Repository navigation
Expand file tree
/
Copy path__init__.py
More file actions
2926 lines (2551 loc) · 127 KB
/
Copy path__init__.py
File metadata and controls
2926 lines (2551 loc) · 127 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
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
import atexit
import copy
import fnmatch
import importlib
import json
import logging
import os
import random
import re
import shutil
import sys
import tarfile
import tempfile
import time
from dataclasses import asdict
from pathlib import PurePath
from typing import TYPE_CHECKING, Dict, Iterator, List, NamedTuple, Optional, Set, Tuple
if TYPE_CHECKING:
from socketsecurity.config import CliConfig
from git import Repo
from socketdev import socketdev
from socketdev.exceptions import APIFailure
from socketdev.fullscans import DiffArtifacts, FullScanParams, SocketArtifact
from socketdev.org import Organization
from socketdev.repos import RepositoryInfo
from socketsecurity import USER_AGENT, __version__
from socketsecurity.core.classes import Alert, Diff, FullScan, Issue, Package, Purl
from socketsecurity.core.exceptions import APIResourceNotFound
from .resource_utils import check_file_count_against_ulimit
from .socket_config import SocketConfig
from .utils import socket_globs
logging_std = importlib.import_module("logging")
__all__ = [
"Core",
"log",
"__version__",
"USER_AGENT",
]
version = __version__
log = logging.getLogger("socketdev")
_ALERT_TYPE_TITLE_OVERRIDES = {
"gptDidYouMean": "Possible typosquat attack (GPT)",
}
_HUMANIZE_BOUNDARY = re.compile(r"(?<=[a-z0-9])(?=[A-Z])|(?<=[A-Z])(?=[A-Z][a-z])")
# How many full scans to request when resolving a diff baseline. The newest scan is
# usually the one we want, but temporary scans have to be skipped (see
# Core.newest_persisted_scan_id), so a single result is not enough.
SCAN_LOOKUP_PAGE_SIZE = 10
# Bounds on the search for a scanned ancestor when the requested baseline commit has
# no full scan of its own. Full-scan listing pages are matched against bounded local
# history, so the local walk cannot grow without limit.
ANCESTOR_SCAN_LOOKUP_LIMIT = 100
ANCESTOR_WALK_MAX_DEPTH = 100
# Reachability facts-file upload compression.
#
# The Socket full-scan endpoint transparently brotli-decompresses any multipart part
# whose basename is exactly ``.socket.facts.json.br`` and stores it as plain
# ``.socket.facts.json``. Compressing the facts file on upload keeps it well under the
# server's per-file size cap (a ~262 MB facts file compresses to roughly 15-30 MB),
# which is required for large full application reachability scans to succeed.
#
# The server matches the *exact* name ``.socket.facts.json.br``, so we only compress
# files whose basename is exactly ``.socket.facts.json`` (a custom ``--reach-output-file``
# name would not be decompressed server-side, so it is left as a plain upload).
SOCKET_FACTS_FILENAME = ".socket.facts.json"
SOCKET_FACTS_BROTLI_FILENAME = ".socket.facts.json.br"
# Brotli quality (0-11); 5 is a good speed/ratio tradeoff for large JSON payloads.
SOCKET_FACTS_BROTLI_QUALITY = 5
# Largest brotli window (2**24 bytes); improves the ratio on large facts files.
SOCKET_FACTS_BROTLI_LGWIN = 24
# Stream the facts file in 1 MiB chunks so large files aren't held fully in memory.
SOCKET_FACTS_BROTLI_CHUNK_SIZE = 1024 * 1024
# Placeholder facts document (see empty_head_scan_file). A zero-byte file does not parse,
# and the API answers an unparseable facts file with the marker artifact below.
SOCKET_FACTS_EMPTY_DOCUMENT = '{"components": []}'
# Synthetic artifact the API adds when an uploaded ``.socket.facts.json`` could not be
# parsed. A diagnostic, not a dependency, so it is dropped from scan results and reported
# as a warning instead.
INVALID_FACTS_MARKER_TYPE = "generic"
INVALID_FACTS_MARKER_NAME = "invalid-socket-facts"
# Full application reachability finalize retry policy. The finalize call links the reachability
# scan to the full scan and can fail transiently (network/API blips); a few backoff retries make it robust.
TIER1_FINALIZE_MAX_ATTEMPTS = 3
TIER1_FINALIZE_BACKOFF_SECONDS = 1.0
# Full scan upload retry policy. An upload can fail transiently at the gateway/connection
# level (an HTTP 502/503/504/408, a dropped or reset connection, or a client-side timeout)
# without the server having created the scan; the SDK classifies those failures via
# APIFailure.is_transient_error() (socketdev>=3.3.0). In these failure modes no scan was
# created, so a retry does not duplicate one. (A duplicate is possible only if a gateway
# timeout races a request the server later completes; that is benign - the retried scan
# supersedes the orphaned one, same as running the CLI twice.)
#
# Each schedule entry is the wait before the next attempt once the current one fails (plus
# a little jitter so a fleet of CI jobs hitting the same failure doesn't retry in
# lock-step); the final None means the last attempt's failure is re-raised, not retried.
FULL_SCAN_UPLOAD_BACKOFF_SCHEDULE_SECONDS = (10.0, 30.0, None)
FULL_SCAN_UPLOAD_MAX_ATTEMPTS = len(FULL_SCAN_UPLOAD_BACKOFF_SCHEDULE_SECONDS)
FULL_SCAN_UPLOAD_BACKOFF_JITTER_SECONDS = 2.0
# Statuses that mean the request is too large to process. Scan metadata travels in the
# query string of the full-scan POST, while manifests travel in its multipart body. A
# 413 can refer to either part; 414 points to the URL, and some proxies report 431 when
# the encoded request target exceeds their header limit. None are transient.
REQUEST_TOO_LARGE_STATUS_CODES = (413, 414, 431)
# Diff-scan polling policy. The legacy scan comparison (fullscans.stream_diff) holds a
# single HTTP connection open, fully idle, while the backend computes the diff; network
# middleboxes with TCP idle timeouts (notably Azure NAT gateways, which default to
# 4 minutes) kill that connection with a RST, surfacing as an intermittent
# ConnectionResetError on large scans. The diff-scans flow instead creates a
# diff-scan resource and polls its cached endpoint with short bounded requests: the API
# answers 202 while the comparison is still computing and 200 with the result once it is
# ready, so no connection is ever idle long enough to be reaped.
#
# Each poll consumes 1 unit of API quota, so the interval backs off toward
# DIFF_SCAN_POLL_MAX_INTERVAL_SECONDS to stay quota-friendly on comparisons that take
# minutes to compute. The timeout is a backstop against a diff scan that never
# completes; on expiry (or any other failure of this flow) the caller falls back to the
# legacy streaming comparison rather than failing the scan outright.
DIFF_SCAN_POLL_INITIAL_INTERVAL_SECONDS = 5.0
DIFF_SCAN_POLL_MAX_INTERVAL_SECONDS = 10.0
DIFF_SCAN_POLL_BACKOFF_MULTIPLIER = 1.5
DIFF_SCAN_POLL_TIMEOUT_SECONDS = 30 * 60.0
# Temp dirs holding placeholder facts files (see Core.empty_head_scan_file). Call sites unlink
# the file itself once the upload finishes; the now-empty directory is removed at process exit
# so a run that raises mid-scan doesn't leak one.
_PLACEHOLDER_FACTS_DIRS: List[str] = []
@atexit.register
def _cleanup_placeholder_facts_dirs() -> None:
for placeholder_dir in _PLACEHOLDER_FACTS_DIRS:
shutil.rmtree(placeholder_dir, ignore_errors=True)
def _humanize_alert_type(alert_type: str) -> str:
"""Convert a camelCase/PascalCase alert type into a Title-Cased label.
Used as a last-resort fallback when the SDK does not have metadata for an
alert type and there is no explicit override. Adjacent capitals are kept
together so acronyms like 'SQL' survive ('SQLInjection' -> 'SQL Injection').
"""
if not alert_type:
return ""
parts = _HUMANIZE_BOUNDARY.split(alert_type)
return " ".join(part[:1].upper() + part[1:] for part in parts if part)
class ManifestPatterns(NamedTuple):
"""Manifest patterns prepared once per scan root, case-folded for matching.
The first three fields are the authoritative matchers used by
Core._matches_manifest_pattern. The candidate_* fields are a prefilter over
basenames alone: a manifest-discovery walk visits every file in the repository
but only a few hundred are manifests, so rejecting a name up front avoids
building a relative path and running a path match for the rest. The prefilter's
globs are pre-compiled into one alternation so the cost per rejected file stays
flat as the API's pattern list grows.
"""
literal_basenames: Set[str]
basename_globs: List[str]
path_globs: List[str]
candidate_basenames: Set[str]
candidate_basename_regex: Optional["re.Pattern"]
@property
def is_empty(self) -> bool:
"""True when every ecosystem was filtered out, so the walk can be skipped."""
return not (self.literal_basenames or self.basename_globs or self.path_globs)
class Core:
"""Main class for interacting with Socket Security API and processing scan results."""
ALERT_TYPE_TO_CAPABILITY = {
"envVars": "Environment Variables",
"networkAccess": "Network Access",
"filesystemAccess": "File System Access",
"shellAccess": "Shell Access",
"usesEval": "Uses Eval",
"unsafe": "Unsafe"
}
config: SocketConfig
sdk: socketdev
cli_config: Optional['CliConfig']
def __init__(self, config: SocketConfig, sdk: socketdev, cli_config: Optional['CliConfig'] = None) -> None:
"""Initialize Core with configuration and SDK instance."""
self.config = config
self.sdk = sdk
self.cli_config = cli_config
self._supported_patterns: Optional[Dict] = None
org_start_time = time.perf_counter()
self.set_org_vars()
log.info(
"Organization initialization completed in "
f"{time.perf_counter() - org_start_time:.2f}s"
)
def set_org_vars(self) -> None:
"""Sets the main shared configuration variables for organization access."""
org_id, org_slug = self.get_org_id_slug()
self.config.org_id = org_id
self.config.org_slug = org_slug
base_path = f"orgs/{org_slug}"
self.config.full_scan_path = f"{base_path}/full-scans"
self.config.repository_path = f"{base_path}/repos"
def get_org_id_slug(self) -> Tuple[str, str]:
"""Gets the Org ID and Org Slug for the API Token."""
response = self.sdk.org.get(use_types=True)
organizations: Dict[str, Organization] = response.get("organizations", {})
if len(organizations) == 1:
org_id = next(iter(organizations))
return org_id, organizations[org_id]['slug']
return None, None
def get_sbom_data(self, full_scan_id: str) -> Dict[str, SocketArtifact]:
"""Returns SBOM artifacts for a full scan keyed by artifact ID."""
response = self.sdk.fullscans.stream(self.config.org_slug, full_scan_id, use_types=True)
if not response.success:
# Raise instead of returning {} so a failed fetch surfaces as an
# API error (exit code 3 by default) rather than empty reports.
log.error(f"Failed to get SBOM data for full-scan {full_scan_id}")
log.error(response.message)
raise APIFailure(
f"Failed to get SBOM data for full-scan {full_scan_id}: {response.message}"
)
if not hasattr(response, "artifacts") or not response.artifacts:
return {}
artifacts = {
artifact_id: artifact
for artifact_id, artifact in response.artifacts.items()
if not Core.is_invalid_facts_marker(artifact)
}
Core.warn_if_invalid_facts_marker(len(artifacts) != len(response.artifacts))
return artifacts
def get_sbom_data_list(self, artifacts_dict: Dict[str, SocketArtifact]) -> list[SocketArtifact]:
"""Converts artifacts dictionary to a list."""
return list(artifacts_dict.values())
@staticmethod
def is_invalid_facts_marker(artifact) -> bool:
"""True for the API's ``generic/invalid-socket-facts`` unparseable-facts marker.
Treated as a package it becomes a blocking alert with an empty "Introduced by" and
"Manifest File" that no developer can act on, so callers drop it and report the parse
failure through ``warn_if_invalid_facts_marker`` instead.
Matches any version; the API pins it to 1.0.0 but the version carries no meaning.
Args:
artifact: A ``SocketArtifact`` or diff artifact (anything with ``type``/``name``).
Returns:
True if the artifact is the marker rather than a real package.
"""
return (
getattr(artifact, "type", None) == INVALID_FACTS_MARKER_TYPE
and getattr(artifact, "name", None) == INVALID_FACTS_MARKER_NAME
)
@staticmethod
def warn_if_invalid_facts_marker(found: bool) -> None:
"""Log the parse failure that ``is_invalid_facts_marker`` stands for.
Dropping the marker silently would hide a real, if non-blocking, problem: the scan ran
without the reachability data it was supposed to carry.
"""
if not found:
return
log.warning(
"Socket could not parse the uploaded .socket.facts.json, so reachability facts "
"were not applied to this scan. Ignoring the "
f"{INVALID_FACTS_MARKER_TYPE}/{INVALID_FACTS_MARKER_NAME} marker returned for it; "
"other scan results are unaffected."
)
def create_sbom_output(self, diff: Diff) -> dict:
"""Creates CycloneDX output for a given diff."""
try:
result = self.sdk.export.cdx_bom(self.config.org_slug, diff.id, use_types=True)
if not result.success:
log.error(f"Failed to get CycloneDX Output for full-scan {diff.id}")
log.error(result.message)
return {}
result.pop("success", None)
return result
except Exception:
log.error(f"Unable to get CycloneDX Output for {diff.id}")
log.error(result.get("message", "No error message provided"))
return {}
@staticmethod
def expand_brace_pattern(pattern: str) -> List[str]:
"""
Recursively expands brace expressions (e.g., {a,b,c}) into separate patterns, supporting nested braces.
"""
def recursive_expand(pat: str) -> List[str]:
stack = []
for i, c in enumerate(pat):
if c == '{':
stack.append(i)
elif c == '}' and stack:
start = stack.pop()
if not stack:
# Found the outermost pair
before = pat[:start]
after = pat[i+1:]
inner = pat[start+1:i]
# Split on commas not inside nested braces
options = []
depth = 0
last = 0
for j, ch in enumerate(inner):
if ch == '{':
depth += 1
elif ch == '}':
depth -= 1
elif ch == ',' and depth == 0:
options.append(inner[last:j])
last = j+1
options.append(inner[last:])
results = []
for opt in options:
expanded = before + opt + after
results.extend(recursive_expand(expanded))
return results
return [pat]
return recursive_expand(pattern)
@staticmethod
def is_excluded(file_path: str, excluded_dirs: Set[str]) -> bool:
parts = os.path.normpath(file_path).split(os.sep)
for part in parts:
if part in excluded_dirs:
return True
return False
@staticmethod
def _exclude_glob_to_regex(pattern: str) -> str:
"""Translate a micromatch-style glob into an anchored regex string.
Mirrors the Node CLI's --exclude-paths matcher (src/commands/scan/exclude-paths.mts):
patterns are matched against scan-root-relative POSIX paths, case-sensitively, where
``*`` does NOT cross ``/`` and ``**`` DOES. Patterns are anchored at the scan root, so
``tests`` matches ``tests`` (not ``src/tests``); use ``**/tests`` to match at any depth.
"""
i, n = 0, len(pattern)
out = ["^"]
while i < n:
c = pattern[i]
if c == "*":
if i + 1 < n and pattern[i + 1] == "*":
if i + 2 < n and pattern[i + 2] == "/":
out.append("(?:[^/]+/)*") # '**/' -> zero or more path segments
i += 3
else:
out.append(".*") # '**' at end / before non-slash -> any, incl '/'
i += 2
else:
out.append("[^/]*") # '*' -> within a single path segment
i += 1
elif c == "?":
out.append("[^/]")
i += 1
else:
out.append(re.escape(c))
i += 1
out.append("$")
return "".join(out)
@staticmethod
def compile_exclude_paths(patterns: Optional[List[str]]) -> List["re.Pattern"]:
"""Compile --exclude-paths globs into anchored regexes (compiled once per scan).
Each pattern ``P`` is expanded the way Node feeds fast-glob's ``ignore``: ``P`` (a file-
or dir-shaped exact match) plus ``P/**`` (its subtree), unless ``P`` already ends with
``/**``. Validation of the patterns happens earlier, in CliConfig.from_args.
"""
compiled: List["re.Pattern"] = []
for raw in patterns or []:
p = (raw or "").strip().replace("\\", "/").rstrip("/")
if not p:
continue
globs = [p] if p.endswith("/**") else [p, f"{p}/**"]
compiled.extend(re.compile(Core._exclude_glob_to_regex(g)) for g in globs)
return compiled
@staticmethod
def path_matches_exclude_regexes(rel_path: str, regexes: List["re.Pattern"]) -> bool:
rp = rel_path.replace(os.sep, "/").replace("\\", "/")
return any(r.match(rp) for r in regexes)
@staticmethod
def matches_exclude_paths(file_path: str, base_path: str, patterns: List[str]) -> bool:
"""Convenience matcher (compiles patterns per call); used in tests/ad-hoc checks."""
rel_path = os.path.relpath(file_path, base_path).replace(os.sep, "/")
return Core.path_matches_exclude_regexes(rel_path, Core.compile_exclude_paths(patterns))
def save_submitted_files_list(self, files: List[str], output_path: str) -> None:
"""
Save the list of submitted file names to a JSON file for debugging.
Args:
files: List of file paths that were submitted for scanning
output_path: Path where to save the JSON file
"""
try:
# Calculate total size of all files
total_size_bytes = 0
valid_files = []
for file_path in files:
try:
if os.path.exists(file_path) and os.path.isfile(file_path):
file_size = os.path.getsize(file_path)
total_size_bytes += file_size
valid_files.append(file_path)
else:
log.warning(f"File not found or not accessible: {file_path}")
valid_files.append(file_path) # Still include in list for debugging
except OSError as e:
log.warning(f"Error accessing file {file_path}: {e}")
valid_files.append(file_path) # Still include in list for debugging
# Convert bytes to human-readable format
def format_bytes(bytes_value):
"""Convert bytes to human readable format"""
for unit in ['B', 'KB', 'MB', 'GB']:
if bytes_value < 1024.0:
return f"{bytes_value:.2f} {unit}"
bytes_value /= 1024.0
return f"{bytes_value:.2f} TB"
file_data = {
"timestamp": time.strftime("%Y-%m-%d %H:%M:%S UTC", time.gmtime()),
"total_files": len(valid_files),
"total_size_bytes": total_size_bytes,
"total_size_human": format_bytes(total_size_bytes),
"files": sorted(valid_files)
}
with open(output_path, 'w', encoding='utf-8') as f:
json.dump(file_data, f, indent=2, ensure_ascii=False)
log.info(f"Saved list of {len(valid_files)} submitted files ({file_data['total_size_human']}) to: {output_path}")
except Exception as e:
log.error(f"Failed to save submitted files list to {output_path}: {e}")
def save_manifest_tar(self, files: List[str], output_path: str, base_dir: str) -> None:
"""
Save all manifest files to a compressed tar.gz archive with original directory structure.
Args:
files: List of file paths to include in the archive
output_path: Path where to save the tar.gz file
base_dir: Base directory to preserve relative structure
"""
try:
# Normalize base directory
base_dir = os.path.abspath(base_dir)
if not base_dir.endswith(os.sep):
base_dir += os.sep
log.info(f"Creating manifest tar.gz file: {output_path}")
log.debug(f"Base directory: {base_dir}")
with tarfile.open(output_path, 'w:gz') as tar:
for file_path in files:
if not os.path.exists(file_path):
log.warning(f"File not found, skipping: {file_path}")
continue
# Calculate relative path within the base directory
abs_file_path = os.path.abspath(file_path)
if abs_file_path.startswith(base_dir):
# File is within base directory - use relative path
arcname = os.path.relpath(abs_file_path, base_dir)
else:
# File is outside base directory - use just the filename
arcname = os.path.basename(abs_file_path)
log.warning(f"File outside base dir, using basename: {file_path} -> {arcname}")
# Normalize archive name to use forward slashes
arcname = arcname.replace(os.sep, '/')
log.debug(f"Adding to tar: {file_path} -> {arcname}")
tar.add(file_path, arcname=arcname)
# Get tar file size for logging
tar_size = os.path.getsize(output_path)
def format_bytes(bytes_value):
"""Convert bytes to human readable format"""
for unit in ['B', 'KB', 'MB', 'GB']:
if bytes_value < 1024.0:
return f"{bytes_value:.2f} {unit}"
bytes_value /= 1024.0
return f"{bytes_value:.2f} TB"
tar_size_human = format_bytes(tar_size)
log.info(f"Successfully created tar.gz with {len(files)} files ({tar_size_human}, {tar_size:,} bytes): {output_path}")
except Exception as e:
log.error(f"Failed to save manifest tar.gz to {output_path}: {e}")
@staticmethod
def _prepare_manifest_patterns(
patterns: Dict,
ecosystems: Optional[List[str]],
excluded_ecosystems: List[str]
) -> "ManifestPatterns":
"""Prepare case-folded manifest patterns for a single filesystem walk.
Literal basenames are kept in a set for the common fast path. Basename
globs and path-shaped globs are kept separately so the latter retain
pathlib's path-segment-aware matching behavior. The candidate basename
collections are derived here so the walker can reject a file on its name
alone; see ManifestPatterns.
"""
included_ecosystems = set(ecosystems) if ecosystems is not None else None
excluded = set(excluded_ecosystems)
literal_basenames: Set[str] = set()
basename_globs: Set[str] = set()
path_globs: Set[str] = set()
for ecosystem, ecosystem_patterns in patterns.items():
if included_ecosystems is not None and ecosystem not in included_ecosystems:
continue
if ecosystem in excluded:
continue
log.debug(f"Scanning ecosystem: {ecosystem}")
for details in ecosystem_patterns.values():
original_pattern = details["pattern"]
for expanded in Core.expand_brace_pattern(original_pattern):
normalized = expanded.replace("\\", "/").casefold()
if "/" in normalized:
path_globs.add(normalized)
elif any(character in normalized for character in "*?["):
basename_globs.add(normalized)
else:
literal_basenames.add(normalized)
# PurePath.match compares pattern segments right to left, so a path-shaped glob
# can only match a file whose basename matches the glob's final segment. Folding
# those final segments into the basename prefilter lets the walk skip the path
# match for everything else. A trailing "/" is directory-only under the legacy
# rglob behavior, so its empty final segment intentionally admits no files.
candidate_basenames = set(literal_basenames)
candidate_basename_globs = set(basename_globs)
for pattern in path_globs:
final_segment = pattern.rsplit("/", 1)[-1]
if any(character in final_segment for character in "*?["):
candidate_basename_globs.add(final_segment)
else:
candidate_basenames.add(final_segment)
return ManifestPatterns(
literal_basenames=literal_basenames,
basename_globs=sorted(basename_globs),
path_globs=sorted(path_globs),
candidate_basenames=candidate_basenames,
candidate_basename_regex=Core._compile_basename_globs(candidate_basename_globs),
)
@staticmethod
def _compile_basename_globs(globs: Set[str]) -> Optional["re.Pattern"]:
"""Compile basename globs into a single alternation, or None if there are none.
fnmatch.translate anchors the tail with ``\\Z`` and re.match anchors the head,
so each alternative matches exactly what fnmatch.fnmatchcase would.
"""
if not globs:
return None
return re.compile(
"|".join(f"(?:{fnmatch.translate(glob)})" for glob in sorted(globs))
)
@staticmethod
def _basename_could_match(normalized_name: str, patterns: "ManifestPatterns") -> bool:
"""Cheap prefilter: could a file with this basename match any manifest pattern?
False is authoritative; True still has to be confirmed by
_matches_manifest_pattern against the scan-root-relative path.
"""
if normalized_name in patterns.candidate_basenames:
return True
return (
patterns.candidate_basename_regex is not None
and patterns.candidate_basename_regex.match(normalized_name) is not None
)
@staticmethod
def _matches_manifest_pattern(relative_path: str, patterns: "ManifestPatterns") -> bool:
"""Return whether a scan-root-relative path matches a manifest pattern."""
normalized_path = relative_path.replace("\\", "/").casefold()
basename = normalized_path.rsplit("/", 1)[-1]
if basename in patterns.literal_basenames:
return True
if any(fnmatch.fnmatchcase(basename, pattern) for pattern in patterns.basename_globs):
return True
if not patterns.path_globs:
return False
candidate = PurePath(normalized_path)
return any(candidate.match(pattern) for pattern in patterns.path_globs)
@staticmethod
def _matches_excluded_directory(directory_name: str, excluded_dirs: Set[str]) -> bool:
"""Match configured directory exclusions, including entries such as ``*.egg-info``."""
return any(fnmatch.fnmatchcase(directory_name, pattern) for pattern in excluded_dirs)
def find_files(self, path: str, ecosystems: Optional[List[str]] = None) -> List[str]:
"""
Finds supported manifest files in the given path.
Args:
path: Path to search for manifest files.
ecosystems: Optional list of ecosystems to include. If None, all ecosystems are included.
Returns:
List of found manifest file paths.
"""
log.debug("Starting manifest discovery")
start_time = time.perf_counter()
files: Set[str] = set()
# Unified --exclude-paths: filter discovered manifests by the same paths/globs that are
# forwarded to coana's --exclude-dirs. Only consulted when the user supplied the flag.
# Patterns are anchored to `path` (the scan root this pass walks), matching coana's
# target and the Node CLI's fast-glob cwd. NOTE: when scanning multiple --sub-path
# targets, find_files runs once per sub-path, so a pattern like `tests` anchors to each
# sub-path independently (Node anchors all patterns to a single scan-root cwd). This only
# differs for the multi-target full-scan + --exclude-paths combo; the reach flow is
# single-target, so it matches Node there.
exclude_paths = getattr(self.cli_config, "exclude_paths", None) if self.cli_config else None
exclude_regexes = Core.compile_exclude_paths(exclude_paths) if exclude_paths else []
patterns = self.get_supported_patterns()
manifest_patterns = self._prepare_manifest_patterns(
patterns,
ecosystems,
self.config.excluded_ecosystems,
)
if manifest_patterns.is_empty:
elapsed = time.perf_counter() - start_time
log.info(
"Manifest discovery completed in "
f"{elapsed:.2f}s: root={os.path.abspath(path)}, "
"directories_visited=0, directories_pruned=0, "
"files_visited=0, manifests_found=0"
)
log.info("Total files found: 0")
return []
directories_visited = 0
directories_pruned = 0
files_visited = 0
excluded_dirs = set(self.config.excluded_dirs)
def handle_walk_error(error: OSError) -> None:
log.debug(f"Unable to inspect path during manifest discovery: {error}")
for current_root, directory_names, file_names in os.walk(
path,
topdown=True,
followlinks=False,
onerror=handle_walk_error,
):
directories_visited += 1
kept_directories = []
for directory_name in directory_names:
if directory_name == ".git" or Core._matches_excluded_directory(
directory_name,
excluded_dirs,
):
directories_pruned += 1
continue
# Only --exclude-paths needs a scan-root-relative path, so build one
# lazily rather than for every directory in the repository.
if exclude_regexes:
relative_directory = os.path.relpath(
os.path.join(current_root, directory_name),
path,
)
if Core.path_matches_exclude_regexes(relative_directory, exclude_regexes):
directories_pruned += 1
continue
kept_directories.append(directory_name)
directory_names[:] = kept_directories
files_visited += len(file_names)
for file_name in file_names:
# Reject on the basename first: os.walk already hands us the name, so
# non-manifests cost one set lookup instead of a relative path plus a
# path match. Exclusions are then only evaluated for real candidates.
if not Core._basename_could_match(file_name.casefold(), manifest_patterns):
continue
file_path = os.path.join(current_root, file_name)
relative_path = os.path.relpath(file_path, path)
if not Core._matches_manifest_pattern(relative_path, manifest_patterns):
continue
if exclude_regexes and Core.path_matches_exclude_regexes(
relative_path,
exclude_regexes,
):
continue
if os.path.isfile(file_path):
files.add(file_path.replace("\\", "/"))
file_list = sorted(files)
file_count = len(file_list)
elapsed = time.perf_counter() - start_time
log.info(
"Manifest discovery completed in "
f"{elapsed:.2f}s: root={os.path.abspath(path)}, "
f"directories_visited={directories_visited}, "
f"directories_pruned={directories_pruned}, "
f"files_visited={files_visited}, manifests_found={file_count}"
)
# Retain the established count-only message for log consumers while they
# transition to the stage-level timing above.
log.info(f"Total files found: {file_count}")
# Check if the number of manifest files might exceed ulimit -n
ulimit_check = check_file_count_against_ulimit(file_count)
if ulimit_check["can_check"]:
if ulimit_check["would_exceed"]:
log.debug(f"Found {file_count} manifest files, which may exceed the file descriptor limit (ulimit -n = {ulimit_check['soft_limit']})")
log.debug(f"Available file descriptors: {ulimit_check['available_fds']} (after {ulimit_check['buffer_size']} buffer)")
log.debug(f"Recommendation: {ulimit_check['recommendation']}")
log.debug("This may cause 'Too many open files' errors during processing")
else:
log.debug(f"File count ({file_count}) is within file descriptor limit ({ulimit_check['soft_limit']})")
else:
log.debug(f"Could not check file descriptor limit: {ulimit_check.get('error', 'Unknown error')}")
return file_list
def find_sbom_files(self, path: str) -> List[str]:
"""
Finds only pre-generated SBOM files (CDX and SPDX) in the given path.
This is used with --reach-use-only-pregenerated-sboms to find only
pre-computed CycloneDX and SPDX manifest files.
Args:
path: Path to search for SBOM files.
Returns:
List of found CDX and SPDX file paths.
"""
log.debug("Starting Find SBOM Files (CDX and SPDX only)")
sbom_ecosystems = ['cdx', 'spdx']
return self.find_files(path, ecosystems=sbom_ecosystems)
def get_supported_patterns(self) -> Dict:
"""
Gets supported file patterns from the Socket API.
Returns:
Dictionary of supported file patterns with 'general' key removed
"""
cached_patterns = getattr(self, "_supported_patterns", None)
if cached_patterns is not None:
log.debug("Using cached supported manifest patterns")
return cached_patterns
start_time = time.perf_counter()
response = self.sdk.report.supported()
source = "api"
if not response:
log.error("Failed to get supported patterns from API")
response = socket_globs
source = "local-fallback"
# Do not mutate the SDK response, which may be shared by its own cache.
patterns = {
ecosystem: ecosystem_patterns
for ecosystem, ecosystem_patterns in response.items()
if ecosystem != "general"
}
# Only cache a successful lookup. The local fallback covers far fewer ecosystems
# than the API, so one transient failure must not pin the rest of the run to it —
# has_manifest_files() runs before find_files() and would poison the cache.
if source == "api":
self._supported_patterns = patterns
elapsed = time.perf_counter() - start_time
log.info(
"Supported manifest patterns loaded in "
f"{elapsed:.2f}s: source={source}, ecosystems={len(patterns)}"
)
return patterns
def has_manifest_files(self, files: list) -> bool:
"""
Checks if any files in the list are supported manifest files.
Args:
files: List of file paths to check
Returns:
True if any files match manifest patterns, False otherwise
"""
# Get supported patterns
try:
patterns = self.get_supported_patterns()
except Exception as e:
log.error(f"Error getting supported patterns from API: {e}")
log.warning("Falling back to local patterns")
from .utils import socket_globs as fallback_patterns
patterns = fallback_patterns
# Normalize all file paths for matching
norm_files = [f.replace('\\', '/').lstrip('./') for f in files]
for ecosystem in patterns:
ecosystem_patterns = patterns[ecosystem]
for file_name in ecosystem_patterns:
pattern_str = ecosystem_patterns[file_name]["pattern"]
# Expand brace patterns for each manifest pattern
expanded_patterns = Core.expand_brace_pattern(pattern_str)
for exp_pat in expanded_patterns:
for file in norm_files:
# Match the pattern as-is first (handles root-level files
# like "package.json" matching pattern "package.json")
if PurePath(file).match(exp_pat):
return True
# Also try with **/ prefix to match files in subdirectories
# (e.g. "src/requirements.txt" matching "*requirements.txt")
if '/' not in exp_pat and PurePath(file).match(f"**/{exp_pat}"):
return True
return False
def check_file_count_limit(self, file_count: int) -> dict:
"""
Check if the given file count would exceed the system's file descriptor limit.
Args:
file_count: Number of files to check
Returns:
Dictionary with check results including recommendations
"""
return check_file_count_against_ulimit(file_count)
@staticmethod
def to_case_insensitive_regex(input_string: str) -> str:
"""
Converts a string into a case-insensitive regex pattern.
Args:
input_string: String to convert
Returns:
Case-insensitive regex pattern
Example:
"pipfile" -> "[Pp][Ii][Pp][Ff][Ii][Ll][Ee]"
"""
return ''.join(f'[{char.lower()}{char.upper()}]' if char.isalpha() else char for char in input_string)
@staticmethod
def empty_head_scan_file() -> List[str]:
"""
Creates a temporary placeholder manifest for scans with no manifest files.
Used for baseline scans when a repository has no head scan yet, and for the new scan
when no supported manifest files were found. The API rejects unsupported filenames, so
the placeholder must be named ``.socket.facts.json`` - which means it must also parse
as a facts document. A zero-byte file does not, and the API answers that by adding a
blocking ``generic/invalid-socket-facts@1.0.0`` artifact to the scan.
Each call gets its own temp directory. The path used to be a fixed
``$TMPDIR/.socket.facts.json``, so two runs sharing a temp dir could delete or
truncate each other's placeholder mid-upload.
Returns:
List containing path to a temporary placeholder facts file
"""
# Own directory per call so concurrent runs can't clobber each other's placeholder;
# the basename must stay exactly SOCKET_FACTS_FILENAME to pass the API's validator.
temp_dir = tempfile.mkdtemp(prefix='socket_baseline_')
_PLACEHOLDER_FACTS_DIRS.append(temp_dir)
temp_path = os.path.join(temp_dir, SOCKET_FACTS_FILENAME)
with open(temp_path, 'w') as f:
f.write(SOCKET_FACTS_EMPTY_DOCUMENT)
log.debug(f"Created temporary placeholder facts file for baseline scan: {temp_path}")
return [temp_path]
def finalize_tier1_scan(self, full_scan_id: str, facts_file_path: str) -> bool:
"""
Finalize a full application reachability scan by associating it with a full scan.
This function reads the tier1ReachabilityScanId from the facts file and
calls the SDK to link it with the specified full scan.
Linking the reachability scan to the full scan helps the Socket team debug potential issues.
Args:
full_scan_id: The ID of the full scan to associate with the reachability scan
facts_file_path: Path to the .socket.facts.json file containing the tier1ReachabilityScanId
Returns:
True if successful, False otherwise
"""
log.debug(f"Finalizing full application reachability scan for full scan {full_scan_id}")
# Read the tier1ReachabilityScanId from the facts file
try:
if not os.path.exists(facts_file_path):
log.debug(f"Facts file not found: {facts_file_path}")
return False
with open(facts_file_path, 'r') as f:
facts = json.load(f)
tier1_scan_id = facts.get('tier1ReachabilityScanId')
if not tier1_scan_id:
log.debug(f"No tier1ReachabilityScanId found in {facts_file_path}")
return False
tier1_scan_id = tier1_scan_id.strip()
log.debug(f"Found tier1ReachabilityScanId: {tier1_scan_id}")
except (json.JSONDecodeError, IOError) as e:
log.debug(f"Failed to read tier1ReachabilityScanId from {facts_file_path}: {e}")
return False
# Call the SDK to finalize the full application reachability scan, retrying transient failures with backoff.
last_error: Optional[Exception] = None
for attempt in range(1, TIER1_FINALIZE_MAX_ATTEMPTS + 1):
try:
success = self.sdk.fullscans.finalize_tier1(
full_scan_id=full_scan_id,
tier1_reachability_scan_id=tier1_scan_id,
)
if success:
log.debug(f"Successfully finalized full application reachability scan {tier1_scan_id} for full scan {full_scan_id}")
return True
log.debug(
f"finalize_tier1 returned a falsy result for scan {tier1_scan_id} "
f"(attempt {attempt}/{TIER1_FINALIZE_MAX_ATTEMPTS})"
)
except Exception as e:
last_error = e
log.debug(
f"Unable to finalize full application reachability scan (attempt {attempt}/{TIER1_FINALIZE_MAX_ATTEMPTS}): {e}"
)
if attempt < TIER1_FINALIZE_MAX_ATTEMPTS:
time.sleep(TIER1_FINALIZE_BACKOFF_SECONDS * (2 ** (attempt - 1)))
if last_error is not None:
log.debug(
f"Giving up finalizing full application reachability scan {tier1_scan_id} after "
f"{TIER1_FINALIZE_MAX_ATTEMPTS} attempts: {last_error}"
)
else:
log.debug(
f"Giving up finalizing full application reachability scan {tier1_scan_id} after "
f"{TIER1_FINALIZE_MAX_ATTEMPTS} attempts"
)
return False
@staticmethod
def _compress_facts_file(source_path: str) -> str:
"""Brotli-compress a ``.socket.facts.json`` file to a sibling ``.socket.facts.json.br``.