@rokelamen/md2html 0.1.2 → 0.1.3

package/README.md

@@ -2,9 +2,11 @@
 
 A simple markdown-html conventer written in Typescript.
 
+> I create this project for learning TS and node dev, not for the purpose of building another better markdown parse engine.
+
 ## Development Log
 
-Why I choose to use `rollup`?
+Why I choose to use [`rollup`](https://rollupjs.org/)?
 
-A: I find that every time I `import` a module, I have to add extension to the module file and it is probably a `.js` ranther than `.ts`. It's wired that I must `import` a future JS file. So I decided to use a build tool(`rollup` of course) to pack all files together, which eliminates all `import` statements.
+A: Every time I `import` a module, I have to add extension to the module file and it is probably a `.js` ranther than `.ts`. It's wired that I must `import` a future JS file. So I decided to use a build tool(`rollup` of course) to pack all files together, which eliminates all `import` statements.
 

package/bin/cli.cjs

@@ -4336,16 +4336,116 @@ const {
   Help,
 } = commander;
 
+const headerReg = /^\s*(#{1,6})(?:\s+|$)(.*)$/;
+
+/**
+ * Since AST-based parsing is too complex and not
+ * suitable for a personal typescript learning project,
+ * So I just let my markdown rendering engine to
+ * receive a line and analysis which type it is,
+ * and then output the corresponding format(with tags).
+ * That means it's just a simple line-by-line parsing.
+ *
+ * Some special scenes still need to be handled carefully:
+ * plain text -> <p>plain text</p>
+ * need to consider whether there is a following empty line.
+ * Cause
+ * ```
+ * plain text 1
+ * plain text 2
+ * ```
+ * is one paragraph NOT two.
+ * The same happens with quote text:
+ * ```
+ * > quote text
+ * plain text
+ * ```
+ * is one paragraph as well.
+ */
 /* The main parse logic */
-function parse(content) {
-    const source = content.trim();
-    return `<p>${source}</p>`;
+function parse(markdown) {
+    /* Split markdown content to many lines */
+    const crlfReg = /\r?\n/;
+    const lines = markdown.split(crlfReg);
+    // console.log(lines);
+    const mdBlocks = parseToBlocks(lines);
+    // console.log(mdBlocks);
+    const html = handleTags(mdBlocks);
+    return html;
 }
+/**
+ * Traverse lines to turn to blocks with different types
+ */
+function parseToBlocks(lines) {
+    let lastTextQuoteBlock = { content: '', type: 'text' };
+    let pushed = true;
+    const mdBlocks = [];
+    for (const line of lines) {
+        // Empty line
+        if (!Boolean(line.trim())) {
+            if (!pushed) {
+                mdBlocks.push(lastTextQuoteBlock);
+                pushed = true;
+            }
+            continue;
+        }
+        // Header
+        const headerM = line.match(headerReg);
+        if (headerM) {
+            mdBlocks.push({
+                type: 'header',
+                level: headerM[1].length,
+                content: headerM[2].trim()
+            });
+            continue;
+        }
+        // Fall back to plain text
+        if (!pushed) {
+            /* last line is also text */
+            lastTextQuoteBlock.content += ' ' + line.trim();
+        }
+        else {
+            lastTextQuoteBlock = {
+                type: 'text',
+                content: line.trim()
+            };
+            pushed = false;
+        }
+    }
+    // Avoid the last block is omitted
+    if (!pushed) {
+        mdBlocks.push(lastTextQuoteBlock);
+        pushed = true;
+    }
+    return mdBlocks;
+}
+/* traverse markdown content blocks and wrap text with tags at proper positions. */
+function handleTags(mdBlocks) {
+    let result = '';
+    for (const block of mdBlocks) {
+        const type = block.type;
+        const content = tagSwtich(block);
+        switch (type) {
+            case "text":
+                result += `<p>${content}</p>` +
+                    '\n';
+                break;
+            case "header":
+                result += `<h${block.level}>${content}</h${block.level}>` +
+                    '\n';
+                break;
+        }
+    }
+    return result;
+}
+function tagSwtich(block) {
+    return block.content;
+}
+
+var name = "@rokelamen/md2html";
+var version = "0.1.3";
+var description = "A simple tool to convert markdown content to html";
 
-// Import these from package.json in the future
-const name = 'md2html';
-const version = '0.1.2';
-const description = 'A simple tool to convert markdown content to html';
 /* Command-line tool logic */
 function command() {
     /* Set basic info */

package/dist/index.js

@@ -1,7 +1,107 @@
+const headerReg = /^\s*(#{1,6})(?:\s+|$)(.*)$/;
+
+/**
+ * Since AST-based parsing is too complex and not
+ * suitable for a personal typescript learning project,
+ * So I just let my markdown rendering engine to
+ * receive a line and analysis which type it is,
+ * and then output the corresponding format(with tags).
+ * That means it's just a simple line-by-line parsing.
+ *
+ * Some special scenes still need to be handled carefully:
+ * plain text -> <p>plain text</p>
+ * need to consider whether there is a following empty line.
+ * Cause
+ * ```
+ * plain text 1
+ * plain text 2
+ * ```
+ * is one paragraph NOT two.
+ * The same happens with quote text:
+ * ```
+ * > quote text
+ * plain text
+ * ```
+ * is one paragraph as well.
+ */
 /* The main parse logic */
-function parse(content) {
-    const source = content.trim();
-    return `<p>${source}</p>`;
+function parse(markdown) {
+    /* Split markdown content to many lines */
+    const crlfReg = /\r?\n/;
+    const lines = markdown.split(crlfReg);
+    // console.log(lines);
+    const mdBlocks = parseToBlocks(lines);
+    // console.log(mdBlocks);
+    const html = handleTags(mdBlocks);
+    return html;
+}
+/**
+ * Traverse lines to turn to blocks with different types
+ */
+function parseToBlocks(lines) {
+    let lastTextQuoteBlock = { content: '', type: 'text' };
+    let pushed = true;
+    const mdBlocks = [];
+    for (const line of lines) {
+        // Empty line
+        if (!Boolean(line.trim())) {
+            if (!pushed) {
+                mdBlocks.push(lastTextQuoteBlock);
+                pushed = true;
+            }
+            continue;
+        }
+        // Header
+        const headerM = line.match(headerReg);
+        if (headerM) {
+            mdBlocks.push({
+                type: 'header',
+                level: headerM[1].length,
+                content: headerM[2].trim()
+            });
+            continue;
+        }
+        // Fall back to plain text
+        if (!pushed) {
+            /* last line is also text */
+            lastTextQuoteBlock.content += ' ' + line.trim();
+        }
+        else {
+            lastTextQuoteBlock = {
+                type: 'text',
+                content: line.trim()
+            };
+            pushed = false;
+        }
+    }
+    // Avoid the last block is omitted
+    if (!pushed) {
+        mdBlocks.push(lastTextQuoteBlock);
+        pushed = true;
+    }
+    return mdBlocks;
+}
+/* traverse markdown content blocks and wrap text with tags at proper positions. */
+function handleTags(mdBlocks) {
+    let result = '';
+    for (const block of mdBlocks) {
+        const type = block.type;
+        const content = tagSwtich(block);
+        switch (type) {
+            case "text":
+                result += `<p>${content}</p>` +
+                    '\n';
+                break;
+            case "header":
+                result += `<h${block.level}>${content}</h${block.level}>` +
+                    '\n';
+                break;
+        }
+    }
+    return result;
+}
+function tagSwtich(block) {
+    return block.content;
 }
 
 export { parse };

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@rokelamen/md2html",
   "type": "module",
-  "version": "0.1.2",
+  "version": "0.1.3",
   "description": "A simple tool to convert markdown content to html",
   "author": "rokelamen <rogerskelamen@gmail.com>",
   "license": "MIT",
@@ -32,6 +32,7 @@
   ],
   "devDependencies": {
     "@rollup/plugin-commonjs": "^29.0.0",
+    "@rollup/plugin-json": "^6.1.0",
     "@rollup/plugin-node-resolve": "^16.0.3",
     "@rollup/plugin-typescript": "^12.3.0",
     "@types/commander": "^2.12.0",