sugar-high 0.0.7 → 0.1.1

package/README.md

@@ -3,7 +3,7 @@
 
 Super lightweight JSX syntax highlighter, around 1KB after minified and gzipped
 
-> ⚠️ Still in experiment! Use it in production with caution!
+![img](https://repository-images.githubusercontent.com/453236442/9416c941-9c40-4dbe-a73e-0e0d21993802)
 
 ### Usage
 

package/index.d.ts

@@ -0,0 +1,2 @@
+export function highlight(code: string): string
+export function tokenize(code: string): Array<[number, string]>

package/lib/index.mjs

@@ -147,15 +147,9 @@ function isRegexStart(str) {
 export function tokenize(code) {
   let current = ''
   let type = -1
+  let nonSpaceType = -1
   /** @type {Array<[number, string]>} */
   const tokens = []
-  // string.type = 0 for string or string template
-  // string.type = 1 for regex
-  const string = { entered: false, type: 0 }
-  
-  // comment.type = 0 for single line comments
-  // comment.type = 1 for multi-line comments
-  const comment = { entered: false, type: 0 }
 
   // jsx.tag for entering open or closed tag
   // jsx.child for entering children
@@ -163,29 +157,36 @@ export function tokenize(code) {
   /** @type {{ tag: boolean; child: boolean; expr: boolean }} */
   const jsx = { tag: false, child: false, expr: false }
 
+  const inJsxLiterals = () => !jsx.tag && jsx.child && !jsx.expr
+
   function classify(token) {
-    if (isCommentStart(token[0] + token[1])) {
+    const [chr0, chr1] = [token[0], token[1]]
+    if (isCommentStart(chr0 + chr1)) {
       return T_COMMENT
     } else if (keywords.has(token)) {
       return T_KEYWORD
     } else if (token === '\n') {
       return T_BREAK
-    } else if (
-       (
-        // is quoted string
-        (isStringQuotation(token[0]) && !isStringQuotation(token[1])) ||
-        // is regex
-        (!jsx.tag && isRegexStart(token[0] + token[1]) && token[token.length - 1] === '/')
-      )
-    ) {
-      return T_STRING
     } else if (token === ' ') {
       return T_SPACE
-    } else if (signs.has(token[0])) {
+    } else if (token.split('').every(ch => signs.has(ch))) {
       return T_SIGN
     } else if (
-      token[0] === token[0].toUpperCase() ||
-      token === 'null'
+      (
+       // 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) &&
+        chr0 === chr0.toUpperCase() ||
+        token === 'null'
+      )
     ) {
       return T_CLS_NUMBER
     } else {
@@ -196,6 +197,7 @@ export function tokenize(code) {
   const append = () => {
     if (current) {
       type = classify(current)
+      if (type !== T_SPACE && type !== T_BREAK) nonSpaceType = type
       tokens.push([type, current])
     }
     current = ''
@@ -205,8 +207,8 @@ 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 isJsxLiterals = !jsx.tag && jsx.child && !jsx.expr
+    const p_c = prev + curr // previous and current    
+    const isJsxLiterals = inJsxLiterals()
 
     if (jsx.tag) {
       const isOpenElementEnd = curr === '>' 
@@ -215,55 +217,57 @@ export function tokenize(code) {
       jsx.tag = !(isOpenElementEnd || isCloseElementEnd)
     }
     // if it's not in a jsx tag declaration or a string, close child if next is jsx close tag
-    if (!jsx.tag && !string.entered && (curr === '<' && isIdentifierChar(next) || c_n === '</')) {
+    if (!jsx.tag && (curr === '<' && isIdentifierChar(next) || c_n === '</')) {
       jsx.tag = true
       jsx.child = false
     }
-    
-    if (jsx.child && curr === '{') {
-      jsx.expr = true
-    }
-    if (jsx.child && jsx.expr && curr === '}') {
-      jsx.expr = false
-    }
 
     if (
-      !string.entered && 
       (isStringQuotation(curr) || !jsx.tag && isRegexStart(c_n))
     ) {
-      string.entered = true
-      string.type = isStringQuotation(curr) ? 0 : 1
       append()
-      current = curr
-    } else if (string.entered) {
-      current += curr
-      if (string.type === 0 && isStringQuotation(curr)) {
-        string.entered = false
-        append()
-      } else if (string.type === 1 && prev !== '\\' && curr === '/') {
-        string.entered = false
+      if (!(nonSpaceType === T_SIGN || nonSpaceType === T_COMMENT)) {
+        current = curr
         append()
+        continue
       }
-    } else if (
-      !comment.entered && 
-      isCommentStart(c_n)
-    ) {
-      comment.type = next === '/' ? 0 : 1
-      comment.entered = true
-      append()
-      current = c_n
-      i++
-    } else if (comment.entered) {
-      current += curr
-      if (comment.type === 0 && next === '\n') {
-        comment.entered = false
+      let j = i + 1
+      let strEnd = i
+      const inString = () => j < code.length && code[j] !== '\n'
+      // string quotation
+      if (isStringQuotation(curr)) {
+        for (; inString(); j++) {
+          if (isStringQuotation(code[j])) {
+            strEnd = j
+            break
+          }
+        }
+      } else {
+        // regex
+        for (; inString(); j++) {
+          if (code[j] === '/' && code[j - 1] !== '\\') {
+            strEnd = j
+            break
+          }
+        }
+      }
+      if (strEnd !== i) {
+        current = code.slice(i, strEnd + 1)
         append()
-      } else if (comment.type === 1 && (c_n === '*/')) {
-        comment.entered = false
-        current += '/'
+        i = j
+      } else {
+        current = curr
         append()
-        i++
       }
+    } else if (isCommentStart(c_n)) {
+      const start = i
+      if (next === '/') {
+        for (; i < code.length && code[i] !== '\n'; i++);
+      } else {
+        for (; i < code.length && code[i - 1] + code[i] !== '*/'; i++);
+      }
+      current = code.slice(start, i + 1)
+      append()
     } else if (curr === ' ' || curr === '\n') {
       if (
         curr === ' ' &&
@@ -296,6 +300,13 @@ export function tokenize(code) {
         else if (jsxBrackets.has(curr)) append()
       }
     }
+
+    if (jsx.child && curr === '{') {
+      jsx.expr = true
+    }
+    if (jsx.child && jsx.expr && curr === '}') {
+      jsx.expr = false
+    }
   }
 
   append()
@@ -320,6 +331,11 @@ function generate(tokens) {
   return output
 }
 
+/**
+ * 
+ * @param {string} code 
+ * @returns {string}
+ */
 export function highlight(code) {
   const tokens = tokenize(code)
   const output = generate(tokens).join('')

package/package.json

@@ -1,11 +1,12 @@
 {
   "name": "sugar-high",
-  "version": "0.0.7",
+  "version": "0.1.1",
   "type": "module",
   "exports": "./lib/index.mjs",
   "description": "Super lightweight JSX syntax highlighter",
   "files": [
     "lib",
+    "index.d.ts",
     "*.md"
   ],
   "license": "MIT",