Files
Lee Chagolla-Christensen 9d62493623 Type checker, reg parsing improvements, UI improvements (#101)
* fix: upgrade python-multipart to 0.0.22 to address CVE-2026-24486 (Dependabot #683)

* update deps+skill, pyright

* pyright + tests in CLI

* Add pyright type checking and enhance registry hive analysis

Registry Hive Analyzer Enhancements:
- Extract machine SID from SAM domain V value with binary SID decoding
- Extract per-user metadata via regipy (ACB flags, timestamps, full
  name, comment via USER_ACCOUNT_V)
- Compute password expiration from domain max password age policy
- Detect empty LM/NT hashes via well-known constants
- Parse DCC cached domain credentials into structured entries
- Store DPAPI system machine_key/user_key as separate fields
- Add structured secret_type field to LSA secrets (dcc, dpapi_system,
  hex_blob, generic)
- Add detailed markdown and plain-text formatters for SAM accounts
  and LSA secrets, replacing _get_lsa_secret_output_string
- Fix SYSTEM hive attribute names: computer_name -> machinename,
  current_control_set -> currentcontrol
- Switch SAM user iteration from sam.users to sam.secrets (pypykatz API)
- Add FILETIME-to-UTC and regipy value-to-bytes helpers

New Tests:
- Add test_registry_hive.py with SAM, SECURITY, and SYSTEM hive fixtures
- Add test_container.py for container analyzer
- Add SAM/SECURITY/SYSTEM binary test fixtures

Pyright Setup:
- Add pyrightconfig.json (basic mode, Python 3.13) to all libs and projects
- Add pyright>=1.1 as dev dependency to all pyproject.toml files
- Update all uv.lock files accordingly
- Update lint.sh to deactivate active venvs and verify pyright availability

Type Annotation Fixes:
- Fix globals initialized as None without Optional type across
  file_enrichment, document_conversion, and agents global_vars
- Fix StorageMinio return types: upload/upload_file/upload_uploadfile
  return str, not uuid.UUID
- Fix MockStorageMinio to match updated StorageMinio return types
- Add explicit type annotations to dict literals in chromekey.py,
  pdf/analyzer.py, registry_hive/analyzer.py, and publish_findings.py
- Fix kubeconfig current_context parameter to accept str | None
- Fix container_contents allowed_extensions: set = None -> set | None = None
- Fix file subscription file_queue: asyncio.Queue = None -> asyncio.Queue | None = None
- Fix web_api upload_file to wrap object_id in uuid.UUID() for response

None-safety Assertions:
- Add assert statements for asyncpg_pool, tracking_service,
  workflow_client, workflow_manager, file_linking_engine, file_queue,
  gotenberg_url, and process.stdout across all activity, subscription,
  route, and workflow files in file_enrichment and document_conversion
- Add assertions for asyncpg_pool in all chromium processors
- Add assertions for File.from_metadata timestamp/expiration fields
- Add assertion for alerting GQL client session type

Pyright Ignore Annotations:
- Suppress third-party type issues in Dapr workflow/activity APIs,
  gRPC subscription imports, ccache/lnk/office_doc attribute access,
  and nemesis_dpapi FlagMixin operators
- Add file-level suppression for office2john.py, pdf2john.py,
  pe/analyzer.py, and test harness files

Bug Fixes:
- Fix office2john.py format string: bare % filename -> % (filename, stream)
- Fix container analyzer 7z iteration: iterate sz.files list instead
  of calling .items()
- Fix file_linking rules_engine: store match result to avoid double call
- Fix logger.exception calls: remove exception object as first arg in
  storage.py, cookies.py, enrichments.py, housekeeping/main.py
- Fix document_conversion lifespan: use stack.callback() for sync shutdown
- Fix NoseyParkerOutput fallback: add missing workflow_id field
- Fix regipy hive_type: handle None return from RegistryHive.hive_type

DPAPI Manager:
- Remove unused guid parameter from get_system_credentials across
  DpapiManager, NullDpapiManager, and DpapiManagerProtocol

Added Missing Dependencies (file_enrichment_modules):
- pypykatz>=0.6.11, pyarrow>=19.0.1, msoffcrypto-tool>=5.4.2,
  oletools>=0.60.2, regipy>=5.2.0, pillow>=11.3.0

* quiet console logs

* feat: lazy file loading with backend range request support

Add offset/length query params to the download endpoint so the frontend
can request partial file content.  FileViewer now fetches data on demand
— hex, transform (Strings, etc.), ZIP, SQLite, and image tabs only load
when activated.  Text-based content is capped at 10 MB previews.

* fix: transform tabs stuck on "Loading content..."

- Prevent stale WS subscription from overwriting fetched content
- Show retry button when transform content fails to load
- Reset fetch guard on non-OK HTTP responses

* fix: hex tab deferred loading with full file content

* feat: truncation dropdown, spinner overlay, and tab render refactor

- Add truncation dropdown to MonacoContentViewer for files > 10MB,
  replacing the old banner alert and hex "Load Hex View" button
- Add spinner overlay on Monaco editor while loading full content
- Wire truncation support to transform tabs (monaco/json types),
  enrichment tabs, and text tabs
- Change hex tab to auto-load first 10MB preview with dropdown for
  full file instead of requiring manual load
- Default word wrap to off
- Remove "File is too large" warning message
- Refactor ~150-line ternary chain into explicit per-tab render
  functions (renderPreviewTab, renderZipTab, renderSqliteTab,
  renderTextTab, renderHexTab, renderTabContent) with shared helpers
  (renderFullFileContent, getTextContent)

* feat: improve Yara Rules UI

- Compact table rows with tighter padding
- Add X button and Esc key to close editor dialog
- Disable Save button when rule content is unchanged
- Disable Create button with inline warning when rule name already exists
- Default source to "Created manually by <user>" for new rules
- Update source placeholder to "e.g. /yara_rules/custom.yara"

* update gitignore

---------

Co-authored-by: Lee Chagolla-Christensen <lee@localhost>
2026-02-05 14:34:25 -08:00

369 lines
12 KiB
Python

"""Tests for cli.submit module - UploadTracker, filters, metadata, file streaming."""
import json
import tempfile
from pathlib import Path
from queue import Queue
import pytest
from cli.submit import (
UploadTracker,
calculate_metadata_path,
create_metadata,
create_session_with_retries,
parse_filters,
stream_files,
validate_filters,
)
# --- UploadTracker ---
class TestUploadTracker:
def test_initial_state(self):
tracker = UploadTracker()
assert tracker.successful == 0
assert tracker.failed == 0
assert tracker.bytes_uploaded == 0
assert tracker.total_files == 0
def test_add_success(self):
tracker = UploadTracker()
tracker.add_success(Path("/tmp/file.txt"), 1024)
assert tracker.successful == 1
assert tracker.bytes_uploaded == 1024
assert tracker.total_files == 1
def test_add_failure(self):
tracker = UploadTracker()
tracker.add_failure(Path("/tmp/file.txt"), "Connection error")
assert tracker.failed == 1
assert tracker.total_files == 1
def test_total_files_combines_success_and_failure(self):
tracker = UploadTracker()
tracker.add_success(Path("/a"), 100)
tracker.add_success(Path("/b"), 200)
tracker.add_failure(Path("/c"), "err")
assert tracker.total_files == 3
def test_format_bytes_bytes(self):
tracker = UploadTracker()
tracker.add_success(Path("/a"), 512)
result = tracker.format_bytes()
assert "512.00 B" == result
def test_format_bytes_kilobytes(self):
tracker = UploadTracker()
tracker.add_success(Path("/a"), 2048)
result = tracker.format_bytes()
assert "2.00 KB" == result
def test_format_bytes_megabytes(self):
tracker = UploadTracker()
tracker.add_success(Path("/a"), 5 * 1024 * 1024)
result = tracker.format_bytes()
assert "5.00 MB" == result
def test_format_bytes_gigabytes(self):
tracker = UploadTracker()
tracker.add_success(Path("/a"), 3 * 1024**3)
result = tracker.format_bytes()
assert "3.00 GB" == result
def test_format_duration(self):
tracker = UploadTracker()
result = tracker.format_duration(3661.5)
assert result == "01:01:01.50 (3661.50s)"
def test_format_duration_short(self):
tracker = UploadTracker()
result = tracker.format_duration(5.25)
assert result == "00:00:05.25 (5.25s)"
def test_get_failures_returns_copy(self):
tracker = UploadTracker()
tracker.add_failure(Path("/a"), "err1")
failures = tracker.get_failures()
assert len(failures) == 1
# Modifying the returned list shouldn't affect the original
failures.clear()
assert len(tracker.get_failures()) == 1
def test_get_successes_returns_copy(self):
tracker = UploadTracker()
tracker.add_success(Path("/a"), 100)
successes = tracker.get_successes()
assert len(successes) == 1
successes.clear()
assert len(tracker.get_successes()) == 1
def test_thread_safety(self):
"""Verify concurrent adds don't corrupt state."""
import threading
tracker = UploadTracker()
threads = []
for i in range(100):
if i % 2 == 0:
t = threading.Thread(target=tracker.add_success, args=(Path(f"/f{i}"), 10))
else:
t = threading.Thread(target=tracker.add_failure, args=(Path(f"/f{i}"), "err"))
threads.append(t)
t.start()
for t in threads:
t.join()
assert tracker.total_files == 100
assert tracker.successful == 50
assert tracker.failed == 50
# --- parse_filters ---
class TestParseFilters:
def test_no_filters_returns_none(self):
result = parse_filters(None, (), (), "glob")
assert result is None
def test_include_patterns(self):
result = parse_filters(None, ("*.exe", "*.dll"), (), "glob")
assert result == {"pattern_type": "glob", "include": ["*.exe", "*.dll"]}
def test_exclude_patterns(self):
result = parse_filters(None, (), ("*/temp/*",), "regex")
assert result == {"pattern_type": "regex", "exclude": ["*/temp/*"]}
def test_both_include_and_exclude(self):
result = parse_filters(None, ("*.exe",), ("*/temp/*",), "glob")
assert result == {"pattern_type": "glob", "include": ["*.exe"], "exclude": ["*/temp/*"]}
def test_filters_file(self):
filters_data = {"include": ["*.exe"], "pattern_type": "glob"}
with tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False) as f:
json.dump(filters_data, f)
f.flush()
result = parse_filters(f.name, (), (), "glob")
assert result == filters_data
def test_filters_file_adds_default_pattern_type(self):
filters_data = {"include": ["*.exe"]}
with tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False) as f:
json.dump(filters_data, f)
f.flush()
result = parse_filters(f.name, (), (), "regex")
assert result["pattern_type"] == "regex"
def test_filters_file_with_inline_patterns_raises(self):
with tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False) as f:
json.dump({"include": ["*.exe"]}, f)
f.flush()
with pytest.raises(ValueError, match="Cannot specify both"):
parse_filters(f.name, ("*.dll",), (), "glob")
def test_invalid_json_file_raises(self):
with tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False) as f:
f.write("not json{{{")
f.flush()
with pytest.raises(ValueError, match="Invalid JSON"):
parse_filters(f.name, (), (), "glob")
# --- validate_filters ---
class TestValidateFilters:
def test_valid_include_filters(self):
validate_filters({"include": ["*.exe"], "pattern_type": "glob"})
def test_valid_exclude_filters(self):
validate_filters({"exclude": ["*/temp/*"], "pattern_type": "regex"})
def test_valid_both_filters(self):
validate_filters({"include": ["*.exe"], "exclude": ["*/temp/*"], "pattern_type": "glob"})
def test_unknown_fields_raises(self):
with pytest.raises(ValueError, match="Unknown filter fields"):
validate_filters({"include": ["*.exe"], "unknown_field": True})
def test_invalid_pattern_type_raises(self):
with pytest.raises(ValueError, match="pattern_type must be"):
validate_filters({"include": ["*.exe"], "pattern_type": "invalid"})
def test_no_include_or_exclude_raises(self):
with pytest.raises(ValueError, match="At least one"):
validate_filters({"pattern_type": "glob"})
def test_include_not_list_raises(self):
with pytest.raises(ValueError, match="must be a list"):
validate_filters({"include": "*.exe"})
def test_include_non_string_elements_raises(self):
with pytest.raises(ValueError, match="must be strings"):
validate_filters({"include": [123]})
def test_non_dict_raises(self):
with pytest.raises(ValueError, match="must be a dictionary"):
validate_filters("not a dict")
def test_default_pattern_type(self):
"""If pattern_type is not specified, default to glob (valid)."""
validate_filters({"include": ["*.exe"]})
# --- create_metadata ---
class TestCreateMetadata:
def test_basic_metadata(self):
result = create_metadata("/path/to/file.txt", "project-1", "agent-1")
assert result["agent_id"] == "agent-1"
assert result["project"] == "project-1"
assert result["path"] == "/path/to/file.txt"
assert "source" not in result
assert "file_filters" not in result
def test_with_source(self):
result = create_metadata("/file.txt", "p", "a", source="host://10.0.0.1")
assert result["source"] == "host://10.0.0.1"
def test_with_file_filters(self):
filters = {"include": ["*.exe"], "pattern_type": "glob"}
result = create_metadata("/file.txt", "p", "a", file_filters=filters)
assert result["file_filters"] == filters
def test_with_invalid_filters_raises(self):
"""create_metadata calls validate_filters internally."""
with pytest.raises(ValueError):
create_metadata("/file.txt", "p", "a", file_filters={"bad_field": True})
# --- calculate_metadata_path ---
class TestCalculateMetadataPath:
def test_no_folder_returns_path_as_is(self):
result = calculate_metadata_path(Path("/tmp/file.txt"), None, None)
assert result == "/tmp/file.txt"
def test_no_base_paths_returns_path_as_is(self):
result = calculate_metadata_path(Path("/tmp/file.txt"), [], "C:\\Users")
assert result == "/tmp/file.txt"
def test_folder_transformation(self, tmp_path):
# Create a test file structure
data_dir = tmp_path / "data"
data_dir.mkdir()
test_file = data_dir / "subdir" / "file.txt"
test_file.parent.mkdir(parents=True)
test_file.touch()
result = calculate_metadata_path(test_file, [data_dir], "C:\\Users\\Admin\\Documents")
assert result == "C:\\Users\\Admin\\Documents/subdir/file.txt"
def test_folder_with_file_base(self, tmp_path):
test_file = tmp_path / "file.txt"
test_file.touch()
result = calculate_metadata_path(test_file, [test_file], "C:\\Users\\Admin")
assert result == "C:\\Users\\Admin/file.txt"
def test_file_not_under_any_base(self, tmp_path):
test_file = tmp_path / "file.txt"
test_file.touch()
other_base = tmp_path / "other"
other_base.mkdir()
result = calculate_metadata_path(test_file, [other_base], "C:\\Users")
assert result == str(test_file)
def test_empty_folder_string(self, tmp_path):
data_dir = tmp_path / "data"
data_dir.mkdir()
test_file = data_dir / "file.txt"
test_file.touch()
result = calculate_metadata_path(test_file, [data_dir], "")
assert result == "file.txt"
# --- stream_files ---
class TestStreamFiles:
def test_single_file(self, tmp_path):
f = tmp_path / "test.txt"
f.write_text("hello")
queue = Queue()
count = stream_files([f], recursive=False, file_queue=queue)
assert count == 1
assert queue.get() == f
def test_directory_non_recursive(self, tmp_path):
f1 = tmp_path / "a.txt"
f1.write_text("a")
sub = tmp_path / "sub"
sub.mkdir()
f2 = sub / "b.txt"
f2.write_text("b")
queue = Queue()
count = stream_files([tmp_path], recursive=False, file_queue=queue)
assert count == 1 # only a.txt, not sub/b.txt
def test_directory_recursive(self, tmp_path):
f1 = tmp_path / "a.txt"
f1.write_text("a")
sub = tmp_path / "sub"
sub.mkdir()
f2 = sub / "b.txt"
f2.write_text("b")
queue = Queue()
count = stream_files([tmp_path], recursive=True, file_queue=queue)
assert count == 2
def test_max_files_limit(self, tmp_path):
for i in range(10):
(tmp_path / f"file_{i}.txt").write_text(str(i))
queue = Queue()
count = stream_files([tmp_path], recursive=False, file_queue=queue, max_files=3)
assert count == 3
def test_nonexistent_path(self):
queue = Queue()
count = stream_files([Path("/nonexistent/path")], recursive=False, file_queue=queue)
assert count == 0
def test_multiple_paths(self, tmp_path):
f1 = tmp_path / "a.txt"
f1.write_text("a")
f2 = tmp_path / "b.txt"
f2.write_text("b")
queue = Queue()
count = stream_files([f1, f2], recursive=False, file_queue=queue)
assert count == 2
# --- create_session_with_retries ---
class TestCreateSessionWithRetries:
def test_returns_session(self):
session = create_session_with_retries()
assert session is not None
session.close()
def test_custom_max_workers(self):
session = create_session_with_retries(max_workers=50)
assert session is not None
session.close()
def test_session_has_adapters(self):
session = create_session_with_retries()
assert "https://" in session.adapters
assert "http://" in session.adapters
session.close()