refactor(AST): 两张 target 表合成一张,加节点类型漏配在结构上不再可能

契约的 AST_NODE_TARGETS_BY_LANGUAGE 是 target → 中文名,judge/ast.ts 的 mappings 是
target → tree-sitter 节点类型,同一批键分在两个包里,靠一句「两边必须同增同减」的注释
维持。只加一边是静默错判:老师给 C 题选到只有 Python 有的 list_comprehension,判题机
拿裸名去比节点类型,C 的语法树里永远不存在它,于是「必须使用列表推导式」永远失败、
「不能使用 f-string」永远通过,两头都不报错,只有学生受着。

现在一个 target 一条 { label, node }:label 给后台下拉和题目页,node 给判题机。加 target
而漏配节点类型在结构上就不可能了。judge/ast.ts 的 mappings 整张删掉,解析统一走契约的
astTargetNodeType()(节点查 node,运算符查运算符表 —— 那张表的值本身就是要比的 token,
判题机原来抄的 and→&& 三条取值逐个相同,纯属重复)。

顺带把同一份数据的四份拷贝收成一份:C 的 14 条原来在契约和判题机里各抄了两遍
(C 一份、C++ 一份),现在 C++ 逐条引用 C_NODE_TARGETS;运算符表的 C++ 改成
{ ...C_OPERATOR_TARGETS, "<<", ">>" }。C++ 那几条仍逐条列出而不是 spread,是为了保住
下拉框的显示顺序(C++ 独有的几条插在中间)。

## 验证

行为零变化,是逐个 target 机械比对过的:把 HEAD 版的两张表原样取出来,对三种语言的
全部 target 比对「label / 运算符文案 / tree-sitter 解析结果 / 下拉框顺序」四项 ——
C 37 个、C++ 47 个、Python3 43 个,全部一致,键集与顺序也一致。

实跑:给题目 1004 配两条 Python3 规则(必须有 for 循环、不能用 f-string),交一发没有
for 循环的正确答案,判成 AST_CHECK_FAILED(10),statistic_info.ast_results 为
「必须使用 for 循环 / 不通过」「不能使用 f-string / 通过」—— label 与 node 两半都走到了。
再交一发带 for 循环的,判成 ACCEPTED(0)。

tsc、vue-tsc、vite build、单二进制编译均通过。

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012j1vgeDqay8wKCh8dPgPcH
This commit is contained in:
2026-09-10 19:30:35 -06:00
parent 8520192a98
commit a6ba5cdf07
3 changed files with 130 additions and 188 deletions

View File

@@ -4,6 +4,7 @@ import {
astNodeLabel,
astOperatorLabel,
astRuleIsMeaningful,
astTargetNodeType,
astRuleSchema,
AST_SUPPORTED_LANGUAGES,
type AstRequirement,
@@ -31,91 +32,11 @@ export interface AstResult {
actual?: number
}
/**
* target → tree-sitter 节点类型。恒等的条目(`+`、`==` 这些运算符)不列,
* 走 `mapping[target] ?? target` 回落。
*
* **这里的键集是契约 AST_NODE_TARGETS_BY_LANGUAGE 的另一半**,两边必须同增同减:
* 那边决定后台下拉能选什么,这边决定判题机认得什么。只加一边就是静默错判。
*/
const mappings: Record<string, Record<string, string>> = {
C: {
for_loop: "for_statement",
while_loop: "while_statement",
do_while: "do_statement",
if_statement: "if_statement",
else_clause: "else_clause",
break: "break_statement",
continue: "continue_statement",
function_definition: "function_definition",
return: "return_statement",
switch_statement: "switch_statement",
case_statement: "case_statement",
assignment: "assignment_expression",
struct: "struct_specifier",
include: "preproc_include",
and: "&&",
or: "||",
not: "!",
},
"C++": {
// C 的那 14 条原样通用tree-sitter-cpp 继承 tree-sitter-c 的语法)
for_loop: "for_statement",
while_loop: "while_statement",
do_while: "do_statement",
if_statement: "if_statement",
else_clause: "else_clause",
break: "break_statement",
continue: "continue_statement",
function_definition: "function_definition",
return: "return_statement",
switch_statement: "switch_statement",
case_statement: "case_statement",
assignment: "assignment_expression",
struct: "struct_specifier",
include: "preproc_include",
// C++ 独有
range_for_loop: "for_range_loop",
class_definition: "class_specifier",
try_except: "try_statement",
throw: "throw_statement",
namespace: "namespace_definition",
template: "template_declaration",
lambda: "lambda_expression",
using: "using_declaration",
and: "&&",
or: "||",
not: "!",
},
Python3: {
for_loop: "for_statement",
while_loop: "while_statement",
if_statement: "if_statement",
else_clause: "else_clause",
elif_clause: "elif_clause",
break: "break_statement",
continue: "continue_statement",
function_definition: "function_definition",
return: "return_statement",
try_except: "try_statement",
with_statement: "with_statement",
list_comprehension: "list_comprehension",
list_literal: "list",
dict_literal: "dictionary",
set_literal: "set",
f_string: "format_string",
import: "import_statement",
import_from: "import_from_statement",
assignment: "assignment",
class_definition: "class_definition",
},
}
let initPromise: Promise<void> | undefined
const languages = new Map<string, Language>()
async function loadLanguage(language: string) {
if (!(language in mappings)) return null
if (!AST_SUPPORTED_LANGUAGES.includes(language)) return null
// locateFile 指到内嵌的 tree-sitter.wasmemscripten 默认按脚本所在目录找,
// 单二进制里那个目录是 /$bunfs/root它自己找不着
if (!initPromise) initPromise = Parser.init({ locateFile: () => treeSitterWasmPath })
@@ -240,7 +161,7 @@ export function astRequirements(value: unknown): AstRequirements | null {
/**
* AST 规则的语义校验。zod 只管形状engine 在枚举里、min 是整数),管不了
* 「给 C 题选了只有 Python 才有的 list_comprehension」这类组合 —— 那种规则存得进去,
* 判题时 `mapping[target] ?? target` 拿裸名去比节点类型,永远失败或永远通过,
* 判题时 astTargetNodeType() 找不到就回落成裸名去比节点类型,永远失败或永远通过,
* 两头都不报错,只有学生受着。
*
* 放这儿而不是 astRulesSchema 的 refine 上:那个 schema 同时用于**读**后台题目详情,
@@ -345,14 +266,9 @@ function methodCalls(root: Node, target: string, language: string) {
})
}
function evaluateRule(
root: Node,
rule: AstRule,
language: string,
mapping: Record<string, string>,
): AstResult | null {
function evaluateRule(root: Node, rule: AstRule, language: string): AstResult | null {
const target = rule.target ?? ""
const nodeType = mapping[target] ?? target
const nodeType = astTargetNodeType(target, language)
switch (rule.engine) {
case "must_exist_node":
@@ -409,8 +325,8 @@ function evaluateRule(
case "must_have_nesting": {
const outer = rule.outer ?? ""
const inner = rule.inner ?? ""
const outerType = mapping[outer] ?? outer
const innerType = mapping[inner] ?? inner
const outerType = astTargetNodeType(outer, language)
const innerType = astTargetNodeType(inner, language)
const passed = collectNodes(root, outerType).some((node) =>
node.children.some((child) => hasNode(child, innerType)),
)
@@ -437,10 +353,9 @@ export async function checkAst(
const tree = parser.parse(code)
if (!tree) return { passed: true, results: [] }
try {
const mapping = mappings[language] ?? {}
const results = rules
.filter(astRuleIsMeaningful)
.map((rule) => evaluateRule(tree.rootNode, rule, language, mapping))
.map((rule) => evaluateRule(tree.rootNode, rule, language))
.filter((result): result is AstResult => result !== null)
return { passed: results.every((result) => result.passed), results }
} finally {