sugar-high 0.1.2 → 0.2.2

package/README.md

@@ -3,7 +3,7 @@
 
 Super lightweight JSX syntax highlighter, around 1KB after minified and gzipped
 
-![img](https://repository-images.githubusercontent.com/453236442/9416c941-9c40-4dbe-a73e-0e0d21993802)
+![img](https://repository-images.githubusercontent.com/453236442/82cf2807-78f0-4009-bbdf-d7e753f73cf4)
 
 ### Usage
 
@@ -33,7 +33,7 @@ Then make your own theme with customized colors by token type and put in global
  * Class, number and null
  * sign
  * comment
- * 
+ * jsxliterals
  */
 .sh__class {
   color: #2d5e9d;
@@ -53,7 +53,9 @@ Then make your own theme with customized colors by token type and put in global
 .sh__comment {
   color: #a19595;
 }
-
+.sh__jsxliterals {
+  color: #03066e;
+}
 ```
 
 ### LICENSE

package/lib/index.mjs

@@ -71,10 +71,11 @@ const signs = new Set([
   '[',
   ']',
   '#',
+  '@',
   '\\',
 ])
 
-const types = [
+export const types = [
   'identifier',
   'keyword',
   'string',
@@ -83,10 +84,11 @@ const types = [
   'comment',
   'break',
   'space',
+  'jsxliterals'
 ]
 
 /**
- * 
+ *
  * 0 - identifier
  * 1 - keyword
  * 2 - string
@@ -95,7 +97,8 @@ const types = [
  * 5 - comment
  * 6 - break
  * 7 - space
- * 
+ * 8 - jsx literals
+ *
  */
 const [
   T_IDENTIFIER,
@@ -106,6 +109,7 @@ const [
   T_COMMENT,
   T_BREAK,
   T_SPACE,
+  T_JSX_LITERALS,
 ] = types.map((_, i) => i)
 
 function isSpaces(str) {
@@ -123,8 +127,16 @@ function encode(str) {
     .replace(/[^\S\r\n]/g, ' ')
 }
 
+function isWord(chr) {
+  return /^[\w_]+$/.test(chr)
+}
+
 function isIdentifierChar(chr) {
-  return /[a-zA-Z0-9_$]/.test(chr)
+  return /^[a-zA-Z_$]$/.test(chr)
+}
+
+function isIdentifier(str) {
+  return isIdentifierChar(str[0]) && (str.length === 1 || isWord(str.slice(1)))
 }
 
 function isStringQuotation(chr) {
@@ -140,8 +152,8 @@ function isRegexStart(str) {
   return str[0] === '/' && !isCommentStart(str[0] + str[1])
 }
 
-/** 
- * @param {string} code 
+/**
+ * @param {string} code
  * @return {Array<[number, string]>}
  */
 export function tokenize(code) {
@@ -160,43 +172,32 @@ export function tokenize(code) {
   const inJsxLiterals = () => !jsx.tag && jsx.child && !jsx.expr
 
   function classify(token) {
-    const [chr0, chr1] = [token[0], token[1]]
-    if (isCommentStart(chr0 + chr1)) {
-      return T_COMMENT
-    } else if (keywords.has(token)) {
+    const chr0 = token[0]
+    if (keywords.has(token)) {
       return T_KEYWORD
     } else if (token === '\n') {
       return T_BREAK
-    } else if (token === ' ') {
+    } else if (isSpaces(token)) {
       return T_SPACE
     } else if (token.split('').every(ch => signs.has(ch))) {
       return T_SIGN
     } else if (
+      !inJsxLiterals() &&
       (
-       // is quoted string
-       (isStringQuotation(chr0) && !isStringQuotation(chr1)) ||
-       // is regex
-       (!jsx.tag && isRegexStart(chr0 + chr1) && token[token.length - 1] === '/')
-     )
-   ) {
-     return T_STRING
-   } else if (
-      !inJsxLiterals() && 
-      (
-        isIdentifierChar(chr0) &&
+        isWord(chr0) &&
         chr0 === chr0.toUpperCase() ||
         token === 'null'
       )
     ) {
       return T_CLS_NUMBER
     } else {
-      return T_IDENTIFIER
+      return isIdentifier(token) ? T_IDENTIFIER : inJsxLiterals() ? T_JSX_LITERALS : T_STRING
     }
   }
 
-  const append = () => {
+  const append = (_type) => {
     if (current) {
-      type = classify(current)
+      type = _type || classify(current)
       /** @type [number, string]  */
       const pair = [type, current]
       if (type !== T_SPACE && type !== T_BREAK) {
@@ -211,11 +212,11 @@ export function tokenize(code) {
     const prev = code[i - 1]
     const next = code[i + 1]
     const c_n = curr + next // current and next
-    const p_c = prev + curr // previous and current    
+    const p_c = prev + curr // previous and current
     const isJsxLiterals = inJsxLiterals()
 
     if (jsx.tag) {
-      const isOpenElementEnd = curr === '>' 
+      const isOpenElementEnd = curr === '>'
       const isCloseElementEnd = p_c === '/>'
       jsx.child = !isCloseElementEnd && !isCloseElementEnd
       jsx.tag = !(isOpenElementEnd || isCloseElementEnd)
@@ -226,45 +227,69 @@ export function tokenize(code) {
       jsx.child = false
     }
 
-    if (
-      (isStringQuotation(curr) || !jsx.tag && isRegexStart(c_n))
-    ) {
+    const isQuotationChar = isStringQuotation(curr)
+    const isRegexChar = !jsx.tag && isRegexStart(c_n)
+    if (isQuotationChar || isRegexChar) {
       append()
       const [lastType, lastToken] = last
-      if (!((lastType === T_SIGN && !'()[]'.includes(lastToken)) || lastType === T_COMMENT)) {
+      // Special cases that are not considered as regex:
+      // * (expr1) / expr2: `()` before `/` operator is still in expression
+      // * <non comment start>/ expr: non comment start before `/` is not regex
+      if (
+        isRegexChar &&
+        lastType &&
+        !(
+          (lastType === T_SIGN && !'()'.includes(lastToken)) ||
+          lastType === T_COMMENT
+        )
+      ) {
         current = curr
         append()
         continue
       }
-      let j = i + 1
-      let strEnd = i
-      const inString = () => j < code.length && code[j] !== '\n'
+      const start = i++
+      const leftQuote = code[start]
+      let foundClose = false
+
+      // end of line of end of file
+      const isEof = () => i >= code.length
+      const isEol = () => isEof() || code[i] === '\n'
       // string quotation
-      if (isStringQuotation(curr)) {
-        for (; inString(); j++) {
-          if (isStringQuotation(code[j])) {
-            strEnd = j
+      if (isQuotationChar) {
+        for (; !isEof(); i++) {
+          // handle single quotes
+          if ((leftQuote === "'" || leftQuote === '"') && code[i] === '\n') break
+          if (isStringQuotation(code[i]) && code[i] === leftQuote) {
+            foundClose = true
             break
           }
         }
       } else {
         // regex
-        for (; inString(); j++) {
-          if (code[j] === '/' && code[j - 1] !== '\\') {
-            strEnd = j
+        for (; !isEol(); i++) {
+          if (code[i] === '/' && code[i - 1] !== '\\') {
+            foundClose = true
+            // append regex flags
+            while (start !== i && /^[a-z]$/.test(code[i + 1]) && !isEol()) {
+              i++
+            }
             break
           }
         }
       }
-      if (strEnd !== i) {
-        current = code.slice(i, strEnd + 1)
-        append()
-        i = j
+      if (start !== i && foundClose) {
+        // If current line is fully closed with string quotes or regex slashes,
+        // add them to tokens
+        current = code.slice(start, i + 1)
+        append(T_STRING)
       } else {
+        // If it doesn't match any of the above, just leave it as operator and move on
         current = curr
         append()
+        i = start
       }
     } else if (isCommentStart(c_n)) {
+      append()
       const start = i
       if (next === '/') {
         for (; i < code.length && code[i] !== '\n'; i++);
@@ -272,7 +297,7 @@ export function tokenize(code) {
         for (; i < code.length && code[i - 1] + code[i] !== '*/'; i++);
       }
       current = code.slice(start, i + 1)
-      append()
+      append(T_COMMENT)
     } else if (curr === ' ' || curr === '\n') {
       if (
         curr === ' ' &&
@@ -290,7 +315,7 @@ export function tokenize(code) {
     } else {
       if (
         (isJsxLiterals && !jsxBrackets.has(curr)) ||
-        isIdentifierChar(curr) === isIdentifierChar(current[current.length - 1]) &&
+        isWord(curr) === isWord(current[current.length - 1]) &&
         !signs.has(curr)
       ) {
         current += curr
@@ -319,8 +344,8 @@ export function tokenize(code) {
   return tokens
 }
 
-/** 
- * @param {Array<[number, string]>} tokens 
+/**
+ * @param {Array<[number, string]>} tokens
  * @return {Array<string>}
  */
 function generate(tokens) {
@@ -328,7 +353,7 @@ function generate(tokens) {
   for (let i = 0; i < tokens.length; i++) {
     const [type, token] = tokens[i]
     output.push(
-      type === T_BREAK 
+      type === T_BREAK
       ? '<br>'
       : `<span class="sh__${types[type]}">${encode(token)}</span>`
     )
@@ -337,8 +362,8 @@ function generate(tokens) {
 }
 
 /**
- * 
- * @param {string} code 
+ *
+ * @param {string} code
  * @returns {string}
  */
 export function highlight(code) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "sugar-high",
-  "version": "0.1.2",
+  "version": "0.2.2",
   "type": "module",
   "exports": "./lib/index.mjs",
   "description": "Super lightweight JSX syntax highlighter",
@@ -11,10 +11,14 @@
   ],
   "license": "MIT",
   "scripts": {
-    "dev": "vite docs",
-    "build": "vite build docs"
+    "test": "vitest run",
+    "dev": "next dev docs",
+    "build": "next build docs"
   },
   "devDependencies": {
-    "vite": "2.7.13"
+    "next": "12.1.4",
+    "react": "18.0.0",
+    "react-dom": "18.0.0",
+    "vitest": "~0.6.0"
   }
 }