File size: 75,716 Bytes
1efd29f 6245b3b b138e3b 274a509 f367387 ae53812 274a509 7209842 adcb6a8 f3ed537 4f3470b 885c1f9 274a509 f367387 3c63f39 f3ed537 6b1539d 123c678 6b1539d f3ed537 123c678 f3ed537 274a509 5b7f342 123c678 39207e4 1dbc8cb 39207e4 3c63f39 274a509 f367387 274a509 e297e4a f367387 e297e4a 274a509 f367387 3c63f39 d27a85c ecb092b 274a509 c9bd851 274a509 ecb092b c9bd851 ecb092b 274a509 e297e4a 274a509 f367387 c9bd851 61c15b7 c9bd851 61c15b7 c9bd851 3330689 274a509 3330689 274a509 5b7f342 274a509 adcb6a8 f3ed537 274a509 3330689 274a509 f3ed537 274a509 f3ed537 274a509 f3ed537 274a509 3330689 f3ed537 274a509 f367387 274a509 f3ed537 274a509 f3ed537 274a509 72a1c3a e297e4a 274a509 61c15b7 274a509 1a943f1 274a509 c9bd851 f3ed537 34139eb 1ac7e9d 34139eb f11fee0 1ac7e9d f11fee0 34139eb 274a509 885c1f9 f367387 274a509 07b6a95 3330689 07b6a95 ecb092b 61c15b7 ecb092b 8aa9109 ecb092b 8aa9109 ecb092b 8aa9109 ecb092b 8aa9109 61c15b7 ecb092b 8aa9109 ecb092b 8aa9109 ecb092b 8aa9109 ecb092b 8aa9109 ecb092b 8aa9109 ecb092b 8aa9109 5ccdfd6 61c15b7 db1867d 61c15b7 db1867d e297e4a 61c15b7 ae53812 5ccdfd6 6b1539d 61c15b7 6b1539d 61c15b7 6b1539d 61c15b7 6b1539d 61c15b7 6b1539d 07b6a95 f367387 274a509 3330689 fd7c5f8 77de677 274a509 07b6a95 c9bd851 9a9c028 c9bd851 9a9c028 c9bd851 9a9c028 c9bd851 274a509 9a9c028 274a509 2689b83 123c678 b3607a6 3c63f39 07b6a95 274a509 2689b83 123c678 3c63f39 3330689 123c678 3330689 123c678 3330689 123c678 07b6a95 ae53812 123c678 274a509 07b6a95 39207e4 123c678 39207e4 8aa9109 39207e4 83f9bef 39207e4 9030be0 39207e4 f8a72ae 9030be0 f8a72ae 9030be0 e297e4a 2689b83 ecb092b 3c63f39 07b6a95 39207e4 274a509 123c678 274a509 123c678 274a509 123c678 3330689 34139eb 6a5d12d 34139eb 3c63f39 274a509 123c678 34139eb 123c678 34139eb aee35a5 34139eb 123c678 34139eb 123c678 3330689 123c678 274a509 9030be0 251a9b3 9030be0 5ccdfd6 9030be0 251a9b3 9030be0 5ccdfd6 9030be0 5ccdfd6 9030be0 5ccdfd6 9030be0 4f3470b 9030be0 4f3470b 9030be0 4f3470b 9030be0 5ccdfd6 9030be0 123c678 9030be0 b3607a6 9030be0 b3607a6 9030be0 b3607a6 9030be0 b3607a6 9030be0 b3607a6 9030be0 b3607a6 9030be0 b3607a6 9030be0 b3607a6 123c678 5ccdfd6 9030be0 ae53812 9030be0 ae53812 e297e4a ae53812 e297e4a ae53812 e297e4a ae53812 39207e4 9d332ff c9bd851 39207e4 9030be0 e297e4a ae53812 e297e4a ae53812 39207e4 ae53812 e297e4a 9030be0 ae53812 9030be0 ae53812 123c678 db1867d 123c678 db1867d 123c678 db1867d 274a509 f367387 6a5d12d 123c678 9030be0 123c678 9030be0 785101b 123c678 b3607a6 9030be0 b3607a6 9030be0 b3607a6 123c678 f3ed537 123c678 ae53812 9030be0 ae53812 f367387 123c678 274a509 123c678 9030be0 123c678 9030be0 274a509 123c678 f367387 274a509 123c678 3330689 123c678 3330689 123c678 9030be0 123c678 9030be0 d27a85c 274a509 b3607a6 9030be0 b3607a6 fd7c5f8 9030be0 5ccdfd6 9030be0 5ccdfd6 9030be0 5ccdfd6 39207e4 9030be0 39207e4 9030be0 4f3470b 9030be0 4f3470b 9030be0 db1867d 123c678 39207e4 9a9c028 f367387 3fa421f f367387 274a509 |
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 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 |
import gradio as gr
import regex as re
import csv
import pandas as pd
from typing import List, Dict, Tuple, Any
import logging
import os
import time
# Import core logic from other modules, as in app_old.py
from analyzer import (
combine_repo_files_for_llm,
parse_llm_json_response,
analyze_combined_file,
handle_load_repository
)
from hf_utils import download_filtered_space_files, search_top_spaces
from chatbot_page import chat_with_user, extract_keywords_from_conversation
from repo_explorer import create_repo_explorer_tab, setup_repo_explorer_events, initialize_repo_chatbot
# --- Configuration ---
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
logger = logging.getLogger(__name__)
CSV_FILE = "repo_ids.csv"
CHATBOT_SYSTEM_PROMPT = (
"You are a helpful assistant whose ONLY job is to gather information about the user's ideal repository requirements. "
"DO NOT suggest any specific repositories or give repository recommendations. "
"Your role is to ask clarifying questions to understand exactly what the user is looking for. "
"Ask about their use case, preferred programming language, specific features needed, project type, etc. "
"When you feel you have gathered enough detailed information about their requirements, "
"tell the user: 'I think I have enough information about your requirements. I'll now search for relevant repositories automatically.' "
"Focus on understanding their needs, not providing solutions."
)
CHATBOT_INITIAL_MESSAGE = "Hello! I'm here to help you find the perfect Hugging Face repository. Tell me about your project - what are you trying to build? I'll ask some questions to understand your needs and then automatically find relevant repositories for you."
# --- Helper Functions (Logic) ---
def is_repo_id_format(text: str) -> bool:
"""Check if text looks like repository IDs (contains forward slashes)."""
lines = [line.strip() for line in re.split(r'[\n,]+', text) if line.strip()]
if not lines:
return False
# If most lines contain forward slashes, treat as repo IDs
slash_count = sum(1 for line in lines if '/' in line)
return slash_count >= len(lines) * 0.5 # At least 50% have slashes
def should_auto_extract_keywords(history: List[Dict[str, str]]) -> bool:
"""Determine if we should automatically extract keywords from conversation."""
if not history or len(history) < 4: # Need at least 2 exchanges
return False
# Check if the last assistant message suggests we have enough info
last_assistant_msg = ""
for msg in reversed(history):
if msg.get('role') == 'assistant':
last_assistant_msg = msg.get('content', '').lower()
break
# Look for key phrases that indicate readiness
ready_phrases = [
"enough information",
"search for repositories",
"find repositories",
"look for repositories",
"automatically",
"ready to search"
]
return any(phrase in last_assistant_msg for phrase in ready_phrases)
def get_top_relevant_repos(df: pd.DataFrame, user_requirements: str, top_n: int = 3) -> pd.DataFrame:
"""
Uses LLM to select the top 3 most relevant repositories based on user requirements and analysis data.
"""
try:
if df.empty:
return pd.DataFrame(columns=["repo id", "strength", "weaknesses", "speciality", "relevance rating"])
# Filter out rows with no analysis data
analyzed_df = df.copy()
analyzed_df = analyzed_df[
(analyzed_df['strength'].str.strip() != '') |
(analyzed_df['weaknesses'].str.strip() != '') |
(analyzed_df['speciality'].str.strip() != '') |
(analyzed_df['relevance rating'].str.strip() != '')
]
if analyzed_df.empty:
logger.warning("No analyzed repositories found for LLM selection")
return pd.DataFrame(columns=["repo id", "strength", "weaknesses", "speciality", "relevance rating"])
# Create a prompt for the LLM
csv_data = ""
for idx, row in analyzed_df.iterrows():
csv_data += f"Repository: {row['repo id']}\n"
csv_data += f"Strengths: {row['strength']}\n"
csv_data += f"Weaknesses: {row['weaknesses']}\n"
csv_data += f"Speciality: {row['speciality']}\n"
csv_data += f"Relevance: {row['relevance rating']}\n\n"
user_context = user_requirements if user_requirements.strip() else "General repository recommendation"
prompt = f"""Based on the user's requirements and the analysis of repositories below, select the top {top_n} most relevant repositories.
User Requirements:
{user_context}
Repository Analysis Data:
{csv_data}
Please analyze all repositories and select the {top_n} most relevant ones based on:
1. How well they match the user's specific requirements
2. Their strengths and capabilities
3. Their relevance rating
4. Their speciality alignment with user needs
Return ONLY a JSON list of the repository IDs in order of relevance (most relevant first). Example format:
["repo1", "repo2", "repo3"]
Selected repositories:"""
try:
from openai import OpenAI
client = OpenAI(api_key=os.getenv("modal_api"))
client.base_url = os.getenv("base_url")
response = client.chat.completions.create(
model="Orion-zhen/Qwen2.5-Coder-7B-Instruct-AWQ",
messages=[
{"role": "system", "content": "You are an expert at analyzing and ranking repositories based on user requirements. Always return valid JSON."},
{"role": "user", "content": prompt}
],
max_tokens=200,
temperature=0.3
)
llm_response = response.choices[0].message.content.strip()
logger.info(f"LLM response for top repos: {llm_response}")
# Extract JSON from response
import json
import re
# Try to find JSON array in the response
json_match = re.search(r'\[.*\]', llm_response)
if json_match:
selected_repos = json.loads(json_match.group())
logger.info(f"LLM selected repositories: {selected_repos}")
# Filter dataframe to only include selected repositories in order
top_repos_list = []
for repo_id in selected_repos[:top_n]:
matching_rows = analyzed_df[analyzed_df['repo id'] == repo_id]
if not matching_rows.empty:
top_repos_list.append(matching_rows.iloc[0])
if top_repos_list:
top_repos = pd.DataFrame(top_repos_list)
logger.info(f"Successfully selected {len(top_repos)} repositories using LLM")
return top_repos
# Fallback: if LLM response parsing fails, use first N analyzed repos
logger.warning("Failed to parse LLM response, using fallback selection")
return analyzed_df.head(top_n)
except Exception as llm_error:
logger.error(f"LLM selection failed: {llm_error}")
# Fallback: return first N repositories with analysis data
return analyzed_df.head(top_n)
except Exception as e:
logger.error(f"Error in LLM-based repo selection: {e}")
return pd.DataFrame(columns=["repo id", "strength", "weaknesses", "speciality", "relevance rating"])
def write_repos_to_csv(repo_ids: List[str]) -> None:
"""Writes a list of repo IDs to the CSV file, overwriting the previous content."""
try:
with open(CSV_FILE, mode="w", newline='', encoding="utf-8") as csvfile:
writer = csv.writer(csvfile)
writer.writerow(["repo id", "strength", "weaknesses", "speciality", "relevance rating"])
for repo_id in repo_ids:
writer.writerow([repo_id, "", "", "", ""])
logger.info(f"Wrote {len(repo_ids)} repo IDs to {CSV_FILE}")
except Exception as e:
logger.error(f"Error writing to CSV: {e}")
def format_text_for_dataframe(text: str, max_length: int = 200) -> str:
"""Format text for better display in dataframe by truncating and cleaning."""
if not text or pd.isna(text):
return ""
# Clean the text
text = str(text).strip()
# Remove excessive whitespace and newlines
text = re.sub(r'\s+', ' ', text)
# Truncate if too long
if len(text) > max_length:
text = text[:max_length-3] + "..."
return text
def read_csv_to_dataframe() -> pd.DataFrame:
"""Reads the CSV file into a pandas DataFrame with full text preserved."""
try:
df = pd.read_csv(CSV_FILE, dtype=str).fillna('')
# Keep the full text intact - don't truncate here
# The truncation will be handled in the UI display layer
return df
except FileNotFoundError:
return pd.DataFrame(columns=["repo id", "strength", "weaknesses", "speciality", "relevance rating"])
except Exception as e:
logger.error(f"Error reading CSV: {e}")
return pd.DataFrame()
def format_dataframe_for_display(df: pd.DataFrame) -> pd.DataFrame:
"""Returns dataframe with full text (no truncation) for display."""
if df.empty:
return df
# Return the dataframe as-is without any text truncation
# This will show the full text content in the CSV display
return df.copy()
def analyze_and_update_single_repo(repo_id: str, user_requirements: str = "") -> Tuple[str, str, pd.DataFrame]:
"""
Downloads, analyzes a single repo, updates the CSV, and returns results.
Now includes user requirements for better relevance rating.
This function combines the logic of downloading, analyzing, and updating the CSV for one repo.
"""
try:
logger.info(f"Starting analysis for repo: {repo_id}")
download_filtered_space_files(repo_id, local_dir="repo_files", file_extensions=['.py', '.md', '.txt'])
txt_path = combine_repo_files_for_llm()
with open(txt_path, "r", encoding="utf-8") as f:
combined_content = f.read()
llm_output = analyze_combined_file(txt_path, user_requirements)
last_start = llm_output.rfind('{')
last_end = llm_output.rfind('}')
final_json_str = llm_output[last_start:last_end+1] if last_start != -1 and last_end != -1 else "{}"
llm_json = parse_llm_json_response(final_json_str)
summary = ""
if isinstance(llm_json, dict) and "error" not in llm_json:
strengths = llm_json.get("strength", "N/A")
weaknesses = llm_json.get("weaknesses", "N/A")
relevance = llm_json.get("relevance rating", "N/A")
summary = f"JSON extraction: SUCCESS\n\nStrengths:\n{strengths}\n\nWeaknesses:\n{weaknesses}\n\nRelevance: {relevance}"
else:
summary = f"JSON extraction: FAILED\nRaw response might not be valid JSON."
# Update CSV
df = read_csv_to_dataframe()
repo_found_in_df = False
for idx, row in df.iterrows():
if row["repo id"] == repo_id:
if isinstance(llm_json, dict):
df.at[idx, "strength"] = llm_json.get("strength", "")
df.at[idx, "weaknesses"] = llm_json.get("weaknesses", "")
df.at[idx, "speciality"] = llm_json.get("speciality", "")
df.at[idx, "relevance rating"] = llm_json.get("relevance rating", "")
repo_found_in_df = True
break
if not repo_found_in_df:
logger.warning(f"Repo ID {repo_id} not found in CSV for updating.")
# Write CSV with better error handling and flushing
try:
df.to_csv(CSV_FILE, index=False)
# Force file system flush
os.sync() if hasattr(os, 'sync') else None
logger.info(f"Successfully updated CSV for {repo_id}")
except Exception as csv_error:
logger.error(f"Failed to write CSV for {repo_id}: {csv_error}")
# Try once more with a small delay
time.sleep(0.2)
try:
df.to_csv(CSV_FILE, index=False)
logger.info(f"Successfully updated CSV for {repo_id} on retry")
except Exception as retry_error:
logger.error(f"Failed to write CSV for {repo_id} on retry: {retry_error}")
logger.info(f"Successfully analyzed and updated CSV for {repo_id}")
return combined_content, summary, df
except Exception as e:
logger.error(f"An error occurred during analysis of {repo_id}: {e}")
error_summary = f"Error analyzing repo: {e}"
return "", error_summary, format_dataframe_for_display(read_csv_to_dataframe())
# --- NEW: Helper for Chat History Conversion ---
def convert_messages_to_tuples(history: List[Dict[str, str]]) -> List[Tuple[str, str]]:
"""
Converts Gradio's 'messages' format to the old 'tuple' format for compatibility.
This robust version correctly handles histories that start with an assistant message.
"""
tuple_history = []
# Iterate through the history to find user messages
for i, msg in enumerate(history):
if msg['role'] == 'user':
# Once a user message is found, check if the next message is from the assistant
if i + 1 < len(history) and history[i+1]['role'] == 'assistant':
user_content = msg['content']
assistant_content = history[i+1]['content']
tuple_history.append((user_content, assistant_content))
return tuple_history
# --- Gradio UI ---
def create_ui() -> gr.Blocks:
"""Creates and configures the entire Gradio interface."""
css = """
/* Modern sleek design */
.gradio-container {
font-family: 'Inter', 'system-ui', sans-serif;
background: linear-gradient(135deg, #0a0a0a 0%, #1a1a1a 100%);
min-height: 100vh;
}
.gr-form {
background: rgba(255, 255, 255, 0.95);
backdrop-filter: blur(10px);
border-radius: 16px;
box-shadow: 0 8px 32px rgba(0, 0, 0, 0.1);
padding: 24px;
margin: 16px;
border: 1px solid rgba(255, 255, 255, 0.2);
}
.gr-button {
background: linear-gradient(45deg, #667eea, #764ba2);
border: none;
border-radius: 12px;
color: white;
font-weight: 600;
padding: 12px 24px;
transition: all 0.3s ease;
box-shadow: 0 4px 15px rgba(102, 126, 234, 0.4);
}
.gr-button:hover {
transform: translateY(-2px);
box-shadow: 0 6px 20px rgba(102, 126, 234, 0.6);
}
.gr-textbox {
border: 2px solid rgba(102, 126, 234, 0.2);
border-radius: 12px;
background: rgba(255, 255, 255, 0.9);
transition: all 0.3s ease;
}
.gr-textbox:focus {
border-color: #667eea;
box-shadow: 0 0 0 3px rgba(102, 126, 234, 0.1);
}
.gr-panel {
background: rgba(255, 255, 255, 0.95);
border-radius: 16px;
box-shadow: 0 8px 32px rgba(0, 0, 0, 0.1);
border: 1px solid rgba(255, 255, 255, 0.2);
}
.gr-tab-nav {
background: rgba(255, 255, 255, 0.95);
border-radius: 12px 12px 0 0;
backdrop-filter: blur(10px);
}
.gr-tab-nav button {
background: transparent;
border: none;
padding: 16px 24px;
font-weight: 600;
color: #666;
transition: all 0.3s ease;
}
.gr-tab-nav button.selected {
background: linear-gradient(45deg, #667eea, #764ba2);
color: white;
border-radius: 8px;
}
.chatbot {
border-radius: 16px;
box-shadow: 0 4px 20px rgba(0, 0, 0, 0.1);
}
/* Hide Gradio footer */
footer {
display: none !important;
}
/* Custom scrollbar */
::-webkit-scrollbar {
width: 8px;
}
::-webkit-scrollbar-track {
background: rgba(255, 255, 255, 0.1);
border-radius: 4px;
}
::-webkit-scrollbar-thumb {
background: linear-gradient(45deg, #667eea, #764ba2);
border-radius: 4px;
}
/* Improved dataframe styling for full text display */
.gr-dataframe {
border-radius: 12px;
overflow: hidden;
box-shadow: 0 4px 20px rgba(0, 0, 0, 0.1);
background: rgba(255, 255, 255, 0.98);
}
.gr-dataframe table {
width: 100%;
table-layout: fixed;
border-collapse: collapse;
}
/* Column width specifications for both dataframes */
.gr-dataframe th,
.gr-dataframe td {
padding: 12px 16px;
text-align: left;
border-bottom: 1px solid rgba(0, 0, 0, 0.1);
font-size: 0.95rem;
line-height: 1.4;
}
/* Specific column widths - applying to both dataframes */
.gr-dataframe th:nth-child(1),
.gr-dataframe td:nth-child(1) { width: 16.67% !important; min-width: 16.67% !important; max-width: 16.67% !important; }
.gr-dataframe th:nth-child(2),
.gr-dataframe td:nth-child(2) { width: 25% !important; min-width: 25% !important; max-width: 25% !important; }
.gr-dataframe th:nth-child(3),
.gr-dataframe td:nth-child(3) { width: 25% !important; min-width: 25% !important; max-width: 25% !important; }
.gr-dataframe th:nth-child(4),
.gr-dataframe td:nth-child(4) { width: 20.83% !important; min-width: 20.83% !important; max-width: 20.83% !important; }
.gr-dataframe th:nth-child(5),
.gr-dataframe td:nth-child(5) { width: 12.5% !important; min-width: 12.5% !important; max-width: 12.5% !important; }
/* Additional specific targeting for both dataframes */
div[data-testid="dataframe"] table th:nth-child(1),
div[data-testid="dataframe"] table td:nth-child(1) { width: 16.67% !important; }
div[data-testid="dataframe"] table th:nth-child(2),
div[data-testid="dataframe"] table td:nth-child(2) { width: 25% !important; }
div[data-testid="dataframe"] table th:nth-child(3),
div[data-testid="dataframe"] table td:nth-child(3) { width: 25% !important; }
div[data-testid="dataframe"] table th:nth-child(4),
div[data-testid="dataframe"] table td:nth-child(4) { width: 20.83% !important; }
div[data-testid="dataframe"] table th:nth-child(5),
div[data-testid="dataframe"] table td:nth-child(5) { width: 12.5% !important; }
/* Make repository names clickable */
.gr-dataframe td:nth-child(1) {
cursor: pointer;
color: #667eea;
font-weight: 600;
transition: all 0.3s ease;
}
.gr-dataframe td:nth-child(1):hover {
background-color: rgba(102, 126, 234, 0.1);
color: #764ba2;
transform: scale(1.02);
}
/* Content columns - readable styling with scroll for long text */
.gr-dataframe td:nth-child(2),
.gr-dataframe td:nth-child(3),
.gr-dataframe td:nth-child(4),
.gr-dataframe td:nth-child(5) {
cursor: default;
font-size: 0.9rem;
}
.gr-dataframe tbody tr:hover {
background-color: rgba(102, 126, 234, 0.05);
}
/* JavaScript for auto-scroll to top on tab change */
<script>
document.addEventListener('DOMContentLoaded', function() {
// Function to scroll to top
function scrollToTop() {
window.scrollTo({
top: 0,
behavior: 'smooth'
});
}
// Observer for tab changes
const observer = new MutationObserver(function(mutations) {
mutations.forEach(function(mutation) {
if (mutation.type === 'attributes' && mutation.attributeName === 'class') {
const target = mutation.target;
if (target.classList && target.classList.contains('selected')) {
// Tab was selected, scroll to top
setTimeout(scrollToTop, 100);
}
}
});
});
// Observe tab navigation buttons
const tabButtons = document.querySelectorAll('.gr-tab-nav button');
tabButtons.forEach(button => {
observer.observe(button, { attributes: true });
// Also add click listener for immediate scroll
button.addEventListener('click', function() {
setTimeout(scrollToTop, 150);
});
});
// Enhanced listener for programmatic tab changes (button-triggered navigation)
let lastSelectedTab = null;
const checkInterval = setInterval(function() {
const currentSelectedTab = document.querySelector('.gr-tab-nav button.selected');
if (currentSelectedTab && currentSelectedTab !== lastSelectedTab) {
lastSelectedTab = currentSelectedTab;
setTimeout(scrollToTop, 100);
}
}, 100);
// Additional scroll trigger for repo explorer navigation
window.addEventListener('repoExplorerNavigation', function() {
setTimeout(scrollToTop, 200);
});
// Watch for specific tab transitions to repo explorer
const repoExplorerObserver = new MutationObserver(function(mutations) {
mutations.forEach(function(mutation) {
if (mutation.type === 'attributes' && mutation.attributeName === 'class') {
const target = mutation.target;
if (target.textContent && target.textContent.includes('🔍 Repo Explorer') && target.classList.contains('selected')) {
setTimeout(scrollToTop, 150);
}
}
});
});
// Start observing for repo explorer specific changes
setTimeout(function() {
const repoExplorerTab = Array.from(document.querySelectorAll('.gr-tab-nav button')).find(btn =>
btn.textContent && btn.textContent.includes('🔍 Repo Explorer')
);
if (repoExplorerTab) {
repoExplorerObserver.observe(repoExplorerTab, { attributes: true });
}
}, 1000);
});
</script>
"""
with gr.Blocks(
theme=gr.themes.Soft(
primary_hue="blue",
secondary_hue="purple",
neutral_hue="gray",
font=["Inter", "system-ui", "sans-serif"]
),
css=css,
title="🚀 HF Repo Analyzer"
) as app:
# --- State Management ---
# Using simple, separate state objects for robustness.
repo_ids_state = gr.State([])
current_repo_idx_state = gr.State(0)
user_requirements_state = gr.State("") # Store user requirements from chatbot
loaded_repo_content_state = gr.State("") # Store loaded repository content
current_repo_id_state = gr.State("") # Store current repository ID
selected_repo_id_state = gr.State("") # Store selected repository ID for modal actions
gr.Markdown(
"""
<div style="text-align: center; padding: 40px 20px; background: rgba(255, 255, 255, 0.1); border-radius: 20px; margin: 20px auto; max-width: 900px; backdrop-filter: blur(10px);">
<h1 style="font-size: 3.5rem; font-weight: 800; margin: 0; background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent; background-clip: text;">
🚀 HF Repo Analyzer
</h1>
<p style="font-size: 1.3rem; color: rgba(255, 255, 255, 0.9); margin: 16px 0 0 0; font-weight: 400; line-height: 1.6;">
Discover, analyze, and evaluate Hugging Face repositories with AI-powered insights
</p>
<div style="height: 4px; width: 80px; background: linear-gradient(45deg, #667eea, #764ba2); margin: 24px auto; border-radius: 2px;"></div>
</div>
"""
)
# Global Reset and Help Buttons - visible on all tabs
with gr.Row():
with gr.Column(scale=3):
pass
with gr.Column(scale=1):
help_btn = gr.Button("❓ Help", variant="secondary", size="lg")
with gr.Column(scale=1):
reset_all_btn = gr.Button("🔄 Reset Everything", variant="stop", size="lg")
with gr.Column(scale=1):
pass
# Help Modal - visible when help button is clicked
with gr.Row():
with gr.Column():
help_modal = gr.Column(visible=False)
with help_modal:
gr.Markdown(
"""
<div style="background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); padding: 20px; border-radius: 16px; text-align: center; margin-bottom: 20px;">
<h2 style="color: white; margin: 0; font-size: 2rem;">📚 How to Use HF Repo Analyzer</h2>
<p style="color: rgba(255,255,255,0.9); margin: 10px 0 0 0;">Step-by-step guide to find and analyze repositories</p>
</div>
"""
)
with gr.Accordion("🚀 Method 1: AI Assistant (Recommended)", open=True):
gr.Markdown(
"""
### **Step 1: Start Conversation**
- Go to the **🤖 AI Assistant** tab
- Describe your project: *"I'm building a sentiment analysis tool"*
- The AI will ask clarifying questions about your needs
### **Step 2: Let AI Work Its Magic**
- Answer the AI's questions about your requirements
- When ready, the AI will automatically:
- Extract keywords from your conversation
- Search for matching repositories
- Analyze and rank them by relevance
### **Step 3: Review Results**
- Interface automatically switches to **🔬 Analysis & Results**
- View **Top 3** most relevant repositories
- Browse detailed analysis with strengths/weaknesses
- Click repository names to visit or explore them
**💡 Tip**: This method gives the best personalized results!
"""
)
with gr.Accordion("📝 Method 2: Smart Search (Direct Input)", open=False):
gr.Markdown(
"""
### **Step 1: Choose Input Type**
Go to **📝 Smart Search** tab and enter either:
**Repository IDs** (with `/`):
```
microsoft/DialoGPT-medium
openai/whisper
huggingface/transformers
```
**Keywords** (no `/`):
```
text generation
image classification
sentiment analysis
```
### **Step 2: Auto-Detection & Processing**
- System automatically detects input type
- Repository IDs → Direct analysis
- Keywords → Search + analysis
- Enable **🚀 Auto-analyze** for instant results
### **Step 3: Get Results**
- Click **🔍 Find & Process Repositories**
- View results in **🔬 Analysis & Results** tab
"""
)
with gr.Accordion("🔬 Understanding Analysis Results", open=False):
gr.Markdown(
"""
### **🏆 Top 3 Repositories**
- AI-selected most relevant for your needs
- Ranked by requirement matching and quality
### **📊 Detailed Analysis Table**
- **Repository**: Click names to visit/explore
- **Strengths**: Key capabilities and advantages
- **Weaknesses**: Limitations and considerations
- **Speciality**: Primary use case and domain
- **Relevance**: How well it matches your needs
### **🔗 Quick Actions**
Click repository names to:
- **🌐 Visit Hugging Face Space**: See live demo
- **🔍 Open in Repo Explorer**: Deep dive analysis
"""
)
with gr.Accordion("🔍 Repository Explorer Deep Dive", open=False):
gr.Markdown(
"""
### **Access Repository Explorer**
- Click **🔍 Open in Repo Explorer** from results
- Or manually enter repo ID in **🔍 Repo Explorer** tab
### **Features Available**
- **Auto-loading**: Repository content analysis
- **AI Chat**: Ask questions about the code
- **File Exploration**: Browse repository structure
- **Code Analysis**: Get explanations and insights
### **Sample Questions to Ask**
- *"How do I use this repository?"*
- *"What are the main functions?"*
- *"Show me example usage"*
- *"Explain the architecture"*
"""
)
with gr.Accordion("🎯 Pro Tips & Best Practices", open=False):
gr.Markdown(
"""
### **🤖 Getting Better AI Results**
- Be specific about your use case
- Mention programming language preferences
- Describe your experience level
- Include performance requirements
### **🔍 Search Optimization**
- Use multiple relevant keywords
- Try different keyword combinations
- Check both general and specific terms
### **📊 Analyzing Results**
- Read both strengths AND weaknesses
- Check speciality alignment with your needs
- Use Repository Explorer for detailed investigation
- Compare multiple options before deciding
### **🔄 Workflow Tips**
- Start with AI Assistant for personalized results
- Use Smart Search for known repositories
- Explore multiple repositories before choosing
- Save interesting repositories for later comparison
"""
)
with gr.Row():
close_help_btn = gr.Button("✅ Got It, Let's Start!", variant="primary", size="lg")
with gr.Tabs() as tabs:
# --- AI Assistant Tab (moved to first) ---
with gr.TabItem("🤖 AI Assistant", id="chatbot_tab"):
gr.Markdown("### 💬 Intelligent Repository Discovery Assistant")
gr.Markdown("🎯 **Tell me what you're building, and I'll automatically find the best repositories for you!**")
chatbot = gr.Chatbot(
label="🤖 AI Assistant",
height=500,
type="messages",
avatar_images=(
"https://cdn-icons-png.flaticon.com/512/149/149071.png",
"https://huggingface.co/datasets/huggingface/brand-assets/resolve/main/hf-logo.png"
),
show_copy_button=True
)
with gr.Row():
msg_input = gr.Textbox(
label="💭 Your Message",
placeholder="Tell me about your project...",
lines=1,
scale=5,
info="Describe what you're building and I'll find the perfect repositories"
)
send_btn = gr.Button("📤", variant="primary", scale=1)
with gr.Row():
extract_analyze_btn = gr.Button("🎯 Extract Keywords & Analyze Now", variant="secondary", size="lg")
# Status and extracted info (auto-updated, no manual buttons needed)
with gr.Row():
with gr.Column():
chat_status = gr.Textbox(
label="🎯 Chat Status",
interactive=False,
lines=2,
info="Conversation progress and auto-actions"
)
with gr.Column():
extracted_keywords_output = gr.Textbox(
label="🏷️ Auto-Extracted Keywords",
interactive=False,
show_copy_button=True,
info="Keywords automatically extracted and used for search"
)
# --- Smart Search Tab (moved to second) ---
with gr.TabItem("📝 Smart Search", id="input_tab"):
gr.Markdown("### 🔍 Intelligent Repository Discovery")
gr.Markdown("💡 **Enter repository IDs (owner/repo) or keywords - I'll automatically detect which type and process accordingly!**")
with gr.Row():
smart_input = gr.Textbox(
label="Repository IDs or Keywords",
lines=6,
placeholder="Examples:\n• Repository IDs: microsoft/DialoGPT-medium, openai/whisper\n• Keywords: text generation, image classification, sentiment analysis",
info="Smart detection: Use / for repo IDs, or enter keywords for search"
)
with gr.Row():
auto_analyze_checkbox = gr.Checkbox(
label="🚀 Auto-analyze repositories",
value=True,
info="Automatically start analysis when repositories are found"
)
smart_submit_btn = gr.Button("🔍 Find & Process Repositories", variant="primary", size="lg", scale=1)
status_box_input = gr.Textbox(label="📊 Status", interactive=False, lines=2)
# --- Analysis & Results Tab (moved to third) ---
with gr.TabItem("🔬 Analysis & Results", id="analysis_tab"):
gr.Markdown("### 🧪 Repository Analysis Results")
# Display current user requirements
with gr.Row():
current_requirements_display = gr.Textbox(
label="📋 Active Requirements Context",
interactive=False,
lines=2,
info="Requirements from AI chat for better relevance scoring"
)
# Manual analysis trigger (hidden by default, shown only when auto-analyze is off)
with gr.Row(visible=False) as manual_analysis_row:
analyze_all_btn = gr.Button("🚀 Analyze All Repositories", variant="primary", size="lg")
status_box_analysis = gr.Textbox(label="📈 Analysis Status", interactive=False, lines=2)
# Progress bar for batch analysis
analysis_progress = gr.Progress()
gr.Markdown("### 📊 Results Dashboard")
# Top 3 Most Relevant Repositories (initially hidden)
with gr.Column(visible=False) as top_repos_section:
gr.Markdown("### 🏆 Top 3 Most Relevant Repositories")
gr.Markdown("🎯 **Click repository names to visit them directly on Hugging Face:**")
top_repos_df = gr.Dataframe(
headers=["Repository", "Strengths", "Weaknesses", "Speciality", "Relevance"],
column_widths=["16.67%", "25%", "25%", "20.83%", "12.5%"],
wrap=True,
interactive=False
)
# Quick links for top repositories
with gr.Row():
top_repo_links = gr.HTML(
value="",
label="🔗 Quick Links",
visible=False
)
# Modal popup for repository action selection (positioned between the two CSV files)
with gr.Row():
with gr.Column():
repo_action_modal = gr.Column(visible=False)
with repo_action_modal:
gr.Markdown("### 🔗 Repository Actions")
selected_repo_display = gr.Textbox(
label="Selected Repository",
interactive=False,
info="Choose what you'd like to do with this repository"
)
with gr.Row():
visit_repo_btn = gr.Button("🌐 Visit Hugging Face Space", variant="primary", size="lg")
explore_repo_btn = gr.Button("🔍 Open in Repo Explorer", variant="secondary", size="lg")
cancel_modal_btn = gr.Button("❌ Cancel", size="lg")
gr.Markdown("### 📋 All Analysis Results")
gr.Markdown("💡 **Click repository names to visit them on Hugging Face**")
df_output = gr.Dataframe(
headers=["Repository", "Strengths", "Weaknesses", "Speciality", "Relevance"],
column_widths=["16.67%", "25%", "25%", "20.83%", "12.5%"],
wrap=True,
interactive=False
)
# Quick links section for all repositories
with gr.Row():
all_repo_links = gr.HTML(
value="",
label="🔗 Repository Quick Links"
)
# --- Repo Explorer Tab (moved to fourth) ---
with gr.TabItem("🔍 Repo Explorer", id="repo_explorer_tab"):
repo_components, repo_states = create_repo_explorer_tab()
# --- Footer ---
gr.Markdown(
"""
<div style="text-align: center; padding: 30px 20px; margin-top: 40px; background: rgba(255, 255, 255, 0.1); border-radius: 16px; backdrop-filter: blur(10px);">
<p style="margin: 0; color: rgba(255, 255, 255, 0.8); font-size: 0.95rem; font-weight: 500;">
🚀 Powered by <span style="background: linear-gradient(45deg, #667eea, #764ba2); -webkit-background-clip: text; -webkit-text-fill-color: transparent; font-weight: 700;">Gradio</span>
& <span style="background: linear-gradient(45deg, #667eea, #764ba2); -webkit-background-clip: text; -webkit-text-fill-color: transparent; font-weight: 700;">Hugging Face</span>
</p>
<div style="height: 2px; width: 60px; background: linear-gradient(45deg, #667eea, #764ba2); margin: 16px auto; border-radius: 1px;"></div>
</div>
"""
)
# --- Event Handler Functions ---
def handle_smart_input(text: str, auto_analyze: bool) -> Tuple[List[str], int, pd.DataFrame, str, Any, str]:
"""Smart input handler that detects if input is repo IDs or keywords and processes accordingly."""
if not text.strip():
return [], 0, pd.DataFrame(), "Status: Please enter repository IDs or keywords.", gr.update(selected="input_tab"), ""
# Determine input type
if is_repo_id_format(text):
# Process as repository IDs
repo_ids = list(dict.fromkeys([repo.strip() for repo in re.split(r'[\n,]+', text) if repo.strip()]))
write_repos_to_csv(repo_ids)
df = format_dataframe_for_display(read_csv_to_dataframe())
status = f"✅ Found {len(repo_ids)} repository IDs. "
if auto_analyze:
status += "Starting automatic analysis..."
return repo_ids, 0, df, status, gr.update(selected="analysis_tab"), "auto_analyze"
else:
status += "Ready for manual analysis."
return repo_ids, 0, df, status, gr.update(selected="analysis_tab"), ""
else:
# Process as keywords
keyword_list = [k.strip() for k in re.split(r'[\n,]+', text) if k.strip()]
repo_ids = []
for kw in keyword_list:
repo_ids.extend(search_top_spaces(kw, limit=5))
unique_repo_ids = list(dict.fromkeys(repo_ids))
write_repos_to_csv(unique_repo_ids)
df = format_dataframe_for_display(read_csv_to_dataframe())
status = f"🔍 Found {len(unique_repo_ids)} repositories from keywords. "
if auto_analyze:
status += "Starting automatic analysis..."
return unique_repo_ids, 0, df, status, gr.update(selected="analysis_tab"), "auto_analyze"
else:
status += "Ready for manual analysis."
return unique_repo_ids, 0, df, status, gr.update(selected="analysis_tab"), ""
def handle_auto_analyze_toggle(auto_analyze: bool) -> Any:
"""Show/hide manual analysis controls based on auto-analyze setting."""
return gr.update(visible=not auto_analyze)
def handle_user_message(user_message: str, history: List[Dict[str, str]]) -> Tuple[List[Dict[str, str]], str]:
"""Appends the user's message to the history, preparing for the bot's response."""
# Initialize chatbot with welcome message if empty
if not history:
history = [{"role": "assistant", "content": CHATBOT_INITIAL_MESSAGE}]
if user_message:
history.append({"role": "user", "content": user_message})
return history, ""
def handle_bot_response(history: List[Dict[str, str]]) -> Tuple[List[Dict[str, str]], str, str, str, List[str], int, pd.DataFrame, Any]:
"""Generates bot response and automatically extracts keywords if conversation is ready."""
if not history or history[-1]["role"] != "user":
return history, "", "", "", [], 0, pd.DataFrame(), gr.update()
user_message = history[-1]["content"]
# Convert all messages *before* the last user message into tuples for the API
tuple_history_for_api = convert_messages_to_tuples(history[:-1])
response = chat_with_user(user_message, tuple_history_for_api)
history.append({"role": "assistant", "content": response})
# Check if we should auto-extract keywords and search
if should_auto_extract_keywords(history):
# Auto-extract keywords
tuple_history = convert_messages_to_tuples(history)
raw_keywords_str = extract_keywords_from_conversation(tuple_history)
# Sanitize keywords
cleaned_keywords = re.findall(r'[\w\s-]+', raw_keywords_str)
cleaned_keywords = [kw.strip() for kw in cleaned_keywords if kw.strip()]
if cleaned_keywords:
final_keywords_str = ", ".join(cleaned_keywords)
# Extract user requirements
user_requirements = extract_user_requirements_from_chat(history)
# Auto-search repositories
repo_ids = []
for kw in cleaned_keywords[:3]: # Use top 3 keywords to avoid too many results
repo_ids.extend(search_top_spaces(kw, limit=5))
unique_repo_ids = list(dict.fromkeys(repo_ids))
write_repos_to_csv(unique_repo_ids)
df = format_dataframe_for_display(read_csv_to_dataframe())
chat_status = f"🎯 Auto-extracted keywords and found {len(unique_repo_ids)} repositories. Analysis starting automatically..."
return history, chat_status, final_keywords_str, user_requirements, unique_repo_ids, 0, df, gr.update(selected="analysis_tab")
return history, "💬 Conversation continuing...", "", "", [], 0, pd.DataFrame(), gr.update()
def handle_dataframe_select(evt: gr.SelectData, df_data) -> Tuple[str, Any, str]:
"""Handle dataframe row selection - show modal for repo ID (column 0) clicks."""
print(f"DEBUG: Selection event triggered!")
print(f"DEBUG: evt = {evt}")
print(f"DEBUG: df_data type = {type(df_data)}")
if evt is None:
return "", gr.update(visible=False), ""
try:
# Get the selected row and column from the event
row_idx = evt.index[0]
col_idx = evt.index[1]
print(f"DEBUG: Selected row {row_idx}, column {col_idx}")
# Handle pandas DataFrame
if isinstance(df_data, pd.DataFrame) and not df_data.empty and row_idx < len(df_data):
if col_idx == 0: # Repository name column - show action modal
repo_id = df_data.iloc[row_idx, 0]
print(f"DEBUG: Extracted repo_id = '{repo_id}'")
if repo_id and str(repo_id).strip() and str(repo_id).strip() != 'nan':
clean_repo_id = str(repo_id).strip()
logger.info(f"Showing modal for repository: {clean_repo_id}")
return clean_repo_id, gr.update(visible=True), clean_repo_id
# For content columns (1,2,3) and relevance (4), do nothing since full text is shown directly
else:
print(f"DEBUG: Clicked on column {col_idx}, full text already shown in table")
return "", gr.update(visible=False), ""
else:
print(f"DEBUG: df_data is not a DataFrame or row_idx {row_idx} out of range")
except Exception as e:
print(f"DEBUG: Exception occurred: {e}")
logger.error(f"Error handling dataframe selection: {e}")
return "", gr.update(visible=False), ""
def handle_visit_repo(repo_id: str) -> Tuple[Any, str]:
"""Handle visiting the Hugging Face Space for the repository."""
if repo_id and repo_id.strip():
hf_url = f"https://huggingface.co/spaces/{repo_id.strip()}"
logger.info(f"User chose to visit: {hf_url}")
return gr.update(visible=False), hf_url
return gr.update(visible=False), ""
def handle_explore_repo(selected_repo_id: str) -> Tuple[Any, Any, Any, str, str]:
"""Handle navigating to the repo explorer and automatically load the repository."""
logger.info(f"DEBUG: handle_explore_repo called with selected_repo_id: '{selected_repo_id}'")
if selected_repo_id and selected_repo_id.strip() and selected_repo_id.strip() != 'nan':
clean_repo_id = selected_repo_id.strip()
return (
gr.update(visible=False), # close modal
gr.update(selected="repo_explorer_tab"), # switch tab
gr.update(value=clean_repo_id), # populate repo explorer input
clean_repo_id, # trigger repository loading with the repo ID
"auto_load" # signal to auto-load the repository
)
else:
return (
gr.update(visible=False), # close modal
gr.update(selected="repo_explorer_tab"), # switch tab
gr.update(), # don't change repo explorer input
"", # no repo ID to load
"" # no auto-load signal
)
def handle_cancel_modal() -> Any:
"""Handle closing the modal."""
return gr.update(visible=False)
def generate_repo_links_html(df: pd.DataFrame) -> str:
"""Generate HTML with clickable links for repositories."""
if df.empty:
return ""
html_links = []
for idx, row in df.iterrows():
repo_id = row.get('repo id', '') if hasattr(row, 'get') else row[0]
if repo_id and str(repo_id).strip() and str(repo_id).strip() != 'nan':
clean_repo_id = str(repo_id).strip()
hf_url = f"https://huggingface.co/spaces/{clean_repo_id}"
html_links.append(f'<a href="{hf_url}" target="_blank" style="display: inline-block; margin: 5px 10px; padding: 8px 16px; background: linear-gradient(45deg, #667eea, #764ba2); color: white; text-decoration: none; border-radius: 8px; font-weight: 600; transition: all 0.3s ease;">{clean_repo_id}</a>')
if html_links:
return f'<div style="margin: 10px 0; padding: 15px; background: rgba(255, 255, 255, 0.1); border-radius: 12px; backdrop-filter: blur(10px);">{"".join(html_links)}</div>'
return ""
def handle_extract_and_analyze(history: List[Dict[str, str]]) -> Tuple[str, str, str, List[str], int, pd.DataFrame, Any, pd.DataFrame, str, Any, str, str]:
"""Extract keywords from chat, search repositories, and immediately start analysis."""
if not history:
return "❌ No conversation to extract from.", "", "", [], 0, pd.DataFrame(), gr.update(), pd.DataFrame(), "", gr.update(visible=False), "", ""
# Convert the full, valid history for the extraction logic
tuple_history = convert_messages_to_tuples(history)
if not tuple_history:
return "❌ No completed conversations to analyze.", "", "", [], 0, pd.DataFrame(), gr.update(), pd.DataFrame(), "", gr.update(visible=False), "", ""
# Get raw keywords string from the LLM
raw_keywords_str = extract_keywords_from_conversation(tuple_history)
# Sanitize the LLM output to extract only keyword-like parts
cleaned_keywords = re.findall(r'[\w\s-]+', raw_keywords_str)
cleaned_keywords = [kw.strip() for kw in cleaned_keywords if kw.strip()]
if not cleaned_keywords:
return f"❌ Could not extract valid keywords. Raw output: '{raw_keywords_str}'", "", "", [], 0, pd.DataFrame(), gr.update(), pd.DataFrame(), "", gr.update(visible=False), "", ""
# Join them into a clean, comma-separated string
final_keywords_str = ", ".join(cleaned_keywords)
# Extract user requirements for analysis
user_requirements = extract_user_requirements_from_chat(history)
# Auto-search repositories
repo_ids = []
for kw in cleaned_keywords[:3]: # Use top 3 keywords to avoid too many results
repo_ids.extend(search_top_spaces(kw, limit=5))
unique_repo_ids = list(dict.fromkeys(repo_ids))
if not unique_repo_ids:
return f"❌ No repositories found for keywords: {final_keywords_str}", final_keywords_str, user_requirements, [], 0, pd.DataFrame(), gr.update(), pd.DataFrame(), "", gr.update(visible=False), "", ""
write_repos_to_csv(unique_repo_ids)
df = format_dataframe_for_display(read_csv_to_dataframe())
# Immediately start analysis
try:
analyzed_df, analysis_status, top_repos, top_section_update, all_links, top_links = handle_analyze_all_repos(unique_repo_ids, user_requirements)
chat_status = f"🎉 Extracted keywords → Found {len(unique_repo_ids)} repositories → Analysis complete!"
return chat_status, final_keywords_str, user_requirements, unique_repo_ids, 0, analyzed_df, gr.update(selected="analysis_tab"), top_repos, analysis_status, top_section_update, all_links, top_links
except Exception as e:
logger.error(f"Error during extract and analyze: {e}")
error_status = f"✅ Found {len(unique_repo_ids)} repositories, but analysis failed: {e}"
return error_status, final_keywords_str, user_requirements, unique_repo_ids, 0, df, gr.update(selected="analysis_tab"), pd.DataFrame(), "", gr.update(visible=False), "", ""
def extract_user_requirements_from_chat(history: List[Dict[str, str]]) -> str:
"""Extract user requirements from chatbot conversation."""
if not history:
return ""
user_messages = []
for msg in history:
if msg.get('role') == 'user':
user_messages.append(msg.get('content', ''))
if not user_messages:
return ""
# Combine all user messages as requirements
requirements = "\n".join([f"- {msg}" for msg in user_messages if msg.strip()])
return requirements
def handle_analyze_all_repos(repo_ids: List[str], user_requirements: str, progress=gr.Progress()) -> Tuple[pd.DataFrame, str, pd.DataFrame, Any, str, str]:
"""Analyzes all repositories in the CSV file with progress tracking."""
if not repo_ids:
return pd.DataFrame(), "Status: No repositories to analyze. Please submit repo IDs first.", pd.DataFrame(), gr.update(visible=False), "", ""
total_repos = len(repo_ids)
try:
# Start the progress tracking
progress(0, desc="Initializing batch analysis...")
successful_analyses = 0
failed_analyses = 0
csv_update_failures = 0
for i, repo_id in enumerate(repo_ids):
# Update progress
progress_percent = (i / total_repos)
progress(progress_percent, desc=f"Analyzing {repo_id} ({i+1}/{total_repos})")
try:
logger.info(f"Batch analysis: Processing {repo_id} ({i+1}/{total_repos})")
# Analyze the repository
content, summary, df = analyze_and_update_single_repo(repo_id, user_requirements)
# Verify the CSV was actually updated by checking if the repo has analysis data
updated_df = read_csv_to_dataframe()
repo_updated = False
for idx, row in updated_df.iterrows():
if row["repo id"] == repo_id:
# Check if any analysis field is populated
if (row.get("strength", "").strip() or
row.get("weaknesses", "").strip() or
row.get("speciality", "").strip() or
row.get("relevance rating", "").strip()):
repo_updated = True
break
if repo_updated:
successful_analyses += 1
else:
# CSV update failed - try once more
logger.warning(f"CSV update failed for {repo_id}, attempting retry...")
time.sleep(0.5) # Wait a bit longer
# Force re-read and re-update
df_retry = read_csv_to_dataframe()
retry_success = False
# Re-parse the analysis if available
if summary and "JSON extraction: SUCCESS" in summary:
# Extract the analysis from summary - this is a fallback
logger.info(f"Attempting to re-update CSV for {repo_id}")
content_retry, summary_retry, df_retry = analyze_and_update_single_repo(repo_id, user_requirements)
# Check again
final_df = read_csv_to_dataframe()
for idx, row in final_df.iterrows():
if row["repo id"] == repo_id:
if (row.get("strength", "").strip() or
row.get("weaknesses", "").strip() or
row.get("speciality", "").strip() or
row.get("relevance rating", "").strip()):
retry_success = True
break
if retry_success:
successful_analyses += 1
else:
csv_update_failures += 1
# Longer delay to prevent file conflicts
time.sleep(0.3)
except Exception as e:
logger.error(f"Error analyzing {repo_id}: {e}")
failed_analyses += 1
# Still wait to prevent rapid failures
time.sleep(0.2)
# Complete the progress
progress(1.0, desc="Batch analysis completed!")
# Get final updated dataframe
updated_df = read_csv_to_dataframe()
# Filter out rows with no analysis data for consistent display with top 3
analyzed_df = updated_df.copy()
analyzed_df = analyzed_df[
(analyzed_df['strength'].str.strip() != '') |
(analyzed_df['weaknesses'].str.strip() != '') |
(analyzed_df['speciality'].str.strip() != '') |
(analyzed_df['relevance rating'].str.strip() != '')
]
# Get top 3 most relevant repositories using full data
top_repos = get_top_relevant_repos(updated_df, user_requirements, top_n=3)
# Generate HTML links for repositories
all_links_html = generate_repo_links_html(analyzed_df)
top_links_html = generate_repo_links_html(top_repos) if not top_repos.empty else ""
# Final status with detailed breakdown
final_status = f"🎉 Batch Analysis Complete!\n✅ Successful: {successful_analyses}/{total_repos}\n❌ Failed: {failed_analyses}/{total_repos}"
if csv_update_failures > 0:
final_status += f"\n⚠️ CSV Update Issues: {csv_update_failures}/{total_repos}"
# Add top repos info if available
if not top_repos.empty:
final_status += f"\n\n🏆 Top {len(top_repos)} most relevant repositories selected!"
# Show top repos section if we have results
show_top_section = gr.update(visible=not top_repos.empty)
logger.info(f"Batch analysis completed: {successful_analyses} successful, {failed_analyses} failed, {csv_update_failures} CSV update issues")
return format_dataframe_for_display(analyzed_df), final_status, format_dataframe_for_display(top_repos), show_top_section, all_links_html, top_links_html
except Exception as e:
logger.error(f"Error in batch analysis: {e}")
error_status = f"❌ Batch analysis failed: {e}"
return format_dataframe_for_display(read_csv_to_dataframe()), error_status, pd.DataFrame(), gr.update(visible=False), "", ""
def handle_reset_everything() -> Tuple[List[str], int, str, pd.DataFrame, pd.DataFrame, Any, List[Dict[str, str]], str, str, str]:
"""Reset everything to initial state - clear all data, CSV, and UI components."""
try:
# Clear the CSV file
if os.path.exists(CSV_FILE):
os.remove(CSV_FILE)
logger.info("CSV file deleted for reset")
# Create empty dataframe
empty_df = pd.DataFrame(columns=["repo id", "strength", "weaknesses", "speciality", "relevance rating"])
# Reset state variables
repo_ids_reset = []
current_idx_reset = 0
user_requirements_reset = ""
# Reset status
status_reset = "Status: Everything has been reset. Ready to start fresh!"
# Reset UI components
current_requirements_reset = "No requirements extracted yet."
extracted_keywords_reset = ""
# Reset chatbot to initial message
chatbot_reset = [{"role": "assistant", "content": CHATBOT_INITIAL_MESSAGE}]
logger.info("Complete system reset performed")
return (
repo_ids_reset, # repo_ids_state
current_idx_reset, # current_repo_idx_state
user_requirements_reset, # user_requirements_state
empty_df, # df_output
empty_df, # top_repos_df
gr.update(visible=False), # top_repos_section
chatbot_reset, # chatbot
status_reset, # status_box_input
current_requirements_reset, # current_requirements_display
extracted_keywords_reset # extracted_keywords_output
)
except Exception as e:
logger.error(f"Error during reset: {e}")
error_status = f"Reset failed: {e}"
return (
[], # repo_ids_state
0, # current_repo_idx_state
"", # user_requirements_state
pd.DataFrame(), # df_output
pd.DataFrame(), # top_repos_df
gr.update(visible=False), # top_repos_section
[{"role": "assistant", "content": CHATBOT_INITIAL_MESSAGE}], # chatbot
error_status, # status_box_input
"No requirements extracted yet.", # current_requirements_display
"" # extracted_keywords_output
)
# --- Component Event Wiring ---
# Initialize chatbot with welcome message on app load
app.load(
fn=lambda: [{"role": "assistant", "content": CHATBOT_INITIAL_MESSAGE}],
outputs=[chatbot]
)
# Smart Input with Auto-processing
smart_input.submit(
fn=handle_smart_input,
inputs=[smart_input, auto_analyze_checkbox],
outputs=[repo_ids_state, current_repo_idx_state, df_output, status_box_input, tabs, status_box_input]
).then(
# If auto_analyze is enabled and we got repos, start analysis automatically
fn=lambda repo_ids, user_reqs, trigger: handle_analyze_all_repos(repo_ids, user_reqs) if trigger == "auto_analyze" and repo_ids else (pd.DataFrame(), "Ready for analysis.", pd.DataFrame(), gr.update(visible=False), "", ""),
inputs=[repo_ids_state, user_requirements_state, status_box_input],
outputs=[df_output, status_box_input, top_repos_df, top_repos_section, all_repo_links, top_repo_links]
)
# Smart Submit Button (same behavior as enter)
smart_submit_btn.click(
fn=handle_smart_input,
inputs=[smart_input, auto_analyze_checkbox],
outputs=[repo_ids_state, current_repo_idx_state, df_output, status_box_input, tabs, status_box_input]
).then(
# If auto_analyze is enabled and we got repos, start analysis automatically
fn=lambda repo_ids, user_reqs, trigger: handle_analyze_all_repos(repo_ids, user_reqs) if trigger == "auto_analyze" and repo_ids else (pd.DataFrame(), "Ready for analysis.", pd.DataFrame(), gr.update(visible=False), "", ""),
inputs=[repo_ids_state, user_requirements_state, status_box_input],
outputs=[df_output, status_box_input, top_repos_df, top_repos_section, all_repo_links, top_repo_links]
)
# Auto-analyze checkbox toggle
auto_analyze_checkbox.change(
fn=handle_auto_analyze_toggle,
inputs=[auto_analyze_checkbox],
outputs=[manual_analysis_row]
)
# Manual analysis button (when auto-analyze is disabled)
analyze_all_btn.click(
fn=handle_analyze_all_repos,
inputs=[repo_ids_state, user_requirements_state],
outputs=[df_output, status_box_analysis, top_repos_df, top_repos_section, all_repo_links, top_repo_links]
)
# Chatbot with Auto-extraction and Auto-search
msg_input.submit(
fn=handle_user_message,
inputs=[msg_input, chatbot],
outputs=[chatbot, msg_input]
).then(
fn=handle_bot_response,
inputs=[chatbot],
outputs=[chatbot, chat_status, extracted_keywords_output, user_requirements_state, repo_ids_state, current_repo_idx_state, df_output, tabs]
).then(
# Update requirements display when they change
fn=lambda req: req if req.strip() else "No specific requirements extracted from conversation.",
inputs=[user_requirements_state],
outputs=[current_requirements_display]
).then(
# If we got repos from chatbot, auto-analyze them
fn=lambda repo_ids, user_reqs: handle_analyze_all_repos(repo_ids, user_reqs) if repo_ids else (pd.DataFrame(), "", pd.DataFrame(), gr.update(visible=False), "", ""),
inputs=[repo_ids_state, user_requirements_state],
outputs=[df_output, chat_status, top_repos_df, top_repos_section, all_repo_links, top_repo_links]
)
send_btn.click(
fn=handle_user_message,
inputs=[msg_input, chatbot],
outputs=[chatbot, msg_input]
).then(
fn=handle_bot_response,
inputs=[chatbot],
outputs=[chatbot, chat_status, extracted_keywords_output, user_requirements_state, repo_ids_state, current_repo_idx_state, df_output, tabs]
).then(
# Update requirements display when they change
fn=lambda req: req if req.strip() else "No specific requirements extracted from conversation.",
inputs=[user_requirements_state],
outputs=[current_requirements_display]
).then(
# If we got repos from chatbot, auto-analyze them
fn=lambda repo_ids, user_reqs: handle_analyze_all_repos(repo_ids, user_reqs) if repo_ids else (pd.DataFrame(), "", pd.DataFrame(), gr.update(visible=False), "", ""),
inputs=[repo_ids_state, user_requirements_state],
outputs=[df_output, chat_status, top_repos_df, top_repos_section, all_repo_links, top_repo_links]
)
# Extract and Analyze Button (one-click solution for chatbot)
extract_analyze_btn.click(
fn=handle_extract_and_analyze,
inputs=[chatbot],
outputs=[chat_status, extracted_keywords_output, user_requirements_state, repo_ids_state, current_repo_idx_state, df_output, tabs, top_repos_df, status_box_analysis, top_repos_section, all_repo_links, top_repo_links]
).then(
# Update requirements display when they change
fn=lambda req: req if req.strip() else "No specific requirements extracted from conversation.",
inputs=[user_requirements_state],
outputs=[current_requirements_display]
)
# Repo Explorer Tab
setup_repo_explorer_events(repo_components, repo_states)
# Direct Repository Clicks - Show Modal (like old_app2.py)
df_output.select(
fn=handle_dataframe_select,
inputs=[df_output],
outputs=[selected_repo_display, repo_action_modal, selected_repo_id_state]
)
top_repos_df.select(
fn=handle_dataframe_select,
inputs=[top_repos_df],
outputs=[selected_repo_display, repo_action_modal, selected_repo_id_state]
)
# Modal button events (like old_app2.py)
visit_repo_btn.click(
fn=handle_visit_repo,
inputs=[selected_repo_display],
outputs=[repo_action_modal, selected_repo_display],
js="(repo_id) => { if(repo_id && repo_id.trim()) { window.open('https://huggingface.co/spaces/' + repo_id.trim(), '_blank'); } }"
)
explore_repo_btn.click(
fn=handle_explore_repo,
inputs=[selected_repo_id_state],
outputs=[
repo_action_modal,
tabs,
repo_components["repo_explorer_input"],
repo_states["current_repo_id"], # Set the current repo ID
status_box_input # Use for auto-load signal
],
js="""(repo_id) => {
console.log('DEBUG: Navigate to repo explorer for:', repo_id);
setTimeout(() => {
window.scrollTo({top: 0, behavior: 'smooth'});
}, 200);
}"""
).then(
# Auto-load the repository if the signal indicates to do so
fn=lambda repo_id, signal: handle_load_repository(repo_id) if signal == "auto_load" and repo_id else ("", ""),
inputs=[repo_states["current_repo_id"], status_box_input],
outputs=[repo_components["repo_status_display"], repo_states["repo_context_summary"]]
).then(
# Initialize the chatbot with welcome message after auto-loading
fn=lambda repo_status, repo_id, repo_context, signal: (
initialize_repo_chatbot(repo_status, repo_id, repo_context)
if signal == "auto_load" and repo_id else []
),
inputs=[repo_components["repo_status_display"], repo_states["current_repo_id"], repo_states["repo_context_summary"], status_box_input],
outputs=[repo_components["repo_chatbot"]]
)
cancel_modal_btn.click(
fn=handle_cancel_modal,
outputs=[repo_action_modal]
)
# Reset button event
reset_all_btn.click(
fn=handle_reset_everything,
outputs=[repo_ids_state, current_repo_idx_state, user_requirements_state, df_output, top_repos_df, top_repos_section, chatbot, status_box_input, current_requirements_display, extracted_keywords_output]
)
# Help modal events
help_btn.click(
fn=lambda: gr.update(visible=True),
outputs=[help_modal]
)
close_help_btn.click(
fn=lambda: gr.update(visible=False),
outputs=[help_modal]
)
return app
if __name__ == "__main__":
app = create_ui()
app.launch(debug=True)
|