Updated stylesheet to add totals to tables, updated gulpfile, updated .gitignore, updated web.config to prevent IIS from accessing documents outside public, updated video and search views to remove WEBVTT noise on the fly.

.gitignore

@@ -136,4 +136,11 @@ dist
 iisnode/
 
 # transpiler artifacts
-app/TikaClient/build/
+app/TikaClient/build/
+
+# cookies
+youtube-cookies.txt
+
+# SSL certificates
+*.cer
+*csr.txt

app/server.js

@@ -86,28 +86,28 @@ glob.globSync('pages/**/*.md', {
   });
 });
 
-console.log("Scanning for documents to create routes");
+// console.log("Scanning for documents to create routes");
-glob.globSync('**/*{.pdf,.docx,.xlsx,.pptx,.doc,.xls,.ppt}', {
+// glob.globSync('**/*{.pdf,.docx,.xlsx,.pptx,.doc,.xls,.ppt}', {
-  cwd: path.join(__dirname, '..', 'public'),
+//   cwd: path.join(__dirname, '..', 'public'),
-  matchBase: true,
+//   matchBase: true,
-  follow: true,
+//   follow: true,
-}).forEach((filePath) => {
+// }).forEach((filePath) => {
-  const expressRoutePathFromFilePath = (filePath) => {
+//   const expressRoutePathFromFilePath = (filePath) => {
-    return filePath.substring(0, filePath.length - path.extname(filePath).length).replaceAll(path.sep, path.posix.sep);
+//     return filePath.substring(0, filePath.length - path.extname(filePath).length).replaceAll(path.sep, path.posix.sep);
-  };
+//   };
-  const route = expressRoutePathFromFilePath(filePath);
+//   const route = expressRoutePathFromFilePath(filePath);
-  const fullFilePath = path.join(__dirname, '..', 'public', filePath);
+//   const fullFilePath = path.join(__dirname, '..', 'public', filePath);
-  let paths = route.split(path.posix.sep);
+//   let paths = route.split(path.posix.sep);
-  paths[0] = 'public';
+//   paths[0] = 'public';
-  console.log(`Setting route for ${route}`);
+//   console.log(`Setting route for ${route}`);
-  app.get(route, async (req, res) => {
+//   app.get(route, async (req, res) => {
-    const fm = matter.read(fullFilePath);
+//     const fm = matter.read(fullFilePath);
-    const fmData = { fm: fm.data, excerpt: fm.excerpt };
+//     const fmData = { fm: fm.data, excerpt: fm.excerpt };
-    const content = helpers.md.render(fm.content, fmData );
+//     const content = helpers.md.render(fm.content, fmData );
-    const renderData = { content, route, filePath, fullFilePath, req, paths, ...fmData };
+//     const renderData = { content, route, filePath, fullFilePath, req, paths, ...fmData };
-    res.render("page", { h: helpers, ...renderData });
+//     res.render("page", { h: helpers, ...renderData });
-  });
+//   });
-});
+// });
 
 // Endpoints for all the site's YouTube videos.
 console.log("Scanning for archived videos to create routes");

gulpfile.js

@@ -25,6 +25,32 @@ gulp.task('index:clear', async () => {
     });
 });
 
+gulp.task('dbfromsolr', async () => {
+    let docs = await request({
+        uri: `${solrUrl}/select`,
+        qs: {
+            q: '*:*',
+            wt: 'json',
+            rows: 10000,
+        },
+        json: true
+    });
+    docs = docs?.response?.docs?.map(({id, sha256sum, url, content_length, content_type, text, _version_}) =>
+    {
+        return {
+            id,
+            url: url.join(''),
+            content_length: parseInt(content_length.join()),
+            sha256sum: sha256sum.join(''),
+            content_type: content_type.join(''),
+            text: text.join(''),
+            _version_,
+        }
+    }).map(doc => {
+        
+    })
+});
+
 async function calculateSHA256Hash(filePath) {
     return new Promise((resolve, reject) => {
         const readStream = fs.createReadStream(filePath);

routes/search.js

@@ -8,6 +8,8 @@ const helpers = require('../views/helpers/functions');
 router.get('/', (req, res) => {
     // Extract paging parameters from request query parameters
     let { q, page = 1, pageSize = 10 } = req.query;
+    // Sanitize query, with particular emphasis on one problem area where soft keyboards are creating fancy quotes but we need basic quotes
+    q = q.replaceAll(/[“”“”„„‟❝❞〝〞〟＂❠⹂🙶🙷🙸]/g, '\"').replaceAll(/[‘’‘’＇‚‛❛❜❟]/g, '\'');
     if (page instanceof String) page = parseInt(page);
     if (pageSize instanceof String) pageSize = parseInt(pageSize);
     // Cap at 100 max per page

static/css/nm3clol.css

@@ -227,4 +227,12 @@ table {
 tbody, td, tfoot, th, thead, tr {
   font-family: 'Sometype Mono';
   padding: 5pt;
+}
+
+thead {
+  border-bottom: 2pt solid #222;
+}
+
+.total {
+  border-top: 5pt double #222;
 }

views/helpers/functions.js

@@ -63,6 +63,23 @@ const printMarkdownFile = ({file}) => {
 const printReadme = ({directory}) => {
   return md.render(fs.readFileSync(resolveReadmeFile({directory})).toString());
 };
+const stripWebVTT = (webvttText) => {
+  const searchHeader = "WEBVTT\nKind: captions\nLanguage: en\n\n";
+  if (webvttText.startsWith(searchHeader)) {
+    webvttText = webvttText.substring(searchHeader.length-1); // remove WEBVTT header
+    webvttText = webvttText.replaceAll(' align:start position:0%', ''); // remove this align and position junk
+    webvttText = webvttText
+                  .split('\n')
+                  .map((line) => { return line.replaceAll(/.*<\d{2}:\d{2}:\d{2}.\d{3}>.*/g, '').trim() }) // remove all the animated subtitles and trim the whitespace on each line
+                  .join('\n');
+    while (webvttText.indexOf('\n\n\n') > -1) {
+      webvttText = webvttText.replace('\n\n\n', '\n\n'); // remove every instance of triple vertical white space, while allowing double vertical white space
+    }
+    webvttText = webvttText.replaceAll(/(\d{2}:\d{2}:\d{2}.\d{3}) --> (\d{2}:\d{2}:\d{2}.\d{3})\n(.*)\n\n(\2) --> (\d{2}:\d{2}:\d{2}.\d{3})\n\3\n/g, '$1 --> $5\n$3\n'); // remove every duplicate entry detected
+    webvttText = webvttText.replaceAll('\n', '<br/>'); // convert \n to <br/>
+  }
+  return webvttText;
+};
 
 module.exports = {
     trimSlashes,
@@ -75,6 +92,7 @@ module.exports = {
     shouldOmitLinkOnLastBreadcrumb,
     directoryContainsReadme,
     printReadme,
+    stripWebVTT,
     md,
     moment,
 };

views/search-results.ejs

@@ -33,7 +33,7 @@
                   <h5><%= doc.title %></h5>
                   <% if (highlighting[doc.id] && highlighting[doc.id].text) { %>
                     <% highlighting[doc.id].text.forEach(snippet => { %>
-                      <p><%- snippet %></p>
+                      <pre><%- h.stripWebVTT(snippet) %></pre>
                     <% }); %>
                   <% } else { %>
                     <!-- <p>No highlight available.</p> -->

views/video-player.ejs

@@ -73,7 +73,7 @@
               Search feature. Please review the <a href="/search-policy" target="_blank">Search Policy</a>
               for details about the site features.
             </p>
-            <pre style="white-space:pre-wrap;overflow-wrap:anywhere"><%- subtitleVTT %></pre>
+            <pre style="white-space:pre-wrap;overflow-wrap:anywhere"><%- h.stripWebVTT(subtitleVTT) %></pre>
           </div>
         </div>
       <% } %>

web.config

@@ -8,9 +8,9 @@
        <rules>
          <rule name="nodejs">
            <match url="(.*)" />
-           <conditions>
+           <!-- <conditions>
              <add input="{REQUEST_FILENAME}" matchType="IsFile" negate="true" />
-           </conditions>
+           </conditions> -->
            <action type="Rewrite" url="app/server.js" />
          </rule>
        </rules>
@@ -22,9 +22,14 @@
            <add segment="node_modules" />
            <add segment="iisnode" />
          </hiddenSegments>
+                <fileExtensions>
+                </fileExtensions>
        </requestFiltering>
      </security>
         <httpErrors errorMode="Detailed" />
         <urlCompression doDynamicCompression="false" />
+        <staticContent>
+            <mimeMap fileExtension=".md" mimeType="text/markdown" />
+        </staticContent>
      </system.webServer> 
  </configuration>