fix: allow unicode to be identifiers, fixes #655

Author: harttle

bin/character-gen.js

@@ -4,18 +4,19 @@ const isQuote = c => c === '"' || c === "'"
 const isOperator = c => '!=<>'.includes(c)
 const isNumber = c => c >= '0' && c <= '9'
 const isCharacter = c => (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')
-const isIdentifier = c => '_-?'.includes(c) || isCharacter(c) || isNumber(c)
-const isBlank = c => c === '\n' || c === '\t' || c === ' ' || c === '\r' || c === '\v' || c === '\f'
+const isWord = c => '_-?'.includes(c) || isCharacter(c) || isNumber(c)
+const isBlank = c => '\n\t \r\v\f'.includes(c)
 const isInlineBlank = c => c === '\t' || c === ' ' || c === '\r'
 const isSign = c => c === '-' || c === '+'
 // See https://developer.mozilla.org/zh-CN/docs/Web/JavaScript/Reference/Global_Objects/RegExp
 const unicodeBlanks = '\u00a0\u1680\u180e\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200a\u2028\u2029\u202f\u205f\u3000'
+const unicodePunctuations = '“”'
 
 const types = []
 for (let i = 0; i < 128; i++) {
   const c = String.fromCharCode(i)
   let n = 0
-  if (isIdentifier(c)) n |= 1
+  if (isWord(c)) n |= 1
   if (isOperator(c)) n |= 2
   if (isBlank(c)) n |= 4
   if (isQuote(c)) n |= 8
@@ -31,13 +32,20 @@ console.log(`
 // This file is generated by bin/character-gen.js
 // bitmask character types to boost performance
 export const TYPES = [${types.join(', ')}]
-export const IDENTIFIER = 1
+export const WORD = 1
 export const OPERATOR = 2
 export const BLANK = 4
 export const QUOTE = 8
 export const INLINE_BLANK = 16
 export const NUMBER = 32
 export const SIGN = 64
+export const PUNCTUATION = 128
+
+export function isWord (char: string): boolean {
+  const code = char.charCodeAt(0)
+  return code >= 128 ? !TYPES[code] : !!(TYPES[code] & WORD)
+}
 `.trim())
 
 console.log([...unicodeBlanks].map(char => `TYPES[${char.charCodeAt(0)}]`).join(' = ') + ' = BLANK')
+console.log([...unicodePunctuations].map(char => `TYPES[${char.charCodeAt(0)}]`).join(' = ') + ' = PUNCTUATION')

bin/perf-diff.sh

@@ -1,11 +1,11 @@
 #!/usr/bin/env bash
 
-VERSION_LATEST=$(cat package.json | grep '"version":' | awk -F'"' '{print $4}')
+VERSION_LATEST=$(cat package.json | grep '"version":' | head -1 | awk -F'"' '{print $4}')
 FILE_LOCAL=dist/liquid.node.cjs.js
 FILE_LATEST=dist/liquid.node.cjs.$VERSION_LATEST.js
 URL_LATEST=https://unpkg.com/liquidjs@$VERSION_LATEST/dist/liquid.node.cjs.js
 
-if [ ! -f $FILE_LATEST ]; then
+if [ ! -f "$FILE_LATEST" ]; then
     curl $URL_LATEST > $FILE_LATEST
 fi
 

docs/themes/navy/layout/partial/all-contributors.swig

@@ -68,6 +68,13 @@
       <td align="center" valign="top" width="0%"><a href="https://github.com/mahyar-pasarzangene"><img src="https://avatars.githubusercontent.com/u/16485039?v=4?s=100" width="100px;" alt="Mahyar Pasarzangene"/></a></td>
       <td align="center" valign="top" width="0%"><a href="https://hubelbauer.net/"><img src="https://avatars.githubusercontent.com/u/6831144?v=4?s=100" width="100px;" alt="Tomáš Hübelbauer"/></a></td>
       <td align="center" valign="top" width="0%"><a href="https://sixtwothree.org"><img src="https://avatars.githubusercontent.com/u/73866?v=4?s=100" width="100px;" alt="Jason Garber"/></a></td>
+      <td align="center" valign="top" width="0%"><a href="https://www.checkoutblocks.com/"><img src="https://avatars.githubusercontent.com/u/114603307?v=4?s=100" width="100px;" alt="Checkout Blocks"/></a></td>
+      <td align="center" valign="top" width="0%"><a href="https://www.dropkiq.com/"><img src="https://avatars.githubusercontent.com/u/69064?v=4?s=100" width="100px;" alt="Adam Darrah"/></a></td>
+      <td align="center" valign="top" width="0%"><a href="https://www.11ty.dev/"><img src="https://avatars.githubusercontent.com/u/35147177?v=4?s=100" width="100px;" alt="Eleventy"/></a></td>
+      <td align="center" valign="top" width="0%"><a href="http://nickreilingh.com/"><img src="https://avatars.githubusercontent.com/u/2458645?v=4?s=100" width="100px;" alt="Nick Reilingh"/></a></td>
+      <td align="center" valign="top" width="0%"><a href="http://ebobby.org"><img src="https://avatars.githubusercontent.com/u/170356?v=4?s=100" width="100px;" alt="Francisco Soto"/></a></td>
+      <td align="center" valign="top" width="0%"><a href="https://www.davidlj95.com"><img src="https://avatars.githubusercontent.com/u/8050648?v=4?s=100" width="100px;" alt="David LJ"/></a></td>
+      <td align="center" valign="top" width="0%"><a href="https://github.com/RasmusWL"><img src="https://avatars.githubusercontent.com/u/1054041?v=4?s=100" width="100px;" alt="Rasmus Wriedt Larsen"/></a></td>
     </tr>
   </tbody>
 </table>

package.json

@@ -26,6 +26,7 @@
     "build:cjs": "BUNDLES=cjs rollup -c rollup.config.mjs",
     "build:min": "BUNDLES=min rollup -c rollup.config.mjs",
     "build:umd": "BUNDLES=umd rollup -c rollup.config.mjs",
+    "build:charmap": "./bin/character-gen.js > src/util/character.ts",
     "build:docs": "bin/build-docs.sh"
   },
   "bin": {

src/parser/tokenizer.ts

@@ -1,6 +1,6 @@
 import { FilteredValueToken, TagToken, HTMLToken, HashToken, QuotedToken, LiquidTagToken, OutputToken, ValueToken, Token, RangeToken, FilterToken, TopLevelToken, PropertyAccessToken, OperatorToken, LiteralToken, IdentifierToken, NumberToken } from '../tokens'
 import { OperatorHandler } from '../render/operator'
-import { TrieNode, LiteralValue, Trie, createTrie, ellipsis, literalValues, TokenizationError, TYPES, QUOTE, BLANK, IDENTIFIER, NUMBER, SIGN } from '../util'
+import { TrieNode, LiteralValue, Trie, createTrie, ellipsis, literalValues, TokenizationError, TYPES, QUOTE, BLANK, NUMBER, SIGN, isWord } from '../util'
 import { Operators, Expression } from '../render'
 import { NormalizedFullOptions, defaultOptions } from '../liquid-options'
 import { FilterArg } from './filter-arg'
@@ -59,7 +59,7 @@ export class Tokenizer {
       if (node['end']) info = node
     }
     if (!info) return -1
-    if (info['needBoundary'] && (this.peekType(i - this.p) & IDENTIFIER)) return -1
+    if (info['needBoundary'] && isWord(this.peek(i - this.p))) return -1
     return i
   }
   readFilteredValue (): FilteredValueToken {
@@ -245,7 +245,7 @@ export class Tokenizer {
   readIdentifier (): IdentifierToken {
     this.skipBlank()
     const begin = this.p
-    while (!this.end() && this.peekType() & IDENTIFIER) ++this.p
+    while (!this.end() && isWord(this.peek())) ++this.p
     return new IdentifierToken(this.input, begin, this.p, this.file)
   }
 
@@ -351,7 +351,7 @@ export class Tokenizer {
         n++
       } else break
     }
-    if (digitFound && !(this.peekType(n) & IDENTIFIER)) {
+    if (digitFound && !isWord(this.peek(n))) {
       const num = new NumberToken(this.input, this.p, this.p + n, this.file)
       this.advance(n)
       return num

src/util/character.ts

@@ -3,11 +3,18 @@
 // This file is generated by bin/character-gen.js
 // bitmask character types to boost performance
 export const TYPES = [0, 0, 0, 0, 0, 0, 0, 0, 0, 20, 4, 4, 4, 20, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 20, 2, 8, 0, 0, 0, 0, 8, 0, 0, 0, 64, 0, 65, 0, 0, 33, 33, 33, 33, 33, 33, 33, 33, 33, 33, 0, 0, 2, 2, 2, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0]
-export const IDENTIFIER = 1
+export const WORD = 1
 export const OPERATOR = 2
 export const BLANK = 4
 export const QUOTE = 8
 export const INLINE_BLANK = 16
 export const NUMBER = 32
 export const SIGN = 64
+export const PUNCTUATION = 128
+
+export function isWord (char: string): boolean {
+  const code = char.charCodeAt(0)
+  return code >= 128 ? !TYPES[code] : !!(TYPES[code] & WORD)
+}
 TYPES[160] = TYPES[5760] = TYPES[6158] = TYPES[8192] = TYPES[8193] = TYPES[8194] = TYPES[8195] = TYPES[8196] = TYPES[8197] = TYPES[8198] = TYPES[8199] = TYPES[8200] = TYPES[8201] = TYPES[8202] = TYPES[8232] = TYPES[8233] = TYPES[8239] = TYPES[8287] = TYPES[12288] = BLANK
+TYPES[8220] = TYPES[8221] = PUNCTUATION

src/util/operator-trie.ts

@@ -1,4 +1,4 @@
-import { IDENTIFIER, TYPES } from '../util/character'
+import { isWord } from '../util/character'
 
 interface TrieInput<T> {
   [key: string]: T
@@ -25,7 +25,7 @@ export function createTrie<T = any> (input: TrieInput<T>): Trie<T> {
       const c = name[i]
       node[c] = node[c] || {}
 
-      if (i === name.length - 1 && (TYPES[name.charCodeAt(i)] & IDENTIFIER)) {
+      if (i === name.length - 1 && isWord(name[i])) {
         node[c].needBoundary = true
       }
 

test/e2e/issues.spec.ts

@@ -464,4 +464,9 @@ describe('Issues', function () {
     }
     expect(engine.parseAndRenderSync(tpl, ctx)).toEqual('FOO')
   })
+  it('#655 Error in the tokenization process due to an invalid value expression', () => {
+    const engine = new Liquid()
+    const result = engine.parseAndRenderSync('{{ÜLKE}}', { ÜLKE: 'Türkiye' })
+    expect(result).toEqual('Türkiye')
+  })
 })