forked from CodeGraphContext/CodeGraphContext
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathserver.py
More file actions
948 lines (825 loc) · 44.3 KB
/
server.py
File metadata and controls
948 lines (825 loc) · 44.3 KB
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
# src/codegraphcontext/server.py
import urllib.parse
import asyncio
import json
import logging
import importlib
import stdlibs
import sys
import traceback
import os
import re
from datetime import datetime
from pathlib import Path
from neo4j.exceptions import CypherSyntaxError
from dataclasses import asdict
from typing import Any, Dict, Coroutine, Optional
from .prompts import LLM_SYSTEM_PROMPT
from .core.database import DatabaseManager
from .core.jobs import JobManager, JobStatus
from .core.watcher import CodeWatcher
from .tools.graph_builder import GraphBuilder
from .tools.code_finder import CodeFinder
from .tools.import_extractor import ImportExtractor
from .utils.debug_log import debug_log
logger = logging.getLogger(__name__)
class MCPServer:
"""
The main MCP Server class.
This class orchestrates all the major components of the application, including:
- Database connection management (`DatabaseManager`)
- Background job tracking (`JobManager`)
- File system watching for live updates (`CodeWatcher`)
- Tool handlers for graph building, code searching, etc.
- The main JSON-RPC communication loop for interacting with an AI assistant.
"""
def __init__(self, loop=None):
"""
Initializes the MCP server and its components.
Args:
loop: The asyncio event loop to use. If not provided, it gets the current
running loop or creates a new one.
"""
try:
# Initialize the database manager and establish a connection early
# to fail fast if credentials are wrong.
self.db_manager = DatabaseManager()
self.db_manager.get_driver()
except ValueError as e:
raise ValueError(f"Database configuration error: {e}")
# Initialize managers for jobs and file watching.
self.job_manager = JobManager()
# Get the current event loop to pass to thread-sensitive components like the graph builder.
if loop is None:
try:
loop = asyncio.get_running_loop()
except RuntimeError:
loop = asyncio.new_event_loop()
asyncio.set_event_loop(loop)
self.loop = loop
# Initialize all the tool handlers, passing them the necessary managers and the event loop.
self.graph_builder = GraphBuilder(self.db_manager, self.job_manager, loop)
self.code_finder = CodeFinder(self.db_manager)
self.import_extractor = ImportExtractor()
self.code_watcher = CodeWatcher(self.graph_builder, self.job_manager)
# Define the tool manifest that will be exposed to the AI assistant.
self._init_tools()
def _init_tools(self):
"""
Defines the complete tool manifest for the LLM.
This dictionary contains the schema for every tool the AI can call,
including its name, description, and input parameters.
"""
self.tools = {
"add_code_to_graph": {
"name": "add_code_to_graph",
"description": "Performs a one-time scan of a local folder to add its code to the graph. Ideal for indexing libraries, dependencies, or projects not being actively modified. Returns a job ID for background processing.",
"inputSchema": {
"type": "object",
"properties": {
"path": {"type": "string", "description": "Path to the directory or file to add."},
"is_dependency": {"type": "boolean", "description": "Whether this code is a dependency.", "default": False}
},
"required": ["path"]
}
},
"check_job_status": {
"name": "check_job_status",
"description": "Check the status and progress of a background job.",
"inputSchema": {
"type": "object",
"properties": { "job_id": {"type": "string", "description": "Job ID from a previous tool call"} },
"required": ["job_id"]
}
},
"list_jobs": {
"name": "list_jobs",
"description": "List all background jobs and their current status.",
"inputSchema": {"type": "object", "properties": {}}
},
"find_code": {
"name": "find_code",
"description": "Find relevant code snippets related to a keyword (e.g., function name, class name, or content).",
"inputSchema": {
"type": "object",
"properties": { "query": {"type": "string", "description": "Keyword or phrase to search for"} },
"required": ["query"]
}
},
"analyze_code_relationships": {
"name": "analyze_code_relationships",
"description": "Analyze code relationships like 'who calls this function' or 'class hierarchy'. Supported query types include: find_callers, find_callees, find_all_callers, find_all_callees, find_importers, who_modifies, class_hierarchy, overrides, dead_code, call_chain, module_deps, variable_scope, find_complexity, find_functions_by_argument, find_functions_by_decorator.",
"inputSchema": {
"type": "object",
"properties": {
"query_type": {"type": "string", "description": "Type of relationship query to run.", "enum": ["find_callers", "find_callees", "find_all_callers", "find_all_callees", "find_importers", "who_modifies", "class_hierarchy", "overrides", "dead_code", "call_chain", "module_deps", "variable_scope", "find_complexity", "find_functions_by_argument", "find_functions_by_decorator"]},
"target": {"type": "string", "description": "The function, class, or module to analyze."},
"context": {"type": "string", "description": "Optional: specific file path for precise results."}
},
"required": ["query_type", "target"]
}
},
"watch_directory": {
"name": "watch_directory",
"description": "Performs an initial scan of a directory and then continuously monitors it for changes, automatically keeping the graph up-to-date. Ideal for projects under active development. Returns a job ID for the initial scan.",
"inputSchema": {
"type": "object",
"properties": { "path": {"type": "string", "description": "Path to directory to watch"} },
"required": ["path"]
}
},
"execute_cypher_query": {
"name": "execute_cypher_query",
"description": "Fallback tool to run a direct, read-only Cypher query against the code graph. Use this for complex questions not covered by other tools. The graph contains nodes representing code structures and relationships between them. **Schema Overview:**\n- **Nodes:** `Repository`, `File`, `Module`, `Class`, `Function`.\n- **Properties:** Nodes have properties like `name`, `path`, `cyclomatic_complexity` (on Function nodes), and `code`.\n- **Relationships:** `CONTAINS` (e.g., File-[:CONTAINS]->Function), `CALLS` (Function-[:CALLS]->Function or File-[:CALLS]->Function), `IMPORTS` (File-[:IMPORTS]->Module), `INHERITS` (Class-[:INHERITS]->Class).",
"inputSchema": {
"type": "object",
"properties": { "cypher_query": {"type": "string", "description": "The read-only Cypher query to execute."} },
"required": ["cypher_query"]
}
},
"add_package_to_graph": {
"name": "add_package_to_graph",
"description": "Add a Python package to Neo4j graph by discovering its location. Returns immediately with job ID.",
"inputSchema": {
"type": "object",
"properties": {
"package_name": {"type": "string", "description": "Name of the Python package to add (e.g., 'requests')"},
"is_dependency": {"type": "boolean", "description": "Mark as a dependency", "default": True}
},
"required": ["package_name"]
}
},
"list_imports": {
"name": "list_imports",
"description": "Extract all package imports from code files in a directory or file",
"inputSchema": {
"type": "object",
"properties": {
"path": {"type": "string", "description": "Path to file or directory to analyze"},
"language": {"type": "string", "description": "Programming language (python, javascript, etc.)", "default": "python"},
"recursive": {"type": "boolean", "description": "Whether to analyze subdirectories recursively", "default": True}
},
"required": ["path"]
}
},
"find_dead_code": {
"name": "find_dead_code",
"description": "Find potentially unused functions (dead code) across the entire indexed codebase, optionally excluding functions with specific decorators.",
"inputSchema": {
"type": "object",
"properties": {
"exclude_decorated_with": {"type": "array", "items": {"type": "string"}, "description": "Optional: A list of decorator names (e.g., '@app.route') to exclude from dead code detection.", "default": []}
}
}
},
"calculate_cyclomatic_complexity": {
"name": "calculate_cyclomatic_complexity",
"description": "Calculate the cyclomatic complexity of a specific function to measure its complexity.",
"inputSchema": {
"type": "object",
"properties": {
"function_name": {"type": "string", "description": "The name of the function to analyze."},
"file_path": {"type": "string", "description": "Optional: The full path to the file containing the function for a more specific query."}
},
"required": ["function_name"]
}
},
"find_most_complex_functions": {
"name": "find_most_complex_functions",
"description": "Find the most complex functions in the codebase based on cyclomatic complexity.",
"inputSchema": {
"type": "object",
"properties": {
"limit": {"type": "integer", "description": "The maximum number of complex functions to return.", "default": 10}
}
}
},
"list_indexed_repositories": {
"name": "list_indexed_repositories",
"description": "List all indexed repositories.",
"inputSchema": {
"type": "object",
"properties": {}
}
},
"delete_repository": {
"name": "delete_repository",
"description": "Delete an indexed repository from the graph.",
"inputSchema": {
"type": "object",
"properties": {
"repo_path": {"type": "string", "description": "The path of the repository to delete."}
},
"required": ["repo_path"]
}
},
"visualize_graph_query": {
"name": "visualize_graph_query",
"description": "Generates a URL to visualize the results of a Cypher query in the Neo4j Browser. The user can open this URL in their web browser to see the graph visualization.",
"inputSchema": {
"type": "object",
"properties": {
"cypher_query": {"type": "string", "description": "The Cypher query to visualize."}
},
"required": ["cypher_query"]
}
},
"list_watched_paths": {
"name": "list_watched_paths",
"description": "Lists all directories currently being watched for live file changes.",
"inputSchema": {"type": "object", "properties": {}}
},
"unwatch_directory": {
"name": "unwatch_directory",
"description": "Stops watching a directory for live file changes.",
"inputSchema": {
"type": "object",
"properties": {
"path": {"type": "string", "description": "The absolute path of the directory to stop watching."}
},
"required": ["path"]
}
}
}
def get_database_status(self) -> dict:
"""Returns the current connection status of the Neo4j database."""
return {"connected": self.db_manager.is_connected()}
def get_local_package_path(self, package_name: str) -> Optional[str]:
"""
Finds the local installation path of a Python package.
This method uses `importlib` to locate a package and determines its root
directory, handling both regular packages (directories with __init__.py)
and single-file modules.
Args:
package_name: The name of the package to locate (e.g., "requests").
Returns:
The absolute path to the package's directory as a string, or None if not found.
"""
try:
debug_log(f"Getting local path for package: {package_name}")
module = importlib.import_module(package_name)
if hasattr(module, '__file__') and module.__file__:
module_file = Path(module.__file__)
debug_log(f"Module file: {module_file}")
if module_file.name == '__init__.py':
# For a package, the path is the parent directory of __init__.py
package_path = str(module_file.parent)
elif package_name in stdlibs.module_names:
# For a standard library single file module, the path is the file itself
package_path = str(module_file)
else:
# For other single-file modules, assume the parent directory is the container
package_path = str(module_file.parent)
debug_log(f"Determined package path: {package_path}")
return package_path
elif hasattr(module, '__path__'):
# This handles namespace packages which may not have an __init__.py
if isinstance(module.__path__, list) and module.__path__:
package_path = str(Path(module.__path__[0]))
debug_log(f"Package path from __path__: {package_path}")
return package_path
else:
# Fallback for other __path__ formats
package_path = str(Path(str(module.__path__)))
debug_log(f"Package path from __path__ (str): {package_path}")
return package_path
debug_log(f"Could not determine path for {package_name}")
return None
except ImportError as e:
debug_log(f"Could not import {package_name}: {e}")
return None
except Exception as e:
debug_log(f"Error getting local path for {package_name}: {e}")
return None
def execute_cypher_query_tool(self, **args) -> Dict[str, Any]:
"""
Tool implementation for executing a read-only Cypher query.
Important: Includes a safety check to prevent any database modification
by disallowing keywords like CREATE, MERGE, DELETE, etc.
"""
cypher_query = args.get("cypher_query")
if not cypher_query:
return {"error": "Cypher query cannot be empty."}
# Safety Check: Prevent any write operations to the database.
# This check first removes all string literals and then checks for forbidden keywords.
forbidden_keywords = ['CREATE', 'MERGE', 'DELETE', 'SET', 'REMOVE', 'DROP', 'CALL apoc']
# Regex to match single or double quoted strings, handling escaped quotes.
string_literal_pattern = r'"(?:\\.|[^"\\])*"|\'(?:\\.|[^\'\\])*\''
# Remove all string literals from the query.
query_without_strings = re.sub(string_literal_pattern, '', cypher_query)
# Now, check for forbidden keywords in the query without strings.
for keyword in forbidden_keywords:
if re.search(r'\b' + keyword + r'\b', query_without_strings, re.IGNORECASE):
return {
"error": "This tool only supports read-only queries. Prohibited keywords like CREATE, MERGE, DELETE, SET, etc., are not allowed."
}
try:
debug_log(f"Executing Cypher query: {cypher_query}")
with self.db_manager.get_driver().session() as session:
result = session.run(cypher_query)
# Convert results to a list of dictionaries for clean JSON serialization.
records = [record.data() for record in result]
return {
"success": True,
"query": cypher_query,
"record_count": len(records),
"results": records
}
except CypherSyntaxError as e:
debug_log(f"Cypher syntax error: {str(e)}")
return {
"error": "Cypher syntax error.",
"details": str(e),
"query": cypher_query
}
except Exception as e:
debug_log(f"Error executing Cypher query: {str(e)}")
return {
"error": "An unexpected error occurred while executing the query.",
"details": str(e)
}
def find_dead_code_tool(self, **args) -> Dict[str, Any]:
"""Tool to find potentially dead code across the entire project."""
exclude_decorated_with = args.get("exclude_decorated_with", [])
try:
debug_log("Finding dead code.")
results = self.code_finder.find_dead_code(exclude_decorated_with=exclude_decorated_with)
return {
"success": True,
"query_type": "dead_code",
"results": results
}
except Exception as e:
debug_log(f"Error finding dead code: {str(e)}")
return {"error": f"Failed to find dead code: {str(e)}"}
def calculate_cyclomatic_complexity_tool(self, **args) -> Dict[str, Any]:
"""Tool to calculate cyclomatic complexity for a given function."""
function_name = args.get("function_name")
file_path = args.get("file_path")
try:
debug_log(f"Calculating cyclomatic complexity for function: {function_name}")
results = self.code_finder.get_cyclomatic_complexity(function_name, file_path)
response = {
"success": True,
"function_name": function_name,
"results": results
}
if file_path:
response["file_path"] = file_path
return response
except Exception as e:
debug_log(f"Error calculating cyclomatic complexity: {str(e)}")
return {"error": f"Failed to calculate cyclomatic complexity: {str(e)}"}
def find_most_complex_functions_tool(self, **args) -> Dict[str, Any]:
"""Tool to find the most complex functions."""
limit = args.get("limit", 10)
try:
debug_log(f"Finding the top {limit} most complex functions.")
results = self.code_finder.find_most_complex_functions(limit)
return {
"success": True,
"limit": limit,
"results": results
}
except Exception as e:
debug_log(f"Error finding most complex functions: {str(e)}")
return {"error": f"Failed to find most complex functions: {str(e)}"}
def list_indexed_repositories_tool(self, **args) -> Dict[str, Any]:
"""Tool to list indexed repositories."""
try:
debug_log("Listing indexed repositories.")
results = self.code_finder.list_indexed_repositories()
return {
"success": True,
"repositories": results
}
except Exception as e:
debug_log(f"Error listing indexed repositories: {str(e)}")
return {"error": f"Failed to list indexed repositories: {str(e)}"}
def delete_repository_tool(self, **args) -> Dict[str, Any]:
"""Tool to delete a repository from the graph."""
repo_path = args.get("repo_path")
try:
debug_log(f"Deleting repository: {repo_path}")
self.graph_builder.delete_repository_from_graph(repo_path)
return {
"success": True,
"message": f"Repository '{repo_path}' deleted successfully."
}
except Exception as e:
debug_log(f"Error deleting repository: {str(e)}")
return {"error": f"Failed to delete repository: {str(e)}"}
def visualize_graph_query_tool(self, **args) -> Dict[str, Any]:
"""Tool to generate a Neo4j browser visualization URL for a Cypher query."""
cypher_query = args.get("cypher_query")
if not cypher_query:
return {"error": "Cypher query cannot be empty."}
try:
encoded_query = urllib.parse.quote(cypher_query)
visualization_url = f"http://localhost:7474/browser/?cmd=edit&arg={encoded_query}"
return {
"success": True,
"visualization_url": visualization_url,
"message": "Open the URL in your browser to visualize the graph query. The query will be pre-filled for editing."
}
except Exception as e:
debug_log(f"Error generating visualization URL: {str(e)}")
return {"error": f"Failed to generate visualization URL: {str(e)}"}
def list_watched_paths_tool(self, **args) -> Dict[str, Any]:
"""Tool to list all currently watched directory paths."""
try:
paths = self.code_watcher.list_watched_paths()
return {"success": True, "watched_paths": paths}
except Exception as e:
return {"error": f"Failed to list watched paths: {str(e)}"}
def unwatch_directory_tool(self, **args) -> Dict[str, Any]:
"""Tool to stop watching a directory."""
path = args.get("path")
if not path:
return {"error": "Path is a required argument."}
# The watcher class handles the logic of checking if the path is watched
# and returns an error dictionary if not, so we can just call it.
return self.code_watcher.unwatch_directory(path)
def watch_directory_tool(self, **args) -> Dict[str, Any]:
"""
Tool implementation to start watching a directory for changes.
This tool is now smart: it checks if the path exists and if it has already been indexed.
"""
path = args.get("path")
if not path:
return {"error": "Path is a required argument."}
path_obj = Path(path).resolve()
path_str = str(path_obj)
# 1. Validate the path before the try...except block
if not path_obj.is_dir():
return {
"success": True,
"status": "path_not_found",
"message": f"Path '{path_str}' does not exist or is not a directory."
}
try:
# Check if already watching
if path_str in self.code_watcher.watched_paths:
return {"success": True, "message": f"Already watching directory: {path_str}"}
# 2. Check if the repository is already indexed
indexed_repos = self.list_indexed_repositories_tool().get("repositories", [])
is_already_indexed = any(Path(repo["path"]).resolve() == path_obj for repo in indexed_repos)
# 3. Decide whether to perform an initial scan
if is_already_indexed:
# If already indexed, just start the watcher without a scan
self.code_watcher.watch_directory(path_str, perform_initial_scan=False)
return {
"success": True,
"message": f"Path '{path_str}' is already indexed. Now watching for live changes."
}
else:
# If not indexed, perform the scan AND start the watcher
scan_job_result = self.add_code_to_graph_tool(path=path_str, is_dependency=False)
if "error" in scan_job_result:
return scan_job_result
self.code_watcher.watch_directory(path_str, perform_initial_scan=True)
return {
"success": True,
"message": f"Path '{path_str}' was not indexed. Started initial scan and now watching for live changes.",
"job_id": scan_job_result.get("job_id"),
"details": "Use check_job_status to monitor the initial scan."
}
B94A
except Exception as e:
logger.error(f"Failed to start watching directory {path}: {e}")
return {"error": f"Failed to start watching directory: {str(e)}"}
def list_imports_tool(self, **args) -> Dict[str, Any]:
"""Tool to list all imports from code files"""
path = args.get("path")
language = args.get("language", "python")
recursive = args.get("recursive", True)
all_imports = set()
file_extensions = {
'python': ['.py'], 'javascript': ['.js', '.jsx', '.mjs'],
'typescript': ['.ts', '.tsx'], 'java': ['.java'],
}
extensions = file_extensions.get(language, ['.py'])
extract_func = {
'python': self.import_extractor.extract_python_imports,
'javascript': self.import_extractor.extract_javascript_imports,
'typescript': self.import_extractor.extract_javascript_imports,
'java': self.import_extractor.extract_java_imports,
}.get(language, self.import_extractor.extract_python_imports)
try:
path_obj = Path(path)
if path_obj.is_file():
if any(str(path_obj).endswith(ext) for ext in extensions):
all_imports.update(extract_func(str(path_obj)))
elif path_obj.is_dir():
pattern = "**/*" if recursive else "*"
for ext in extensions:
for file_path in path_obj.glob(f"{pattern}{ext}"):
if file_path.is_file():
all_imports.update(extract_func(str(file_path)))
else:
return {"error": f"Path {path} does not exist"}
# Removed standard library filtering as per user request.
# if language == 'python':
# # Get the list of stdlib modules for the current Python version
# stdlib_modules = set(stdlibs.module_names)
# all_imports = all_imports - stdlib_modules
return {
"imports": sorted(list(all_imports)), "language": language,
"path": path, "count": len(all_imports)
}
except Exception as e:
return {"error": f"Failed to analyze imports: {str(e)}"}
def add_code_to_graph_tool(self, **args) -> Dict[str, Any]:
"""
Tool implementation to index a directory of code.
This creates a background job and runs the indexing asynchronously
so the AI assistant can continue to be responsive.
"""
path = args.get("path")
is_dependency = args.get("is_dependency", False)
try:
path_obj = Path(path).resolve()
if not path_obj.exists():
return {
"success": True,
"status": "path_not_found",
"message": f"Path '{path}' does not exist."
}
# Prevent re-indexing the same repository.
indexed_repos = self.list_indexed_repositories_tool().get("repositories", [])
for repo in indexed_repos:
if Path(repo["path"]).resolve() == path_obj:
return {
"success": False,
"message": f"Repository '{path}' is already indexed."
}
# Estimate time and create a job for the user to track.
total_files, estimated_time = self.graph_builder.estimate_processing_time(path_obj)
job_id = self.job_manager.create_job(str(path_obj), is_dependency)
self.job_manager.update_job(job_id, total_files=total_files, estimated_duration=estimated_time)
# Create the coroutine for the background task and schedule it on the main event loop.
coro = self.graph_builder.build_graph_from_path_async(
path_obj, is_dependency, job_id
)
asyncio.run_coroutine_threadsafe(coro, self.loop)
debug_log(f"Started background job {job_id} for path: {str(path_obj)}, is_dependency: {is_dependency}")
return {
"success": True, "job_id": job_id,
"message": f"Background processing started for {str(path_obj)}",
"estimated_files": total_files,
"estimated_duration_seconds": round(estimated_time, 2),
"estimated_duration_human": f"{int(estimated_time // 60)}m {int(estimated_time % 60)}s" if estimated_time >= 60 else f"{int(estimated_time)}s",
"instructions": f"Use 'check_job_status' with job_id '{job_id}' to monitor progress"
}
except Exception as e:
debug_log(f"Error creating background job: {str(e)}")
return {"error": f"Failed to start background processing: {str(e)}"}
def add_package_to_graph_tool(self, **args) -> Dict[str, Any]:
"""Tool to add a Python package to Neo4j graph by auto-discovering its location"""
package_name = args.get("package_name")
is_dependency = args.get("is_dependency", True)
try:
# Check if the package is already indexed
indexed_repos = self.list_indexed_repositories_tool().get("repositories", [])
for repo in indexed_repos:
if repo.get("is_dependency") and (repo.get("name") == package_name or repo.get("name") == f"{package_name}.py"):
return {
"success": False,
"message": f"Package '{package_name}' is already indexed."
}
package_path = self.get_local_package_path(package_name)
if not package_path:
return {"error": f"Could not find package '{package_name}'. Make sure it's installed."}
if not os.path.exists(package_path):
return {"error": f"Package path '{package_path}' does not exist"}
path_obj = Path(package_path)
total_files, estimated_time = self.graph_builder.estimate_processing_time(path_obj)
job_id = self.job_manager.create_job(package_path, is_dependency)
self.job_manager.update_job(job_id, total_files=total_files, estimated_duration=estimated_time)
coro = self.graph_builder.build_graph_from_path_async(
path_obj, is_dependency, job_id
)
asyncio.run_coroutine_threadsafe(coro, self.loop)
debug_log(f"Started background job {job_id} for package: {package_name} at {package_path}, is_dependency: {is_dependency}")
return {
"success": True, "job_id": job_id, "package_name": package_name,
"discovered_path": package_path,
"message": f"Background processing started for package '{package_name}'",
"estimated_files": total_files,
"estimated_duration_seconds": round(estimated_time, 2),
"estimated_duration_human": f"{int(estimated_time // 60)}m {int(estimated_time % 60)}s" if estimated_time >= 60 else f"{int(estimated_time)}s",
"instructions": f"Use 'check_job_status' with job_id '{job_id}' to monitor progress"
}
except Exception as e:
debug_log(f"Error creating background job for package {package_name}: {str(e)}")
return {"error": f"Failed to start background processing for package '{package_name}': {str(e)}"}
def check_job_status_tool(self, **args) -> Dict[str, Any]:
"""Tool to check job status"""
job_id = args.get("job_id")
if not job_id:
return {"error": "Job ID is a required argument."}
try:
job = self.job_manager.get_job(job_id)
if not job:
return {
"success": True, # Return success to avoid generic error wrapper
"
5276
status": "not_found",
"message": f"Job with ID '{job_id}' not found. The ID may be incorrect or the job may have been cleared after a server restart."
}
job_dict = asdict(job)
if job.status == JobStatus.RUNNING:
if job.estimated_time_remaining:
remaining = job.estimated_time_remaining
job_dict["estimated_time_remaining_human"] = (
f"{int(remaining // 60)}m {int(remaining % 60)}s"
if remaining >= 60 else f"{int(remaining)}s"
)
if job.start_time:
elapsed = (datetime.now() - job.start_time).total_seconds()
job_dict["elapsed_time_human"] = (
f"{int(elapsed // 60)}m {int(elapsed % 60)}s"
if elapsed >= 60 else f"{int(elapsed)}s"
)
elif job.status == JobStatus.COMPLETED and job.start_time and job.end_time:
duration = (job.end_time - job.start_time).total_seconds()
job_dict["actual_duration_human"] = (
f"{int(duration // 60)}m {int(duration % 60)}s"
if duration >= 60 else f"{int(duration)}s"
)
job_dict["start_time"] = job.start_time.strftime("%Y-%m-%d %H:%M:%S")
if job.end_time:
job_dict["end_time"] = job.end_time.strftime("%Y-%m-%d %H:%M:%S")
job_dict["status"] = job.status.value
return {"success": True, "job": job_dict}
except Exception as e:
debug_log(f"Error checking job status: {str(e)}")
return {"error": f"Failed to check job status: {str(e)}"}
def list_jobs_tool(self) -> Dict[str, Any]:
"""Tool to list all jobs"""
try:
jobs = self.job_manager.list_jobs()
jobs_data = []
for job in jobs:
job_dict = asdict(job)
job_dict["status"] = job.status.value
job_dict["start_time"] = job.start_time.strftime("%Y-%m-%d %H:%M:%S")
if job.end_time:
job_dict["end_time"] = job.end_time.strftime("%Y-%m-%d %H:%M:%S")
jobs_data.append(job_dict)
jobs_data.sort(key=lambda x: x["start_time"], reverse=True)
return {"success": True, "jobs": jobs_data, "total_jobs": len(jobs_data)}
except Exception as e:
debug_log(f"Error listing jobs: {str(e)}")
return {"error": f"Failed to list jobs: {str(e)}"}
def analyze_code_relationships_tool(self, **args) -> Dict[str, Any]:
"""Tool to analyze code relationships"""
query_type = args.get("query_type")
target = args.get("target")
context = args.get("context")
if not query_type or not target:
return {
"error": "Both 'query_type' and 'target' are required",
"supported_query_types": [
"find_callers", "find_callees", "find_importers", "who_modifies",
"class_hierarchy", "overrides", "dead_code", "call_chain",
"module_deps", "variable_scope", "find_complexity"
]
}
try:
debug_log(f"Analyzing relationships: {query_type} for {target}")
results = self.code_finder.analyze_code_relationships(query_type, target, context)
return {
"success": True, "query_type": query_type, "target": target,
"context": context, "results": results
}
except Exception as e:
debug_log(f"Error analyzing relationships: {str(e)}")
return {"error": f"Failed to analyze relationships: {str(e)}"}
def find_code_tool(self, **args) -> Dict[str, Any]:
"""Tool to find relevant code snippets"""
query = args.get("query")
try:
debug_log(f"Finding code for query: {query}")
results = self.code_finder.find_related_code(query)
return {"success": True, "query": query, "results": results}
except Exception as e:
debug_log(f"Error finding code: {str(e)}")
return {"error": f"Failed to find code: {str(e)}"}
async def handle_tool_call(self, tool_name: str, args: Dict[str, Any]) -> Dict[str, Any]:
"""
Routes a tool call from the AI assistant to the appropriate handler function.
Args:
tool_name: The name of the tool to execute.
args: A dictionary of arguments for the tool.
Returns:
A dictionary containing the result of the tool execution.
"""
tool_map: Dict[str, Coroutine] = {
"list_imports": self.list_imports_tool,
"add_package_to_graph": self.add_package_to_graph_tool,
"find_dead_code": self.find_dead_code_tool,
"find_code": self.find_code_tool,
"analyze_code_relationships": self.analyze_code_relationships_tool,
"watch_directory": self.watch_directory_tool,
"execute_cypher_query": self.execute_cypher_query_tool,
"add_code_to_graph": self.add_code_to_graph_tool,
"check_job_status": self.check_job_status_tool,
"list_jobs": self.list_jobs_tool,
"calculate_cyclomatic_complexity": self.calculate_cyclomatic_complexity_tool,
"find_most_complex_functions": self.find_most_complex_functions_tool,
"list_indexed_repositories": self.list_indexed_repositories_tool,
"delete_repository": self.delete_repository_tool,
"visualize_graph_query": self.visualize_graph_query_tool,
"list_watched_paths": self.list_watched_paths_tool,
"unwatch_directory": self.unwatch_directory_tool
}
handler = tool_map.get(tool_name)
if handler:
# Run the synchronous tool function in a separate thread to avoid
# blocking the main asyncio event loop.
return await asyncio.to_thread(handler, **args)
else:
return {"error": f"Unknown tool: {tool_name}"}
async def run(self):
"""
Runs the main server loop, listening for JSON-RPC requests from stdin.
This loop continuously reads lines from stdin, parses them as JSON-RPC
requests, and routes them to the appropriate handlers (e.g., initialize,
tools/list, tools/call). The response is then printed to stdout.
"""
logger.info("MCP Server is running. Waiting for requests...")
self.code_watcher.start()
loop = asyncio.get_event_loop()
while True:
try:
# Read a request from the standard input.
line = await loop.run_in_executor(None, sys.stdin.readline)
if not line:
logger.info("Client disconnected (EOF received). Shutting down.")
break
request = json.loads(line.strip())
method = request.get('method')
params = request.get('params', {})
request_id = request.get('id')
response = {}
# Route the request based on the JSON-RPC method.
if method == 'initialize':
response = {
"jsonrpc": "2.0", "id": request_id,
"result": {
"protocolVersion": "2025-03-26",
"serverInfo": {
"name": "CodeGraphContext", "version": "0.1.0",
"systemPrompt": LLM_SYSTEM_PROMPT
},
"capabilities": {"tools": {"listTools": True}},
}
}
elif method == 'tools/list':
# Return the list of tools defined in _init_tools.
response = {
"jsonrpc": "2.0", "id": request_id,
"result": {"tools": list(self.tools.values())}
}
elif method == 'tools/call':
# Execute a tool call and return the result.
tool_name = params.get('name')
args = params.get('arguments', {})
result = await self.handle_tool_call(tool_name, args)
if "error" in result:
response = {
"jsonrpc": "2.0", "id": request_id,
"error": {"code": -32000, "message": "Tool execution error", "data": result}
}
else:
response = {
"jsonrpc": "2.0", "id": request_id,
"result": {"content": [{"type": "text", "text": json.dumps(result, indent=2)}]}
}
elif method == 'notifications/initialized':
# This is a notification, no response needed.
pass
else:
# Handle unknown methods.
if request_id is not None:
response = {
"jsonrpc": "2.0", "id": request_id,
"error": {"code": -32601, "message": f"Method not found: {method}"}
}
# Send the response to standard output if it's not a notification.
if request_id is not None and response:
print(json.dumps(response), flush=True)
except Exception as e:
logger.error(f"Error processing request: {e}\n{traceback.format_exc()}")
request_id = "unknown"
if 'request' in locals() and isinstance(request, dict):
request_id = request.get('id', "unknown")
error_response = {
"jsonrpc": "2.0", "id": request_id,
"error": {"code": -32603, "message": f"Internal error: {str(e)}", "data": traceback.format_exc()}
}
print(json.dumps(error_response), flush=True)
def shutdown(self):
"""Gracefully shuts down the server and its components."""
logger.info("Shutting down server...")
self.code_watcher.stop()
self.db_manager.close_driver()