intentumdiff-python 0.0.1__py3-none-manylinux_2_39_x86_64.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- intentumdiff/__init__.py +134 -0
- intentumdiff/_differ_gate.py +210 -0
- intentumdiff/_differ_presentation.py +786 -0
- intentumdiff/_differ_runtime.py +320 -0
- intentumdiff/analysis/__init__.py +1 -0
- intentumdiff/analysis/compile_commands.py +151 -0
- intentumdiff/analysis/cross_file.py +71 -0
- intentumdiff/analysis/diagnostics.py +309 -0
- intentumdiff/analysis/guardrail_reports.py +254 -0
- intentumdiff/analysis/guardrails.py +318 -0
- intentumdiff/analysis/invariances.py +1207 -0
- intentumdiff/analysis/keyed_profiles.py +25 -0
- intentumdiff/analysis/resource_profiles.py +14 -0
- intentumdiff/analysis/schema_resolver.py +816 -0
- intentumdiff/analysis/text_review.py +46 -0
- intentumdiff/analysis/user_schemas.py +430 -0
- intentumdiff/assets.py +45 -0
- intentumdiff/cache/__init__.py +37 -0
- intentumdiff/cache/duckdb_store.py +98 -0
- intentumdiff/cache/sqlite_store.py +149 -0
- intentumdiff/cache/store.py +190 -0
- intentumdiff/cli/__init__.py +199 -0
- intentumdiff/cli/_commands.py +2141 -0
- intentumdiff/cli/_parser.py +1061 -0
- intentumdiff/cli/_shared.py +644 -0
- intentumdiff/content_type.py +59 -0
- intentumdiff/core/__init__.py +1 -0
- intentumdiff/core/commit_differ.py +599 -0
- intentumdiff/core/config.py +51 -0
- intentumdiff/core/cst_serializer.py +101 -0
- intentumdiff/core/diffignore.py +87 -0
- intentumdiff/core/index.py +189 -0
- intentumdiff/core/indexer.py +792 -0
- intentumdiff/core/models.py +837 -0
- intentumdiff/differ.py +2195 -0
- intentumdiff/github_app.py +287 -0
- intentumdiff/intentumdiff_rust_core/__init__.py +7 -0
- intentumdiff/intentumdiff_rust_core/ffi.py +8 -0
- intentumdiff/intentumdiff_rust_core/libintentumdiff_rust_core.so +0 -0
- intentumdiff/invariances/__init__.py +2 -0
- intentumdiff/invariances/rules.schema.json +283 -0
- intentumdiff/invariances/rules.yaml +349 -0
- intentumdiff/live_server.py +1314 -0
- intentumdiff/lsp/__init__.py +67 -0
- intentumdiff/lsp/client.py +544 -0
- intentumdiff/lsp/config.py +96 -0
- intentumdiff/lsp/enricher.py +227 -0
- intentumdiff/lsp/exceptions.py +27 -0
- intentumdiff/lsp/launcher.py +273 -0
- intentumdiff/lsp/servers.py +356 -0
- intentumdiff/lsp_server/__init__.py +14 -0
- intentumdiff/lsp_server/_codelens.py +67 -0
- intentumdiff/lsp_server/_diagnostics.py +62 -0
- intentumdiff/lsp_server/_handlers.py +175 -0
- intentumdiff/lsp_server/server.py +205 -0
- intentumdiff/plugins/__init__.py +1 -0
- intentumdiff/plugins/adapter.py +397 -0
- intentumdiff/plugins/builtins.py +323 -0
- intentumdiff/plugins/exceptions.py +93 -0
- intentumdiff/plugins/hub.py +1155 -0
- intentumdiff/plugins/language_metadata.py +250 -0
- intentumdiff/plugins/loader.py +1252 -0
- intentumdiff/plugins/registry.py +1079 -0
- intentumdiff/plugins/registry.schema.json +99 -0
- intentumdiff/plugins/registry_schema.py +280 -0
- intentumdiff/plugins/wit/plugin.wit +351 -0
- intentumdiff/rust_core.py +2370 -0
- intentumdiff/serve/__init__.py +19 -0
- intentumdiff/serve/_app.py +485 -0
- intentumdiff/serve/_static/index.html +326 -0
- intentumdiff/serve/_static/playground.js +567 -0
- intentumdiff/sources/__init__.py +1 -0
- intentumdiff/sources/base.py +28 -0
- intentumdiff/sources/file_source.py +48 -0
- intentumdiff/sources/git_source.py +375 -0
- intentumdiff/sources/live_buffer_source.py +113 -0
- intentumdiff/sources/patch_source.py +126 -0
- intentumdiff/sources/string_source.py +44 -0
- intentumdiff/testing.py +165 -0
- intentumdiff/vcs/__init__.py +37 -0
- intentumdiff/vcs/_safe_path.py +50 -0
- intentumdiff/vcs/base.py +136 -0
- intentumdiff/vcs/git_backend.py +97 -0
- intentumdiff/vcs/git_cli.py +100 -0
- intentumdiff/vcs/hg_backend.py +94 -0
- intentumdiff/vcs/perforce_backend.py +109 -0
- intentumdiff/vcs/svn_backend.py +105 -0
- intentumdiff/wasm/.gitkeep +2 -0
- intentumdiff/wasm/abap_parser.wasm +0 -0
- intentumdiff/wasm/adf_parser.wasm +0 -0
- intentumdiff/wasm/asciidoc_parser.wasm +0 -0
- intentumdiff/wasm/asm_parser.wasm +0 -0
- intentumdiff/wasm/assemblyscript_parser.wasm +0 -0
- intentumdiff/wasm/astro_parser.wasm +0 -0
- intentumdiff/wasm/bash_parser.wasm +0 -0
- intentumdiff/wasm/clojure_parser.wasm +0 -0
- intentumdiff/wasm/cmake_parser.wasm +0 -0
- intentumdiff/wasm/cpp_parser.wasm +0 -0
- intentumdiff/wasm/csharp_parser.wasm +0 -0
- intentumdiff/wasm/css_parser.wasm +0 -0
- intentumdiff/wasm/dart_parser.wasm +0 -0
- intentumdiff/wasm/databricks_parser.wasm +0 -0
- intentumdiff/wasm/dax_parser.wasm +0 -0
- intentumdiff/wasm/delphi_parser.wasm +0 -0
- intentumdiff/wasm/dockerfile_parser.wasm +0 -0
- intentumdiff/wasm/elixir_parser.wasm +0 -0
- intentumdiff/wasm/freebasic_parser.wasm +0 -0
- intentumdiff/wasm/generic_parser.wasm +0 -0
- intentumdiff/wasm/gitignore_parser.wasm +0 -0
- intentumdiff/wasm/go_parser.wasm +0 -0
- intentumdiff/wasm/gomod_parser.wasm +0 -0
- intentumdiff/wasm/graphql_parser.wasm +0 -0
- intentumdiff/wasm/groovy_parser.wasm +0 -0
- intentumdiff/wasm/haskell_parser.wasm +0 -0
- intentumdiff/wasm/html_parser.wasm +0 -0
- intentumdiff/wasm/ini_parser.wasm +0 -0
- intentumdiff/wasm/java_parser.wasm +0 -0
- intentumdiff/wasm/js_ts_parser.wasm +0 -0
- intentumdiff/wasm/json_parser.wasm +0 -0
- intentumdiff/wasm/kotlin_parser.wasm +0 -0
- intentumdiff/wasm/latex_parser.wasm +0 -0
- intentumdiff/wasm/lua_parser.wasm +0 -0
- intentumdiff/wasm/make_parser.wasm +0 -0
- intentumdiff/wasm/markdown_parser.wasm +0 -0
- intentumdiff/wasm/mdx_parser.wasm +0 -0
- intentumdiff/wasm/ocaml_parser.wasm +0 -0
- intentumdiff/wasm/odin_parser.wasm +0 -0
- intentumdiff/wasm/parser_manifest.json +590 -0
- intentumdiff/wasm/perl_parser.wasm +0 -0
- intentumdiff/wasm/php_parser.wasm +0 -0
- intentumdiff/wasm/plsql_parser.wasm +0 -0
- intentumdiff/wasm/po_parser.wasm +0 -0
- intentumdiff/wasm/postscript_parser.wasm +0 -0
- intentumdiff/wasm/powershell_parser.wasm +0 -0
- intentumdiff/wasm/proto_parser.wasm +0 -0
- intentumdiff/wasm/puppet_parser.wasm +0 -0
- intentumdiff/wasm/python_parser.wasm +0 -0
- intentumdiff/wasm/qsharp_parser.wasm +0 -0
- intentumdiff/wasm/r_parser.wasm +0 -0
- intentumdiff/wasm/reasonml_parser.wasm +0 -0
- intentumdiff/wasm/ruby_parser.wasm +0 -0
- intentumdiff/wasm/rust_parser.wasm +0 -0
- intentumdiff/wasm/sas_parser.wasm +0 -0
- intentumdiff/wasm/scala_parser.wasm +0 -0
- intentumdiff/wasm/scss_parser.wasm +0 -0
- intentumdiff/wasm/sql_parser.wasm +0 -0
- intentumdiff/wasm/squirrel_parser.wasm +0 -0
- intentumdiff/wasm/svelte_parser.wasm +0 -0
- intentumdiff/wasm/swift_parser.wasm +0 -0
- intentumdiff/wasm/terraform_parser.wasm +0 -0
- intentumdiff/wasm/toml_parser.wasm +0 -0
- intentumdiff/wasm/tsql_parser.wasm +0 -0
- intentumdiff/wasm/vbnet_parser.wasm +0 -0
- intentumdiff/wasm/vue_parser.wasm +0 -0
- intentumdiff/wasm/wat_parser.wasm +0 -0
- intentumdiff/wasm/xml_parser.wasm +0 -0
- intentumdiff/wasm/yaml_parser.wasm +0 -0
- intentumdiff/wasm/zig_parser.wasm +0 -0
- intentumdiff/watcher.py +259 -0
- intentumdiff_python-0.0.1.dist-info/METADATA +112 -0
- intentumdiff_python-0.0.1.dist-info/RECORD +165 -0
- intentumdiff_python-0.0.1.dist-info/WHEEL +4 -0
- intentumdiff_python-0.0.1.dist-info/entry_points.txt +80 -0
- intentumdiff_python-0.0.1.dist-info/licenses/LICENSE +21 -0
- intentumdiff_python-0.0.1.dist-info/sboms/intentumdiff-rust-core.cyclonedx.json +10968 -0
|
@@ -0,0 +1,320 @@
|
|
|
1
|
+
"""File lifecycle, phase profiling, fuel and telemetry helpers for intentumdiff.differ."""
|
|
2
|
+
|
|
3
|
+
from __future__ import annotations
|
|
4
|
+
|
|
5
|
+
import logging
|
|
6
|
+
import time
|
|
7
|
+
from collections.abc import Iterator
|
|
8
|
+
from contextlib import contextmanager
|
|
9
|
+
from typing import Any
|
|
10
|
+
|
|
11
|
+
from intentumdiff.analysis.diagnostics import (
|
|
12
|
+
DiagnosticsRecorder,
|
|
13
|
+
)
|
|
14
|
+
from intentumdiff.core.models import (
|
|
15
|
+
ChangeGroupKind,
|
|
16
|
+
SemanticDiff,
|
|
17
|
+
)
|
|
18
|
+
|
|
19
|
+
logger = logging.getLogger(__name__)
|
|
20
|
+
|
|
21
|
+
_ADDED_FILE_STATUSES = frozenset({"added", "add", "a", "new", "untracked"})
|
|
22
|
+
_DELETED_FILE_STATUSES = frozenset({"deleted", "delete", "d", "removed", "remove"})
|
|
23
|
+
|
|
24
|
+
|
|
25
|
+
def _infer_file_lifecycle(
|
|
26
|
+
old_source: str,
|
|
27
|
+
new_source: str,
|
|
28
|
+
staging_status: str | None = None,
|
|
29
|
+
) -> str:
|
|
30
|
+
"""Infer the file lifecycle from source presence and git status facts.
|
|
31
|
+
|
|
32
|
+
This is shell-owned source metadata, not semantic classification. The engine
|
|
33
|
+
still owns the raw additions/deletions and review groups.
|
|
34
|
+
"""
|
|
35
|
+
|
|
36
|
+
status = (staging_status or "").strip().lower()
|
|
37
|
+
if status in _ADDED_FILE_STATUSES:
|
|
38
|
+
return "added"
|
|
39
|
+
if status in _DELETED_FILE_STATUSES:
|
|
40
|
+
return "deleted"
|
|
41
|
+
if old_source == "" and new_source != "":
|
|
42
|
+
return "added"
|
|
43
|
+
if old_source != "" and new_source == "":
|
|
44
|
+
return "deleted"
|
|
45
|
+
return "modified"
|
|
46
|
+
|
|
47
|
+
|
|
48
|
+
def _apply_file_lifecycle_to_diff(
|
|
49
|
+
diff: SemanticDiff,
|
|
50
|
+
lifecycle: str,
|
|
51
|
+
) -> SemanticDiff:
|
|
52
|
+
metadata = dict(diff.metadata)
|
|
53
|
+
metadata["file_lifecycle"] = lifecycle
|
|
54
|
+
if lifecycle == "modified":
|
|
55
|
+
return diff.model_copy(update={"metadata": metadata})
|
|
56
|
+
|
|
57
|
+
change_groups = [
|
|
58
|
+
group for group in diff.change_groups if group.kind != ChangeGroupKind.IGNORED_STYLE
|
|
59
|
+
]
|
|
60
|
+
return diff.model_copy(
|
|
61
|
+
update={
|
|
62
|
+
"metadata": metadata,
|
|
63
|
+
"change_groups": change_groups,
|
|
64
|
+
"is_style_only": False,
|
|
65
|
+
"has_semantic_changes": bool(diff.changes),
|
|
66
|
+
}
|
|
67
|
+
)
|
|
68
|
+
|
|
69
|
+
|
|
70
|
+
class _PhaseProfiler:
|
|
71
|
+
"""Tiny opt-in wall-clock phase timer used by benchmark/debug paths."""
|
|
72
|
+
|
|
73
|
+
def __init__(self, enabled: bool) -> None:
|
|
74
|
+
self.enabled = enabled
|
|
75
|
+
self._started = time.perf_counter()
|
|
76
|
+
self._phases: list[dict[str, Any]] = []
|
|
77
|
+
|
|
78
|
+
@contextmanager
|
|
79
|
+
def phase(self, name: str) -> Iterator[None]:
|
|
80
|
+
if not self.enabled:
|
|
81
|
+
yield
|
|
82
|
+
return
|
|
83
|
+
started = time.perf_counter()
|
|
84
|
+
try:
|
|
85
|
+
yield
|
|
86
|
+
finally:
|
|
87
|
+
self.record(name, (time.perf_counter() - started) * 1000)
|
|
88
|
+
|
|
89
|
+
def record(self, name: str, duration_ms: float, *, shared: bool = False) -> None:
|
|
90
|
+
if not self.enabled:
|
|
91
|
+
return
|
|
92
|
+
entry: dict[str, Any] = {
|
|
93
|
+
"name": name,
|
|
94
|
+
"duration_ms": round(duration_ms, 3),
|
|
95
|
+
}
|
|
96
|
+
if shared:
|
|
97
|
+
entry["shared"] = True
|
|
98
|
+
self._phases.append(entry)
|
|
99
|
+
|
|
100
|
+
def snapshot(self) -> dict[str, Any]:
|
|
101
|
+
total_ms = (time.perf_counter() - self._started) * 1000
|
|
102
|
+
return {
|
|
103
|
+
"schema_version": 1,
|
|
104
|
+
"total_ms": round(total_ms, 3),
|
|
105
|
+
"phases": list(self._phases),
|
|
106
|
+
}
|
|
107
|
+
|
|
108
|
+
|
|
109
|
+
_DEFAULT_PLUGIN_FUEL = 100_000_000
|
|
110
|
+
_EXPLICIT_FUEL_EXHAUSTION_TEST_CAP = 1_000_000
|
|
111
|
+
# Hotspot thresholds are calibrated to the measured worst legitimate parser plus ~30%
|
|
112
|
+
# headroom. Two effects moved the floor in 2026-07: literal-container text capture in the
|
|
113
|
+
# tree-sitter converters (issue #46 — a deliberate correctness cost), and whole-binary LTO
|
|
114
|
+
# (lto=true, codegen-units=1) letting ANY crate change reshuffle the tree-sitter parse
|
|
115
|
+
# loop's codegen by ±10-15% between rebuilds. Measured post-rebuild worst cases:
|
|
116
|
+
# powershell 30-func 16.4M/KB, typescript tiny-file 20.5M absolute. Thresholds exist to
|
|
117
|
+
# catch pathological (quadratic) parsers, not codegen jitter on the heaviest grammars.
|
|
118
|
+
_FUEL_HOTSPOT_ABSOLUTE = 30_000_000
|
|
119
|
+
_FUEL_HOTSPOT_PER_KB = 22_000_000.0
|
|
120
|
+
_FUEL_HOTSPOT_PER_LINE = 1_500_000.0
|
|
121
|
+
|
|
122
|
+
|
|
123
|
+
def _attach_content_type_metadata(
|
|
124
|
+
diff: SemanticDiff, old_content: str, new_content: str
|
|
125
|
+
) -> SemanticDiff:
|
|
126
|
+
"""Enrich a diff's metadata with the detected content type (mime + category).
|
|
127
|
+
|
|
128
|
+
Uses the Rust magic-byte detector on the changed content so downstream
|
|
129
|
+
surfaces can show what kind of file was diffed. Never raises — the detector
|
|
130
|
+
bridge falls back gracefully when the native core is unavailable.
|
|
131
|
+
"""
|
|
132
|
+
from intentumdiff.content_type import detect_content_type
|
|
133
|
+
|
|
134
|
+
head = (new_content or old_content)[:8192].encode("utf-8", errors="replace")
|
|
135
|
+
metadata = dict(diff.metadata or {})
|
|
136
|
+
metadata["content_type"] = detect_content_type(head)
|
|
137
|
+
return diff.model_copy(update={"metadata": metadata})
|
|
138
|
+
|
|
139
|
+
|
|
140
|
+
def _fuel_budget(base: int, extra: int) -> int:
|
|
141
|
+
"""Return an adaptive budget without hiding explicit low-fuel tests.
|
|
142
|
+
|
|
143
|
+
Real editor/CLI fuel settings are a floor, not a guarantee that ordinary
|
|
144
|
+
source files will fail once they cross that exact number. Tiny budgets are
|
|
145
|
+
still respected so fuel-pressure tests can prove runaway plugins fail
|
|
146
|
+
explicitly instead of being silently promoted to a larger cap.
|
|
147
|
+
"""
|
|
148
|
+
if base == -1:
|
|
149
|
+
return base
|
|
150
|
+
if base < _EXPLICIT_FUEL_EXHAUSTION_TEST_CAP:
|
|
151
|
+
return base
|
|
152
|
+
return max(base, extra)
|
|
153
|
+
|
|
154
|
+
|
|
155
|
+
def _drain_plugin_telemetry(plugin_adapter: Any) -> list[dict[str, Any]]:
|
|
156
|
+
drain = getattr(plugin_adapter, "drain_telemetry", None)
|
|
157
|
+
if not callable(drain):
|
|
158
|
+
return []
|
|
159
|
+
records = drain()
|
|
160
|
+
return records if isinstance(records, list) else []
|
|
161
|
+
|
|
162
|
+
|
|
163
|
+
def _summarize_engine_telemetry(records: list[dict[str, Any]]) -> dict[str, Any]:
|
|
164
|
+
by_plugin: dict[tuple[str, str], dict[str, Any]] = {}
|
|
165
|
+
for record in records:
|
|
166
|
+
plugin = str(record.get("plugin") or "?")
|
|
167
|
+
function = str(record.get("function") or "?")
|
|
168
|
+
key = (plugin, function)
|
|
169
|
+
entry = by_plugin.setdefault(
|
|
170
|
+
key,
|
|
171
|
+
{
|
|
172
|
+
"plugin": plugin,
|
|
173
|
+
"function": function,
|
|
174
|
+
"engine_owner": record.get("engine_owner") or "unknown",
|
|
175
|
+
"engine": record.get("engine") or "unknown",
|
|
176
|
+
"provenance": record.get("provenance") or "unknown",
|
|
177
|
+
"call_count": 0,
|
|
178
|
+
"elapsed_ms": 0.0,
|
|
179
|
+
"fuel_budget": record.get("fuel_budget"),
|
|
180
|
+
"fuel_consumed": 0,
|
|
181
|
+
"total_fuel_consumed": 0,
|
|
182
|
+
"max_fuel_used_percent": None,
|
|
183
|
+
"language": record.get("language"),
|
|
184
|
+
"filename": record.get("filename"),
|
|
185
|
+
"input_bytes": 0,
|
|
186
|
+
"input_lines": 0,
|
|
187
|
+
"statuses": {},
|
|
188
|
+
"trusted": bool(record.get("trusted")),
|
|
189
|
+
},
|
|
190
|
+
)
|
|
191
|
+
entry["call_count"] += 1
|
|
192
|
+
entry["elapsed_ms"] = round(
|
|
193
|
+
float(entry["elapsed_ms"]) + float(record.get("elapsed_ms") or 0.0),
|
|
194
|
+
3,
|
|
195
|
+
)
|
|
196
|
+
consumed = record.get("fuel_consumed")
|
|
197
|
+
if isinstance(consumed, int):
|
|
198
|
+
entry["fuel_consumed"] = max(entry["fuel_consumed"], consumed)
|
|
199
|
+
entry["total_fuel_consumed"] += consumed
|
|
200
|
+
input_bytes = record.get("input_bytes")
|
|
201
|
+
if isinstance(input_bytes, int):
|
|
202
|
+
entry["input_bytes"] += input_bytes
|
|
203
|
+
input_lines = record.get("input_lines")
|
|
204
|
+
if isinstance(input_lines, int):
|
|
205
|
+
entry["input_lines"] += input_lines
|
|
206
|
+
if entry.get("language") is None and record.get("language") is not None:
|
|
207
|
+
entry["language"] = record.get("language")
|
|
208
|
+
if entry.get("filename") is None and record.get("filename") is not None:
|
|
209
|
+
entry["filename"] = record.get("filename")
|
|
210
|
+
percent = record.get("fuel_used_percent")
|
|
211
|
+
if isinstance(percent, int | float):
|
|
212
|
+
current = entry["max_fuel_used_percent"]
|
|
213
|
+
entry["max_fuel_used_percent"] = percent if current is None else max(current, percent)
|
|
214
|
+
status = str(record.get("status") or "unknown")
|
|
215
|
+
statuses = entry["statuses"]
|
|
216
|
+
statuses[status] = statuses.get(status, 0) + 1
|
|
217
|
+
calls = sorted(
|
|
218
|
+
by_plugin.values(),
|
|
219
|
+
key=lambda item: (
|
|
220
|
+
item.get("fuel_consumed") or 0,
|
|
221
|
+
item.get("elapsed_ms") or 0.0,
|
|
222
|
+
),
|
|
223
|
+
reverse=True,
|
|
224
|
+
)
|
|
225
|
+
hotspots = [_fuel_hotspot_for_call(call) for call in calls]
|
|
226
|
+
hotspots = [hotspot for hotspot in hotspots if hotspot is not None]
|
|
227
|
+
return {"schema_version": 1, "calls": calls, "fuel_hotspots": hotspots}
|
|
228
|
+
|
|
229
|
+
|
|
230
|
+
def _fuel_hotspot_for_call(call: dict[str, Any]) -> dict[str, Any] | None:
|
|
231
|
+
consumed = call.get("fuel_consumed")
|
|
232
|
+
if not isinstance(consumed, int) or consumed <= 0:
|
|
233
|
+
return None
|
|
234
|
+
input_bytes = call.get("input_bytes")
|
|
235
|
+
input_lines = call.get("input_lines")
|
|
236
|
+
kb = max((input_bytes or 0) / 1024.0, 1.0)
|
|
237
|
+
lines = max(input_lines or 0, 1)
|
|
238
|
+
fuel_per_kb = consumed / kb
|
|
239
|
+
fuel_per_line = consumed / lines
|
|
240
|
+
exceeded: list[str] = []
|
|
241
|
+
if consumed > _FUEL_HOTSPOT_ABSOLUTE:
|
|
242
|
+
exceeded.append("absolute")
|
|
243
|
+
if fuel_per_kb > _FUEL_HOTSPOT_PER_KB:
|
|
244
|
+
exceeded.append("per_kb")
|
|
245
|
+
if fuel_per_line > _FUEL_HOTSPOT_PER_LINE:
|
|
246
|
+
exceeded.append("per_line")
|
|
247
|
+
normalized_exceeded = any(threshold in exceeded for threshold in ("per_kb", "per_line"))
|
|
248
|
+
if not exceeded or ("absolute" in exceeded and not normalized_exceeded):
|
|
249
|
+
return None
|
|
250
|
+
return {
|
|
251
|
+
"plugin": call.get("plugin"),
|
|
252
|
+
"function": call.get("function"),
|
|
253
|
+
"language": call.get("language"),
|
|
254
|
+
"filename": call.get("filename"),
|
|
255
|
+
"fuel_consumed": consumed,
|
|
256
|
+
"fuel_budget": call.get("fuel_budget"),
|
|
257
|
+
"fuel_used_percent": call.get("max_fuel_used_percent"),
|
|
258
|
+
"input_bytes": input_bytes,
|
|
259
|
+
"input_lines": input_lines,
|
|
260
|
+
"fuel_per_kb": round(fuel_per_kb, 3),
|
|
261
|
+
"fuel_per_line": round(fuel_per_line, 3),
|
|
262
|
+
"thresholds_exceeded": exceeded,
|
|
263
|
+
}
|
|
264
|
+
|
|
265
|
+
|
|
266
|
+
def _attach_run_telemetry(
|
|
267
|
+
diff: SemanticDiff,
|
|
268
|
+
engine_telemetry: list[dict[str, Any]],
|
|
269
|
+
diagnostics: DiagnosticsRecorder,
|
|
270
|
+
) -> SemanticDiff:
|
|
271
|
+
"""Final metadata attach (engine_telemetry + diagnostics snapshot) shared by the
|
|
272
|
+
normal pipeline tail and the Rust finalize-routed return (issue #57) — the routed
|
|
273
|
+
short-circuit skipped it, dropping the wasm parser telemetry contract."""
|
|
274
|
+
if engine_telemetry:
|
|
275
|
+
telemetry_metadata = dict(diff.metadata)
|
|
276
|
+
telemetry_metadata["engine_telemetry"] = _summarize_engine_telemetry(
|
|
277
|
+
engine_telemetry
|
|
278
|
+
)
|
|
279
|
+
diff = diff.model_copy(update={"metadata": telemetry_metadata})
|
|
280
|
+
if diagnostics.enabled:
|
|
281
|
+
diagnostics_metadata = dict(diff.metadata)
|
|
282
|
+
diagnostics.summary.update(
|
|
283
|
+
{
|
|
284
|
+
"final_change_count": len(diff.changes),
|
|
285
|
+
"final_group_count": len(diff.change_groups),
|
|
286
|
+
"has_semantic_changes": diff.has_semantic_changes,
|
|
287
|
+
"is_style_only": diff.is_style_only,
|
|
288
|
+
}
|
|
289
|
+
)
|
|
290
|
+
diagnostics_metadata["diagnostics"] = diagnostics.snapshot()
|
|
291
|
+
diff = diff.model_copy(update={"metadata": diagnostics_metadata})
|
|
292
|
+
return diff
|
|
293
|
+
|
|
294
|
+
|
|
295
|
+
def _record_engine_telemetry(
|
|
296
|
+
diagnostics: DiagnosticsRecorder,
|
|
297
|
+
records: list[dict[str, Any]],
|
|
298
|
+
) -> None:
|
|
299
|
+
if not records:
|
|
300
|
+
return
|
|
301
|
+
summary = _summarize_engine_telemetry(records)
|
|
302
|
+
diagnostics.summary["engine_telemetry"] = summary
|
|
303
|
+
diagnostics.record(
|
|
304
|
+
stage="engine.telemetry",
|
|
305
|
+
action="wasm_fuel_summary",
|
|
306
|
+
rule_id="engine.wasm_fuel_telemetry",
|
|
307
|
+
reason="Wasm plugin calls recorded fuel and duration telemetry",
|
|
308
|
+
metadata=summary,
|
|
309
|
+
)
|
|
310
|
+
hotspots = summary.get("fuel_hotspots")
|
|
311
|
+
if hotspots:
|
|
312
|
+
diagnostics.record(
|
|
313
|
+
stage="engine.telemetry",
|
|
314
|
+
action="wasm_fuel_hotspot",
|
|
315
|
+
rule_id="engine.wasm_fuel_hotspot",
|
|
316
|
+
reason="Wasm plugin fuel use exceeded the excessive-fuel policy",
|
|
317
|
+
metadata={"fuel_hotspots": hotspots},
|
|
318
|
+
)
|
|
319
|
+
|
|
320
|
+
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
# intentumdiff/analysis
|
|
@@ -0,0 +1,151 @@
|
|
|
1
|
+
"""Compile database metadata for C/C++ review context."""
|
|
2
|
+
|
|
3
|
+
from __future__ import annotations
|
|
4
|
+
|
|
5
|
+
import hashlib
|
|
6
|
+
import json
|
|
7
|
+
import shlex
|
|
8
|
+
from pathlib import Path
|
|
9
|
+
from typing import Any
|
|
10
|
+
|
|
11
|
+
_CXX_LANGUAGES = frozenset({"c", "cpp", "c++", "cxx"})
|
|
12
|
+
_COMPILE_DATABASE = "compile_commands.json"
|
|
13
|
+
|
|
14
|
+
|
|
15
|
+
def compile_commands_metadata(
|
|
16
|
+
*,
|
|
17
|
+
filename: str,
|
|
18
|
+
language: str,
|
|
19
|
+
cwd: Path | None = None,
|
|
20
|
+
) -> dict[str, Any] | None:
|
|
21
|
+
"""Return normalized compile-command context for a C/C++ file if available."""
|
|
22
|
+
|
|
23
|
+
if language.lower() not in _CXX_LANGUAGES:
|
|
24
|
+
return None
|
|
25
|
+
logical_path = Path(filename)
|
|
26
|
+
if str(logical_path).startswith("<"):
|
|
27
|
+
return None
|
|
28
|
+
base_dir = cwd or Path.cwd()
|
|
29
|
+
file_path = logical_path if logical_path.is_absolute() else base_dir / logical_path
|
|
30
|
+
database = _find_compile_database(file_path)
|
|
31
|
+
if database is None:
|
|
32
|
+
return None
|
|
33
|
+
entry = _matching_entry(database, file_path)
|
|
34
|
+
if entry is None:
|
|
35
|
+
return None
|
|
36
|
+
arguments = _entry_arguments(entry)
|
|
37
|
+
if not arguments:
|
|
38
|
+
return None
|
|
39
|
+
metadata = {
|
|
40
|
+
"database": _repo_relative(database, base_dir),
|
|
41
|
+
"file": _repo_relative(_entry_file_path(entry, database.parent), base_dir),
|
|
42
|
+
"directory": _repo_relative(Path(str(entry.get("directory") or database.parent)), base_dir),
|
|
43
|
+
"arguments": arguments,
|
|
44
|
+
"defines": _flag_values(arguments, "-D"),
|
|
45
|
+
"include_dirs": _include_dirs(arguments),
|
|
46
|
+
"standard": _standard(arguments),
|
|
47
|
+
}
|
|
48
|
+
metadata["fingerprint"] = _fingerprint(metadata)
|
|
49
|
+
return metadata
|
|
50
|
+
|
|
51
|
+
|
|
52
|
+
def _find_compile_database(file_path: Path) -> Path | None:
|
|
53
|
+
for directory in [file_path.parent, *file_path.parents]:
|
|
54
|
+
candidate = directory / _COMPILE_DATABASE
|
|
55
|
+
if candidate.is_file():
|
|
56
|
+
return candidate
|
|
57
|
+
return None
|
|
58
|
+
|
|
59
|
+
|
|
60
|
+
def _matching_entry(database: Path, file_path: Path) -> dict[str, Any] | None:
|
|
61
|
+
try:
|
|
62
|
+
raw = json.loads(database.read_text(encoding="utf8"))
|
|
63
|
+
except (OSError, json.JSONDecodeError):
|
|
64
|
+
return None
|
|
65
|
+
if not isinstance(raw, list):
|
|
66
|
+
return None
|
|
67
|
+
target = _normalized_path(file_path)
|
|
68
|
+
basename = file_path.name.lower()
|
|
69
|
+
basename_match: dict[str, Any] | None = None
|
|
70
|
+
for item in raw:
|
|
71
|
+
if not isinstance(item, dict):
|
|
72
|
+
continue
|
|
73
|
+
entry_file = _entry_file_path(item, database.parent)
|
|
74
|
+
if _normalized_path(entry_file) == target:
|
|
75
|
+
return item
|
|
76
|
+
if entry_file.name.lower() == basename and basename_match is None:
|
|
77
|
+
basename_match = item
|
|
78
|
+
return basename_match
|
|
79
|
+
|
|
80
|
+
|
|
81
|
+
def _entry_file_path(entry: dict[str, Any], database_dir: Path) -> Path:
|
|
82
|
+
raw_file = Path(str(entry.get("file") or ""))
|
|
83
|
+
directory = Path(str(entry.get("directory") or database_dir))
|
|
84
|
+
return raw_file if raw_file.is_absolute() else directory / raw_file
|
|
85
|
+
|
|
86
|
+
|
|
87
|
+
def _entry_arguments(entry: dict[str, Any]) -> list[str]:
|
|
88
|
+
arguments = entry.get("arguments")
|
|
89
|
+
if isinstance(arguments, list):
|
|
90
|
+
return [str(arg) for arg in arguments if str(arg)]
|
|
91
|
+
command = entry.get("command")
|
|
92
|
+
if isinstance(command, str) and command.strip():
|
|
93
|
+
try:
|
|
94
|
+
return shlex.split(command, posix=False)
|
|
95
|
+
except ValueError:
|
|
96
|
+
return command.split()
|
|
97
|
+
return []
|
|
98
|
+
|
|
99
|
+
|
|
100
|
+
def _flag_values(arguments: list[str], prefix: str) -> list[str]:
|
|
101
|
+
values: list[str] = []
|
|
102
|
+
index = 0
|
|
103
|
+
while index < len(arguments):
|
|
104
|
+
arg = arguments[index]
|
|
105
|
+
if arg == prefix and index + 1 < len(arguments):
|
|
106
|
+
values.append(arguments[index + 1])
|
|
107
|
+
index += 2
|
|
108
|
+
continue
|
|
109
|
+
if arg.startswith(prefix) and len(arg) > len(prefix):
|
|
110
|
+
values.append(arg[len(prefix):])
|
|
111
|
+
index += 1
|
|
112
|
+
return values
|
|
113
|
+
|
|
114
|
+
|
|
115
|
+
def _include_dirs(arguments: list[str]) -> list[str]:
|
|
116
|
+
values = _flag_values(arguments, "-I")
|
|
117
|
+
values.extend(_flag_values(arguments, "/I"))
|
|
118
|
+
return values
|
|
119
|
+
|
|
120
|
+
|
|
121
|
+
def _standard(arguments: list[str]) -> str | None:
|
|
122
|
+
for arg in arguments:
|
|
123
|
+
if arg.startswith("-std="):
|
|
124
|
+
return arg.removeprefix("-std=")
|
|
125
|
+
if arg.startswith("/std:"):
|
|
126
|
+
return arg.removeprefix("/std:")
|
|
127
|
+
return None
|
|
128
|
+
|
|
129
|
+
|
|
130
|
+
def _fingerprint(metadata: dict[str, Any]) -> str:
|
|
131
|
+
payload = {
|
|
132
|
+
key: metadata[key]
|
|
133
|
+
for key in ("file", "arguments", "defines", "include_dirs", "standard")
|
|
134
|
+
if metadata.get(key)
|
|
135
|
+
}
|
|
136
|
+
encoded = json.dumps(payload, sort_keys=True, separators=(",", ":")).encode("utf8")
|
|
137
|
+
return hashlib.sha256(encoded).hexdigest()[:16]
|
|
138
|
+
|
|
139
|
+
|
|
140
|
+
def _normalized_path(path: Path) -> str:
|
|
141
|
+
try:
|
|
142
|
+
return str(path.resolve()).replace("\\", "/").lower()
|
|
143
|
+
except OSError:
|
|
144
|
+
return str(path.absolute()).replace("\\", "/").lower()
|
|
145
|
+
|
|
146
|
+
|
|
147
|
+
def _repo_relative(path: Path, base_dir: Path) -> str:
|
|
148
|
+
try:
|
|
149
|
+
return path.resolve().relative_to(base_dir.resolve()).as_posix()
|
|
150
|
+
except (OSError, ValueError):
|
|
151
|
+
return path.as_posix()
|
|
@@ -0,0 +1,71 @@
|
|
|
1
|
+
"""
|
|
2
|
+
intentumdiff.analysis.cross_file
|
|
3
|
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
|
4
|
+
|
|
5
|
+
Detect cross-file semantic changes by comparing two ``SemanticIndex``
|
|
6
|
+
snapshots.
|
|
7
|
+
|
|
8
|
+
Rust-authoritative (#91): the detection algorithm — ``MOVE_TO_MODULE`` (a
|
|
9
|
+
definition appears in a different file under the same qualified name),
|
|
10
|
+
``SPLIT_MODULE`` (one file's symbols fan out across several new files), and
|
|
11
|
+
``CROSS_FILE_RENAME`` (a same-typed, same-scope definition reappears under a new
|
|
12
|
+
name) — runs in the Rust core (``index-engine-lib`` via ``diff_symbol_tables_json``).
|
|
13
|
+
The former pure-Python mirror was deleted; only the thin DTO-marshalling wrapper
|
|
14
|
+
survives here, so "if Python didn't exist, IntentumDiff still works".
|
|
15
|
+
"""
|
|
16
|
+
|
|
17
|
+
from __future__ import annotations
|
|
18
|
+
|
|
19
|
+
import json
|
|
20
|
+
|
|
21
|
+
from intentumdiff.core.index import SemanticIndex
|
|
22
|
+
from intentumdiff.core.models import ChangeType, CrossFileChange
|
|
23
|
+
|
|
24
|
+
|
|
25
|
+
def detect_cross_file_changes(
|
|
26
|
+
old_index: SemanticIndex,
|
|
27
|
+
new_index: SemanticIndex,
|
|
28
|
+
) -> list[CrossFileChange]:
|
|
29
|
+
"""
|
|
30
|
+
Compare *old_index* and *new_index* via the Rust core and return the
|
|
31
|
+
resulting ``CrossFileChange`` instances.
|
|
32
|
+
|
|
33
|
+
Both indices must have had ``build()`` called first. The comparison itself
|
|
34
|
+
is performed by the Rust core over the flat symbol tables; this function only
|
|
35
|
+
marshals the tables in and the change list back out. Returns an empty list
|
|
36
|
+
when the core is unavailable.
|
|
37
|
+
"""
|
|
38
|
+
from intentumdiff.rust_core import try_rust_diff_symbol_tables
|
|
39
|
+
|
|
40
|
+
old_json = json.dumps(
|
|
41
|
+
{k: [d.model_dump() for d in defs] for k, defs in old_index.symbols.items()}
|
|
42
|
+
)
|
|
43
|
+
new_json = json.dumps(
|
|
44
|
+
{k: [d.model_dump() for d in defs] for k, defs in new_index.symbols.items()}
|
|
45
|
+
)
|
|
46
|
+
|
|
47
|
+
result = try_rust_diff_symbol_tables(old_json, new_json)
|
|
48
|
+
if result is None:
|
|
49
|
+
return []
|
|
50
|
+
|
|
51
|
+
changes: list[CrossFileChange] = []
|
|
52
|
+
for item in result:
|
|
53
|
+
changes.append(
|
|
54
|
+
CrossFileChange(
|
|
55
|
+
change_type=ChangeType(item["change_type"]),
|
|
56
|
+
symbol_name=item["symbol_name"],
|
|
57
|
+
old_file=item["old_file"],
|
|
58
|
+
new_file=item["new_file"],
|
|
59
|
+
old_node_id=item.get("old_node_id"),
|
|
60
|
+
new_node_id=item.get("new_node_id"),
|
|
61
|
+
old_position=item.get("old_position"),
|
|
62
|
+
new_position=item.get("new_position"),
|
|
63
|
+
old_language=item.get("old_language", ""),
|
|
64
|
+
new_language=item.get("new_language", ""),
|
|
65
|
+
node_type=item.get("node_type", ""),
|
|
66
|
+
symbol_kind=item.get("symbol_kind", ""),
|
|
67
|
+
confidence=item.get("confidence", 1.0),
|
|
68
|
+
description=item.get("description", ""),
|
|
69
|
+
)
|
|
70
|
+
)
|
|
71
|
+
return changes
|