coffeehaml 0.6.0 → 0.7.1

package/dist/src/lexer.js

@@ -0,0 +1,504 @@
+// Generated by CoffeeScript 2.7.0
+// lexer.coffee — CoffeeHaml tokenizer
+// Converts raw .chaml source into a token stream with INDENT/DEDENT
+var countIndent, dedentToken, extractBracketed, indentToken, isHamlConstruct, parseModifiersAndAttrs, tokenizeLine;
+
+import {
+  SourceLocation
+} from './types.js';
+
+// ─── Token Types ───────────────────────────────────────────
+export var TokenType = {
+  TAG: 'TAG',
+  CLASS: 'CLASS',
+  ID: 'ID',
+  ATTRS_PAREN: 'ATTRS_PAREN',
+  ATTRS_BRACE: 'ATTRS_BRACE',
+  OUTPUT: 'OUTPUT',
+  OUTPUT_UNESC: 'OUTPUT_UNESC',
+  CONTROL: 'CONTROL',
+  COMMENT: 'COMMENT',
+  HTML_COMMENT: 'HTML_COMMENT',
+  FILTER: 'FILTER',
+  DOCTYPE: 'DOCTYPE',
+  TEXT: 'TEXT',
+  PROLOGUE: 'PROLOGUE',
+  INDENT: 'INDENT',
+  DEDENT: 'DEDENT',
+  NEWLINE: 'NEWLINE',
+  SELF_CLOSE: 'SELF_CLOSE'
+};
+
+// ─── Token (plain object convention) ───────────────────────
+// { type, value, location, indent? }
+
+// ─── Main entry ────────────────────────────────────────────
+export var tokenize = function(source, filename = null) {
+  var content, currentIndent, i, inPrologue, indent, indentStack, len, line, lineIndex, lineLength, lineStartOffset, lineTokens, lines, offset, rawLine, tokens;
+  tokens = [];
+  lines = source.split('\n');
+  indentStack = [0];
+  offset = 0;
+  lineIndex = 0;
+  inPrologue = true;
+  for (i = 0, len = lines.length; i < len; i++) {
+    rawLine = lines[i];
+    line = rawLine.replace(/\r$/, '');
+    lineLength = rawLine.length + 1;
+    // Skip empty lines
+    if (line.trim() === '') {
+      offset += lineLength;
+      lineIndex++;
+      continue;
+    }
+    indent = countIndent(line);
+    content = line.slice(indent);
+    if (content === '') {
+      offset += lineLength;
+      lineIndex++;
+      continue;
+    }
+    // Prologue detection: non-indented JS before first HAML construct
+    if (inPrologue && indent === 0 && !isHamlConstruct(content)) {
+      tokens.push({
+        type: TokenType.PROLOGUE,
+        value: content,
+        location: {
+          start: {
+            line: lineIndex,
+            column: indent
+          },
+          end: {
+            line: lineIndex,
+            column: indent + content.length
+          },
+          offset: offset + indent,
+          length: lineLength - indent,
+          file: filename
+        }
+      });
+      offset += lineLength;
+      lineIndex++;
+      continue;
+    }
+    inPrologue = false;
+    // Handle indentation changes
+    currentIndent = indentStack[indentStack.length - 1];
+    if (indent > currentIndent) {
+      indentStack.push(indent);
+      tokens.push(indentToken(indent, offset, indent, filename, lineIndex));
+    } else if (indent < currentIndent) {
+      while (indentStack.length > 1 && indent < indentStack[indentStack.length - 1]) {
+        indentStack.pop();
+        tokens.push(dedentToken(offset, filename, lineIndex));
+      }
+      if (indent !== indentStack[indentStack.length - 1]) {
+        indentStack.push(indent);
+        tokens.push(indentToken(indent, offset, indent, filename, lineIndex));
+      }
+    }
+    // Parse line content
+    lineStartOffset = offset + indent;
+    lineTokens = tokenizeLine(content, lineStartOffset, lineLength - indent, filename, lineIndex);
+    tokens.push(...lineTokens);
+    offset += lineLength;
+    lineIndex++;
+  }
+  // Emit remaining DEDENT tokens at EOF
+  while (indentStack.length > 1) {
+    indentStack.pop();
+    tokens.push(dedentToken(offset, filename, lineIndex));
+  }
+  return tokens;
+};
+
+// ─── Line Tokenization ─────────────────────────────────────
+tokenizeLine = function(content, lineStartOffset, _lineLength, filename = null, lineIndex = 0) {
+  var end, expr, filterContent, filterName, firstChar, loc, modResult, pos, remaining, spaceIdx, tagName, tokens, trimmedOffset, val;
+  tokens = [];
+  pos = 0;
+  loc = function(start, end) {
+    return {
+      start: {
+        line: lineIndex != null ? lineIndex : 0,
+        column: lineStartOffset + start
+      },
+      end: {
+        line: lineIndex != null ? lineIndex : 0,
+        column: lineStartOffset + end
+      },
+      offset: lineStartOffset + start,
+      length: end - start,
+      file: filename
+    };
+  };
+  firstChar = content[0];
+  // %tag
+  if (firstChar === '%') {
+    end = 1;
+    while (end < content.length && /[\w-]/.test(content[end])) {
+      end++;
+    }
+    tagName = content.slice(1, end);
+    tokens.push({
+      type: TokenType.TAG,
+      value: tagName,
+      location: loc(0, end)
+    });
+    pos = end;
+    // Self-closing /
+    if (content[pos] === '/') {
+      tokens.push({
+        type: TokenType.SELF_CLOSE,
+        value: '/',
+        location: loc(pos, pos + 1)
+      });
+      pos++;
+    }
+    modResult = parseModifiersAndAttrs(content, pos, lineStartOffset, filename, lineIndex);
+    tokens.push(...modResult.tokens);
+    pos = modResult.pos;
+    remaining = content.slice(pos).trimStart();
+    trimmedOffset = content.length - remaining.length;
+    if (remaining) {
+      if (remaining.startsWith('= ') || remaining === '=') {
+        tokens.push({
+          type: TokenType.OUTPUT,
+          value: remaining.slice(1).trimStart(),
+          location: loc(trimmedOffset + 1, content.length)
+        });
+      } else if (remaining.startsWith('!= ')) {
+        tokens.push({
+          type: TokenType.OUTPUT_UNESC,
+          value: remaining.slice(2).trimStart(),
+          location: loc(trimmedOffset + 2, content.length)
+        });
+      } else if (!(remaining.startsWith('{') || remaining.startsWith('('))) {
+        tokens.push({
+          type: TokenType.TEXT,
+          value: remaining,
+          location: loc(trimmedOffset, content.length)
+        });
+      }
+    }
+    return tokens;
+  }
+  // .class / #id (implicit div) — only if followed by valid identifier
+  if ((firstChar === '.' || firstChar === '#') && /^[.#][\w-]/.test(content)) {
+    modResult = parseModifiersAndAttrs(content, 0, lineStartOffset, filename, lineIndex);
+    tokens.push(...modResult.tokens);
+    pos = modResult.pos;
+    remaining = content.slice(pos).trimStart();
+    trimmedOffset = content.length - remaining.length;
+    if (remaining) {
+      if (remaining.startsWith('= ') || remaining === '=') {
+        tokens.push({
+          type: TokenType.OUTPUT,
+          value: remaining.slice(1).trimStart(),
+          location: loc(trimmedOffset + 1, content.length)
+        });
+      } else if (remaining.startsWith('!= ')) {
+        tokens.push({
+          type: TokenType.OUTPUT_UNESC,
+          value: remaining.slice(2).trimStart(),
+          location: loc(trimmedOffset + 2, content.length)
+        });
+      } else if (!(remaining.startsWith('{') || remaining.startsWith('('))) {
+        tokens.push({
+          type: TokenType.TEXT,
+          value: remaining,
+          location: loc(trimmedOffset, content.length)
+        });
+      }
+    }
+    return tokens;
+  }
+  // = expression
+  if (firstChar === '=' && content[1] !== '=') {
+    expr = content.slice(1).trimStart();
+    return [
+      {
+        type: TokenType.OUTPUT,
+        value: expr,
+        location: loc(1,
+      content.length)
+      }
+    ];
+  }
+  // != expression
+  if (firstChar === '!' && content[1] === '=') {
+    expr = content.slice(2).trimStart();
+    return [
+      {
+        type: TokenType.OUTPUT_UNESC,
+        value: expr,
+        location: loc(2,
+      content.length)
+      }
+    ];
+  }
+  // - control / comment
+  if (firstChar === '-') {
+    if (content[1] === '#') {
+      return [
+        {
+          type: TokenType.COMMENT,
+          value: content.slice(2).trimStart(),
+          location: loc(2,
+        content.length)
+        }
+      ];
+    }
+    expr = content.slice(1).trimStart();
+    return [
+      {
+        type: TokenType.CONTROL,
+        value: expr,
+        location: loc(1,
+      content.length)
+      }
+    ];
+  }
+  // / HTML comment
+  if (firstChar === '/') {
+    return [
+      {
+        type: TokenType.HTML_COMMENT,
+        value: content.slice(1).trimStart(),
+        location: loc(1,
+      content.length)
+      }
+    ];
+  }
+  // :filter
+  if (firstChar === ':') {
+    spaceIdx = content.search(/\s/);
+    filterName = spaceIdx === -1 ? content.slice(1) : content.slice(1, spaceIdx);
+    filterContent = spaceIdx === -1 ? '' : content.slice(spaceIdx + 1);
+    return [
+      {
+        type: TokenType.FILTER,
+        value: `${filterName}\n${filterContent}`,
+        location: loc(0,
+      content.length)
+      }
+    ];
+  }
+  // !!! doctype
+  if (content.startsWith('!!!')) {
+    val = content.slice(3).trimStart();
+    return [
+      {
+        type: TokenType.DOCTYPE,
+        value: val || 'html',
+        location: loc(0,
+      content.length)
+      }
+    ];
+  }
+  return [
+    {
+      // Plain text
+      type: TokenType.TEXT,
+      value: content,
+      location: loc(0,
+    content.length)
+    }
+  ];
+};
+
+// ─── Modifiers and Attributes ──────────────────────────────
+parseModifiersAndAttrs = function(content, startPos, lineStartOffset, filename = null, lineIndex = 0) {
+  var block, ch, className, end, idName, loc, pos, tokens;
+  tokens = [];
+  pos = startPos;
+  loc = function(start, end) {
+    return {
+      start: {
+        line: lineIndex != null ? lineIndex : 0,
+        column: lineStartOffset + start
+      },
+      end: {
+        line: lineIndex != null ? lineIndex : 0,
+        column: lineStartOffset + end
+      },
+      offset: lineStartOffset + start,
+      length: end - start,
+      file: filename
+    };
+  };
+  while (pos < content.length) {
+    ch = content[pos];
+    if (ch === ' ' || ch === '\t') {
+      pos++;
+      continue;
+    }
+    // .class
+    if (ch === '.') {
+      end = pos + 1;
+      while (end < content.length && /[\w-]/.test(content[end])) {
+        end++;
+      }
+      className = content.slice(pos + 1, end);
+      tokens.push({
+        type: TokenType.CLASS,
+        value: className,
+        location: loc(pos, end)
+      });
+      pos = end;
+      continue;
+    }
+    // #id
+    if (ch === '#') {
+      end = pos + 1;
+      while (end < content.length && /[\w-]/.test(content[end])) {
+        end++;
+      }
+      idName = content.slice(pos + 1, end);
+      tokens.push({
+        type: TokenType.ID,
+        value: idName,
+        location: loc(pos, end)
+      });
+      pos = end;
+      continue;
+    }
+    // {attribute block}
+    if (ch === '{') {
+      block = extractBracketed(content, pos, '{', '}');
+      if (block !== null) {
+        tokens.push({
+          type: TokenType.ATTRS_BRACE,
+          value: block,
+          location: loc(pos, pos + block.length + 2)
+        });
+        pos += block.length + 2;
+      } else {
+        tokens.push({
+          type: TokenType.ATTRS_BRACE,
+          value: content.slice(pos + 1),
+          location: loc(pos, content.length)
+        });
+        pos = content.length;
+      }
+      continue;
+    }
+    // (attribute block)
+    if (ch === '(') {
+      block = extractBracketed(content, pos, '(', ')');
+      if (block !== null) {
+        tokens.push({
+          type: TokenType.ATTRS_PAREN,
+          value: block,
+          location: loc(pos, pos + block.length + 2)
+        });
+        pos += block.length + 2;
+      } else {
+        tokens.push({
+          type: TokenType.ATTRS_PAREN,
+          value: content.slice(pos + 1),
+          location: loc(pos, content.length)
+        });
+        pos = content.length;
+      }
+      continue;
+    }
+    // Anything else → stop
+    break;
+  }
+  return {tokens, pos};
+};
+
+// ─── Helpers ───────────────────────────────────────────────
+isHamlConstruct = function(content) {
+  var fc;
+  fc = content[0];
+  return (fc === '%' || fc === '.' || fc === '#' || fc === '=' || fc === '-' || fc === '/' || fc === ':') || (fc === '!' && content[1] === '=') || content.startsWith('!!!');
+};
+
+countIndent = function(line) {
+  var ch, count, i, len;
+  count = 0;
+  for (i = 0, len = line.length; i < len; i++) {
+    ch = line[i];
+    if (ch === ' ') {
+      count++;
+    } else if (ch === '\t') {
+      count += 2;
+    } else {
+      break;
+    }
+  }
+  return count;
+};
+
+indentToken = function(_indent, offset, level, filename = null, line = 0) {
+  return {
+    type: TokenType.INDENT,
+    value: '',
+    indent: level,
+    location: {
+      start: {
+        line: line != null ? line : 0,
+        column: 0
+      },
+      end: {
+        line: line != null ? line : 0,
+        column: 0
+      },
+      offset,
+      length: 0,
+      file: filename
+    }
+  };
+};
+
+dedentToken = function(offset, filename = null, line = 0) {
+  return {
+    type: TokenType.DEDENT,
+    value: '',
+    indent: void 0,
+    location: {
+      start: {
+        line: line != null ? line : 0,
+        column: 0
+      },
+      end: {
+        line: line != null ? line : 0,
+        column: 0
+      },
+      offset,
+      length: 0,
+      file: filename
+    }
+  };
+};
+
+extractBracketed = function(content, start, open, close) {
+  var depth, pos, q, ref;
+  depth = 0;
+  pos = start;
+  while (pos < content.length) {
+    if (content[pos] === open) {
+      depth++;
+    } else if (content[pos] === close) {
+      depth--;
+      if (depth === 0) {
+        return content.slice(start + 1, pos);
+      }
+    }
+    // Skip string literals
+    if ((ref = content[pos]) === '"' || ref === "'") {
+      q = content[pos];
+      pos++;
+      while (pos < content.length && content[pos] !== q) {
+        if (content[pos] === '\\') {
+          pos++;
+        }
+        pos++;
+      }
+    }
+    pos++;
+  }
+  return null;
+};