Merge pull request #404 from rokubop/feat/explorer-tag-curation

Updates for "Repository Explorer"

Author: nriley

.github/workflows/deploy-ghpages.yml

@@ -5,6 +5,10 @@ on:
     branches:
       - main
       - emily/gh-pages-deploy
+  schedule:
+    # Rebuild weekly (Saturday night / Sunday 00:00 UTC)
+    - cron: "0 0 * * 0"
+  workflow_dispatch: # Allow manual trigger
 
 jobs:
   build_and_deploy:

plugins/repo-data-omit-list.json

@@ -1,4 +1,37 @@
 {
-  "description": "List of repositories to exclude from the Repository Explorer. Add repository full names (e.g., 'owner/repo-name') to the omitRepos array.",
-  "omitRepos": ["voqal/voqal"]
+  "description": "Configuration for Repository Explorer exclusions and tag normalization.",
+  "omitRepos": ["voqal/voqal", "Jailsonfs/community"],
+  "implicitTags": [
+    "talonvoice",
+    "talon",
+    "voice",
+    "voice-recognition",
+    "speech-recognition",
+    "voice-commands",
+    "voice-control"
+  ],
+  "tagAliases": {
+    "a11y": "accessibility",
+    "macos-accessibility": "accessibility",
+    "maths": "math",
+    "python3": "python",
+    "hci": "human-computer-interaction",
+    "games": "game",
+    "garrysmod": "game",
+    "gmod": "game",
+    "slaythespire": "game",
+    "slaythespire-mod": "game",
+    "gameboyadvance": "game",
+    "mgba": "game",
+    "blazor-server": "blazor",
+    "blazor-components": "blazor",
+    "awesome": "list",
+    "awesome-list": "list",
+    "gpt": "ai",
+    "openai": "ai",
+    "chatgpt": "ai",
+    "llm": "ai",
+    "copilot": "ai"
+  },
+  "matchNamesToExistingTags": true
 }

plugins/repo-data-plugin.js

@@ -11,8 +11,11 @@ module.exports = function (context, options) {
         "../.docusaurus/repo-data-plugin/default/repos.json",
       );
 
-      // Load omit configuration from repo-explorer-omit-list.json
+      // Load omit configuration from repo-data-omit-list.json
       let omitRepos = [];
+      let implicitTags = [];
+      let tagAliases = {};
+      let matchNamesToExistingTags = false;
 
       try {
         const omitListFile = path.join(__dirname, "repo-data-omit-list.json");
@@ -21,6 +24,21 @@ module.exports = function (context, options) {
           if (omitConfig.omitRepos && Array.isArray(omitConfig.omitRepos)) {
             omitRepos = omitConfig.omitRepos;
           }
+          if (
+            omitConfig.implicitTags &&
+            Array.isArray(omitConfig.implicitTags)
+          ) {
+            implicitTags = omitConfig.implicitTags;
+          }
+          if (
+            omitConfig.tagAliases &&
+            typeof omitConfig.tagAliases === "object"
+          ) {
+            tagAliases = omitConfig.tagAliases;
+          }
+          if (omitConfig.matchNamesToExistingTags === true) {
+            matchNamesToExistingTags = true;
+          }
         }
       } catch (error) {
         console.warn("Failed to load repo-data-omit-list.json:", error.message);
@@ -33,7 +51,70 @@ module.exports = function (context, options) {
         console.log(
           `Repository omit list loaded: ${omitRepos.length} repositories will be excluded`,
         );
-      } // Determine if we should fetch fresh data
+      }
+
+      // canonicalTags is built after repos are loaded, since we need
+      // to include tags that actually exist across repos.
+      let canonicalTags = null;
+
+      function buildCanonicalTags(repos) {
+        const tags = new Set(Object.values(tagAliases));
+        repos.forEach((repo) => {
+          repo.topics.forEach((t) => {
+            const canonical = tagAliases[t] || t;
+            if (!implicitTags.includes(canonical)) {
+              tags.add(canonical);
+            }
+          });
+        });
+        return tags;
+      }
+
+      /**
+       * Split a repo name into words, handling kebab-case, snake_case,
+       * camelCase, PascalCase, and mixed conventions.
+       * e.g. "VoiceLauncherBlazor" -> ["voice", "launcher", "blazor"]
+       *      "talon-mouse-rig"     -> ["talon", "mouse", "rig"]
+       *      "talon_mgba_http"     -> ["talon", "mgba", "http"]
+       */
+      function splitRepoName(name) {
+        return (
+          name
+            // Insert boundary before uppercase runs: "VoiceLauncher" -> "Voice Launcher"
+            .replace(/([a-z])([A-Z])/g, "$1 $2")
+            // Split acronym from next word: "HTTPServer" -> "HTTP Server"
+            .replace(/([A-Z]+)([A-Z][a-z])/g, "$1 $2")
+            .split(/[-_.\s]+/)
+            .map((w) => w.toLowerCase())
+            .filter((w) => w.length > 2)
+        );
+      }
+
+      /**
+       * Infer tags from a repo's name by matching words against known
+       * canonical tags. Only adds tags the repo doesn't already have
+       * (after alias resolution).
+       */
+      function inferTags(repo) {
+        if (!matchNamesToExistingTags || !canonicalTags) return;
+        const words = splitRepoName(repo.name);
+        const existingCanonical = new Set(
+          repo.topics.map((t) => tagAliases[t] || t),
+        );
+        for (const word of words) {
+          const canonical = tagAliases[word] || word;
+          if (
+            canonicalTags.has(canonical) &&
+            !existingCanonical.has(canonical) &&
+            !implicitTags.includes(canonical)
+          ) {
+            repo.topics.push(word);
+            existingCanonical.add(canonical);
+          }
+        }
+      }
+
+      // Determine if we should fetch fresh data
       const isUpdateRepos =
         (process.env.npm_config_argv &&
           JSON.parse(process.env.npm_config_argv).original.includes(
@@ -74,12 +155,17 @@ module.exports = function (context, options) {
               );
             }
 
+            canonicalTags = buildCanonicalTags(filteredRepos);
+            filteredRepos.forEach(inferTags);
+
             return {
               ...cachedData,
               repositories: filteredRepos,
               filtered_count: filteredRepos.length,
               omitted_count:
                 (cachedData.repositories.length || 0) - filteredRepos.length,
+              implicitTags,
+              tagAliases,
             };
           }
         } catch (error) {
@@ -157,12 +243,17 @@ module.exports = function (context, options) {
           `After filtering: ${filteredRepos.length} repositories (${allRepos.length - filteredRepos.length} omitted)`,
         );
 
+        canonicalTags = buildCanonicalTags(filteredRepos);
+        filteredRepos.forEach(inferTags);
+
         return {
           repositories: filteredRepos,
           total_count: totalCount,
           filtered_count: filteredRepos.length,
           omitted_count: allRepos.length - filteredRepos.length,
           generated_at: new Date().toISOString(),
+          implicitTags,
+          tagAliases,
         };
       } catch (error) {
         console.error("Failed to fetch repository data:", error);
@@ -179,13 +270,18 @@ module.exports = function (context, options) {
               return !omitRepos.includes(fullName);
             });
 
+            canonicalTags = buildCanonicalTags(filteredRepos);
+            filteredRepos.forEach(inferTags);
+
             return {
               ...cachedData,
               repositories: filteredRepos,
               filtered_count: filteredRepos.length,
               omitted_count:
                 (cachedData.repositories.length || 0) - filteredRepos.length,
               error: `Build-time fetch failed: ${error.message}. Using cached data.`,
+              implicitTags,
+              tagAliases,
             };
           }
         } catch (cacheError) {
@@ -198,6 +294,8 @@ module.exports = function (context, options) {
           total_count: 0,
           generated_at: new Date().toISOString(),
           error: error.message,
+          implicitTags,
+          tagAliases,
         };
       }
     },