feat: stripped tags from end of document

eleventy.config.js

@@ -40,15 +40,12 @@ const extractTags = (content, mdInstance) => {
   return [...new Set(tags)];
 }
 
-const getPostTags = (post, mdInstance) => {
+const getPostTags = (post, tagMdInstance) => {
   const filePath = post.inputPath;
   try {
     const content = fs.readFileSync(filePath, 'utf-8');
-    const tags = extractTags(content, mdInstance);
-    return tags.map(tag => {
-      const normalizedTag = tag.toLowerCase();
-      return normalizedTag
-    });
+    const tags = extractTags(content, tagMdInstance);
+    return tags.map(tag => tag.toLowerCase());
   } catch (e) {
     // Skip if file can't be read
     return []
@@ -60,13 +57,13 @@ const isReleased = (post) => {
 }
 
 const markdownItHashtag = (md) => {
-  const hashtagRegex = /^#([a-zA-Z][a-zA-Z0-9_\/]*)(?![a-zA-Z0-9_-])/;
+  const hashtagRegex = /^#([a-zA-Z][a-zA-Z0-9_\\\\/]*)(?![a-zA-Z0-9_-])/;
   
   const HASH_CODE = '#'.charCodeAt(0);
   const SPACE_CODE = ' '.charCodeAt(0);
-  const TAB_CODE = '\t'.charCodeAt(0);
-  const NEWLINE_CODE = '\n'.charCodeAt(0);
-  const CARRIAGE_RETURN_CODE = '\r'.charCodeAt(0);
+  const TAB_CODE = '\\\\t'.charCodeAt(0);
+  const NEWLINE_CODE = '\\\\n'.charCodeAt(0);
+  const CARRIAGE_RETURN_CODE = '\\\\r'.charCodeAt(0);
 
   md.inline.ruler.push('hashtag', function(state, silent) {
     const pos = state.pos;
@@ -99,42 +96,94 @@ const markdownItHashtag = (md) => {
     const slug = tagName.toLowerCase();
     return `<a href="/tags/${slug}/" class="inline-tag">#${md.utils.escapeHtml(tagName)}</a>`;
   };
-}
+};
 
-const md = markdownIt({
-  html: true,
-  breaks: true,
-  linkify: true
-});
+// Plugin: Strip trailing hashtag-only paragraphs from rendered output
+// Must be applied AFTER footnote plugin since footnotes are moved to the end
+const markdownItStripTrailingHashtags = (md) => {
+  md.core.ruler.push('strip_trailing_hashtags', function(state) {
+    const tokens = state.tokens;
     
-md.use(markdownItContainer, 'details', {
-  validate: function (params) {
-    return params.trim().match(/^(.*)$/);
-  },
-  render: function (tokens, idx) {
+    const isHashtagOnlyParagraph = (inlineToken) => 
+      inlineToken?.type === 'inline' && 
+      inlineToken.children?.every(child => 
+        child.type === 'hashtag' || 
+        (child.type === 'text' && child.content.trim() === '') ||
+        child.type === 'softbreak'
+      ) && 
+      inlineToken.children?.some(child => child.type === 'hashtag');
+    
+    const isHashtagParagraphAt = (idx) =>
+      tokens[idx]?.type === 'paragraph_open' &&
+      tokens[idx + 1]?.type === 'inline' &&
+      tokens[idx + 2]?.type === 'paragraph_close' &&
+      isHashtagOnlyParagraph(tokens[idx + 1]);
+    
+    const footnoteIdx = tokens.findIndex(t => t.type === 'footnote_block_open');
+    const footnoteSectionStart = footnoteIdx === -1 ? tokens.length : footnoteIdx;
+    
+    const hashtagSectionStart = Array.from(
+      { length: Math.floor(footnoteSectionStart / 3) },
+      (_, i) => footnoteSectionStart - 3 * (i + 1)
+    ).reduce(
+      (start, idx) => isHashtagParagraphAt(idx) ? idx : start,
+      footnoteSectionStart
+    );
+    
+    state.tokens = tokens.filter((_, idx) => 
+      idx < hashtagSectionStart || idx >= footnoteSectionStart
+    );
+    
+    return true;
+  });
+};
+
+const markdownItDetails = (md) => {
+  md.use(markdownItContainer, 'details', {
+    validate: (params) => params.trim().match(/^(.*)$/),
+    render: (tokens, idx) => {
       const m = tokens[idx].info.trim().match(/^(.*)$/);
       if (tokens[idx].nesting === 1) {
         const title = md.utils.escapeHtml(m[1]);
-      return `<details class="expandable">\n<summary>${title}</summary>\n`;
-    } else {
-      return '</details>\n';
+        return `<details class="expandable">\\n<summary>${title}</summary>\\n`;
       }
+      return '</details>\\n';
     }
-});
+  });
+};
 
-md.use(markdownItFootnote);
-md.use(markdownItHashtag);
-md.use(markdownItMermaid);
-md.use(markdownItTaskLists, { enabled: true, label: true, labelAfter: true });
+const sharedPlugins = [
+  markdownItFootnote,
+  markdownItHashtag,
+  markdownItMermaid,
+  [markdownItTaskLists, { enabled: true, label: true, labelAfter: true }],
+  markdownItDetails,
+];
+
+const applyPlugins = (md, plugins) => 
+  plugins.reduce((instance, plugin) => {
+    const [pluginFn, options] = Array.isArray(plugin) ? plugin : [plugin];
+    return instance.use(pluginFn, options), instance;
+  }, md);
+
+const createMarkdownInstance = (extraPlugins = []) => {
+  const md = markdownIt({ html: true, breaks: true, linkify: true });
+  applyPlugins(md, [...sharedPlugins, ...extraPlugins]);
+  return md;
+};
+
+const md = createMarkdownInstance([markdownItStripTrailingHashtags]);
+
+const tagExtractorMd = createMarkdownInstance();
 
 module.exports = (eleventyConfig) => {
   eleventyConfig.addPlugin(syntaxHighlight);
 
-  eleventyConfig.addFilter("extractTags", (content) => extractTags(content, md));
+  eleventyConfig.addFilter("extractTags", (content) => extractTags(content, tagExtractorMd));
   eleventyConfig.addFilter("extractTagsFromFile", (filePath) => {
     try {
       const content = fs.readFileSync(filePath, 'utf-8');
-      return extractTags(content, md);
+      return extractTags(content, tagExtractorMd);
     } catch (e) {
       return [];
     }
@@ -216,7 +265,7 @@ module.exports = (eleventyConfig) => {
     const tagMap = {};
     
     posts.forEach(post => {
-      const tags = getPostTags(post, md)
+      const tags = getPostTags(post, tagExtractorMd)
       tags.forEach((tag) => {
         tagMap[tag] = {
           name: tag,
@@ -275,7 +324,7 @@ module.exports = (eleventyConfig) => {
     const filePath = recipe.inputPath;
     try {
       const content = fs.readFileSync(filePath, 'utf-8');
-      const tags = extractTags(content, md);
+      const tags = extractTags(content, tagExtractorMd);
       return tags.map(tag => tag.toLowerCase());
     } catch (e) {
       return [];
@@ -287,7 +336,7 @@ module.exports = (eleventyConfig) => {
     const recipes = collectionApi.getFilteredByGlob("recipes/**/*.md")
       .filter(r => r.data.isNewestVersion && r.data.draft !== true);
     
-    const postTags = posts.flatMap(post => getPostTags(post, md));
+    const postTags = posts.flatMap(post => getPostTags(post, tagExtractorMd));
     const recipeTags = recipes.flatMap(recipe => getRecipeTags(recipe));
     
     return [...new Set([...postTags, ...recipeTags])].sort();
@@ -306,7 +355,7 @@ module.exports = (eleventyConfig) => {
     
     // Build tag map from posts
     const postTagMap = posts.reduce((acc, post) => {
-      const tags = getPostTags(post, md);
+      const tags = getPostTags(post, tagExtractorMd);
       return tags.reduce((innerAcc, tag) => ({
         ...innerAcc,
         [tag]: {

posts/drafts/hyper-logLog-tombstone-garbage-collection.md

@@ -255,3 +255,5 @@ The HyperLogLog approach trades exactness (~3% error) for dramatic scalability
 ## References
 
 A working simulation is available at [simulations/hyperloglog-tombstone/simulation.ts](/simulations/hyperloglog-tombstone/simulation.ts).
+
+#algorithm #computer #distributed #peer_to_peer