File size: 47,498 Bytes
21b2cd1
 
 
 
 
 
25ea748
21b2cd1
7c2dc3d
cba807f
7c2dc3d
 
 
cba807f
21b2cd1
 
c08b5c8
 
 
 
 
21b2cd1
 
 
7c2dc3d
21b2cd1
 
7c2dc3d
21b2cd1
 
7c2dc3d
21b2cd1
 
 
 
 
44d194a
c2c7991
 
 
 
 
 
 
 
 
 
44d194a
 
c2c7991
 
 
 
 
 
 
 
 
 
 
 
44d194a
 
 
21b2cd1
 
44d194a
 
 
 
 
21b2cd1
 
 
 
 
 
cba807f
 
 
 
 
7c2dc3d
cba807f
 
7c2dc3d
 
 
 
cba807f
 
 
 
7c2dc3d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21b2cd1
 
 
 
cba807f
21b2cd1
cba807f
7c2dc3d
cba807f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7c2dc3d
cba807f
7c2dc3d
 
 
 
 
cba807f
7c2dc3d
cba807f
 
 
 
 
 
 
21b2cd1
 
 
 
7c2dc3d
21b2cd1
 
 
 
 
7c2dc3d
 
 
21b2cd1
7c2dc3d
21b2cd1
 
 
 
 
7c2dc3d
 
 
21b2cd1
 
 
 
7c2dc3d
21b2cd1
7c2dc3d
 
ef3d758
 
 
 
7c2dc3d
 
 
 
 
 
 
 
 
 
ef3d758
7c2dc3d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ef3d758
 
 
 
 
7c2dc3d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ef3d758
 
 
 
 
 
 
 
 
7c2dc3d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ef3d758
 
 
7c2dc3d
ef3d758
7c2dc3d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ef3d758
7c2dc3d
 
 
 
cba807f
7c2dc3d
 
1e884ce
7c2dc3d
 
1e884ce
 
 
 
 
 
7c2dc3d
 
 
 
 
 
 
 
 
 
 
 
1e884ce
7c2dc3d
 
 
 
 
1e884ce
7c2dc3d
 
 
1e884ce
7c2dc3d
 
1e884ce
7c2dc3d
 
 
1e884ce
7c2dc3d
 
 
 
 
 
21b2cd1
 
cba807f
 
7c2dc3d
cba807f
0b69ffd
7c2dc3d
 
cba807f
 
 
7c2dc3d
cba807f
 
 
 
 
 
 
0b69ffd
 
cba807f
1e884ce
cba807f
0b69ffd
 
 
cba807f
 
 
7c2dc3d
cba807f
 
 
 
1e884ce
cba807f
 
 
7c2dc3d
cba807f
 
 
 
 
 
 
1e884ce
 
cba807f
 
 
 
 
 
 
 
 
1e884ce
cba807f
 
 
7c2dc3d
 
 
 
 
0b69ffd
1e884ce
 
7c2dc3d
 
 
 
 
 
 
 
0b69ffd
7c2dc3d
cba807f
7c2dc3d
 
 
1e884ce
7c2dc3d
 
 
 
 
0b69ffd
1e884ce
 
7c2dc3d
 
 
 
cba807f
0b69ffd
 
 
 
 
cba807f
 
0b69ffd
 
 
 
 
cba807f
 
 
0b69ffd
cba807f
 
 
 
 
7c2dc3d
cba807f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
44d194a
cba807f
44d194a
cba807f
 
 
 
0b69ffd
 
cba807f
 
 
7c2dc3d
 
0b69ffd
 
cba807f
 
0b69ffd
cba807f
 
 
 
 
 
0b69ffd
1e884ce
0b69ffd
cba807f
 
 
 
 
 
7c2dc3d
 
cba807f
 
 
0b69ffd
 
cba807f
 
21b2cd1
7c2dc3d
 
 
 
 
 
 
 
 
 
 
 
cba807f
21b2cd1
7c2dc3d
cba807f
 
 
21b2cd1
 
7c2dc3d
21b2cd1
 
 
7c2dc3d
21b2cd1
 
 
44d194a
21b2cd1
7c2dc3d
 
 
 
44d194a
21b2cd1
7c2dc3d
44d194a
7c2dc3d
 
 
 
 
 
44d194a
7c2dc3d
 
44d194a
7c2dc3d
 
 
 
 
 
44d194a
7c2dc3d
 
44d194a
7c2dc3d
 
 
 
44d194a
 
cba807f
21b2cd1
 
cba807f
 
 
 
 
21b2cd1
cba807f
 
21b2cd1
cba807f
21b2cd1
 
 
 
cba807f
 
 
 
21b2cd1
 
 
 
7c2dc3d
21b2cd1
 
 
 
 
 
 
 
0e0742c
 
 
 
 
 
 
 
 
 
 
 
5412d71
0e0742c
 
 
 
5412d71
0e0742c
 
 
 
5412d71
0e0742c
 
21b2cd1
5412d71
21b2cd1
 
 
5412d71
21b2cd1
 
 
5412d71
21b2cd1
 
 
5412d71
 
21b2cd1
 
 
 
 
5412d71
21b2cd1
 
ff81734
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21b2cd1
5412d71
21b2cd1
 
 
5412d71
 
21b2cd1
 
 
 
 
 
 
ff81734
 
 
 
21b2cd1
 
ff81734
 
 
 
21b2cd1
 
ff81734
 
 
 
21b2cd1
 
ff81734
 
 
 
 
 
 
 
 
 
21b2cd1
 
ff81734
 
21b2cd1
 
 
ff81734
 
21b2cd1
 
ff81734
 
21b2cd1
ff81734
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21b2cd1
ff81734
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21b2cd1
 
 
ff81734
21b2cd1
 
5412d71
21b2cd1
 
 
5412d71
21b2cd1
5412d71
21b2cd1
 
5412d71
21b2cd1
 
 
5412d71
21b2cd1
 
 
5412d71
21b2cd1
 
5412d71
 
21b2cd1
 
 
 
5412d71
21b2cd1
 
5412d71
 
21b2cd1
 
 
5412d71
21b2cd1
 
 
5412d71
21b2cd1
 
 
5412d71
21b2cd1
 
 
 
07b7648
 
 
 
 
ff81734
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21b2cd1
 
f58e822
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cba807f
7c2dc3d
cba807f
 
 
21b2cd1
7044d5e
 
 
 
 
 
 
 
 
 
 
 
 
7705dea
7044d5e
 
f58e822
cba807f
ac1ecca
cba807f
7705dea
cba807f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7c2dc3d
 
 
 
 
 
cba807f
7c2dc3d
 
 
 
 
cba807f
 
 
 
ae2b1d1
7305294
21b2cd1
cba807f
c08b5c8
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
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
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
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
408
409
410
411
412
413
414
415
416
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
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
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
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
import gradio as gr
from neo4j import GraphDatabase
import logging
from typing import List, Dict, Tuple
import pandas as pd
from datetime import datetime
import os

# Set up logging with more detailed format for debugging
logging.basicConfig(
    level=logging.INFO,
    format='%(asctime)s - %(levelname)s - %(message)s',
    datefmt='%Y-%m-%d %H:%M:%S'
)
logger = logging.getLogger(__name__)

# Get database credentials from environment variables
NEO4J_URL = os.getenv("NEO4J_URL")
NEO4J_USER = os.getenv("NEO4J_USER")
NEO4J_PASSWORD = os.getenv("NEO4J_PASSWORD")

def format_neo4j_datetime(dt) -> str:
    """Convert Neo4j datetime to string format."""
    if dt is None:
        logger.info("Received None datetime")
        return 'Unknown date'
    try:
        logger.info(f"Formatting datetime: {dt} of type {type(dt)}")
        if hasattr(dt, 'to_native'):
            dt = dt.to_native()
            logger.info(f"Converted to native: {dt} of type {type(dt)}")
        return dt.strftime('%Y-%m-%d')
    except Exception as e:
        logger.warning(f"Error formatting datetime: {e}")
        return 'Unknown date'

def is_displayable_keyword(keyword: str) -> bool:
    """
    Check if a keyword should be displayed (not just numbers and separators).
    Filters out:
    - Pure numbers (1234)
    - Numbers with dots (15.10)
    - Numbers with dashes (2023-01)
    - Numbers with spaces (15 10)
    - Numbers with slashes (15/10)
    - Any combination of above
    """
    if not keyword:
        return False
    
    # Remove all common number separators and spaces
    cleaned = keyword.replace('.', '') \
                    .replace('-', '') \
                    .replace('/', '') \
                    .replace('\\', '') \
                    .replace(' ', '') \
                    .replace(':', '') \
                    .replace(',', '')
    
    # Check if what remains is just digits
    return not cleaned.isdigit()

def format_interest_list_for_display(interests: set, max_items: int = 10) -> str:
    """Format a list of interests for display, hiding numeric-only keywords."""
    if not interests:
        return 'None'
    # Filter numeric keywords only for display
    displayable_interests = {interest for interest in interests if is_displayable_keyword(interest)}
    if not displayable_interests:
        return 'None'
    sorted_interests = sorted(displayable_interests)
    if len(sorted_interests) <= max_items:
        return ', '.join(sorted_interests)
    return f"{', '.join(sorted_interests[:max_items])} (+{len(sorted_interests) - max_items} more)"

class QuestionRecommender:
    def __init__(self):
        try:
            self.driver = GraphDatabase.driver(
                NEO4J_URL,
                auth=(NEO4J_USER, NEO4J_PASSWORD)
            )
            logger.info("Initializing QuestionRecommender with debug database")
            # Test connection immediately
            self.driver.verify_connectivity()
            logger.info("Successfully connected to Neo4j database")
            self.verify_connection()
            # Inspect question types on initialization
            self.inspect_question_types()
        except Exception as e:
            logger.error(f"Failed to initialize database connection: {str(e)}")
            raise

    def verify_connection(self):
        """Verify database connection and log basic statistics."""
        try:
            with self.driver.session() as session:
                # First try a simple query to verify connection
                test_result = session.run("MATCH (n) RETURN count(n) as count").single()
                if not test_result:
                    raise Exception("Could not execute test query")
                logger.info(f"Database contains {test_result['count']} total nodes")
                
                # Get database statistics with relationship counts
                stats = session.run("""
                    // Count nodes
                    MATCH (u:User)
                    WITH COUNT(u) as user_count
                    MATCH (k:Keyword)
                    WITH user_count, COUNT(k) as keyword_count
                    MATCH (q:Question)
                    WITH user_count, keyword_count, COUNT(q) as question_count
                    MATCH (t:Topic)
                    WITH user_count, keyword_count, question_count, COUNT(t) as topic_count
                    
                    // Count relationships
                    OPTIONAL MATCH ()-[r:INTERESTED_IN_KEYWORD]->()
                    WITH user_count, keyword_count, question_count, topic_count, COUNT(r) as keyword_rel_count
                    OPTIONAL MATCH ()-[r:INTERESTED_IN_TOPIC]->()
                    WITH user_count, keyword_count, question_count, topic_count, keyword_rel_count, COUNT(r) as topic_rel_count
                    OPTIONAL MATCH ()-[r:HAS_KEYWORD]->()
                    WITH user_count, keyword_count, question_count, topic_count, keyword_rel_count, topic_rel_count, COUNT(r) as question_keyword_count
                    OPTIONAL MATCH ()-[r:HAS_TOPIC]->()
                    RETURN 
                        user_count, keyword_count, question_count, topic_count,
                        keyword_rel_count, topic_rel_count,
                        question_keyword_count, COUNT(r) as question_topic_count
                """).single()

                if not stats:
                    raise Exception("Could not retrieve database statistics")
                
                logger.info("=== Database Statistics ===")
                logger.info(f"Nodes:")
                logger.info(f"  Users: {stats['user_count']}")
                logger.info(f"  Keywords: {stats['keyword_count']}")
                logger.info(f"  Questions: {stats['question_count']}")
                logger.info(f"  Topics: {stats['topic_count']}")
                logger.info(f"\nRelationships:")
                logger.info(f"  User->Keyword (INTERESTED_IN_KEYWORD): {stats['keyword_rel_count']}")
                logger.info(f"  User->Topic (INTERESTED_IN_TOPIC): {stats['topic_rel_count']}")
                logger.info(f"  Question->Keyword (HAS_KEYWORD): {stats['question_keyword_count']}")
                logger.info(f"  Question->Topic (HAS_TOPIC): {stats['question_topic_count']}")
                
        except Exception as e:
            logger.error(f"Database verification failed: {str(e)}")
            logger.error(f"URL: {NEO4J_URL}")
            logger.error(f"User: {NEO4J_USER}")
            raise Exception(f"Failed to verify database connection: {str(e)}")

    def inspect_question_types(self):
        """Inspect different types of questions and their attributes in the database."""
        with self.driver.session() as session:
            try:
                # Get all distinct question types and their properties
                result = session.run("""
                    MATCH (q:Question)
                    WITH DISTINCT keys(q) as props, labels(q) as types
                    RETURN types, props, count(*) as count
                    ORDER BY count DESC
                """)
                
                logger.info("\n=== Question Types and Properties ===")
                for record in result:
                    types = record["types"]
                    props = record["props"]
                    count = record["count"]
                    logger.info(f"\nType: {types}")
                    logger.info(f"Count: {count}")
                    logger.info("Properties:")
                    for prop in props:
                        # Get a sample value for this property
                        sample = session.run("""
                            MATCH (q:Question)
                            WHERE $prop in keys(q)
                            RETURN q[$prop] as value
                            LIMIT 1
                        """, prop=prop).single()
                        
                        value = sample["value"] if sample else None
                        value_type = type(value).__name__ if value is not None else "None"
                        logger.info(f"  - {prop}: {value_type} (example: {str(value)[:100]}{'...' if str(value)[100:] else ''})")
                
                # Get relationships specific to different question types
                result = session.run("""
                    MATCH (q:Question)-[r]->(target)
                    WITH DISTINCT type(r) as rel_type, labels(target) as target_labels, count(*) as count
                    RETURN rel_type, target_labels, count
                    ORDER BY count DESC
                """)
                
                logger.info("\n=== Question Relationships ===")
                for record in result:
                    rel_type = record["rel_type"]
                    target_labels = record["target_labels"]
                    count = record["count"]
                    logger.info(f"Relationship: {rel_type} -> {target_labels} (Count: {count})")
                
            except Exception as e:
                logger.error(f"Error inspecting question types: {str(e)}")
                raise

    def close(self):
        self.driver.close()

    def get_all_users(self) -> List[str]:
        """Get list of all users with interest counts."""
        with self.driver.session() as session:
            try:
                # Get users with their interest counts using proper relationship patterns
                result = session.run("""
                    MATCH (u:User)
                    OPTIONAL MATCH (u)-[r:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]->(interest)
                    WITH u, 
                         COUNT(DISTINCT CASE WHEN type(r) = 'INTERESTED_IN_KEYWORD' THEN interest END) as keyword_count,
                         COUNT(DISTINCT CASE WHEN type(r) = 'INTERESTED_IN_TOPIC' THEN interest END) as topic_count
                    WHERE keyword_count > 0 OR topic_count > 0
                    RETURN 
                        u.name as username,
                        keyword_count,
                        topic_count,
                        keyword_count + topic_count as total_interests
                    ORDER BY total_interests DESC, username
                """)
                
                users_with_counts = [(
                    record["username"],
                    record["keyword_count"],
                    record["topic_count"]
                ) for record in result if record["username"]]
                
                if not users_with_counts:
                    logger.warning("No users found with interests")
                    return []
                    
                logger.info(f"Retrieved {len(users_with_counts)} users with interests")
                logger.info("Top 5 users by interest count:")
                for username, kw_count, topic_count in users_with_counts[:5]:
                    logger.info(f"  - {username}: {kw_count} keywords, {topic_count} topics")
                
                # Format usernames with their counts
                return [
                    f"{username} ({kw_count} keywords, {topic_count} topics)"
                    for username, kw_count, topic_count in users_with_counts
                ]
            except Exception as e:
                logger.error(f"Error fetching users: {str(e)}")
                return []

    def get_user_interests(self, username: str) -> Dict[str, set]:
        """Get keywords and topics a user is interested in."""
        with self.driver.session() as session:
            # Get keywords the user is interested in
            keyword_result = session.run("""
                MATCH (u:User {name: $username})-[:INTERESTED_IN_KEYWORD]->(k:Keyword)
                RETURN DISTINCT k.keyword as keyword
            """, username=username)
            keywords = {str(record["keyword"]) for record in keyword_result if record["keyword"]}
            
            # Log keyword count for debugging
            logger.debug(f"Found {len(keywords)} keywords for user {username}")

            # Get topics the user is interested in
            topic_result = session.run("""
                MATCH (u:User {name: $username})-[:INTERESTED_IN_TOPIC]->(t:Topic)
                RETURN DISTINCT t.topic as topic
            """, username=username)
            topics = {str(record["topic"]) for record in topic_result if record["topic"]}
            
            # Log topic count for debugging
            logger.debug(f"Found {len(topics)} topics for user {username}")

            return {"keywords": keywords or set(), "topics": topics or set()}

    def find_common_questions(self, user1: str, user2: str, max_questions: int = 5) -> List[Dict]:
        """Find questions to recommend based on common interests using advanced Neo4j features."""
        with self.driver.session() as session:
            # Debug: Check if users exist and have interests
            user_check = session.run("""
                MATCH (u1:User {name: $user1})
                MATCH (u2:User {name: $user2})
                OPTIONAL MATCH (u1)-[r1:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]->(interest1)
                OPTIONAL MATCH (u2)-[r2:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]->(interest2)
                RETURN 
                    COUNT(DISTINCT u1) as user1_exists,
                    COUNT(DISTINCT u2) as user2_exists,
                    COUNT(DISTINCT interest1) as user1_interests,
                    COUNT(DISTINCT interest2) as user2_interests
            """, user1=user1, user2=user2).single()
            
            if not (user_check and user_check['user1_exists'] and user_check['user2_exists']):
                logger.error(f"One or both users not found: {user1}, {user2}")
                return []
                
            logger.info(f"User {user1} has {user_check['user1_interests']} total interests")
            logger.info(f"User {user2} has {user_check['user2_interests']} total interests")
            
            # Advanced question recommendation query using Neo4j path finding and scoring
            questions_query = """
            // Find all interests (both keywords and topics) for both users
            MATCH (u1:User {name: $user1})
            MATCH (u2:User {name: $user2})
            
            // Get all interests for both users
            OPTIONAL MATCH (u1)-[r1:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]->(interest1)
            OPTIONAL MATCH (u2)-[r2:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]->(interest2)
            WITH u1, u2, 
                 COLLECT(DISTINCT interest1) as u1_interests,
                 COLLECT(DISTINCT interest2) as u2_interests
            
            // Find questions related to either user's interests for each source
            CALL {
                WITH u1, u2, u1_interests, u2_interests
                UNWIND u1_interests + u2_interests as interest
                MATCH (q:Question)-[r:HAS_KEYWORD|HAS_TOPIC]->(interest)
                WHERE 
                    q.author <> $user1 AND 
                    q.author <> $user2 AND
                    q.source = 'stack_exchange' AND
                    (
                        (interest IN u1_interests AND interest IN u2_interests) OR
                        (interest IN u1_interests AND EXISTS((q)-[:HAS_KEYWORD|HAS_TOPIC]->()<-[:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]-(u2))) OR
                        (interest IN u2_interests AND EXISTS((q)-[:HAS_KEYWORD|HAS_TOPIC]->()<-[:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]-(u1)))
                    )
                WITH q, interest, type(r) as rel_type,
                     CASE WHEN interest IN u1_interests AND interest IN u2_interests THEN 2.0 ELSE 1.0 END as interest_weight
                WITH q, collect({interest: interest, weight: interest_weight, type: rel_type}) as interests,
                     sum(interest_weight) as base_score
                RETURN q, interests, base_score
                ORDER BY base_score * rand() DESC
                LIMIT 15 // Increased from 10 to get more variety
            
                UNION
                
                WITH u1, u2, u1_interests, u2_interests
                UNWIND u1_interests + u2_interests as interest
                MATCH (q:Question)-[r:HAS_KEYWORD|HAS_TOPIC]->(interest)
                WHERE 
                    q.source = 'trivia' AND
                    (
                        (interest IN u1_interests AND interest IN u2_interests) OR
                        (interest IN u1_interests AND EXISTS((q)-[:HAS_KEYWORD|HAS_TOPIC]->()<-[:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]-(u2))) OR
                        (interest IN u2_interests AND EXISTS((q)-[:HAS_KEYWORD|HAS_TOPIC]->()<-[:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]-(u1)))
                    )
                WITH q, interest, type(r) as rel_type,
                     CASE WHEN interest IN u1_interests AND interest IN u2_interests THEN 2.0 ELSE 1.0 END as interest_weight
                WITH q, collect({interest: interest, weight: interest_weight, type: rel_type}) as interests,
                     sum(interest_weight) as base_score
                RETURN q, interests, base_score
                ORDER BY base_score * rand() DESC
                LIMIT 15 // Increased from 10 to get more variety
                
                UNION
                
                WITH u1, u2, u1_interests, u2_interests
                UNWIND u1_interests + u2_interests as interest
                MATCH (q:Question)-[r:HAS_KEYWORD|HAS_TOPIC]->(interest)
                WHERE 
                    q.source = 'wikipedia' AND
                    (
                        (interest IN u1_interests AND interest IN u2_interests) OR
                        (interest IN u1_interests AND EXISTS((q)-[:HAS_KEYWORD|HAS_TOPIC]->()<-[:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]-(u2))) OR
                        (interest IN u2_interests AND EXISTS((q)-[:HAS_KEYWORD|HAS_TOPIC]->()<-[:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]-(u1)))
                    )
                WITH q, interest, type(r) as rel_type,
                     CASE WHEN interest IN u1_interests AND interest IN u2_interests THEN 2.0 ELSE 1.0 END as interest_weight
                WITH q, collect({interest: interest, weight: interest_weight, type: rel_type}) as interests,
                     sum(interest_weight) as base_score
                RETURN q, interests, base_score
                ORDER BY base_score * rand() DESC
                LIMIT 15 // Increased from 10 to get more variety
                
                UNION
                
                WITH u1, u2, u1_interests, u2_interests
                UNWIND u1_interests + u2_interests as interest
                MATCH (q:Question)-[r:HAS_KEYWORD|HAS_TOPIC]->(interest)
                WHERE 
                    q.source = 'reddit' AND
                    (
                        (interest IN u1_interests AND interest IN u2_interests) OR
                        (interest IN u1_interests AND EXISTS((q)-[:HAS_KEYWORD|HAS_TOPIC]->()<-[:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]-(u2))) OR
                        (interest IN u2_interests AND EXISTS((q)-[:HAS_KEYWORD|HAS_TOPIC]->()<-[:INTERESTED_IN_KEYWORD|INTERESTED_IN_TOPIC]-(u1)))
                    )
                WITH q, interest, type(r) as rel_type,
                     CASE WHEN interest IN u1_interests AND interest IN u2_interests THEN 2.0 ELSE 1.0 END as interest_weight
                WITH q, collect({interest: interest, weight: interest_weight, type: rel_type}) as interests,
                     sum(interest_weight) as base_score
                RETURN q, interests, base_score
                ORDER BY base_score * rand() DESC
                LIMIT 15 // Increased from 10 to get more variety
            }
            
            // Calculate temporal relevance for the combined results
            WITH q, interests, base_score,
                 CASE 
                     WHEN q.created_utc_ts IS NOT NULL 
                     THEN base_score * (1.0 + 0.1 * (1.0 - duration.between(q.created_utc_ts, datetime()).days / 365.0))
                     ELSE base_score 
                 END as temporal_score,
                 // Add source-specific random boost to ensure better mixing
                 CASE q.source
                     WHEN 'stack_exchange' THEN rand() * 0.4
                     WHEN 'trivia' THEN rand() * 0.4
                     WHEN 'wikipedia' THEN rand() * 0.4
                     WHEN 'reddit' THEN rand() * 0.4
                     ELSE rand() * 0.4
                 END as source_random_boost
            
            // Return results with all metadata
            WITH q, interests, temporal_score, source_random_boost,
                 temporal_score * (0.6 + 0.8 * rand()) + source_random_boost as final_score
                RETURN DISTINCT 
                    q.title as title,
                    q.body as body,
                q.created_utc_ts as created_utc_ts,
                    q.author as author,
                q.source as source,
                q.correct_answer as correct_answer,
                q.incorrect_answers as incorrect_answers,
                q.upvotes as upvotes,
                q.num_comments as num_comments,
                q.subreddit as subreddit,
                [i in interests | CASE 
                    WHEN i.type = 'HAS_KEYWORD' THEN i.interest.keyword 
                    ELSE i.interest.topic 
                END] as matching_interests,
                [i in interests | CASE 
                    WHEN i.type = 'HAS_KEYWORD' THEN 'keyword'
                    ELSE 'topic'
                END] as interest_types,
                final_score as relevance_score
            ORDER BY final_score DESC
            LIMIT $max_questions
            """
            
            questions = [dict(record) for record in session.run(questions_query, 
                                                              user1=user1, 
                                                              user2=user2, 
                                                              max_questions=max_questions)]
            
            if questions:
                first_q = questions[0]
                logger.info(f"Sample question:")
                logger.info(f"Title: {first_q.get('title', 'No title')}")
                logger.info(f"Author: {first_q.get('author', 'No author')}")
                logger.info(f"Score: {first_q.get('relevance_score', 0)}")
                logger.info(f"Interests: {first_q.get('matching_interests', [])}")
            
            logger.info(f"Found {len(questions)} questions with common interests")
            return questions

def process_body(text, title):
    """Process question body to handle images and HTML."""
    if not text:
        logger.warning(f"Empty body for question: {title}")
        return ""
        
    try:
        from bs4 import BeautifulSoup
        
        # Parse the HTML content
        soup = BeautifulSoup(str(text), 'html.parser')
        
        # Function to fix Stack Exchange URLs
        def fix_stack_exchange_url(url):
            if not url:
                return url
            if url.startswith(('http://', 'https://')):
                return url
            if url.startswith('//'):
                return 'https:' + url
            if url.startswith('/'):
                return 'https://i.stack.imgur.com' + url
            return 'https://i.stack.imgur.com/' + url
        
        # Find all img tags and replace with preview cards
        for img in soup.find_all('img'):
            src = img.get('src', '')
            if not src:
                continue
                
            fixed_src = fix_stack_exchange_url(src)
            alt_text = img.get('alt', '').strip()
            if not alt_text or alt_text.lower() == 'enter image description here':
                alt_text = 'Question image'
            
            # Create an image preview card
            preview_html = f"""
    <div class="image-preview" style="margin: 10px 0; padding: 10px; background: #f9fafc; border-radius: 6px;">
        <div style="display: flex; align-items: center; margin-bottom: 8px;">
            <span style="font-size: 20px; margin-right: 8px;">πŸ–ΌοΈ</span>
            <span style="color: #219ebc;">{alt_text}</span>
        </div>
        <a href="{fixed_src}" target="_blank" rel="noopener noreferrer" 
            style="color: #219ebc; text-decoration: none;">View image</a>
    </div>
    """
            
            new_soup = BeautifulSoup(preview_html, 'html.parser')
            img.replace_with(new_soup)
        
        # Style other elements
        for link in soup.find_all('a'):
            if 'View Image' not in (link.get_text() or ''):
                href = link.get('href', '')
                if href and not href.startswith(('http://', 'https://')):
                    link['href'] = fix_stack_exchange_url(href)
                link['target'] = '_blank'
                link['rel'] = 'noopener noreferrer'
                link['style'] = 'color: #219ebc; text-decoration: none;'
        
        # Add paragraph styling
        for p in soup.find_all(['p', 'div']):
            if not any(cls in (p.get('class', []) or []) for cls in ['image-preview', 'question-card']):
                current_style = p.get('style', '')
                p['style'] = f"{current_style}; margin: 0.8em 0; line-height: 1.6; color: #333333;"
        
        # Add list styling
        for ul in soup.find_all(['ul', 'ol']):
            ul['style'] = 'margin: 0.8em 0; padding-left: 1.5em; color: #333333;'
        
        for li in soup.find_all('li'):
            li['style'] = 'margin: 0.4em 0; line-height: 1.6; color: #333333;'
        
        # Add code block styling
        for code in soup.find_all(['code', 'pre']):
            code['style'] = 'background: #f9fafc; padding: 0.2em 0.4em; border-radius: 4px; font-family: monospace; color: #333333;'
        
        return str(soup)
        
    except Exception as e:
        logger.error(f"Error processing question body: {str(e)}")
        return str(text) if text else ""

def format_question(q: Dict) -> str:
    """Format a question for display based on its source."""
    try:
        # Extract and validate basic question data
        title = q.get('title', 'Untitled')
        source = q.get('source', '').lower()
        
        # Format metadata section based on source
        metadata_html = ""
        content_html = ""
        
        # Default metadata for questions with author/date
        if 'author' in q or 'created_utc_ts' in q:
            author = q.get('author', 'Unknown author')
            created_date = format_neo4j_datetime(q.get('created_utc_ts'))
            upvotes = q.get('upvotes', 0)
            num_comments = q.get('num_comments', 0)
            
            metadata_html = f"""
            <div class="question-meta" style="font-size: 0.9em; margin-bottom: 15px;">
                <span style="color: #219ebc; font-weight: 500;">{author}</span>
                {' asked' if source == 'stack_exchange' else ' posted'} on 
                <span style="color: #023047;">{created_date}</span>
                <div class="stats" style="margin-top: 5px;">
                    <span title="Upvotes"><span style="color: #219ebc;">β–²</span> {upvotes}</span>
                    <span style="margin-left: 15px;" title="Comments"><span style="color: #219ebc;">πŸ’¬</span> {num_comments}</span>
                </div>
            </div>
            """
        
        # Handle content based on source and available fields
        if source == "stack_exchange":
            body = q.get('body', '')
            if body:
                content_html = f"""
                <div class="question-content" style="margin-top: 20px; font-family: 'Segoe UI', system-ui, -apple-system, sans-serif; color: #023047; line-height: 1.6;">
                    {process_body(body, title)}
                </div>
                """
                
        elif source == "trivia":
            correct_answer = q.get('correct_answer', '')
            incorrect_answers = q.get('incorrect_answers', [])
            
            answers = [correct_answer] + incorrect_answers if incorrect_answers else [correct_answer]
            answers_html = "".join([
                f"""
                <div class="answer-option" style="margin: 8px 0; padding: 10px; background: #f9fafc; border-radius: 6px; border: 1px solid #8ecae6; border-left: 3px solid {'#4caf50' if answer == correct_answer else '#8ecae6'};">
                    <span style="color: {'#4caf50' if answer == correct_answer else '#023047'}; font-weight: {'500' if answer == correct_answer else 'normal'};">
                        {answer}
                    </span>
                </div>
                """
                for answer in answers
            ])
            
            content_html = f"""
            <div class="answers-container" style="margin-top: 15px;">
                <div style="color: #023047; margin-bottom: 10px; font-weight: 500;">Answer options:</div>
                {answers_html}
            </div>
            """
            
        elif source == "wikipedia":
            correct_answer = q.get('correct_answer', '')
            if correct_answer:
                content_html = f"""
                <div class="answer" style="margin-top: 15px; padding: 15px; background: #f9fafc; border-radius: 6px; border: 1px solid #8ecae6;">
                    <div style="color: #023047; margin-bottom: 10px; font-weight: 500;">Answer:</div>
                    <div style="color: #4caf50; font-weight: 500;">{correct_answer}</div>
                </div>
                """
            
        elif source == "reddit":
            if 'subreddit' in q:
                subreddit = q.get('subreddit', '')
                metadata_html = metadata_html.replace(
                    'posted on',
                    f'posted in <span style="color: #219ebc; font-weight: 500;">r/{subreddit}</span> on'
                )
        
        if not content_html:
            if 'body' in q:
                content_html = f"""
                <div class="question-content" style="margin-top: 20px; font-family: 'Segoe UI', system-ui, -apple-system, sans-serif; color: #023047; line-height: 1.6;">
                    {process_body(q['body'], title)}
                </div>
                """
            elif 'correct_answer' in q:
                content_html = f"""
                <div class="answer" style="margin-top: 15px; padding: 15px; background: #f9fafc; border-radius: 6px; border: 1px solid #8ecae6;">
                    <div style="color: #023047; margin-bottom: 10px; font-weight: 500;">Answer:</div>
                    <div style="color: #4caf50; font-weight: 500;">{q['correct_answer']}</div>
                </div>
                """
        
        # Get source-specific icon and color
        source_icon = {
            'stack_exchange': '⚑',
            'reddit': 'πŸ”Έ',
            'wikipedia': 'πŸ“š',
            'trivia': '🎯',
        }.get(source, '❔')
        
        source_color = {
            'stack_exchange': '#219ebc',
            'reddit': '#fb8500',
            'wikipedia': '#4caf50',
            'trivia': '#ffb703',
        }.get(source, '#219ebc')
        
        source_display = source.title() if source else "Unknown"
        source_badge = f"""
        <div class="source-badge" style="display: inline-flex; align-items: center; padding: 4px 8px; background: #f9fafc; border-radius: 4px; margin-right: 10px; border: 1px solid {source_color};">
            <span style="margin-right: 6px; font-size: 1.1em;">{source_icon}</span>
            <span style="color: {source_color}; font-size: 0.9em; font-weight: 500;">{source_display}</span>
        </div>
        """
        
        # Handle matching interests display
        matching_interests = q.get('matching_interests', [])
        interest_types = q.get('interest_types', [])
        interests_with_types = []
        for interest, type_ in zip(matching_interests, interest_types):
            if interest and type_:
                interests_with_types.append({
                    'name': interest,
                    'type': type_
                })
        
        keywords = [i['name'] for i in interests_with_types if i['type'] == 'keyword']
        topics = [i['name'] for i in interests_with_types if i['type'] == 'topic']
        
        interests_display = []
        if keywords:
            interests_display.append(f"Keywords: {format_interest_list_for_display(set(keywords), max_items=3)}")
        if topics:
            interests_display.append(f"Topics: {format_interest_list_for_display(set(topics), max_items=3)}")
        interests_str = " | ".join(interests_display) if interests_display else "No common interests found"
        
        relevance_score = q.get('relevance_score', 0)
        score_display = f"""
        <div class="relevance-score" style="display: inline-block; padding: 4px 8px; background: #f9fafc; border-radius: 4px; margin-left: 10px; border: 1px solid #8ecae6;">
            <span style="color: #219ebc; font-size: 0.9em;">Relevance: {relevance_score:.2f}</span>
        </div>
        """ if relevance_score > 0 else ""
        
        # Create the question card HTML
        question_html = f"""
        <div class="question-card" style="background: #ffffff; padding: 20px; border-radius: 8px; margin: 15px 0; border: 1px solid #219ebc; box-shadow: 0 4px 6px rgba(0, 0, 0, 0.1);">
            <div class="question-header" style="display: flex; justify-content: space-between; align-items: flex-start; margin-bottom: 15px; border-bottom: 1px solid #8ecae6; padding-bottom: 10px;">
                <div style="flex: 1; display: flex; align-items: center;">
                    {source_badge}
                    <h3 style="color: #023047; margin: 0; font-size: 1.4em; display: inline;">{title}</h3>
                </div>
                {score_display}
            </div>
            
            {metadata_html}
            
            <div class="interests-bar" style="margin: 15px 0; padding: 10px; background: #f9fafc; border-radius: 6px; border: 1px solid #8ecae6; border-left: 3px solid #219ebc;">
                <div style="color: #023047; font-size: 0.9em; font-weight: 500;">Common Interests:</div>
                <div style="color: #219ebc; font-weight: 500; margin-top: 5px;">{interests_str}</div>
            </div>
            
            {content_html}
        </div>
        """
        
        return question_html
        
    except Exception as e:
        logger.error(f"Error formatting question: {str(e)}")
        return f"""
        <div style="background: #fee2e2; padding: 15px; border-radius: 8px; margin: 10px 0; border: 1px solid #dc2626;">
            <div style="color: #dc2626;">Error displaying question: {str(e)}</div>
        </div>
        """

def loading_message() -> Tuple[str, str, str]:
    """Return loading message in proper HTML format."""
    loading_html = """
    <div class="loading-spinner">
        <div style="text-align: center;">
            <div style="border: 4px solid #60a5fa; border-top: 4px solid transparent; border-radius: 50%; width: 40px; height: 40px; animation: spin 1s linear infinite; margin: 20px auto;"></div>
            <div style="color: #60a5fa; margin-top: 10px;">Analyzing interests and finding recommendations...</div>
        </div>
    </div>
    """
    return loading_html, loading_html, loading_html

def recommend_questions(user1: str, user2: str) -> Tuple[str, str, str, List[Dict]]:
    """Main function to get recommendations and user interests."""
    # Extract actual usernames from the formatted strings
    user1 = user1.split(" (")[0] if " (" in user1 else user1
    user2 = user2.split(" (")[0] if " (" in user2 else user2
    
    recommender = QuestionRecommender()
    try:
        # Get interests for both users
        user1_interests = recommender.get_user_interests(user1)
        user2_interests = recommender.get_user_interests(user2)
        
        # Find common interests
        common_keywords = user1_interests['keywords'] & user2_interests['keywords']
        common_topics = user1_interests['topics'] & user2_interests['topics']
        
        # Format interests summary using display formatter
        interests_summary = f"""
<div class="interests-summary">
    <div class="user-interests">
        <h3>{user1}'s Interests</h3>
        <div class="interest-section">
            <strong>Keywords:</strong> {format_interest_list_for_display(user1_interests['keywords'], max_items=8)}
        </div>
        <div class="interest-section">
            <strong>Topics:</strong> {format_interest_list_for_display(user1_interests['topics'], max_items=5)}
        </div>
    </div>
    
    <div class="user-interests">
        <h3>{user2}'s Interests</h3>
        <div class="interest-section">
            <strong>Keywords:</strong> {format_interest_list_for_display(user2_interests['keywords'], max_items=8)}
        </div>
        <div class="interest-section">
            <strong>Topics:</strong> {format_interest_list_for_display(user2_interests['topics'], max_items=5)}
        </div>
    </div>
    
    <div class="common-interests">
        <h3>Common Interests</h3>
        <div class="interest-section">
            <strong>Keywords:</strong> {format_interest_list_for_display(common_keywords, max_items=8)}
        </div>
        <div class="interest-section">
            <strong>Topics:</strong> {format_interest_list_for_display(common_topics, max_items=5)}
        </div>
    </div>
</div>
"""

        # Get all recommended questions - using original interests for matching
        questions = recommender.find_common_questions(user1, user2, max_questions=50)
        
        if questions:
            questions_text = '<div class="questions-container">\n' + \
                           '\n'.join(format_question(q) for q in questions) + \
                           '\n</div>'
            
            recommendation_type = '<h2 class="recommendation-header">Recommendations Based on Common Interests</h2>'
        else:
            questions_text = '<div class="no-questions">No questions found based on common interests.</div>'
            recommendation_type = '<h2 class="recommendation-header">No Recommendations Available</h2>'
        
        return interests_summary, recommendation_type, questions_text, questions
    
    except Exception as e:
        logger.error(f"Error in recommend_questions: {str(e)}")
        return (
            '<div class="error">Error fetching user interests. Please try again.</div>',
            '<h2 class="error-header">Error</h2>',
            f'<div class="error-message">An error occurred: {str(e)}</div>',
            []
        )
    finally:
        recommender.close()

# Custom CSS for better styling
custom_css = """
.gradio-container {
    max-width: 1200px !important;
    margin: auto !important;
    padding: 20px !important;
    font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif !important;
}

/* Dropdown styling */
.gradio-dropdown > ul {
    max-height: 300px !important;
    overflow-y: auto !important;
    scrollbar-width: thin !important;
}

.gradio-dropdown > ul::-webkit-scrollbar {
    width: 6px !important;
}

.gradio-dropdown > ul::-webkit-scrollbar-track {
    background: #f9fafc !important;
    border-radius: 3px !important;
}

.gradio-dropdown > ul::-webkit-scrollbar-thumb {
    background: #219ebc !important;
    border-radius: 3px !important;
}

.gradio-dropdown > ul::-webkit-scrollbar-thumb:hover {
    background: #023047 !important;
}

.interests-summary {
    background: #ffffff;
    padding: 20px;
    border-radius: 10px;
    margin-bottom: 20px;
    border: 1px solid #219ebc;
}

.user-interests, .common-interests {
    background: #f9fafc;
    padding: 15px;
    border-radius: 8px;
    margin: 10px 0;
    box-shadow: 0 4px 6px rgba(0, 0, 0, 0.1);
    border: 1px solid #8ecae6;
}

.interest-section {
    margin: 10px 0;
    line-height: 1.5;
    color: #333333;
}

/* Progress bar and loading animation */
.progress-bar > div {
    background: #ffb703 !important;
}

.progress-bar {
    background: #f9fafc !important;
}

.loading {
    color: #219ebc !important;
}

/* Question section styling */
.questions-container {
    margin-top: 20px;
}

.question-card {
    background: #ffffff;
    padding: 20px;
    border-radius: 8px;
    margin: 15px 0;
    box-shadow: 0 4px 6px rgba(0, 0, 0, 0.1);
    border: 1px solid #219ebc;
    transition: transform 0.2s ease;
}

.question-card:hover {
    transform: translateY(-2px);
}

.question-header {
    border-bottom: 1px solid #8ecae6;
    padding-bottom: 10px;
    margin-bottom: 15px;
}

.question-header h3 {
    color: #023047 !important;
    margin: 0;
    font-size: 1.4em;
}

.source-badge {
    background: #f9fafc !important;
    border: 1px solid #219ebc !important;
    color: #219ebc !important;
}

.interests-bar {
    background: #f9fafc !important;
    border: 1px solid #8ecae6 !important;
    border-left: 3px solid #219ebc !important;
    padding: 10px;
    margin: 15px 0;
}

.interests-bar div {
    color: #333333 !important;
}

.question-meta {
    color: #333333 !important;
    margin: 10px 0;
}

.question-meta span {
    color: #219ebc !important;
}

.question-content {
    color: #333333 !important;
    line-height: 1.6;
}

.question-content a {
    color: #219ebc !important;
    text-decoration: none;
}

.question-content code, .question-content pre {
    background: #f9fafc !important;
    border: 1px solid #8ecae6 !important;
    color: #333333 !important;
    padding: 0.2em 0.4em;
    border-radius: 4px;
}

.answer-option {
    background: #f9fafc !important;
    border: 1px solid #8ecae6 !important;
    margin: 8px 0;
    padding: 10px;
    border-radius: 6px;
}

.answer {
    background: #f9fafc !important;
    border: 1px solid #8ecae6 !important;
    padding: 15px;
    border-radius: 6px;
    margin-top: 15px;
}

.image-preview {
    background: #f9fafc !important;
    border: 1px solid #8ecae6 !important;
    padding: 10px;
    border-radius: 6px;
    margin: 10px 0;
}

.relevance-score {
    background: #f9fafc !important;
    border: 1px solid #8ecae6 !important;
    padding: 4px 8px;
    border-radius: 4px;
}

.relevance-score span {
    color: #219ebc !important;
}

.recommendation-header {
    color: #023047 !important;
    padding: 10px 0;
    margin: 20px 0;
    border-bottom: 2px solid #219ebc;
}

.error {
    color: #dc2626;
    padding: 15px;
    background: #fee2e2;
    border-radius: 8px;
    margin: 10px 0;
    border: 1px solid #dc2626;
}

.error-header {
    color: #dc2626;
}

.error-message {
    background: #fee2e2;
    padding: 15px;
    border-radius: 8px;
    color: #dc2626;
    border: 1px solid #dc2626;
}

.no-questions {
    padding: 20px;
    background: #f9fafc;
    border-radius: 8px;
    text-align: center;
    color: #333333;
    border: 1px solid #219ebc;
}

h1, h2, h3 {
    color: #023047 !important;
}

strong {
    color: #219ebc;
}

.user-interests h3, .common-interests h3 {
    color: #023047;
    margin-top: 0;
    margin-bottom: 15px;
    font-size: 1.2rem;
}

@keyframes spin {
    0% { transform: rotate(0deg); }
    100% { transform: rotate(360deg); }
}

/* Additional Gradio-specific overrides */
.gr-button-primary {
    background: #ffb703 !important;
    color: #ffffff !important;
}

.gr-button-secondary {
    background: #219ebc !important;
    color: #ffffff !important;
}

.gr-form {
    background: #ffffff !important;
    border: 1px solid #219ebc !important;
}

.gr-input {
    border-color: #219ebc !important;
}

.gr-input:focus {
    border-color: #023047 !important;
}

.gr-box {
    background: #ffffff !important;
    border: 1px solid #219ebc !important;
}

.gr-panel {
    background: #ffffff !important;
    border: 1px solid #219ebc !important;
}
"""

force_light_mode_js = """
function() {
    // Remove dark mode class from html element
    document.documentElement.classList.remove('dark');
    
    // Set data-theme attribute to light
    document.documentElement.setAttribute('data-theme', 'light');
    
    // Override any system preference
    document.documentElement.style.colorScheme = 'light';
    
    // Also try to set it on the body
    document.body.classList.remove('dark');
    document.body.setAttribute('data-theme', 'light');
}
"""

def main():
    # Create Gradio interface
    recommender = QuestionRecommender()
    users = recommender.get_all_users()
    recommender.close()

    theme = gr.themes.Base().set(
        body_background_fill="#f9fafc",
        block_background_fill="#ffffff",
        block_border_width="1px",
        block_border_color="#219ebc",
        block_title_text_color="#023047",
        block_label_text_color="#333333",
        input_background_fill="#ffffff",
        input_border_color="#219ebc",
        input_border_width="1px",
        button_primary_background_fill="#ffb703",
        button_secondary_background_fill="#219ebc",
        background_fill_primary="#ffffff",
        background_fill_secondary="#f9fafc"
    )

    with gr.Blocks(title="Let's Talk - Question Recommender", theme=theme, css=custom_css, mode="light", js=force_light_mode_js) as iface:
        gr.Markdown("""
        # 🀝 Let's Talk - Question Recommender
        Find questions that two users might be interested in discussing together based on their common interests.
                """)
        
        with gr.Row(equal_height=True):
            with gr.Column(scale=1):
                user1_dropdown = gr.Dropdown(
                    choices=users, 
                    label="πŸ‘€ First User",
                    interactive=True,
                    elem_id="user1-input"
                )
            with gr.Column(scale=1):
                user2_dropdown = gr.Dropdown(
                    choices=users,
                    label="πŸ‘€ Second User",
                    interactive=True,
                    elem_id="user2-input"
                )
        
        recommend_btn = gr.Button(
            "πŸ” Get Recommendations",
            variant="primary",
            size="lg"
        )
        
        with gr.Row():
            interests_output = gr.HTML(label="Common Interests")
        
        recommendation_type = gr.HTML()
        questions_output = gr.HTML()
        
        def recommend_and_store(user1, user2):
            """Get recommendations and store questions."""
            interests, rec_type, questions_html, questions_data = recommend_questions(user1, user2)
            return interests, rec_type, questions_html
        
        # Wire up the components
        recommend_btn.click(
            fn=loading_message,
            outputs=[interests_output, recommendation_type, questions_output],
            queue=False
        ).then(
            fn=recommend_and_store,
            inputs=[user1_dropdown, user2_dropdown],
            outputs=[interests_output, recommendation_type, questions_output]
        )

    iface.launch(allowed_paths=["*"], 
            show_error=True)

if __name__ == "__main__":
    main()