ata-validator 1.6.2 → 1.7.0

package/lib/safe-regex.js

@@ -226,60 +226,240 @@ function compileProg (ast) {
   return prog
 }
 
-function matchClass (instr, c) {
-  let inside = false
+// Numeric opcodes for the runner. The program is compiled once into flat
+// typed arrays so the inner loop does no property lookups or string compares.
+const OP_CHAR = 0
+const OP_ANY = 1
+const OP_CLASS = 2
+const OP_SPLIT = 3
+const OP_JMP = 4
+const OP_BOL = 5
+const OP_EOL = 6
+const OP_MATCH = 7
+
+function classMatcher (instr) {
+  // ASCII is answered from a bitmap; anything above 0x7f walks the ranges.
+  const bits = new Uint8Array(128)
   const r = instr.ranges
-  for (let k = 0; k < r.length; k++) { if (c >= r[k][0] && c <= r[k][1]) { inside = true; break } }
-  return instr.neg ? !inside : inside
+  for (let k = 0; k < r.length; k++) {
+    const hi = Math.min(r[k][1], 127)
+    for (let c = r[k][0]; c <= hi; c++) bits[c] = 1
+  }
+  return { bits, ranges: r, neg: instr.neg }
+}
+
+function matchClass (cls, c) {
+  let inside
+  if (c < 128) {
+    inside = cls.bits[c] === 1
+  } else {
+    inside = false
+    const r = cls.ranges
+    for (let k = 0; k < r.length; k++) { if (c >= r[k][0] && c <= r[k][1]) { inside = true; break } }
+  }
+  return cls.neg ? !inside : inside
 }
 
 function makeRunner (prog) {
   const n = prog.length
+  const ops = new Uint8Array(n)
+  const xs = new Int32Array(n)
+  const ys = new Int32Array(n)
+  const cs = new Int32Array(n)
+  const classes = new Array(n)
+  for (let i = 0; i < n; i++) {
+    const I = prog[i]
+    switch (I.op) {
+      case 'char': ops[i] = OP_CHAR; cs[i] = I.c; break
+      case 'any': ops[i] = OP_ANY; break
+      case 'class': ops[i] = OP_CLASS; classes[i] = classMatcher(I); break
+      case 'split': ops[i] = OP_SPLIT; xs[i] = I.x; ys[i] = I.y; break
+      case 'jmp': ops[i] = OP_JMP; xs[i] = I.x; break
+      case 'bol': ops[i] = OP_BOL; break
+      case 'eol': ops[i] = OP_EOL; break
+      case 'match': ops[i] = OP_MATCH; break
+    }
+  }
+
   const lastGen = new Int32Array(n).fill(-1)
   let gen = 0
-  const stack = []
+  // Each unvisited instruction is popped once and pushes at most two, so the
+  // stack never holds more than 2n + 1 entries.
+  const stack = new Int32Array(2 * n + 2)
+  // Thread lists hold at most one entry per instruction per step.
+  let clist = new Int32Array(n)
+  let nlist = new Int32Array(n)
+  let clen = 0
+  let nlen = 0
 
-  function addThread (list, pc, pos, len) {
-    stack.length = 0
-    stack.push(pc)
-    while (stack.length) {
-      const p = stack.pop()
+  // Follows epsilon edges from `pc` and records every consuming instruction
+  // (or match) reached in `list`. `lastGen` dedupes per step.
+  function addThread (list, len0, pc, pos, len) {
+    // Most transitions land directly on a consuming instruction; skip the
+    // stack walk for those.
+    if (ops[pc] <= OP_CLASS || ops[pc] === OP_MATCH) {
+      if (lastGen[pc] === gen) return len0
+      lastGen[pc] = gen
+      list[len0] = pc
+      return len0 + 1
+    }
+    let sp = 0
+    stack[sp++] = pc
+    let count = len0
+    while (sp > 0) {
+      const p = stack[--sp]
       if (lastGen[p] === gen) continue
       lastGen[p] = gen
-      const I = prog[p]
-      switch (I.op) {
-        case 'jmp': stack.push(I.x); break
-        case 'split': stack.push(I.y); stack.push(I.x); break
-        case 'bol': if (pos === 0) stack.push(p + 1); break
-        case 'eol': if (pos === len) stack.push(p + 1); break
-        default: list.push(p)
+      switch (ops[p]) {
+        case OP_JMP: stack[sp++] = xs[p]; break
+        case OP_SPLIT: stack[sp++] = ys[p]; stack[sp++] = xs[p]; break
+        case OP_BOL: if (pos === 0) stack[sp++] = p + 1; break
+        case OP_EOL: if (pos === len) stack[sp++] = p + 1; break
+        default: list[count++] = p
       }
     }
+    return count
   }
 
-  return function test (s) {
+  // A pattern is anchored when starting it anywhere but position 0 yields no
+  // thread, which is the case for `^...` and its alternations. The probe sits
+  // at position 1 of a length-1 string so that only `^` can fail. For anchored
+  // patterns the per-position restart below is skipped, and an empty thread
+  // list means the match has already failed.
+  gen++
+  const anchored = addThread(nlist, 0, 0, 1, 1) === 0
+
+  function testNFA (s) {
     const len = s.length
-    let clist = []
-    let nlist = []
     gen++
-    addThread(clist, 0, 0, len)
+    clen = addThread(clist, 0, 0, 0, len)
     for (let pos = 0; pos <= len; pos++) {
       const c = pos < len ? s.charCodeAt(pos) : -1
       gen++
-      nlist.length = 0
-      for (let k = 0; k < clist.length; k++) {
+      nlen = 0
+      for (let k = 0; k < clen; k++) {
         const pc = clist[k]
-        const I = prog[pc]
-        if (I.op === 'match') return true
-        else if (I.op === 'char') { if (c === I.c) addThread(nlist, pc + 1, pos + 1, len) }
-        else if (I.op === 'any') { if (c !== -1 && c !== 10) addThread(nlist, pc + 1, pos + 1, len) }
-        else if (I.op === 'class') { if (c !== -1 && matchClass(I, c)) addThread(nlist, pc + 1, pos + 1, len) }
+        switch (ops[pc]) {
+          case OP_MATCH: return true
+          case OP_CHAR: if (c === cs[pc]) nlen = addThread(nlist, nlen, pc + 1, pos + 1, len); break
+          case OP_ANY: if (c !== -1 && c !== 10) nlen = addThread(nlist, nlen, pc + 1, pos + 1, len); break
+          case OP_CLASS: if (c !== -1 && matchClass(classes[pc], c)) nlen = addThread(nlist, nlen, pc + 1, pos + 1, len); break
+        }
+      }
+      if (pos < len) {
+        if (!anchored) nlen = addThread(nlist, nlen, 0, pos + 1, len)
+        else if (nlen === 0) return false
       }
-      if (pos < len) addThread(nlist, 0, pos + 1, len)
       const tmp = clist; clist = nlist; nlist = tmp
+      clen = nlen
     }
     return false
   }
+
+  // Lazy DFA on top of the NFA. A DFA state is the set of consuming
+  // instructions live at a position; transitions are computed on first use
+  // and cached per ASCII character. `^` and `$` depend on position, so the
+  // closure is taken with flags for "at start" and "at end", which gives two
+  // start states and two transition tables per state. The state count is
+  // capped; past the cap the matcher falls back to the NFA walk above, so the
+  // time bound stays linear either way.
+  const MAX_STATES = 256
+  const states = []
+  const stateIds = new Map()
+  let overflow = false
+
+  function closure (list, count, pc, atStart, atEnd) {
+    // Same walk as addThread, with the position replaced by the two flags.
+    let sp = 0
+    stack[sp++] = pc
+    while (sp > 0) {
+      const p = stack[--sp]
+      if (lastGen[p] === gen) continue
+      lastGen[p] = gen
+      switch (ops[p]) {
+        case OP_JMP: stack[sp++] = xs[p]; break
+        case OP_SPLIT: stack[sp++] = ys[p]; stack[sp++] = xs[p]; break
+        case OP_BOL: if (atStart) stack[sp++] = p + 1; break
+        case OP_EOL: if (atEnd) stack[sp++] = p + 1; break
+        default: list[count++] = p
+      }
+    }
+    return count
+  }
+
+  function internState (list, count) {
+    const pcs = Array.from(list.subarray(0, count)).sort((a, b) => a - b)
+    const key = pcs.join(',')
+    let id = stateIds.get(key)
+    if (id !== undefined) return id
+    if (states.length >= MAX_STATES) { overflow = true; return -1 }
+    id = states.length
+    let isMatch = false
+    for (let k = 0; k < pcs.length; k++) if (ops[pcs[k]] === OP_MATCH) { isMatch = true; break }
+    states.push({ pcs: Int32Array.from(pcs), isMatch, next: new Int32Array(128).fill(-2), nextEnd: new Int32Array(128).fill(-2) })
+    stateIds.set(key, id)
+    return id
+  }
+
+  function step (state, c, atEnd) {
+    gen++
+    let count = 0
+    const pcs = state.pcs
+    for (let k = 0; k < pcs.length; k++) {
+      const pc = pcs[k]
+      switch (ops[pc]) {
+        case OP_CHAR: if (c === cs[pc]) count = closure(nlist, count, pc + 1, false, atEnd); break
+        case OP_ANY: if (c !== 10) count = closure(nlist, count, pc + 1, false, atEnd); break
+        case OP_CLASS: if (matchClass(classes[pc], c)) count = closure(nlist, count, pc + 1, false, atEnd); break
+      }
+    }
+    if (!anchored) count = closure(nlist, count, 0, false, atEnd)
+    return internState(nlist, count)
+  }
+
+  let startEmpty = -2
+  let startNonEmpty = -2
+
+  function startState (atEnd) {
+    gen++
+    const count = closure(nlist, 0, 0, true, atEnd)
+    return internState(nlist, count)
+  }
+
+  function testDFA (s) {
+    const len = s.length
+    let id
+    if (len === 0) {
+      if (startEmpty === -2) startEmpty = startState(true)
+      id = startEmpty
+    } else {
+      if (startNonEmpty === -2) startNonEmpty = startState(false)
+      id = startNonEmpty
+    }
+    if (id < 0) return testNFA(s)
+    let state = states[id]
+    for (let pos = 0; pos < len; pos++) {
+      if (state.isMatch) return true
+      const c = s.charCodeAt(pos)
+      const atEnd = pos + 1 === len
+      let nid
+      if (c < 128) {
+        const table = atEnd ? state.nextEnd : state.next
+        nid = table[c]
+        if (nid === -2) { nid = step(state, c, atEnd); table[c] = nid }
+      } else {
+        nid = step(state, c, atEnd)
+      }
+      if (nid < 0) return testNFA(s)
+      state = states[nid]
+      if (anchored && state.pcs.length === 0) return false
+    }
+    return state.isMatch
+  }
+
+  return function test (s) {
+    return overflow ? testNFA(s) : testDFA(s)
+  }
 }
 
 function compileSafe (pattern) {

package/lib/version.js

@@ -7,4 +7,4 @@
 //
 // Kept in lockstep with package.json by `tests/test_version_sync.js`.
 
-module.exports = '1.6.2';
+module.exports = '1.7.0';

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ata-validator",
-  "version": "1.6.2",
+  "version": "1.7.0",
   "description": "JSON Schema validation with first-class TypeScript and zero runtime cost. AOT compile to per-schema ESM modules with zero validator dependency. Generic Validator<T> for TypeBox/Zod/Valibot composition. Optional runtime API. Standard Schema V1 compatible.",
   "main": "index.js",
   "module": "index.mjs",
@@ -44,7 +44,7 @@
     "release:check": "node scripts/regen-safe-regex-source.js && node tests/test_pack_purity.js && node scripts/check-doc-coverage.js && node tests/test_error_codes_lock.js && node tests/test_safe_regex_source_sync.js && node tests/test_version_sync.js",
     "build": "cmake-js build --target ata",
     "rebuild": "cmake-js rebuild --target ata",
-    "test": "node test.js && node tests/test_removed_aot_methods.js && node tests/test_no_native.js && node tests/test_no_eval.js && node tests/test_property_dependencies.js && node tests/test_v1_dialect.js && node tests/test_buffer_path_parity.js && node tests/test_pure_js_unsupported.js && node tests/test_native_load_order.js && node tests/test_pack_purity.js && node tests/test_make_native_package.js && node tests/test_browser_nofs.js && node tests/test_browser_imports_guard.js && node tests/test_version_sync.js && node tests/test_safe_regex_source_sync.js && node tests/test_t_builder.js && node tests/test_async_refine.js && node tests/test_safe_regex.js && node tests/test_safe_regex_integration.js && node tests/test_aot_build.js && node tests/test_aot_differential.js && node tests/test_aot_cli_build.js && node tests/test_aot_cli_smoke.js && node tests/test_bundle_standalone.js && node tests/test_standalone_anyof.js && node tests/test_standalone_formats.js && node tests/test_aot_additional_props_errors.js && node tests/test_id_anchor_refs.js && node tests/test_engine_routing.js && node tests/test_format_engine_parity.js && node tests/test_defs_pointer_alias.js && node tests/test_cross_doc_root_ref.js && node tests/test_no_input_mutation.js && node tests/test_typed_validator_runner.js && node tests/test_define_schema.js && node tests/test_error_codes_lock.js && node tests/test_error_code_lookup.js && node tests/test_error_order.js && node tests/test_nullable.js && node tests/test_validate_and_parse.js && node tests/test_validate_data.js && node tests/test_enrich_error.js && node tests/test_enrich_received.js && node tests/test_rich_errors_optout.js && node tests/test_error_messages.js && node tests/test_source_positions.js && node tests/fuzz_positions.js && node tests/test_data_positions.js && node tests/test_render_shared.js && node tests/test_renderers.js && node tests/test_runtime_error_dx.js && node tests/test_aot_error_dx.js && node tests/test_abort_early.js && node tests/test_branch_collapse.js && node tests/test_suggestions.js && node tests/test_cli_validate.js && node tests/test_cli_version.js && node benchmark/bench_aot_size.mjs",
+    "test": "node test.js && node tests/test_removed_aot_methods.js && node tests/test_no_native.js && node tests/test_no_eval.js && node tests/test_property_dependencies.js && node tests/test_v1_dialect.js && node tests/test_buffer_path_parity.js && node tests/test_buffer_gate.js && node tests/test_draft7_semantics.js && node tests/test_metaschema_ref.js && node tests/test_pure_js_unsupported.js && node tests/test_native_load_order.js && node tests/test_pack_purity.js && node tests/test_make_native_package.js && node tests/test_browser_nofs.js && node tests/test_browser_imports_guard.js && node tests/test_version_sync.js && node tests/test_native_loaded.js && node tests/test_safe_regex_source_sync.js && node tests/test_t_builder.js && node tests/test_async_refine.js && node tests/test_safe_regex.js && node tests/test_safe_regex_integration.js && node tests/test_aot_build.js && node tests/test_aot_differential.js && node tests/test_aot_cli_build.js && node tests/test_aot_cli_smoke.js && node tests/test_bundle_standalone.js && node tests/test_standalone_anyof.js && node tests/test_standalone_formats.js && node tests/test_aot_format_mode.js && node tests/test_aot_additional_props_errors.js && node tests/test_id_anchor_refs.js && node tests/test_engine_routing.js && node tests/test_engine_diagnostic.js && node tests/test_format_engine_parity.js && node tests/test_defs_pointer_alias.js && node tests/test_cross_doc_root_ref.js && node tests/test_codegen_entrypoint_agreement.js && node tests/test_pattern_properties_errors.js && node tests/test_no_input_mutation.js && node tests/test_typed_validator_runner.js && node tests/test_define_schema.js && node tests/test_error_codes_lock.js && node tests/test_error_code_lookup.js && node tests/test_error_order.js && node tests/test_nullable.js && node tests/test_validate_and_parse.js && node tests/test_validate_data.js && node tests/test_enrich_error.js && node tests/test_enrich_received.js && node tests/test_rich_errors_optout.js && node tests/test_error_messages.js && node tests/test_source_positions.js && node tests/fuzz_positions.js && node tests/test_data_positions.js && node tests/test_render_shared.js && node tests/test_renderers.js && node tests/test_runtime_error_dx.js && node tests/test_aot_error_dx.js && node tests/test_abort_early.js && node tests/test_branch_collapse.js && node tests/test_suggestions.js && node tests/test_cli_validate.js && node tests/test_cli_version.js && node benchmark/bench_aot_size.mjs",
     "bench:size": "node benchmark/bench_aot_size.mjs",
     "test:suite": "node tests/run_suite.js && node tests/run_suite.js draft7 && node tests/run_suite.js v1",
     "test:compat": "node tests/test_compat.js",
@@ -112,12 +112,12 @@
     "LICENSE"
   ],
   "optionalDependencies": {
-    "@ata-validator/native-darwin-arm64": "1.6.2",
-    "@ata-validator/native-linux-x64-gnu": "1.6.2",
-    "@ata-validator/native-linux-arm64-gnu": "1.6.2",
-    "@ata-validator/native-linux-x64-musl": "1.6.2",
-    "@ata-validator/native-linux-arm64-musl": "1.6.2",
-    "@ata-validator/native-win32-x64": "1.6.2"
+    "@ata-validator/native-darwin-arm64": "1.7.0",
+    "@ata-validator/native-linux-x64-gnu": "1.7.0",
+    "@ata-validator/native-linux-arm64-gnu": "1.7.0",
+    "@ata-validator/native-linux-x64-musl": "1.7.0",
+    "@ata-validator/native-linux-arm64-musl": "1.7.0",
+    "@ata-validator/native-win32-x64": "1.7.0"
   },
   "peerDependencies": {
     "yaml": "^2.0.0"