server: (webui) no more gzip compression

.gitattributes

@@ -0,0 +1,4 @@
+# Treat the generated single-file WebUI build as binary for diff purposes.
+# Git's pack-file delta compression still works (byte-level), but this prevents
+# git diff from printing the entire minified file on every change.
+tools/server/public/index.html -diff

.gitignore

@@ -95,6 +95,8 @@
 # Server Web UI temporary files
 /tools/server/webui/node_modules
 /tools/server/webui/dist
+# we no longer use gz for index.html 
+/tools/server/public/index.html.gz
 
 # Python
 

tools/server/CMakeLists.txt

@@ -42,7 +42,7 @@ option(LLAMA_BUILD_WEBUI "Build the embedded Web UI" ON)
 
 if (LLAMA_BUILD_WEBUI)
     set(PUBLIC_ASSETS
-        index.html.gz
+        index.html
         loading.html
     )
 

tools/server/README-dev.md

@@ -259,6 +259,6 @@ npm run test
 npm run build
 ```
 
-After `public/index.html.gz` has been generated, rebuild `llama-server` as described in the [build](#build) section to include the updated UI.
+After `public/index.html` has been generated, rebuild `llama-server` as described in the [build](#build) section to include the updated UI.
 
 **Note:** The Vite dev server automatically proxies API requests to `http://localhost:8080`. Make sure `llama-server` is running on that port during development.

tools/server/server-http.cpp

@@ -10,7 +10,7 @@
 
 #ifdef LLAMA_BUILD_WEBUI
 // auto generated files (see README.md for details)
-#include "index.html.gz.hpp"
+#include "index.html.hpp"
 #include "loading.html.hpp"
 #endif
 
@@ -262,16 +262,11 @@ bool server_http_context::init(const common_params & params) {
         } else {
 #ifdef LLAMA_BUILD_WEBUI
             // using embedded static index.html
-            srv->Get(params.api_prefix + "/", [](const httplib::Request & req, httplib::Response & res) {
-                if (req.get_header_value("Accept-Encoding").find("gzip") == std::string::npos) {
-                    res.set_content("Error: gzip is not supported by this browser", "text/plain");
-                } else {
-                    res.set_header("Content-Encoding", "gzip");
-                    // COEP and COOP headers, required by pyodide (python interpreter)
-                    res.set_header("Cross-Origin-Embedder-Policy", "require-corp");
-                    res.set_header("Cross-Origin-Opener-Policy", "same-origin");
-                    res.set_content(reinterpret_cast<const char*>(index_html_gz), index_html_gz_len, "text/html; charset=utf-8");
-                }
+            srv->Get(params.api_prefix + "/", [](const httplib::Request & /*req*/, httplib::Response & res) {
+                // COEP and COOP headers, required by pyodide (python interpreter)
+                res.set_header("Cross-Origin-Embedder-Policy", "require-corp");
+                res.set_header("Cross-Origin-Opener-Policy", "same-origin");
+                res.set_content(reinterpret_cast<const char*>(index_html), index_html_len, "text/html; charset=utf-8");
                 return false;
             });
 #endif

tools/server/webui/package.json

@@ -50,7 +50,6 @@
 		"eslint-config-prettier": "^10.0.1",
 		"eslint-plugin-storybook": "^10.2.4",
 		"eslint-plugin-svelte": "^3.0.0",
-		"fflate": "^0.8.2",
 		"globals": "^16.0.0",
 		"http-server": "^14.1.1",
 		"mdast": "^3.0.0",

tools/server/webui/scripts/install-git-hooks.sh

@@ -83,7 +83,7 @@ if [ -n "$WEBUI_CHANGES" ]; then
     fi
 
     # Check if build output exists and is newer than source files
-    BUILD_FILE="../public/index.html.gz"
+    BUILD_FILE="../public/index.html"
     NEEDS_BUILD=false
 
     if [ ! -f "$BUILD_FILE" ]; then
@@ -127,9 +127,9 @@ if [ -n "$WEBUI_CHANGES" ]; then
         cd ../../..
 
         # Check if build output was created/updated
-        if [ -f "tools/server/public/index.html.gz" ]; then
+        if [ -f "tools/server/public/index.html" ]; then
             # Add the build output and commit it
-            git add tools/server/public/index.html.gz
+            git add tools/server/public/index.html
             if ! git diff --cached --quiet; then
                 echo "Committing updated build output..."
                 git commit -m "chore: update webui build output"

tools/server/webui/scripts/post-build.sh

@@ -1,3 +1,3 @@
 rm -rf ../public/_app;
 rm ../public/favicon.svg;
-rm ../public/index.html;
+rm -f ../public/index.html.gz; # deprecated, but may still be generated by older versions of the build process

tools/server/webui/vite.config.ts

@@ -1,6 +1,5 @@
 import tailwindcss from '@tailwindcss/vite';
 import { sveltekit } from '@sveltejs/kit/vite';
-import * as fflate from 'fflate';
 import { readFileSync, writeFileSync, existsSync } from 'fs';
 import { dirname, resolve } from 'path';
 import { fileURLToPath } from 'url';
@@ -20,15 +19,13 @@ const GUIDE_FOR_FRONTEND = `
 -->
 `.trim();
 
-const MAX_BUNDLE_SIZE = 2 * 1024 * 1024;
-
 /**
  * the maximum size of an embedded asset in bytes,
  * e.g. maximum size of embedded font (see node_modules/katex/dist/fonts/*.woff2)
  */
 const MAX_ASSET_SIZE = 32000;
 
-/** public/index.html.gz minified flag */
+/** public/index.html minified flag */
 const ENABLE_JS_MINIFICATION = true;
 
 function llamaCppBuildPlugin() {
@@ -40,7 +37,6 @@ function llamaCppBuildPlugin() {
 			setTimeout(() => {
 				try {
 					const indexPath = resolve('../public/index.html');
-					const gzipPath = resolve('../public/index.html.gz');
 
 					if (!existsSync(indexPath)) {
 						return;
@@ -62,25 +58,10 @@ function llamaCppBuildPlugin() {
 					content = content.replace(/\r/g, '');
 					content = GUIDE_FOR_FRONTEND + '\n' + content;
 
-					const compressed = fflate.gzipSync(Buffer.from(content, 'utf-8'), { level: 9 });
-
-					compressed[0x4] = 0;
-					compressed[0x5] = 0;
-					compressed[0x6] = 0;
-					compressed[0x7] = 0;
-					compressed[0x9] = 0;
-
-					if (compressed.byteLength > MAX_BUNDLE_SIZE) {
-						throw new Error(
-							`Bundle size is too large (${Math.ceil(compressed.byteLength / 1024)} KB).\n` +
-								`Please reduce the size of the frontend or increase MAX_BUNDLE_SIZE in vite.config.ts.\n`
-						);
-					}
-
-					writeFileSync(gzipPath, compressed);
-					console.log('✓ Created index.html.gz');
+					writeFileSync(indexPath, content, 'utf-8');
+					console.log('✓ Updated index.html');
 				} catch (error) {
-					console.error('Failed to create gzip file:', error);
+					console.error('Failed to update index.html:', error);
 				}
 			}, 100);
 		}