docs: add errata, and documentation for BCP47 parser

README.md

@@ -9,7 +9,7 @@
 
 `isogloss` is a Python–based command–line tool designed for looking up language details based on [ISO 639](https://www.iso.org/iso-639-language-code) codes and IETF ([BCP-47](https://www.rfc-editor.org/info/bcp47)) language tags. It provides comprehensive information about languages, including their names, native names, and additional details associated with each code or tag.
 
-There is also a [web–based version here](https://thunderpoot.github.io/isogloss).
+There is also a [web–based version here](https://thunderpoot.github.io/isogloss).  The [BCP47 parser](https://thunderpoot.github.io/isogloss/bcp-index.html) has some known issues, documented below in the "Errata" section.
 
 Elsewhere, [the word isogloss](https://en.wikipedia.org/wiki/Isogloss) means a boundary line on a map denoting the regional use of a particular linguistic characteristic, but in this case it just seemed to fit.
 
@@ -224,6 +224,61 @@ $ isogloss/isogloss.py -i ar-ajp-apc-apd-Arab-CV-arevela-g-231243-r-sdarre-x-pri
 - `data/script_codes.json`: Contains script code data in JSON format used for the BCP47 lookup.
 - `data/deprecated-639-3.csv`: Contains deprecated ISO 639-3 codes in CSV format, for quick reference.
 
+## Errata
+
+There are known issues with the BCP47 parser in the web interface.  It uses regular expressions to validate input, such that:
+
+### Examples of valid tags:
+
+- `en`
+
+- `fr-CA`
+
+- `i-klingon`
+
+- `az-Arab-IR`
+
+- `sr-Cyrl-RS`
+
+- `zh-cmn-Hans`
+
+- `ja-JP-x-tokyo`
+
+- `uz-Cyrl-UZ-1992`
+
+- `bo-Tibt-x-dialect`
+
+- `zh-cmn-Hans-CN-x-private1`
+
+- `hy-Latn-IT-arevela-x-test`
+
+
+### Examples of invalid tags (malformed):
+
+- `en-GB-oed-x-private`
+
+- `de-CH-1901-co-phonebk-sc-gothic-x-bavaria`
+
+(and more)
+
+### Examples of inputs that reveal parsing bugs:
+
+- `ca-valencia-nedis`
+    (Highlighted input section is missing "valencia")
+
+- `en-US-u-islamcal`
+    (Variant "u" and Extension "islamcal", Extension section says "u - islamcal")
+
+- `es-419-fonipa`
+    (Extended languages blank)
+
+- `de-Latf-1901`
+    (Region undefined)
+
+- `sl-rozaj`
+    (rozaj is coloured differently in the result container to how it is in the highlighted input section)
+
+
 ## Contributing
 
 Contributions, issues, and feature requests are welcome!

docs/bcp-app.js

@@ -78,8 +78,8 @@ function parseIetfTag(tag) {
         extLangs: [],
         script: null,
         region: null,
-        variant: null,
-        extension: null,
+        variants: [],
+        extensions: [],
         private: null
     };
     let idx = 1;
@@ -103,26 +103,17 @@ function parseIetfTag(tag) {
     }
 
     // Variants and extensions
-    const variantList = [];
-    const extensionList = [];
     while (idx < components.length) {
         if (components[idx].startsWith('x')) {
             result.private = components.slice(idx).join('-');
             break;
-        } else if (components[idx].startsWith('r') || components[idx].startsWith('g')) {
-            extensionList.push(components[idx]);
+        } else if (/^[a-wy-z0-9]$/.test(components[idx])) {
+            result.extensions.push(components[idx] + '-' + components[idx + 1]);
+            idx += 2;
         } else {
-            if (!result.variant) {
-                result.variant = components[idx];
-            } else {
-                extensionList.push(components[idx]);
-            }
+            result.variants.push(components[idx]);
+            idx += 1;
         }
-        idx += 1;
-    }
-
-    if (extensionList.length > 0) {
-        result.extension = extensionList.join('-');
     }
 
     return result;
@@ -258,14 +249,16 @@ function processTag() {
     if (parsedTag.region) {
         html += `<div class="section region"><strong>Region:</strong> ${regionData[parsedTag.region]}</div>`;
     }
-    // Process Variant
-    if (parsedTag.variant) {
-        html += `<div class="section variant"><strong>Variant:</strong> ${parsedTag.variant}</div>`;
+    // Process Variants
+    if (parsedTag.variants.length > 0) {
+        const variantsHtml = parsedTag.variants.map(variant => `<li>${variant}</li>`).join('');
+        html += `<div class="section variant"><strong>Variants:</strong> <ul>${variantsHtml}</ul></div>`;
     }
 
-    // Process Extension
-    if (parsedTag.extension) {
-        html += `<div class="section extension"><strong>Extension:</strong> ${parsedTag.extension}</div>`;
+    // Process Extensions
+    if (parsedTag.extensions.length > 0) {
+        const extensionsHtml = parsedTag.extensions.map(extension => `<li>${extension}</li>`).join('');
+        html += `<div class="section extension"><strong>Extensions:</strong> <ul>${extensionsHtml}</ul></div>`;
     }
 
     // Process Private Use
@@ -292,7 +285,7 @@ async function fetchLanguages() {
 
 function loadPage() {
     fetchLanguages();
-    // fetchLatestCommit();
+    fetchLatestCommit();
 }
 
 window.onload = loadPage;

docs/bcp-index.html

@@ -20,6 +20,7 @@ <h2>BCP47 Parser</h2>
                 <span><a rel="nofollow" href="https://github.com/thunderpoot/isogloss?tab=MIT-1-ov-file#readme" target="_blank">Licence</a></span> |
                 <span><a rel="nofollow" href="https://www.iso.org/iso-639-language-code" target="_blank">ISO</a></span> |
                 <span><a rel="nofollow" href="https://sil.org/" target="_blank">SIL</a></span> |
+                <span><a rel="nofollow" href="./" target="_blank">639</a></span> |
                 <span id="dark-mode-toggle" class="dark-mode-toggle" onclick="toggleDarkMode()">Dark</span>
             </div>
         </footer>

docs/bcp-style.css

@@ -232,8 +232,8 @@ button:hover {
     gap: 5px;
 }
 
-.warning { color: #b44545; }
-body.dark-mode .warning { color: #e87474; }
+.warning { color: #b49f45; }
+body.dark-mode .warning { color: #e8d974; }
 
 .primary-language {
     background-color: var(--primary-color);

docs/index.html

@@ -65,6 +65,7 @@
                 <span><a rel="nofollow" href="https://github.com/thunderpoot/isogloss?tab=MIT-1-ov-file#readme" target="_blank">Licence</a></span> |
                 <span><a rel="nofollow" href="https://www.iso.org/iso-639-language-code" target="_blank">ISO</a></span> |
                 <span><a rel="nofollow" href="https://sil.org/" target="_blank">SIL</a></span> |
+                <span><a rel="nofollow" href="./bcp-index.html" target="_blank">BCP47</a></span> |
                 <span id="dark-mode-toggle" class="dark-mode-toggle">Dark</span>
             </div>
         </footer>