From 832ab6e2f4ae3d92fa2b8cfa78012c28c18a76c2 Mon Sep 17 00:00:00 2001
From: Leon van Kammen <leonvankammen@gmail.com>
Date: Tue, 11 Nov 2025 23:21:21 +0100
Subject: [PATCH] > scrape a room URL and rewrite JML to serve local assets
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

```bash
$ ./janusxr --scrape https://www.janusxr.org/newlobby/index.html mydir
🔗 http://dizzket.com/archive/dotmatrix/
🔗 https://vesta.janusvr.com/nazrin/minecraft-sandbox
✅ http://www.janusvr.com/newlobby/scripts/home.txt
...

$ ls -la mydir
index.html
home.txt
...
```
---
 README.md | 26 ++++++++++++++++---
 janusxr   | 74 +++++++++++++++++++++++++++++++++++++++++++------------
 2 files changed, 81 insertions(+), 19 deletions(-)

diff --git a/README.md b/README.md
index ca3d106..40d3625 100644
--- a/README.md
+++ b/README.md
@@ -4,12 +4,17 @@ Portable swiss-army knife to automate [janusXR](https://janusxr.org) / JML thing
 
 > *What is [janusXR](https://janusxr.org)?* It's an open, user-operated immersive web layer, open-sourced by the JanusVR company, that seamlessly lives within websites, even 12 years later thanks to [janusweb](https://github.com/jbaicoianu/janusweb). It provides a highly viable and easy-to-adopt ecosystem of portals, enabling immersive experiences that challenge the traditional app store paradigm. Get started with [this guide](https://madjin.github.io/janus-guide/#/).
 
-
 # Usage
 
+```
+Usage: ./janusxr --health <room_url> [--max-time-per-asset 5] 
+       ./janusxr --scrape <room_url> <outdir>
+```
+## Examples
+
 > scan a room URL for broken links in JML+HTML
 
-```
+```bash
 $ ./janusxr --health http://localhost:8790/models/m5gr26w0wqqs
 
 ✅ http://localhost:8791/templates/xrfragment/%232/website.glb
@@ -28,7 +33,22 @@ $ ./janusxr --health http://localhost:8790/models/m5gr26w0wqqs
 
 ```
 
-# Awk?
+> scrape a room URL and rewrite JML to serve local assets
+
+```bash
+$ ./janusxr --scrape https://www.janusxr.org/newlobby/index.html mydir
+🔗 http://dizzket.com/archive/dotmatrix/
+🔗 https://vesta.janusvr.com/nazrin/minecraft-sandbox
+✅ http://www.janusvr.com/newlobby/scripts/home.txt
+...
+
+$ ls -la mydir
+index.html
+home.txt
+...
+```
+
+## Awk?
 
 Why not some superfancy scripting for this task?
 
diff --git a/janusxr b/janusxr
index 837825f..1751a1b 100755
--- a/janusxr
+++ b/janusxr
@@ -1,7 +1,8 @@
 #!/usr/bin/env -S awk -f
 
 function usage() {
-    print "Usage: ./janusxr --health <room_url> \n"
+    print "Usage: ./janusxr --health <room_url> [--max-time-per-asset 5] \n" \
+          "       ./janusxr --scrape <room_url> <outdir>\n"
     exit 1
 }
 
@@ -11,15 +12,15 @@ function usage() {
 BEGIN {
     if (ARGC < 2) usage()
     command = ARGV[1]
-    if (command == "--health") {
-        health()
+    if (command == "--health" || command == "--scrape") {
+        scrape(command)
     } else {
         printf("Unknown command: %s\n", command)
         usage()
     }
 }
 
-function health( tmpfile, line, attr, check, u) {
+function scrape( arg, tmpfile, line, attr, check, u) {
     url = ARGV[2]
     if (url == "") {
         print "❌ Missing URL argument."
@@ -46,18 +47,59 @@ function health( tmpfile, line, attr, check, u) {
     close(tmpfile)
 
     # Check each extracted links
-    nlinks = 0
-    nlinksok = 0
-    for (u in links) {
-        if( substr(u,1,1) == "/" ) u = rooturl""u
-        check = "curl -I -s \"" u "\" > /dev/null"
-        if (system(check) == 0){
-            nlinksok++
-            printf("✅ %s\n", u)
-        }else printf("❌ %s\n", u)
-        nlinks+=1
+    if( arg == "--health" ){
+      maxtime = ARGV[3]
+      if ( maxtime == "" ) maxtime = 5
+      nlinks = 0
+      nlinksok = 0
+      for (u in links) {
+          if( substr(u,1,1) == "/" ) u = rooturl""u
+          check = "curl -L --max-time "maxtime" -I -s \"" u "\" > /dev/null"
+          if (system(check) == 0){
+              nlinksok++
+              printf("✅ %s\n", u)
+          }else printf("❌ %s\n", u)
+          nlinks+=1
+      }
+      print "⚕️ health: "(( 100/nlinks )*nlinksok)"%"
+      if( nlinks != nlinksok ) exit(1)
+    }
+
+    if( arg == "--scrape" ) {
+      outdir = ARGV[3]
+      if ( outdir == "" ) outdir = "."
+      system("mkdir "outdir" || true ")
+      system("cp "tmpfile" "outdir"/index.html")
+
+      for (u in links) {
+          if( substr(u,1,1) == "/" ) u = rooturl""u
+          check = "curl -L --max-time 20 -I -s \"" u "\" > /dev/null"
+          if (system(check) == 0 && has_non_html_ext(u) ){
+              n = split(u, fileparts, "/")
+              filepart = fileparts[n]
+              outfile  = outdir"/"filepart
+              system("curl -L --max-time 20 -s \"" u "\" > "outfile)
+              system("sed -i 's|"u"|"filepart"|g' "outdir"/index.html")
+              nlinksok++
+              printf("✅ %s\n", u)
+          }else printf("🔗 %s\n", u)
+          nlinks+=1
+      }
     }
-    print "⚕️ health: "(( 100/nlinks )*nlinksok)"%"
-    if( nlinks != nlinksok ) exit(1)
 }
 
+# Function: has_non_html_ext
+# Returns 1 if the file has an extension after at least one slash and it is not "html"
+# Returns 0 otherwise
+function has_non_html_ext(file,   arr) {
+    # Check for at least one slash
+    if (file !~ /\//) return 0
+
+    # Match extension after last dot that is not a slash
+    if (match(file, /\.([^.\/]+)$/, arr)) {
+        ext = arr[1]
+        if (ext != "html") return 1
+    }
+
+    return 0
+}