@danielblomma/cortex-mcp 0.4.5 → 1.0.0

package/scaffold/scripts/parsers/rust.mjs

@@ -0,0 +1,515 @@
+#!/usr/bin/env node
+/**
+ * Regex-based Rust parser for Cortex.
+ *
+ * Extracts semantic chunks from Rust source files: functions, structs, enums,
+ * traits, impl blocks (with methods), inline modules, macro_rules! definitions,
+ * use imports, and call relationships.
+ *
+ * No external dependencies — pure regex, always available.
+ */
+
+const CALL_KEYWORDS = new Set([
+  "if", "for", "while", "loop", "match", "return",
+  "Some", "None", "Ok", "Err", "Box", "Vec", "String",
+  "println", "eprintln", "format", "write", "writeln",
+  "panic", "todo", "unimplemented", "unreachable",
+  "assert", "assert_eq", "assert_ne", "debug_assert",
+  "debug_assert_eq", "debug_assert_ne",
+  "cfg", "derive", "allow", "warn", "deny"
+]);
+
+const VIS_PREFIX = /(?:pub(?:\s*\([^)]*\))?\s+)?/;
+const VIS_PREFIX_SRC = VIS_PREFIX.source;
+const LINE_START = "^[^\\S\\n]*";
+
+const FN_PATTERN = new RegExp(
+  `${LINE_START}${VIS_PREFIX_SRC}(?:default\\s+)?(?:async\\s+)?(?:unsafe\\s+)?(?:const\\s+)?(?:extern\\s+"[^"]*"\\s+)?fn\\s+([A-Za-z_]\\w*)`,
+  "gm"
+);
+
+const STRUCT_PATTERN = new RegExp(
+  `${LINE_START}${VIS_PREFIX_SRC}struct\\s+([A-Za-z_]\\w*)`,
+  "gm"
+);
+
+const ENUM_PATTERN = new RegExp(
+  `${LINE_START}${VIS_PREFIX_SRC}enum\\s+([A-Za-z_]\\w*)`,
+  "gm"
+);
+
+const TRAIT_PATTERN = new RegExp(
+  `${LINE_START}${VIS_PREFIX_SRC}(?:unsafe\\s+)?trait\\s+([A-Za-z_]\\w*)`,
+  "gm"
+);
+
+const IMPL_PATTERN = /^[^\S\n]*(?:unsafe\s+)?impl(?:<[^>]*>)?\s+(?:([A-Za-z_]\w*(?:<[^>]*>)?)\s+for\s+)?([A-Za-z_]\w*)/gm;
+
+const MOD_PATTERN = new RegExp(
+  `${LINE_START}${VIS_PREFIX_SRC}mod\\s+([A-Za-z_]\\w*)`,
+  "gm"
+);
+
+const MACRO_PATTERN = new RegExp(
+  `${LINE_START}${VIS_PREFIX_SRC}macro_rules!\\s+([A-Za-z_]\\w*)`,
+  "gm"
+);
+
+const USE_PATTERN = new RegExp(
+  `^\\s*${VIS_PREFIX_SRC}use\\s+(.+?)\\s*;`,
+  "gm"
+);
+
+function countLinesBefore(text, index) {
+  let line = 1;
+  for (let i = 0; i < index; i += 1) {
+    if (text[i] === "\n") {
+      line += 1;
+    }
+  }
+  return line;
+}
+
+function normalizeWhitespace(value) {
+  return value.replace(/\s+/g, " ").trim();
+}
+
+function findMatchingBrace(text, openBraceIndex) {
+  if (openBraceIndex < 0 || text[openBraceIndex] !== "{") {
+    return -1;
+  }
+
+  let depth = 0;
+  let inSingleLineComment = false;
+  let inBlockComment = false;
+  let inString = false;
+  let stringChar = "";
+  let inRawString = false;
+  let rawHashCount = 0;
+
+  for (let index = openBraceIndex; index < text.length; index += 1) {
+    const current = text[index];
+    const next = text[index + 1];
+
+    if (inSingleLineComment) {
+      if (current === "\n") {
+        inSingleLineComment = false;
+      }
+      continue;
+    }
+
+    if (inBlockComment) {
+      if (current === "*" && next === "/") {
+        inBlockComment = false;
+        index += 1;
+      }
+      continue;
+    }
+
+    if (inRawString) {
+      if (current === '"') {
+        let hashes = 0;
+        while (hashes < rawHashCount && text[index + 1 + hashes] === "#") {
+          hashes += 1;
+        }
+        if (hashes === rawHashCount) {
+          inRawString = false;
+          index += hashes;
+        }
+      }
+      continue;
+    }
+
+    if (inString) {
+      if (current === "\\" && next) {
+        index += 1;
+        continue;
+      }
+      if (current === stringChar) {
+        inString = false;
+        stringChar = "";
+      }
+      continue;
+    }
+
+    if (current === "/" && next === "/") {
+      inSingleLineComment = true;
+      index += 1;
+      continue;
+    }
+
+    if (current === "/" && next === "*") {
+      inBlockComment = true;
+      index += 1;
+      continue;
+    }
+
+    // Rust raw strings: r#"..."#, r##"..."##, etc.
+    if (current === "r" && (next === '"' || next === "#")) {
+      let hashes = 0;
+      let pos = index + 1;
+      while (text[pos] === "#") {
+        hashes += 1;
+        pos += 1;
+      }
+      if (text[pos] === '"') {
+        inRawString = true;
+        rawHashCount = hashes;
+        index = pos;
+        continue;
+      }
+    }
+
+    if (current === '"' || current === "'") {
+      // Rust lifetime annotations ('a) should not trigger string mode
+      if (current === "'" && next && /[a-zA-Z_]/.test(next)) {
+        // Check if this is a lifetime like 'a or a char like 'x'
+        const afterIdent = text.indexOf("'", index + 2);
+        const nextNewline = text.indexOf("\n", index + 1);
+        if (afterIdent === -1 || (nextNewline !== -1 && afterIdent > nextNewline) || afterIdent > index + 4) {
+          // Lifetime — skip the tick and identifier
+          continue;
+        }
+      }
+      inString = true;
+      stringChar = current;
+      continue;
+    }
+
+    if (current === "{") {
+      depth += 1;
+      continue;
+    }
+
+    if (current === "}") {
+      depth -= 1;
+      if (depth === 0) {
+        return index;
+      }
+    }
+  }
+
+  return -1;
+}
+
+function findOpenBraceAfterMatch(code, matchEnd) {
+  let inLineComment = false;
+  let inBlockComment = false;
+  let inString = false;
+  let stringChar = "";
+
+  for (let i = matchEnd; i < code.length; i += 1) {
+    const ch = code[i];
+    const next = code[i + 1];
+
+    if (inLineComment) {
+      if (ch === "\n") inLineComment = false;
+      continue;
+    }
+    if (inBlockComment) {
+      if (ch === "*" && next === "/") { inBlockComment = false; i += 1; }
+      continue;
+    }
+    if (inString) {
+      if (ch === "\\" && next) { i += 1; continue; }
+      if (ch === stringChar) { inString = false; stringChar = ""; }
+      continue;
+    }
+    if (ch === "/" && next === "/") { inLineComment = true; i += 1; continue; }
+    if (ch === "/" && next === "*") { inBlockComment = true; i += 1; continue; }
+    if (ch === '"' || ch === "'") { inString = true; stringChar = ch; continue; }
+
+    if (ch === "{") return i;
+    if (ch === ";") return -1; // Declaration without body
+  }
+  return -1;
+}
+
+function buildSignature(source) {
+  const snippet = normalizeWhitespace(source);
+  const braceIndex = snippet.indexOf("{");
+  return (braceIndex === -1 ? snippet : snippet.slice(0, braceIndex)).trim();
+}
+
+function extractUseImports(code) {
+  const imports = [];
+  let match;
+  USE_PATTERN.lastIndex = 0;
+  while ((match = USE_PATTERN.exec(code)) !== null) {
+    imports.push(match[1].trim());
+  }
+  return [...new Set(imports)];
+}
+
+function collectCallNames(body, chunkName) {
+  const refs = new Set();
+  const ownTailName = chunkName.split("::").pop() || chunkName;
+  const pattern = /\b([A-Za-z_]\w*(?:::\w+)*)\s*[!(]\s*/g;
+  let match;
+  while ((match = pattern.exec(body)) !== null) {
+    let name = match[1];
+    const tailName = name.split("::").pop() || name;
+    if (CALL_KEYWORDS.has(tailName) || tailName === ownTailName) {
+      continue;
+    }
+    // Skip if it matched a macro invocation keyword
+    if (CALL_KEYWORDS.has(name)) {
+      continue;
+    }
+    refs.add(tailName);
+  }
+  return [...refs];
+}
+
+function extractBlockChunks(code, pattern, kind, language) {
+  const chunks = [];
+  pattern.lastIndex = 0;
+  let match;
+  while ((match = pattern.exec(code)) !== null) {
+    const name = match[1];
+    const openBraceIndex = findOpenBraceAfterMatch(code, match.index + match[0].length);
+    if (openBraceIndex === -1) {
+      // Could be a unit struct like `struct Foo;` — extract as single-line chunk
+      if (kind === "struct") {
+        const lineEnd = code.indexOf("\n", match.index);
+        const endIdx = lineEnd === -1 ? code.length : lineEnd;
+        const body = code.slice(match.index, endIdx).trimEnd();
+        if (body.includes(";")) {
+          const startLine = countLinesBefore(code, match.index);
+          chunks.push({
+            name,
+            kind,
+            signature: normalizeWhitespace(body),
+            body,
+            startLine,
+            endLine: startLine,
+            language,
+            calls: [],
+            imports: []
+          });
+        }
+      }
+      continue;
+    }
+    const closeBraceIndex = findMatchingBrace(code, openBraceIndex);
+    if (closeBraceIndex === -1) continue;
+
+    const bodyEndIndex = closeBraceIndex + 1;
+    const body = code.slice(match.index, bodyEndIndex);
+    const startLine = countLinesBefore(code, match.index);
+    const endLine = countLinesBefore(code, Math.max(match.index, bodyEndIndex - 1));
+
+    chunks.push({
+      name,
+      kind,
+      signature: buildSignature(body),
+      body,
+      startLine,
+      endLine,
+      language,
+      calls: kind === "function" ? collectCallNames(body, name) : [],
+      imports: []
+    });
+  }
+  return chunks;
+}
+
+function extractImplBlocks(code, language, imports) {
+  const chunks = [];
+  IMPL_PATTERN.lastIndex = 0;
+  let match;
+  while ((match = IMPL_PATTERN.exec(code)) !== null) {
+    const traitName = match[1] || null;
+    const typeName = match[2];
+    const openBraceIndex = findOpenBraceAfterMatch(code, match.index + match[0].length);
+    if (openBraceIndex === -1) continue;
+    const closeBraceIndex = findMatchingBrace(code, openBraceIndex);
+    if (closeBraceIndex === -1) continue;
+
+    const implBody = code.slice(match.index, closeBraceIndex + 1);
+    const implStartLine = countLinesBefore(code, match.index);
+    const implEndLine = countLinesBefore(code, closeBraceIndex);
+    const implName = traitName ? `${traitName} for ${typeName}` : typeName;
+
+    // Add the impl block itself
+    chunks.push({
+      name: implName,
+      kind: "impl",
+      signature: buildSignature(implBody),
+      body: implBody,
+      startLine: implStartLine,
+      endLine: implEndLine,
+      language,
+      calls: [],
+      imports: []
+    });
+
+    // Extract methods within the impl block
+    const innerCode = code.slice(openBraceIndex + 1, closeBraceIndex);
+    const innerOffset = openBraceIndex + 1;
+    FN_PATTERN.lastIndex = 0;
+    let fnMatch;
+    while ((fnMatch = FN_PATTERN.exec(innerCode)) !== null) {
+      const fnName = fnMatch[1];
+      const qualifiedName = `${typeName}::${fnName}`;
+      const fnOpenBrace = findOpenBraceAfterMatch(innerCode, fnMatch.index + fnMatch[0].length);
+      if (fnOpenBrace === -1) continue;
+      const fnCloseBrace = findMatchingBrace(innerCode, fnOpenBrace);
+      if (fnCloseBrace === -1) continue;
+
+      const fnBodyEndIndex = fnCloseBrace + 1;
+      const fnBody = innerCode.slice(fnMatch.index, fnBodyEndIndex);
+      const fnStartLine = countLinesBefore(code, innerOffset + fnMatch.index);
+      const fnEndLine = countLinesBefore(code, innerOffset + Math.max(fnMatch.index, fnBodyEndIndex - 1));
+
+      chunks.push({
+        name: qualifiedName,
+        kind: "method",
+        signature: buildSignature(fnBody),
+        body: fnBody,
+        startLine: fnStartLine,
+        endLine: fnEndLine,
+        language,
+        calls: collectCallNames(fnBody, qualifiedName),
+        imports
+      });
+    }
+  }
+  return chunks;
+}
+
+function extractMacroChunks(code, language) {
+  const chunks = [];
+  MACRO_PATTERN.lastIndex = 0;
+  let match;
+  while ((match = MACRO_PATTERN.exec(code)) !== null) {
+    const name = match[1];
+    // macro_rules! uses { } or ( ) or [ ] as delimiters
+    const afterMatch = code.slice(match.index + match[0].length).trimStart();
+    let openChar, closeChar;
+    if (afterMatch[0] === "{") {
+      openChar = "{";
+    } else if (afterMatch[0] === "(") {
+      openChar = "(";
+    } else if (afterMatch[0] === "[") {
+      openChar = "[";
+    } else {
+      continue;
+    }
+
+    // For braces, use findMatchingBrace; for parens/brackets, do simple depth counting
+    let endIndex;
+    if (openChar === "{") {
+      const openBraceIndex = code.indexOf("{", match.index + match[0].length);
+      const closeBraceIndex = findMatchingBrace(code, openBraceIndex);
+      if (closeBraceIndex === -1) continue;
+      endIndex = closeBraceIndex + 1;
+    } else {
+      closeChar = openChar === "(" ? ")" : "]";
+      const startSearch = match.index + match[0].length + afterMatch.indexOf(openChar);
+      let depth = 0;
+      endIndex = -1;
+      for (let i = startSearch; i < code.length; i += 1) {
+        if (code[i] === openChar) depth += 1;
+        else if (code[i] === closeChar) {
+          depth -= 1;
+          if (depth === 0) {
+            endIndex = i + 1;
+            break;
+          }
+        }
+      }
+      if (endIndex === -1) continue;
+    }
+
+    const body = code.slice(match.index, endIndex);
+    const startLine = countLinesBefore(code, match.index);
+    const endLine = countLinesBefore(code, Math.max(match.index, endIndex - 1));
+
+    chunks.push({
+      name,
+      kind: "macro",
+      signature: `macro_rules! ${name}`,
+      body,
+      startLine,
+      endLine,
+      language,
+      calls: [],
+      imports: []
+    });
+  }
+  return chunks;
+}
+
+function extractTopLevelFunctions(code, language, implChunks, imports) {
+  const chunks = [];
+  FN_PATTERN.lastIndex = 0;
+  let match;
+  while ((match = FN_PATTERN.exec(code)) !== null) {
+    const name = match[1];
+    const openBraceIndex = findOpenBraceAfterMatch(code, match.index + match[0].length);
+    if (openBraceIndex === -1) continue;
+    const closeBraceIndex = findMatchingBrace(code, openBraceIndex);
+    if (closeBraceIndex === -1) continue;
+
+    const startLine = countLinesBefore(code, match.index);
+    const endLine = countLinesBefore(code, closeBraceIndex);
+
+    // Skip functions that are inside impl blocks (already extracted as methods)
+    const insideImpl = implChunks.some(
+      (impl) => impl.kind === "impl" && startLine >= impl.startLine && endLine <= impl.endLine
+    );
+    if (insideImpl) continue;
+
+    const bodyEndIndex = closeBraceIndex + 1;
+    const body = code.slice(match.index, bodyEndIndex);
+
+    chunks.push({
+      name,
+      kind: "function",
+      signature: buildSignature(body),
+      body,
+      startLine,
+      endLine,
+      language,
+      calls: collectCallNames(body, name),
+      imports
+    });
+  }
+  return chunks;
+}
+
+export function parseCode(code, filePath, language = "rust") {
+  const imports = extractUseImports(code);
+  const implChunks = extractImplBlocks(code, language, imports);
+  const structChunks = extractBlockChunks(code, STRUCT_PATTERN, "struct", language);
+  const enumChunks = extractBlockChunks(code, ENUM_PATTERN, "enum", language);
+  const traitChunks = extractBlockChunks(code, TRAIT_PATTERN, "trait", language);
+  const modChunks = extractBlockChunks(code, MOD_PATTERN, "module", language);
+  const macroChunks = extractMacroChunks(code, language);
+  const fnChunks = extractTopLevelFunctions(code, language, implChunks, imports);
+
+  const seen = new Set();
+  const chunks = [...structChunks, ...enumChunks, ...traitChunks, ...implChunks, ...modChunks, ...macroChunks, ...fnChunks].filter((chunk) => {
+    const key = `${chunk.kind}|${chunk.name}|${chunk.startLine}|${chunk.endLine}`;
+    if (seen.has(key)) return false;
+    seen.add(key);
+    return true;
+  });
+
+  return { chunks, errors: [] };
+}
+
+if (import.meta.url === `file://${process.argv[1]}`) {
+  const fs = await import("node:fs");
+  const filePath = process.argv[2];
+
+  if (!filePath) {
+    console.error("Usage: rust.mjs <file.rs>");
+    process.exit(1);
+  }
+
+  const code = fs.readFileSync(filePath, "utf8");
+  const result = parseCode(code, filePath, "rust");
+  console.log(JSON.stringify(result, null, 2));
+}

package/scaffold/scripts/parsers/sql.mjs

@@ -0,0 +1,137 @@
+#!/usr/bin/env node
+/**
+ * SQL parser for Cortex.
+ * Extracts stored procedures, views, functions, tables, and triggers as chunks.
+ */
+
+const SQL_OBJECT_PATTERN =
+  /create\s+(?:or\s+alter\s+)?(procedure|proc|view|function|table|trigger)\s+([^\s(]+)/gi;
+
+const SQL_REFERENCE_PATTERNS = [
+  /\bexec(?:ute)?\s+([#@]?[A-Za-z0-9_[\].]+)/gi,
+  /\bfrom\s+([#@]?[A-Za-z0-9_[\].]+)/gi,
+  /\bjoin\s+([#@]?[A-Za-z0-9_[\].]+)/gi,
+  /\bupdate\s+([#@]?[A-Za-z0-9_[\].]+)/gi,
+  /\binsert\s+into\s+([#@]?[A-Za-z0-9_[\].]+)/gi,
+  /\bdelete\s+from\s+([#@]?[A-Za-z0-9_[\].]+)/gi,
+  /\bmerge\s+into\s+([#@]?[A-Za-z0-9_[\].]+)/gi
+];
+
+const OBJECT_KIND_MAP = new Map([
+  ["proc", "procedure"],
+  ["procedure", "procedure"],
+  ["view", "view"],
+  ["function", "function"],
+  ["table", "table"],
+  ["trigger", "trigger"]
+]);
+
+function countLinesBefore(text, index) {
+  let line = 1;
+  for (let i = 0; i < index; i += 1) {
+    if (text[i] === "\n") {
+      line += 1;
+    }
+  }
+  return line;
+}
+
+function normalizeSqlName(value) {
+  if (!value) {
+    return "";
+  }
+
+  return value
+    .trim()
+    .replace(/[;"`]/g, "")
+    .replace(/\[(.+?)\]/g, "$1")
+    .replace(/\s+/g, "")
+    .replace(/^\.+|\.+$/g, "")
+    .replace(/\.\.+/g, ".")
+    .toLowerCase();
+}
+
+function sqlNameAliases(name) {
+  const normalized = normalizeSqlName(name);
+  if (!normalized) {
+    return [];
+  }
+
+  const aliases = new Set([normalized]);
+  const parts = normalized.split(".").filter(Boolean);
+  if (parts.length > 1) {
+    aliases.add(parts[parts.length - 1]);
+  }
+  return [...aliases];
+}
+
+function extractReferenceNames(body, selfAliases) {
+  const refs = new Set();
+
+  for (const pattern of SQL_REFERENCE_PATTERNS) {
+    let match;
+    while ((match = pattern.exec(body)) !== null) {
+      const name = normalizeSqlName(match[1]);
+      if (!name || name.startsWith("@") || name.startsWith("#")) {
+        continue;
+      }
+
+      const aliases = sqlNameAliases(name);
+      if (aliases.some((alias) => selfAliases.has(alias))) {
+        continue;
+      }
+
+      refs.add(name);
+    }
+  }
+
+  return [...refs];
+}
+
+export function parseCode(code, filePath, language = "sql") {
+  const matches = [...code.matchAll(SQL_OBJECT_PATTERN)];
+  const chunks = [];
+
+  for (let index = 0; index < matches.length; index += 1) {
+    const match = matches[index];
+    const kind = OBJECT_KIND_MAP.get((match[1] || "").toLowerCase()) || "sql_object";
+    const objectName = normalizeSqlName(match[2] || "");
+    if (!objectName) {
+      continue;
+    }
+
+    const start = match.index ?? 0;
+    const end = index + 1 < matches.length ? matches[index + 1].index ?? code.length : code.length;
+    const body = code.slice(start, end).trimEnd();
+    const firstLine = body.split(/\r?\n/, 1)[0]?.trim() || `${kind} ${objectName}`;
+    const selfAliases = new Set(sqlNameAliases(objectName));
+
+    chunks.push({
+      name: objectName,
+      kind,
+      signature: firstLine,
+      body,
+      startLine: countLinesBefore(code, start),
+      endLine: countLinesBefore(code, Math.max(start, end - 1)),
+      language,
+      calls: extractReferenceNames(body, selfAliases),
+      imports: []
+    });
+  }
+
+  return { chunks, errors: [] };
+}
+
+if (import.meta.url === `file://${process.argv[1]}`) {
+  const fs = await import("node:fs");
+  const filePath = process.argv[2];
+
+  if (!filePath) {
+    console.error("Usage: sql.mjs <file.sql>");
+    process.exit(1);
+  }
+
+  const code = fs.readFileSync(filePath, "utf8");
+  const result = parseCode(code, filePath, "sql");
+  console.log(JSON.stringify(result, null, 2));
+}