز شباهت': f"{result['similarity_score']:.3f}",
|
1254 |
+
'میزان ارتباط': result['relevance_category'],
|
1255 |
+
'پیشنمایش محتوا': result['content_preview'][:100] + '...'
|
1256 |
+
})
|
1257 |
+
|
1258 |
+
df = pd.DataFrame(df_data)
|
1259 |
+
|
1260 |
+
return summary, df
|
1261 |
+
|
1262 |
+
except Exception as e:
|
1263 |
+
return f"❌ خطا در جستجو: {str(e)}", pd.DataFrame()
|
1264 |
+
|
1265 |
+
def export_advanced_data(self) -> Tuple[str, Optional[str]]:
|
1266 |
+
"""صدور دادههای پیشرفته"""
|
1267 |
+
try:
|
1268 |
+
# دریافت آمار
|
1269 |
+
stats = self.db_manager.get_advanced_stats()
|
1270 |
+
cache_stats = self.cache_system.get_stats()
|
1271 |
+
|
1272 |
+
# تولید گزارش جامع
|
1273 |
+
report = {
|
1274 |
+
'system_info': {
|
1275 |
+
'export_time': datetime.now().isoformat(),
|
1276 |
+
'total_documents': stats['total_documents'],
|
1277 |
+
'cache_performance': cache_stats
|
1278 |
+
},
|
1279 |
+
'database_statistics': stats,
|
1280 |
+
'performance_metrics': {
|
1281 |
+
'cache_hit_ratio': cache_stats['total_accesses'] / max(cache_stats['database_entries'], 1),
|
1282 |
+
'avg_processing_time': stats['average_processing_time']
|
1283 |
+
}
|
1284 |
+
}
|
1285 |
+
|
1286 |
+
# ذخیره در فایل موقت
|
1287 |
+
temp_file = tempfile.NamedTemporaryFile(mode='w', suffix='.json', delete=False, encoding='utf-8')
|
1288 |
+
json.dump(report, temp_file, ensure_ascii=False, indent=2)
|
1289 |
+
temp_file.close()
|
1290 |
+
|
1291 |
+
status = f"""
|
1292 |
+
📊 **گزارش صدور دادهها**
|
1293 |
+
|
1294 |
+
✅ **موفقیت آمیز**
|
1295 |
+
📄 **تعداد اسناد**: {stats['total_documents']}
|
1296 |
+
⚡ **کارایی کش**: {cache_stats['memory_cache_size']} items در حافظه
|
1297 |
+
📈 **میانگین زمان پردازش**: {stats['average_processing_time']} ثانیه
|
1298 |
+
📅 **زمان صدور**: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}
|
1299 |
+
"""
|
1300 |
+
|
1301 |
+
return status, temp_file.name
|
1302 |
+
|
1303 |
+
except Exception as e:
|
1304 |
+
return f"❌ خطا در صدور دادهها: {str(e)}", None
|
1305 |
+
|
1306 |
+
def get_comprehensive_system_status(self) -> str:
|
1307 |
+
"""وضعیت جامع سیستم"""
|
1308 |
+
try:
|
1309 |
+
# آمار پایگاه داده
|
1310 |
+
db_stats = self.db_manager.get_advanced_stats()
|
1311 |
+
|
1312 |
+
# آمار کش
|
1313 |
+
cache_stats = self.cache_system.get_stats()
|
1314 |
+
|
1315 |
+
# آمار مدلها
|
1316 |
+
models_ready = self.classification_system.is_ready
|
1317 |
+
search_ready = self.search_engine.is_ready
|
1318 |
+
|
1319 |
+
status_parts = [
|
1320 |
+
"## 🏠 وضعیت سیستم پیشرفته اسناد حقوقی",
|
1321 |
+
"",
|
1322 |
+
"### 📊 آمار کلی",
|
1323 |
+
f"• **تعداد کل اسناد**: {db_stats['total_documents']}",
|
1324 |
+
f"• **میانگین زمان پردازش**: {db_stats['average_processing_time']} ثانیه",
|
1325 |
+
f"• **آخرین بروزرسانی**: {db_stats['last_updated'][:19]}",
|
1326 |
+
"",
|
1327 |
+
"### ⚡ عملکرد کش",
|
1328 |
+
f"• **حافظه فعال**: {cache_stats['memory_cache_size']} آیتم",
|
1329 |
+
f"• **پایگاه داده کش**: {cache_stats['database_entries']} ورودی",
|
1330 |
+
f"• **تعداد دسترسیها**: {cache_stats['total_accesses']}",
|
1331 |
+
f"• **میانگین استفاده**: {cache_stats['average_accesses']}",
|
1332 |
+
"",
|
1333 |
+
"### 🧠 وضعیت مدلها",
|
1334 |
+
f"• **سیستم طبقهبندی**: {'🟢 آماده' if models_ready else '🔴 غیرفعال'}",
|
1335 |
+
f"• **موتور جستجو**: {'🟢 آماده' if search_ready else '🔴 غیرفعال'}",
|
1336 |
+
"",
|
1337 |
+
"### 📈 آمار منابع"
|
1338 |
+
]
|
1339 |
+
|
1340 |
+
# اضافه کردن آمار منابع
|
1341 |
+
for source, count, avg_words in db_stats.get('source_statistics', []):
|
1342 |
+
if source:
|
1343 |
+
status_parts.append(f"• **{source}**: {count} سند (میانگین {avg_words:.0f} کلمه)")
|
1344 |
+
|
1345 |
+
# وضعیت پردازش
|
1346 |
+
status_parts.extend([
|
1347 |
+
"",
|
1348 |
+
"### 🔧 وضعیت پردازش"
|
1349 |
+
])
|
1350 |
+
|
1351 |
+
for status, count in db_stats.get('status_statistics', []):
|
1352 |
+
status_parts.append(f"• **{status}**: {count} سند")
|
1353 |
+
|
1354 |
+
return "\n".join(status_parts)
|
1355 |
+
|
1356 |
+
except Exception as e:
|
1357 |
+
return f"❌ خطا در دریافت وضعیت سیستم: {str(e)}"
|
1358 |
+
|
1359 |
+
def create_advanced_interface(self):
|
1360 |
+
"""ایجاد رابط کاربری پیشرفته"""
|
1361 |
+
# تنظیمات CSS سفارشی برای بهبود ظاهر
|
1362 |
+
custom_css = """
|
1363 |
+
.rtl { direction: rtl; text-align: right; }
|
1364 |
+
.status-success { color: #10b981; font-weight: bold; }
|
1365 |
+
.status-error { color: #ef4444; font-weight: bold; }
|
1366 |
+
.metric-card {
|
1367 |
+
background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
|
1368 |
+
padding: 20px;
|
1369 |
+
border-radius: 10px;
|
1370 |
+
color: white;
|
1371 |
+
margin: 10px 0;
|
1372 |
+
}
|
1373 |
+
.gradio-container { font-family: 'Vazir', 'Tahoma', sans-serif; }
|
1374 |
+
"""
|
1375 |
|
1376 |
+
with gr.Blocks(
|
1377 |
+
theme=gr.themes.Soft(),
|
1378 |
+
css=custom_css,
|
1379 |
+
title="🏛️ سیستم پیشرفته اسناد حقوقی ایران"
|
1380 |
+
) as interface:
|
1381 |
|
1382 |
+
# هدر اصلی
|
1383 |
+
gr.HTML("""
|
1384 |
+
<div style="text-align: center; padding: 20px; background: linear-gradient(90deg, #1e40af, #7c3aed); border-radius: 15px; margin-bottom: 20px;">
|
1385 |
+
<h1 style="color: white; margin: 0; font-size: 2.5em;">🏛️ سیستم پیشرفته اسناد حقوقی ایران</h1>
|
1386 |
+
<p style="color: #e0e7ff; margin: 10px 0 0 0; font-size: 1.2em;">
|
1387 |
+
پلتفرم هوشمند تحلیل و دستهبندی متون حقوقی با قابلیتهای پیشرفته
|
1388 |
+
</p>
|
1389 |
+
</div>
|
1390 |
+
""")
|
1391 |
|
1392 |
+
# تبهای اصلی
|
1393 |
+
with gr.Tabs():
|
1394 |
+
|
1395 |
+
# تب خانه و داشبورد
|
1396 |
+
with gr.Tab("🏠 داشبورد اصلی"):
|
1397 |
+
with gr.Row():
|
1398 |
+
with gr.Column(scale=2):
|
1399 |
+
gr.Markdown("### 🚀 مقداردهی سیستم")
|
1400 |
+
|
1401 |
+
initialize_btn = gr.Button(
|
1402 |
+
"⚡ بارگذاری مدلهای هوش مصنوعی",
|
1403 |
+
variant="primary",
|
1404 |
+
size="lg"
|
1405 |
+
)
|
1406 |
+
|
1407 |
+
initialization_status = gr.Textbox(
|
1408 |
+
label="وضعیت مقداردهی",
|
1409 |
+
interactive=False,
|
1410 |
+
elem_classes=["rtl"]
|
1411 |
+
)
|
1412 |
+
|
1413 |
+
with gr.Column(scale=1):
|
1414 |
+
gr.Markdown("### 📊 آمار سریع")
|
1415 |
+
quick_stats = gr.HTML()
|
1416 |
+
|
1417 |
+
# تب پردازش هوشمند
|
1418 |
+
with gr.Tab("🤖 پردازش هوشمند اسناد"):
|
1419 |
+
gr.Markdown("### 🧠 پردازش و تحلیل هوشمند اسناد حقوقی")
|
1420 |
+
|
1421 |
+
with gr.Row():
|
1422 |
+
urls_input = gr.Textbox(
|
1423 |
+
label="📝 آدرس اسناد (هر خط یک URL)",
|
1424 |
+
placeholder="https://rc.majlis.ir/fa/law/show/123456\nhttps://www.dotic.ir/portal/law/789",
|
1425 |
+
lines=5,
|
1426 |
+
elem_classes=["rtl"]
|
1427 |
+
)
|
1428 |
+
|
1429 |
+
with gr.Row():
|
1430 |
+
process_intelligent_btn = gr.Button(
|
1431 |
+
"🚀 شروع پردازش هوشمند",
|
1432 |
+
variant="primary",
|
1433 |
+
size="lg"
|
1434 |
+
)
|
1435 |
+
clear_cache_btn = gr.Button("🗑️ پاک کردن کش", variant="secondary")
|
1436 |
+
|
1437 |
+
with gr.Row():
|
1438 |
+
intelligent_summary = gr.Textbox(
|
1439 |
+
label="📊 گزارش جامع پردازش",
|
1440 |
+
interactive=False,
|
1441 |
+
lines=8,
|
1442 |
+
elem_classes=["rtl"]
|
1443 |
+
)
|
1444 |
+
intelligent_details = gr.Textbox(
|
1445 |
+
label="📋 جزئیات و امتیازها",
|
1446 |
+
interactive=False,
|
1447 |
+
lines=8,
|
1448 |
+
elem_classes=["rtl"]
|
1449 |
+
)
|
1450 |
+
|
1451 |
+
# تب جستجوی هوشمند
|
1452 |
+
with gr.Tab("🔍 جستجوی معنایی پیشرفته"):
|
1453 |
+
gr.Markdown("### 🎯 جستجوی معنایی با الگوریتمهای پیشرفته")
|
1454 |
+
|
1455 |
+
with gr.Row():
|
1456 |
+
build_advanced_index_btn = gr.Button("🔧 ساخت ایندکس هوشمند", variant="secondary")
|
1457 |
+
advanced_index_status = gr.Textbox(
|
1458 |
+
label="📈 وضعیت ایندکس",
|
1459 |
+
interactive=False,
|
1460 |
+
elem_classes=["rtl"]
|
1461 |
+
)
|
1462 |
+
|
1463 |
+
with gr.Row():
|
1464 |
+
search_query = gr.Textbox(
|
1465 |
+
label="🔍 عبارت جستجو",
|
1466 |
+
placeholder="مسئولیت کیفری اشخاص حقوقی",
|
1467 |
+
elem_classes=["rtl"],
|
1468 |
+
scale=3
|
1469 |
+
)
|
1470 |
+
search_limit = gr.Slider(
|
1471 |
+
minimum=5, maximum=20, value=10, step=1,
|
1472 |
+
label="📊 تعداد نتایج", scale=1
|
1473 |
+
)
|
1474 |
+
|
1475 |
+
advanced_search_btn = gr.Button("🎯 جستجوی هوشمند", variant="primary")
|
1476 |
+
|
1477 |
+
advanced_search_results_text = gr.Markdown(elem_classes=["rtl"])
|
1478 |
+
advanced_search_results_df = gr.DataFrame(
|
1479 |
+
label="📋 نتایج تفصیلی",
|
1480 |
+
interactive=False
|
1481 |
+
)
|
1482 |
+
|
1483 |
+
# تب مدیریت سیستم
|
1484 |
+
with gr.Tab("📊 مدیریت سیستم پیشرفته"):
|
1485 |
+
with gr.Row():
|
1486 |
+
with gr.Column():
|
1487 |
+
gr.Markdown("### 🏥 وضعیت جامع سیستم")
|
1488 |
+
comprehensive_status = gr.Markdown(elem_classes=["rtl"])
|
1489 |
+
|
1490 |
+
with gr.Row():
|
1491 |
+
refresh_status_btn = gr.Button("🔄 بروزرسانی")
|
1492 |
+
optimize_system_btn = gr.Button("⚡ بهینهسازی سیستم")
|
1493 |
+
|
1494 |
+
with gr.Column():
|
1495 |
+
gr.Markdown("### 📤 صدور و پشتیبانگیری")
|
1496 |
+
|
1497 |
+
advanced_export_btn = gr.Button("📊 صدور دادههای پیشرفته", variant="primary")
|
1498 |
+
|
1499 |
+
advanced_export_status = gr.Textbox(
|
1500 |
+
label="📋 وضعیت صدور",
|
1501 |
+
interactive=False,
|
1502 |
+
elem_classes=["rtl"]
|
1503 |
+
)
|
1504 |
+
|
1505 |
+
advanced_export_file = gr.File(label="📁 فایل صادر شده")
|
1506 |
|
1507 |
+
# === اتصال Event Handlers ===
|
1508 |
+
|
1509 |
+
# مقداردهی سیستم
|
1510 |
+
initialize_btn.click(
|
1511 |
+
fn=self.initialize_models,
|
1512 |
+
outputs=[initialization_status],
|
1513 |
+
show_progress=True
|
1514 |
)
|
1515 |
|
1516 |
+
# پردازش هوشمند
|
1517 |
+
process_intelligent_btn.click(
|
1518 |
+
fn=self.process_urls_intelligent,
|
1519 |
+
inputs=[urls_input],
|
1520 |
+
outputs=[intelligent_summary, intelligent_details],
|
1521 |
+
show_progress=True
|
1522 |
)
|
|
|
|
|
|
|
|
|
1523 |
|
1524 |
+
# پاک کردن کش
|
1525 |
+
clear_cache_btn.click(
|
1526 |
+
fn=lambda: self.cache_system.cleanup_expired() or "🗑️ کش پاکسازی شد",
|
1527 |
+
outputs=[intelligent_summary]
|
1528 |
+
)
|
1529 |
|
1530 |
+
# ساخت ایندکس هوشمند
|
1531 |
+
build_advanced_index_btn.click(
|
1532 |
+
fn=self.build_intelligent_search_index,
|
1533 |
+
outputs=[advanced_index_status],
|
1534 |
+
show_progress=True
|
1535 |
+
)
|
1536 |
|
1537 |
+
# جستجوی هوشمند
|
1538 |
+
advanced_search_btn.click(
|
1539 |
+
fn=self.intelligent_semantic_search,
|
1540 |
+
inputs=[search_query, search_limit],
|
1541 |
+
outputs=[advanced_search_results_text, advanced_search_results_df]
|
1542 |
+
)
|
1543 |
|
1544 |
+
# مدیریت سیستم
|
1545 |
+
refresh_status_btn.click(
|
1546 |
+
fn=self.get_comprehensive_system_status,
|
1547 |
+
outputs=[comprehensive_status]
|
1548 |
+
)
|
1549 |
|
1550 |
+
optimize_system_btn.click(
|
1551 |
+
fn=lambda: (gc.collect(), self.cache_system.cleanup_expired(), "⚡ سیستم بهینهسازی شد")[2],
|
1552 |
+
outputs=[advanced_export_status]
|
1553 |
+
)
|
1554 |
|
1555 |
+
# صدور پیشرفته
|
1556 |
+
advanced_export_btn.click(
|
1557 |
+
fn=self.export_advanced_data,
|
1558 |
+
outputs=[advanced_export_status, advanced_export_file]
|
1559 |
+
)
|
1560 |
|
1561 |
+
# بارگذاری اولیه
|
1562 |
+
interface.load(
|
1563 |
+
fn=self.get_comprehensive_system_status,
|
1564 |
+
outputs=[comprehensive_status]
|
1565 |
+
)
|
1566 |
+
|
1567 |
+
return interface
|
1568 |
|
1569 |
+
# === تابع اصلی ===
|
1570 |
def main():
|
1571 |
+
"""تابع اصلی اجرای برنامه"""
|
1572 |
+
try:
|
1573 |
+
# ایجاد اپلیکیشن
|
1574 |
+
app = AdvancedLegalScrapingApp()
|
1575 |
+
|
1576 |
+
# ایجاد رابط کاربری
|
1577 |
+
interface = app.create_advanced_interface()
|
1578 |
+
|
1579 |
+
# اجرای برنامه
|
1580 |
+
interface.launch(
|
1581 |
+
server_name="0.0.0.0",
|
1582 |
+
server_port=7860,
|
1583 |
+
share=True,
|
1584 |
+
show_error=True,
|
1585 |
+
favicon_path=None,
|
1586 |
+
ssl_verify=False
|
1587 |
+
)
|
1588 |
+
|
1589 |
+
except Exception as e:
|
1590 |
+
logger.error(f"خطا در اجرای برنامه: {e}")
|
1591 |
+
print(f"❌ خطا در راهاندازی: {e}")
|
1592 |
|
1593 |
if __name__ == "__main__":
|
1594 |
main()
|