featmap 0.1.1__py3-none-any.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.
- featmap-0.1.1.dist-info/METADATA +95 -0
- featmap-0.1.1.dist-info/RECORD +130 -0
- featmap-0.1.1.dist-info/WHEEL +5 -0
- featmap-0.1.1.dist-info/entry_points.txt +2 -0
- featmap-0.1.1.dist-info/licenses/LICENSE +24 -0
- featmap-0.1.1.dist-info/licenses/fmap/vendor/typescript/LICENSE.txt +55 -0
- featmap-0.1.1.dist-info/licenses/fmap/vendor/typescript/ThirdPartyNoticeText.txt +193 -0
- featmap-0.1.1.dist-info/top_level.txt +1 -0
- fmap/__init__.py +1 -0
- fmap/__main__.py +3 -0
- fmap/axum.py +485 -0
- fmap/cache.py +25 -0
- fmap/cli.py +907 -0
- fmap/goext/go.mod +3 -0
- fmap/goext/main.go +547 -0
- fmap/goext.py +43 -0
- fmap/graph.py +920 -0
- fmap/impact.py +171 -0
- fmap/index.py +66 -0
- fmap/pyx.py +1133 -0
- fmap/routes.py +163 -0
- fmap/rsx.py +381 -0
- fmap/scip.py +80 -0
- fmap/trace_site.py +63 -0
- fmap/tsx.cjs +1258 -0
- fmap/tsx.py +98 -0
- fmap/vendor/typescript/LICENSE.txt +55 -0
- fmap/vendor/typescript/ThirdPartyNoticeText.txt +193 -0
- fmap/vendor/typescript/lib/lib.d.ts +22 -0
- fmap/vendor/typescript/lib/lib.decorators.d.ts +384 -0
- fmap/vendor/typescript/lib/lib.decorators.legacy.d.ts +22 -0
- fmap/vendor/typescript/lib/lib.dom.asynciterable.d.ts +41 -0
- fmap/vendor/typescript/lib/lib.dom.d.ts +39429 -0
- fmap/vendor/typescript/lib/lib.dom.iterable.d.ts +571 -0
- fmap/vendor/typescript/lib/lib.es2015.collection.d.ts +147 -0
- fmap/vendor/typescript/lib/lib.es2015.core.d.ts +597 -0
- fmap/vendor/typescript/lib/lib.es2015.d.ts +28 -0
- fmap/vendor/typescript/lib/lib.es2015.generator.d.ts +77 -0
- fmap/vendor/typescript/lib/lib.es2015.iterable.d.ts +605 -0
- fmap/vendor/typescript/lib/lib.es2015.promise.d.ts +81 -0
- fmap/vendor/typescript/lib/lib.es2015.proxy.d.ts +128 -0
- fmap/vendor/typescript/lib/lib.es2015.reflect.d.ts +144 -0
- fmap/vendor/typescript/lib/lib.es2015.symbol.d.ts +46 -0
- fmap/vendor/typescript/lib/lib.es2015.symbol.wellknown.d.ts +326 -0
- fmap/vendor/typescript/lib/lib.es2016.array.include.d.ts +116 -0
- fmap/vendor/typescript/lib/lib.es2016.d.ts +21 -0
- fmap/vendor/typescript/lib/lib.es2016.full.d.ts +23 -0
- fmap/vendor/typescript/lib/lib.es2016.intl.d.ts +31 -0
- fmap/vendor/typescript/lib/lib.es2017.arraybuffer.d.ts +21 -0
- fmap/vendor/typescript/lib/lib.es2017.d.ts +26 -0
- fmap/vendor/typescript/lib/lib.es2017.date.d.ts +31 -0
- fmap/vendor/typescript/lib/lib.es2017.full.d.ts +23 -0
- fmap/vendor/typescript/lib/lib.es2017.intl.d.ts +44 -0
- fmap/vendor/typescript/lib/lib.es2017.object.d.ts +49 -0
- fmap/vendor/typescript/lib/lib.es2017.sharedmemory.d.ts +135 -0
- fmap/vendor/typescript/lib/lib.es2017.string.d.ts +45 -0
- fmap/vendor/typescript/lib/lib.es2017.typedarrays.d.ts +53 -0
- fmap/vendor/typescript/lib/lib.es2018.asyncgenerator.d.ts +77 -0
- fmap/vendor/typescript/lib/lib.es2018.asynciterable.d.ts +53 -0
- fmap/vendor/typescript/lib/lib.es2018.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.es2018.full.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.es2018.intl.d.ts +83 -0
- fmap/vendor/typescript/lib/lib.es2018.promise.d.ts +30 -0
- fmap/vendor/typescript/lib/lib.es2018.regexp.d.ts +37 -0
- fmap/vendor/typescript/lib/lib.es2019.array.d.ts +79 -0
- fmap/vendor/typescript/lib/lib.es2019.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.es2019.full.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.es2019.intl.d.ts +23 -0
- fmap/vendor/typescript/lib/lib.es2019.object.d.ts +33 -0
- fmap/vendor/typescript/lib/lib.es2019.string.d.ts +37 -0
- fmap/vendor/typescript/lib/lib.es2019.symbol.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.es2020.bigint.d.ts +765 -0
- fmap/vendor/typescript/lib/lib.es2020.d.ts +27 -0
- fmap/vendor/typescript/lib/lib.es2020.date.d.ts +42 -0
- fmap/vendor/typescript/lib/lib.es2020.full.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.es2020.intl.d.ts +474 -0
- fmap/vendor/typescript/lib/lib.es2020.number.d.ts +28 -0
- fmap/vendor/typescript/lib/lib.es2020.promise.d.ts +47 -0
- fmap/vendor/typescript/lib/lib.es2020.sharedmemory.d.ts +99 -0
- fmap/vendor/typescript/lib/lib.es2020.string.d.ts +44 -0
- fmap/vendor/typescript/lib/lib.es2020.symbol.wellknown.d.ts +41 -0
- fmap/vendor/typescript/lib/lib.es2021.d.ts +23 -0
- fmap/vendor/typescript/lib/lib.es2021.full.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.es2021.intl.d.ts +166 -0
- fmap/vendor/typescript/lib/lib.es2021.promise.d.ts +48 -0
- fmap/vendor/typescript/lib/lib.es2021.string.d.ts +33 -0
- fmap/vendor/typescript/lib/lib.es2021.weakref.d.ts +78 -0
- fmap/vendor/typescript/lib/lib.es2022.array.d.ts +121 -0
- fmap/vendor/typescript/lib/lib.es2022.d.ts +25 -0
- fmap/vendor/typescript/lib/lib.es2022.error.d.ts +75 -0
- fmap/vendor/typescript/lib/lib.es2022.full.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.es2022.intl.d.ts +145 -0
- fmap/vendor/typescript/lib/lib.es2022.object.d.ts +26 -0
- fmap/vendor/typescript/lib/lib.es2022.regexp.d.ts +39 -0
- fmap/vendor/typescript/lib/lib.es2022.string.d.ts +25 -0
- fmap/vendor/typescript/lib/lib.es2023.array.d.ts +924 -0
- fmap/vendor/typescript/lib/lib.es2023.collection.d.ts +21 -0
- fmap/vendor/typescript/lib/lib.es2023.d.ts +22 -0
- fmap/vendor/typescript/lib/lib.es2023.full.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.es2023.intl.d.ts +56 -0
- fmap/vendor/typescript/lib/lib.es2024.arraybuffer.d.ts +65 -0
- fmap/vendor/typescript/lib/lib.es2024.collection.d.ts +29 -0
- fmap/vendor/typescript/lib/lib.es2024.d.ts +26 -0
- fmap/vendor/typescript/lib/lib.es2024.full.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.es2024.object.d.ts +29 -0
- fmap/vendor/typescript/lib/lib.es2024.promise.d.ts +35 -0
- fmap/vendor/typescript/lib/lib.es2024.regexp.d.ts +25 -0
- fmap/vendor/typescript/lib/lib.es2024.sharedmemory.d.ts +68 -0
- fmap/vendor/typescript/lib/lib.es2024.string.d.ts +29 -0
- fmap/vendor/typescript/lib/lib.es5.d.ts +4601 -0
- fmap/vendor/typescript/lib/lib.es6.d.ts +23 -0
- fmap/vendor/typescript/lib/lib.esnext.array.d.ts +35 -0
- fmap/vendor/typescript/lib/lib.esnext.collection.d.ts +96 -0
- fmap/vendor/typescript/lib/lib.esnext.d.ts +29 -0
- fmap/vendor/typescript/lib/lib.esnext.decorators.d.ts +28 -0
- fmap/vendor/typescript/lib/lib.esnext.disposable.d.ts +193 -0
- fmap/vendor/typescript/lib/lib.esnext.error.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.esnext.float16.d.ts +445 -0
- fmap/vendor/typescript/lib/lib.esnext.full.d.ts +24 -0
- fmap/vendor/typescript/lib/lib.esnext.intl.d.ts +21 -0
- fmap/vendor/typescript/lib/lib.esnext.iterator.d.ts +148 -0
- fmap/vendor/typescript/lib/lib.esnext.promise.d.ts +34 -0
- fmap/vendor/typescript/lib/lib.esnext.sharedmemory.d.ts +25 -0
- fmap/vendor/typescript/lib/lib.scripthost.d.ts +322 -0
- fmap/vendor/typescript/lib/lib.webworker.asynciterable.d.ts +41 -0
- fmap/vendor/typescript/lib/lib.webworker.d.ts +13150 -0
- fmap/vendor/typescript/lib/lib.webworker.importscripts.d.ts +23 -0
- fmap/vendor/typescript/lib/lib.webworker.iterable.d.ts +340 -0
- fmap/vendor/typescript/lib/typescript.js +200276 -0
- fmap/vendor/typescript/package.json +120 -0
fmap/axum.py
ADDED
|
@@ -0,0 +1,485 @@
|
|
|
1
|
+
"""Conservative axum registration scan; SCIP supplies symbol identity, not name guesses."""
|
|
2
|
+
import json
|
|
3
|
+
import re
|
|
4
|
+
|
|
5
|
+
TOKEN = re.compile(r'//[^\n]*|/\*[\s\S]*?\*/|r(\#*)"[\s\S]*?"\1|"(?:\\.|[^"\\])*"|\'(?:\\.|[^\'\\\n])\'|\b[A-Za-z_]\w*\b|::|->|[^\s]')
|
|
6
|
+
ROUTER_METHODS = {"route", "route_service", "nest", "nest_service", "merge", "layer", "route_layer",
|
|
7
|
+
"with_state", "fallback", "fallback_service", "without_v07_checks"}
|
|
8
|
+
METHODS = {"get", "post", "put", "patch", "delete", "head", "options", "trace", "connect", "any"}
|
|
9
|
+
FRAMEWORK_WORD = re.compile(r"\b(?:axum|actix_web|rocket|warp)\b")
|
|
10
|
+
FRAMEWORKS = {"axum": "axum", "actix_web": "actix-web", "rocket": "rocket", "warp": "warp"}
|
|
11
|
+
|
|
12
|
+
|
|
13
|
+
def tokens(text):
|
|
14
|
+
result = []
|
|
15
|
+
line, line_start, previous = 1, 0, 0
|
|
16
|
+
for match in TOKEN.finditer(text):
|
|
17
|
+
start = match.start()
|
|
18
|
+
line += text.count("\n", previous, start)
|
|
19
|
+
last_newline = text.rfind("\n", previous, start)
|
|
20
|
+
if last_newline >= 0:
|
|
21
|
+
line_start = last_newline + 1
|
|
22
|
+
previous = start
|
|
23
|
+
if not match.group().startswith(("//", "/*")):
|
|
24
|
+
result.append((match.group(), line, start - line_start))
|
|
25
|
+
return result
|
|
26
|
+
|
|
27
|
+
|
|
28
|
+
def literal(value):
|
|
29
|
+
if value.startswith('r'):
|
|
30
|
+
match = re.fullmatch(r'r(\#*)"([\s\S]*)"\1', value)
|
|
31
|
+
return match[2] if match else None
|
|
32
|
+
try:
|
|
33
|
+
return json.loads(value) if value.startswith('"') else None
|
|
34
|
+
except ValueError:
|
|
35
|
+
return None
|
|
36
|
+
|
|
37
|
+
|
|
38
|
+
ITEM_WORDS = {"const", "static", "fn", "mod", "struct", "enum", "union", "type", "trait", "as", "!"}
|
|
39
|
+
ITEM_START = {"fn", "mod", "struct", "enum", "union", "type", "trait", "impl", "use", "const", "static",
|
|
40
|
+
"extern", "async", "unsafe", "macro_rules"}
|
|
41
|
+
|
|
42
|
+
|
|
43
|
+
def strip_cfg_test(stream):
|
|
44
|
+
"""Drop test-only items: `#[cfg(test)]`, `#[cfg(all(test, ..))]` and `#[test]`/`#[x::test]` (absent from
|
|
45
|
+
production routers).
|
|
46
|
+
|
|
47
|
+
Returns the kept tokens, top-level `mod x;` names gated by cfg(test), plain top-level `mod x;` names
|
|
48
|
+
and whether the whole file is `#![cfg(test)]`."""
|
|
49
|
+
values = [token[0] for token in stream]
|
|
50
|
+
pairs, stack, depth = {}, [], [0] * len(values)
|
|
51
|
+
for index, value in enumerate(values):
|
|
52
|
+
depth[index] = len(stack)
|
|
53
|
+
if value in ("(", "{", "["):
|
|
54
|
+
stack.append(index)
|
|
55
|
+
elif value in (")", "}", "]") and stack:
|
|
56
|
+
pairs[stack.pop()] = index
|
|
57
|
+
opener = {close: open_ for open_, close in pairs.items()}
|
|
58
|
+
|
|
59
|
+
def test_only(start, close, inner=False):
|
|
60
|
+
"""Attribute body values[start:close] is test-only: cfg(test), cfg(all(.., test, ..)) or a `..::test` path."""
|
|
61
|
+
body = values[start:close]
|
|
62
|
+
if body[:2] == ["cfg", "("]:
|
|
63
|
+
if body[2:] == ["test", ")"]:
|
|
64
|
+
return True
|
|
65
|
+
group = start + 3 # `all(` only: `any(test, ..)` and `not(test)` still compile outside tests
|
|
66
|
+
return (body[2:4] == ["all", "("] and pairs.get(group) == close - 2
|
|
67
|
+
and any(values[k] == "test" and depth[k] == depth[group] + 1 and values[k - 1] in ("(", ",")
|
|
68
|
+
and values[k + 1] in (",", ")") for k in range(group + 1, close - 2)))
|
|
69
|
+
path = body[:body.index("(")] if "(" in body else body
|
|
70
|
+
return (not inner and path[-1:] == ["test"] and all(t == "::" for t in path[1::2])
|
|
71
|
+
and all(re.fullmatch(r"[A-Za-z_]\w*", t) for t in path[::2]))
|
|
72
|
+
|
|
73
|
+
drop, gated, index = set(), set(), 0
|
|
74
|
+
while index < len(values):
|
|
75
|
+
if values[index:index + 2] != ["#", "["] or index + 1 not in pairs or not test_only(index + 2, pairs[index + 1]):
|
|
76
|
+
index += 1
|
|
77
|
+
continue
|
|
78
|
+
start, end, attrs = index, pairs[index + 1] + 1, []
|
|
79
|
+
while values[end:end + 2] == ["#", "["] and end + 1 in pairs:
|
|
80
|
+
attrs.append(values[end + 2])
|
|
81
|
+
end = pairs[end + 1] + 1
|
|
82
|
+
item = end
|
|
83
|
+
if values[item:item + 2] == ["pub", "("]:
|
|
84
|
+
item = pairs.get(item + 1, item) + 1
|
|
85
|
+
elif values[item:item + 1] == ["pub"]:
|
|
86
|
+
item += 1
|
|
87
|
+
keyword, attr_end, last, unclear = values[item:item + 1] and values[item] in ITEM_START, end, None, False
|
|
88
|
+
# An item ends at its `;` or body; a field/argument/variant ends at `,` or the enclosing closer.
|
|
89
|
+
while end < len(values) and values[end] not in ("{", ";", ",", ")", "}", "]"):
|
|
90
|
+
# A statement without `;` (`return`, `unreachable!()`) must not swallow the next line's statement.
|
|
91
|
+
unclear = (not keyword and last is not None and stream[end][1] > stream[last][1]
|
|
92
|
+
and re.match(r"\w", values[end]) and re.match(r"[\w\"')\]]", values[last]))
|
|
93
|
+
if unclear:
|
|
94
|
+
break
|
|
95
|
+
last = pairs.get(end, end)
|
|
96
|
+
end = last + 1
|
|
97
|
+
if unclear:
|
|
98
|
+
drop.update(range(start, attr_end)) # statement end unclear: drop only the attributes
|
|
99
|
+
index = attr_end
|
|
100
|
+
continue
|
|
101
|
+
if values[end:end + 1] == ["{"]:
|
|
102
|
+
end = pairs.get(end, end)
|
|
103
|
+
elif values[end:end + 1] not in ([";"], [","]):
|
|
104
|
+
end -= 1
|
|
105
|
+
if depth[start] == 0 and "path" not in attrs and values[end:end + 1] == [";"] and values[item:end - 1] == ["mod"]:
|
|
106
|
+
gated.add(values[end - 1])
|
|
107
|
+
drop.update(range(start, end + 1))
|
|
108
|
+
index = end + 1
|
|
109
|
+
|
|
110
|
+
def path_attr(i):
|
|
111
|
+
"""`mod` at values[i] carries a `#[path]` attribute (its file is not the default child path)."""
|
|
112
|
+
if values[i - 1:i] == ["pub"]:
|
|
113
|
+
i -= 1
|
|
114
|
+
elif values[i - 1:i] == [")"] and values[opener.get(i - 1, 0) - 1:opener.get(i - 1, 0)] == ["pub"]:
|
|
115
|
+
i = opener[i - 1] - 1
|
|
116
|
+
while values[i - 1:i] == ["]"] and i - 1 in opener and values[opener[i - 1] - 1:opener[i - 1]] == ["#"]:
|
|
117
|
+
if values[opener[i - 1] + 1:opener[i - 1] + 2] == ["path"]:
|
|
118
|
+
return True
|
|
119
|
+
i = opener[i - 1] - 1
|
|
120
|
+
return False
|
|
121
|
+
|
|
122
|
+
plain = {values[i + 1] for i in range(len(values) - 2) if values[i] == "mod" and depth[i] == 0
|
|
123
|
+
and values[i + 2] == ";" and not path_attr(i) and i not in drop}
|
|
124
|
+
index = 0 # whole file: any `#![cfg(test)]` in the leading run of inner attributes
|
|
125
|
+
while values[index:index + 3] == ["#", "!", "["] and index + 2 in pairs:
|
|
126
|
+
if test_only(index + 3, pairs[index + 2], inner=True):
|
|
127
|
+
return [], gated, plain, True
|
|
128
|
+
index = pairs[index + 2] + 1
|
|
129
|
+
return [token for i, token in enumerate(stream) if i not in drop], gated, plain, False
|
|
130
|
+
|
|
131
|
+
|
|
132
|
+
def child_module(path, name):
|
|
133
|
+
"""Files that may hold `mod name;` declared in `path` (no `#[path]`)."""
|
|
134
|
+
base = path.rsplit("/", 1)
|
|
135
|
+
directory = base[0] + "/" if len(base) == 2 else ""
|
|
136
|
+
if base[-1] not in ("mod.rs", "lib.rs", "main.rs"):
|
|
137
|
+
directory = path[:-3] + "/"
|
|
138
|
+
return (f"{directory}{name}.rs", f"{directory}{name}/mod.rs")
|
|
139
|
+
|
|
140
|
+
|
|
141
|
+
def normalize(path):
|
|
142
|
+
return "/".join("*" if re.fullmatch(r'\*\w+|\{\*\w+\}', part)
|
|
143
|
+
else re.sub(r'^\{(\w+)\}$', r':\1', part) for part in path.split("/"))
|
|
144
|
+
|
|
145
|
+
|
|
146
|
+
class Scanner:
|
|
147
|
+
def __init__(self, path, text, nodes, references, tokenized=None, owners=None):
|
|
148
|
+
self.path, self.tokens, self.nodes, self.references = path, tokens(text) if tokenized is None else tokenized, nodes, references
|
|
149
|
+
self.owners = owners if owners is not None else [(nid, node) for nid, node in nodes.items()
|
|
150
|
+
if nid.startswith(path + ":") and node["kind"] != "module"]
|
|
151
|
+
self.values = [token[0] for token in self.tokens]
|
|
152
|
+
self.imported = set()
|
|
153
|
+
for index in range(len(self.values) - 2):
|
|
154
|
+
if self.values[index:index + 3] == ["use", "axum", "::"]:
|
|
155
|
+
end = next((i for i in range(index, len(self.values)) if self.values[i] == ";"), index)
|
|
156
|
+
self.imported.update(self.values[index + 3:end])
|
|
157
|
+
self.pairs, self.reverse, stack, top = {}, {}, [], set()
|
|
158
|
+
for index, value in enumerate(self.values):
|
|
159
|
+
if not stack:
|
|
160
|
+
top.add(index)
|
|
161
|
+
if value in ("(", "{", "["):
|
|
162
|
+
stack.append(index)
|
|
163
|
+
elif value in (")", "}", "]") and stack:
|
|
164
|
+
start = stack.pop()
|
|
165
|
+
self.pairs[start] = index
|
|
166
|
+
self.reverse[index] = start
|
|
167
|
+
# A module-level `const|static NAME: &str = "...";` stands for its literal. Rust forbids let/parameter
|
|
168
|
+
# bindings that shadow it, so only another item or import named NAME in this file can; then it is unknown.
|
|
169
|
+
defs, imports, using = {}, set(), False
|
|
170
|
+
for index, value in enumerate(self.values):
|
|
171
|
+
using = value == "use" or using and value != ";"
|
|
172
|
+
if using:
|
|
173
|
+
imports.add(index)
|
|
174
|
+
for index in sorted(top):
|
|
175
|
+
if self.values[index] not in ("const", "static") or index + 1 >= len(self.values):
|
|
176
|
+
continue
|
|
177
|
+
rest = self.values[index + 2:index + 11]
|
|
178
|
+
for shape in ([":", "&", "str", "="], [":", "&", "'", "static", "str", "="]):
|
|
179
|
+
text = literal(rest[len(shape)]) if rest[:len(shape)] == shape and len(rest) > len(shape) + 1 else None
|
|
180
|
+
if text is not None and rest[len(shape) + 1] == ";":
|
|
181
|
+
defs.setdefault(self.values[index + 1], []).append((index + 1, text))
|
|
182
|
+
for index, value in enumerate(self.values):
|
|
183
|
+
if value in defs and index != defs[value][0][0] and (index in imports or self.values[index - 1] in ITEM_WORDS):
|
|
184
|
+
defs[value].append(None)
|
|
185
|
+
self.consts = {name: found[0][1] for name, found in defs.items() if len(found) == 1}
|
|
186
|
+
self.inline = []
|
|
187
|
+
for index in range(1, len(self.values) - 1):
|
|
188
|
+
if self.values[index - 1:index + 2] != [".", "nest", "("]:
|
|
189
|
+
continue
|
|
190
|
+
args = self.arguments(index + 1)
|
|
191
|
+
if len(args) == 2 and "Router" in self.values[args[1][0]:args[1][0] + 3]:
|
|
192
|
+
self.inline.append(args[1])
|
|
193
|
+
self.bindings, self.records = [], []
|
|
194
|
+
for index, value in enumerate(self.values):
|
|
195
|
+
if value != "let" or index + 2 >= len(self.values):
|
|
196
|
+
continue
|
|
197
|
+
name = index + 2 if self.values[index + 1] == "mut" else index + 1
|
|
198
|
+
end = self.statement_end(name)
|
|
199
|
+
equal = next((i for i in range(name, end) if self.values[i] == "="), None)
|
|
200
|
+
if equal is not None:
|
|
201
|
+
self.bindings.append((self.values[name], equal + 1, end, self.owner(index)))
|
|
202
|
+
|
|
203
|
+
def string(self, span):
|
|
204
|
+
"""A path argument: one string literal token or a same-file `&str` const."""
|
|
205
|
+
if span[1] != span[0] + 1:
|
|
206
|
+
return None
|
|
207
|
+
value = self.values[span[0]]
|
|
208
|
+
return self.consts.get(value) if re.fullmatch(r"[A-Za-z_]\w*", value) else literal(value)
|
|
209
|
+
|
|
210
|
+
def statement_end(self, start):
|
|
211
|
+
while start < len(self.values) and self.values[start] != ";":
|
|
212
|
+
start = self.pairs.get(start, start) + 1
|
|
213
|
+
return start
|
|
214
|
+
|
|
215
|
+
def owner(self, index):
|
|
216
|
+
line = self.tokens[index][1]
|
|
217
|
+
candidates = [(node["end_line"] - node["line"], nid) for nid, node in self.owners
|
|
218
|
+
if node["line"] <= line <= node["end_line"]]
|
|
219
|
+
return min(candidates)[1] if candidates else self.path + ":<module>"
|
|
220
|
+
|
|
221
|
+
def app(self, index):
|
|
222
|
+
inline = [span for span in self.inline if span[0] <= index < span[1]]
|
|
223
|
+
if inline:
|
|
224
|
+
return f"{self.path}:inline:{max(inline)[0]}"
|
|
225
|
+
binding = next((b for b in self.bindings if b[1] <= index < b[2]), None)
|
|
226
|
+
return f"{self.path}:let:{binding[1]}" if binding else self.owner(index)
|
|
227
|
+
|
|
228
|
+
def resolve(self, start, end, binding=False):
|
|
229
|
+
values = self.values[start:end]
|
|
230
|
+
if binding and (start, end) in self.inline:
|
|
231
|
+
return f"{self.path}:inline:{start}"
|
|
232
|
+
if not values:
|
|
233
|
+
return None
|
|
234
|
+
if binding and len(values) == 1:
|
|
235
|
+
hits = [b for b in self.bindings if b[0] == values[0] and b[2] < start and b[3] == self.owner(start)]
|
|
236
|
+
if len(hits) == 1:
|
|
237
|
+
return f"{self.path}:let:{hits[0][1]}"
|
|
238
|
+
# Router factories retain their identity even when state is passed as an argument.
|
|
239
|
+
if binding:
|
|
240
|
+
opening = next((i for i in range(start, end) if self.values[i] == "("), None)
|
|
241
|
+
if opening is not None and self.pairs.get(opening) == end - 1:
|
|
242
|
+
values = self.values[start:opening]
|
|
243
|
+
if not values or not re.fullmatch(r'[A-Za-z_]\w*(::[A-Za-z_]\w*)*', "".join(values)):
|
|
244
|
+
return None
|
|
245
|
+
last = start + len(values) - 1
|
|
246
|
+
position = self.tokens[last][1:]
|
|
247
|
+
if position in self.references:
|
|
248
|
+
return self.references[position]
|
|
249
|
+
if len(values) == 1:
|
|
250
|
+
if any(b[0] == values[0] and b[2] < start and b[3] == self.owner(start) for b in self.bindings):
|
|
251
|
+
return None
|
|
252
|
+
hits = [nid for nid, node in self.nodes.items() if nid == self.path + ":" + values[0]
|
|
253
|
+
and node["kind"] == "function"]
|
|
254
|
+
return hits[0] if len(hits) == 1 else None
|
|
255
|
+
return None
|
|
256
|
+
|
|
257
|
+
def marker(self, index, expr):
|
|
258
|
+
self.records.append({"t": "edge", "from": self.owner(index), "to": "?", "kind": "?route",
|
|
259
|
+
"at": f"{self.path}:{self.tokens[index][1]}", "expr": expr})
|
|
260
|
+
|
|
261
|
+
def arguments(self, opening):
|
|
262
|
+
end, parts, start = self.pairs.get(opening, opening), [], opening + 1
|
|
263
|
+
index = start
|
|
264
|
+
while index < end:
|
|
265
|
+
if self.values[index] == ",":
|
|
266
|
+
parts.append((start, index))
|
|
267
|
+
start = index + 1
|
|
268
|
+
index = self.pairs.get(index, index) + 1
|
|
269
|
+
if start < end:
|
|
270
|
+
parts.append((start, end))
|
|
271
|
+
return parts
|
|
272
|
+
|
|
273
|
+
def receiver(self, index, seen=()):
|
|
274
|
+
"""Walk a fluent receiver back to an axum constructor or an immutable binding."""
|
|
275
|
+
end = index - 1
|
|
276
|
+
start = end - 1
|
|
277
|
+
if start < 0:
|
|
278
|
+
return False
|
|
279
|
+
if self.values[start] == ")":
|
|
280
|
+
opening = self.reverse.get(start)
|
|
281
|
+
if opening is None or opening == 0:
|
|
282
|
+
return False
|
|
283
|
+
name = opening - 1
|
|
284
|
+
if name and self.values[name - 1] == ".":
|
|
285
|
+
return self.values[name] in ROUTER_METHODS and self.receiver(name, seen)
|
|
286
|
+
first = name
|
|
287
|
+
while first >= 2 and self.values[first - 1] == "::":
|
|
288
|
+
first -= 2
|
|
289
|
+
path = self.values[first:opening]
|
|
290
|
+
if path == ["axum", "::", "Router", "::", "new"]:
|
|
291
|
+
return True
|
|
292
|
+
if path == ["Router", "::", "new"] and "Router" in self.imported:
|
|
293
|
+
return self.path + ":Router" not in self.nodes
|
|
294
|
+
target = self.resolve(first, start + 1, binding=True)
|
|
295
|
+
return target is not None and any(r.get("app") == target for r in self.records)
|
|
296
|
+
hits = [b for b in self.bindings if b[0] == self.values[start] and b[2] < index and b[3] == self.owner(index)]
|
|
297
|
+
if len(hits) != 1 or hits[0][1] in seen:
|
|
298
|
+
return False
|
|
299
|
+
binding = hits[0]
|
|
300
|
+
# Reassignment makes the syntactic binding unsuitable for exact extraction.
|
|
301
|
+
if any(self.values[i:i + 2] == [binding[0], "="] for i in range(binding[2], index)):
|
|
302
|
+
return False
|
|
303
|
+
first = binding[1]
|
|
304
|
+
return "Router" in self.imported and self.values[first:first + 5] == ["Router", "::", "new", "(", ")"]
|
|
305
|
+
|
|
306
|
+
def method_range(self, start, end):
|
|
307
|
+
if end == start + 1:
|
|
308
|
+
hits = [b for b in self.bindings if b[0] == self.values[start] and b[2] < start and b[3] == self.owner(start)]
|
|
309
|
+
if len(hits) == 1:
|
|
310
|
+
return hits[0][1:3]
|
|
311
|
+
return start, end
|
|
312
|
+
|
|
313
|
+
def route(self, index, args):
|
|
314
|
+
path = self.string(args[0]) if args else None
|
|
315
|
+
if path is None or len(args) != 2:
|
|
316
|
+
self.marker(index, "route(non-literal path)")
|
|
317
|
+
return
|
|
318
|
+
start, end = self.method_range(*args[1])
|
|
319
|
+
handlers = self.handlers(start, end)
|
|
320
|
+
if handlers is None:
|
|
321
|
+
self.marker(index, f"route({path}: unresolved MethodRouter)")
|
|
322
|
+
return
|
|
323
|
+
for method, target in handlers:
|
|
324
|
+
if not target:
|
|
325
|
+
self.marker(index, f"route({path}, {method}: unresolved handler)")
|
|
326
|
+
continue
|
|
327
|
+
self.records.append({"t": "entry", "kind": "route", "value": normalize(path),
|
|
328
|
+
"method": "ALL" if method == "any" else method.upper(), "node": target,
|
|
329
|
+
"app": self.app(index), "at": f"{self.path}:{self.tokens[index][1]}"})
|
|
330
|
+
|
|
331
|
+
def closure_owner(self, start, end):
|
|
332
|
+
"""`get(|| async { .. })`: an inline closure runs inside the router function, which owns its calls."""
|
|
333
|
+
i = start
|
|
334
|
+
while i < end and self.values[i] in ("move", "async"):
|
|
335
|
+
i += 1
|
|
336
|
+
return self.owner(start) if i < end and self.values[i] in ("|", "||") else None
|
|
337
|
+
|
|
338
|
+
def unlayer(self, start, end):
|
|
339
|
+
"""`create.layer(mw)` / `.with_state(s)`: the wrapped handler still runs; strip the trailing calls."""
|
|
340
|
+
while end - start > 3 and self.values[end - 1] == ")":
|
|
341
|
+
opening = self.reverse.get(end - 1)
|
|
342
|
+
if opening is None or opening - 2 < start or self.values[opening - 2] != "." \
|
|
343
|
+
or self.values[opening - 1] not in {"layer", "with_state"}:
|
|
344
|
+
break
|
|
345
|
+
end = opening - 2
|
|
346
|
+
return start, end
|
|
347
|
+
|
|
348
|
+
def handlers(self, start, end):
|
|
349
|
+
handlers = []
|
|
350
|
+
is_first = True
|
|
351
|
+
while start < end:
|
|
352
|
+
name = start
|
|
353
|
+
while name + 2 < end and self.values[name + 1] == "::":
|
|
354
|
+
name += 2
|
|
355
|
+
method = self.values[name]
|
|
356
|
+
opening = name + 1
|
|
357
|
+
if opening >= end or self.values[opening] != "(" or opening not in self.pairs:
|
|
358
|
+
return None
|
|
359
|
+
args = self.arguments(opening)
|
|
360
|
+
if method in METHODS:
|
|
361
|
+
path = self.values[start:name + 1]
|
|
362
|
+
if is_first and not (path == ["axum", "::", "routing", "::", method]
|
|
363
|
+
or path == ["routing", "::", method] and "routing" in self.imported
|
|
364
|
+
or path == [method] and method in self.imported):
|
|
365
|
+
return None
|
|
366
|
+
# A project function named `get` is not axum's routing constructor.
|
|
367
|
+
if start == name and self.path + ":" + method in self.nodes:
|
|
368
|
+
return None
|
|
369
|
+
handlers.append((method, (self.closure_owner(*args[0]) or self.resolve(*self.unlayer(*args[0])))
|
|
370
|
+
if len(args) == 1 else None))
|
|
371
|
+
elif not (method == "new" and (self.values[start:name + 1] == ["axum", "::", "routing", "::", "MethodRouter", "::", "new"]
|
|
372
|
+
or self.values[start:name + 1] == ["MethodRouter", "::", "new"] and "MethodRouter" in self.imported)) and method not in {"layer", "route_layer", "with_state"}:
|
|
373
|
+
return None
|
|
374
|
+
is_first = False
|
|
375
|
+
start = self.pairs[opening] + 1
|
|
376
|
+
if start < end:
|
|
377
|
+
if self.values[start] != ".":
|
|
378
|
+
return None
|
|
379
|
+
start += 1
|
|
380
|
+
return handlers or None
|
|
381
|
+
|
|
382
|
+
def scan(self):
|
|
383
|
+
for index in range(len(self.values) - 4):
|
|
384
|
+
if self.values[index:index + 5] == ["Router", "::", "new", "(", ")"] and "Router" in self.imported:
|
|
385
|
+
self.records.append({"t": "router", "app": self.app(index)})
|
|
386
|
+
for index in range(1, len(self.values) - 1):
|
|
387
|
+
method = self.values[index]
|
|
388
|
+
if self.values[index - 1] != "." or self.values[index + 1] != "(" or method not in {"route", "nest", "merge"}:
|
|
389
|
+
continue
|
|
390
|
+
args = self.arguments(index + 1)
|
|
391
|
+
if not self.receiver(index):
|
|
392
|
+
self.marker(index, f"{method}(unresolved router receiver)")
|
|
393
|
+
continue
|
|
394
|
+
if method == "route":
|
|
395
|
+
self.route(index, args)
|
|
396
|
+
elif method in ("nest", "merge"):
|
|
397
|
+
prefix = "" if method == "merge" else (self.string(args[0]) if args else None)
|
|
398
|
+
child = self.resolve(*args[-1], binding=True) if args else None
|
|
399
|
+
if prefix is None or child is None:
|
|
400
|
+
if child:
|
|
401
|
+
self.records.append({"t": "blocked_mount", "child": child})
|
|
402
|
+
self.marker(index, f"{method}(unresolved router or prefix)")
|
|
403
|
+
else:
|
|
404
|
+
self.records.append({"t": "mount", "child": child, "parent": self.app(index),
|
|
405
|
+
"prefix": normalize(prefix), "at": f"{self.path}:{self.tokens[index][1]}"})
|
|
406
|
+
for name, start, end, owner in self.bindings:
|
|
407
|
+
child = self.resolve(start, end, binding=True)
|
|
408
|
+
if child:
|
|
409
|
+
self.records.append({"t": "router_alias", "child": child, "parent": f"{self.path}:let:{start}",
|
|
410
|
+
"prefix": "", "at": f"{self.path}:{self.tokens[start][1]}"})
|
|
411
|
+
node = self.nodes.get(owner)
|
|
412
|
+
if not node:
|
|
413
|
+
continue
|
|
414
|
+
tail = [i for i in range(end + 1, len(self.values)) if self.tokens[i][1] <= node["end_line"]]
|
|
415
|
+
if tail and self.values[tail[0]] == name and all(self.values[i] == "}" for i in tail[1:]):
|
|
416
|
+
self.records.append({"t": "mount", "child": f"{self.path}:let:{start}", "parent": owner,
|
|
417
|
+
"prefix": "", "at": f"{self.path}:{self.tokens[start][1]}"})
|
|
418
|
+
return self.records
|
|
419
|
+
|
|
420
|
+
|
|
421
|
+
def frameworks(sources, tokenized=None):
|
|
422
|
+
"""Known framework paths in Rust code, excluding comments and strings."""
|
|
423
|
+
imports = set()
|
|
424
|
+
for path, text in sources.items():
|
|
425
|
+
if not FRAMEWORK_WORD.search(text):
|
|
426
|
+
continue
|
|
427
|
+
stream = tokens(text)
|
|
428
|
+
if tokenized is not None:
|
|
429
|
+
tokenized[path] = stream
|
|
430
|
+
values = [token[0] for token in stream]
|
|
431
|
+
for index, value in enumerate(values[:-1]):
|
|
432
|
+
if value in FRAMEWORKS and (values[index + 1] == "::" or index > 0 and values[index - 1] in {"use", "crate"}):
|
|
433
|
+
imports.add(value)
|
|
434
|
+
return imports
|
|
435
|
+
|
|
436
|
+
|
|
437
|
+
def extract(sources, nodes, references, framework_sources=None):
|
|
438
|
+
"""Sources are one crate's indexed files; references map exact (line, column) sites to node ids."""
|
|
439
|
+
tokenized = {}
|
|
440
|
+
imports = frameworks(sources if framework_sources is None else framework_sources, tokenized)
|
|
441
|
+
records = [{"t": "fw", "name": FRAMEWORKS[name]} for name in sorted(imports - {"axum"})]
|
|
442
|
+
if "axum" in imports:
|
|
443
|
+
owners = {}
|
|
444
|
+
for nid, node in nodes.items():
|
|
445
|
+
if node["kind"] != "module":
|
|
446
|
+
owners.setdefault(nid.split(":", 1)[0], []).append((nid, node))
|
|
447
|
+
stripped, modules, pending = {}, {}, []
|
|
448
|
+
for path, text in sources.items():
|
|
449
|
+
stripped[path], gated, plain, whole = strip_cfg_test(tokenized[path] if path in tokenized else tokens(text))
|
|
450
|
+
modules[path] = [child for name in sorted(gated | plain) for child in child_module(path, name)]
|
|
451
|
+
pending += [path] if whole else [child for name in sorted(gated) for child in child_module(path, name)]
|
|
452
|
+
tests = set() # files compiled only under cfg(test), including their submodules
|
|
453
|
+
while pending:
|
|
454
|
+
path = pending.pop()
|
|
455
|
+
if path in sources and path not in tests:
|
|
456
|
+
tests.add(path)
|
|
457
|
+
pending += modules[path]
|
|
458
|
+
for path, text in sorted(sources.items()):
|
|
459
|
+
if path not in tests:
|
|
460
|
+
records += Scanner(path, text, nodes, references.get(path, {}), stripped[path], owners.get(path, ())).scan()
|
|
461
|
+
known = {r["app"] for r in records if r["t"] in {"entry", "router"}}
|
|
462
|
+
while True:
|
|
463
|
+
parents = {r["parent"] for r in records if r["t"] in {"mount", "router_alias"} and r["child"] in known}
|
|
464
|
+
if parents <= known:
|
|
465
|
+
break
|
|
466
|
+
known |= parents
|
|
467
|
+
records = [dict(r, t="mount") if r["t"] == "router_alias" else r for r in records
|
|
468
|
+
if r["t"] != "router_alias" or r["child"] in known]
|
|
469
|
+
for record in records[:]:
|
|
470
|
+
if record["t"] == "mount" and record["child"] not in known:
|
|
471
|
+
records.append({"t": "edge", "from": record["parent"] if record["parent"] in nodes else record["at"].rsplit(":", 1)[0] + ":<module>",
|
|
472
|
+
"to": "?", "kind": "?route", "at": record["at"], "expr": "unresolved router target"})
|
|
473
|
+
records = [r for r in records if r["t"] != "router"]
|
|
474
|
+
blocked = {r["child"] for r in records if r["t"] == "blocked_mount"}
|
|
475
|
+
mounted = {r["child"] for r in records if r["t"] == "mount"}
|
|
476
|
+
blocked -= mounted
|
|
477
|
+
while True:
|
|
478
|
+
children = {r["child"] for r in records if r["t"] == "mount" and r["parent"] in blocked}
|
|
479
|
+
if children <= blocked:
|
|
480
|
+
break
|
|
481
|
+
blocked |= children
|
|
482
|
+
records = [r for r in records if r["t"] != "blocked_mount" and r.get("app") not in blocked]
|
|
483
|
+
if "axum" in imports and not any(r["t"] == "entry" for r in records):
|
|
484
|
+
records.append({"t": "fw", "name": "axum"})
|
|
485
|
+
return records
|
fmap/cache.py
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
1
|
+
"""Repository cache paths and file signatures."""
|
|
2
|
+
import hashlib
|
|
3
|
+
import os
|
|
4
|
+
import subprocess
|
|
5
|
+
|
|
6
|
+
HERE = os.path.dirname(os.path.abspath(__file__))
|
|
7
|
+
|
|
8
|
+
|
|
9
|
+
def cache_dir(root):
|
|
10
|
+
r = subprocess.run(["git", "rev-parse", "--git-common-dir"], cwd=root, capture_output=True, text=True)
|
|
11
|
+
d = os.path.join(root, r.stdout.strip(), "fmap", "cache")
|
|
12
|
+
os.makedirs(d, exist_ok=True)
|
|
13
|
+
return d
|
|
14
|
+
|
|
15
|
+
|
|
16
|
+
def file_sig(root, rel):
|
|
17
|
+
p = rel if os.path.isabs(rel) else os.path.join(root, rel)
|
|
18
|
+
try:
|
|
19
|
+
if "/node_modules/" in p or p.startswith(HERE):
|
|
20
|
+
st = os.stat(p) # shortcut: dependency files keyed by size+mtime, upgrade if installs keep mtimes
|
|
21
|
+
return f"{st.st_size}:{st.st_mtime_ns}"
|
|
22
|
+
with open(p, "rb") as fh:
|
|
23
|
+
return hashlib.sha256(fh.read()).hexdigest()
|
|
24
|
+
except OSError:
|
|
25
|
+
return "missing"
|