How to read a HTML file from the browser cache only

[Marc Gueury]

Hello all,
I would like to write a Firefox extension that validates
the HTML of the browser in real time. The Html validation
code is already written and works fine.
I need to get the HTML text of a page when browsing.
I found all I needed except one thing. I have no idea how
to get the HTML text of the current page from the cache only.
I need opinion and feedback, since I am completely lost.
I tested this:
1.
    // - but get the content a second time !!!!
    const oldURL  = window.content.document.URL;
    const request = new XMLHttpRequest();
    // This must be done to make generated content render
    request.open("GET", oldURL, false);
    request.send("");
This work fine but the problem of XMLHttpRequest is that in
the internal code,there is a load flags with SKIP_CACHE. Then
the page is always fetched again from the webserver, what I
want to avoid assolutely.
I could rewritte the whole XMLHttpRequest. But
it is a really long code (about 2000 lines)
2. Then I did something like this:
    try
    {
      const nsICacheService = Components.interfaces.nsICacheService;
      const cacheService = 
Components.classes["@mozilla.org/network/cache-service;1"].getService(nsICacheService);
      var httpCacheSession = cacheService.createSession("HTTP", 0, true);
      httpCacheSession.doomEntriesIfExpired = false;
      var cacheEntryDescriptor = httpCacheSession.openCacheEntry(url, 
Components.interfaces.nsICache.ACCESS_READ, false);

      if (cacheEntryDescriptor)
      {
        this.lastFetched = cacheEntryDescriptor.lastFetched;
        alert( "device = " + cacheEntryDescriptor.deviceID + " / size = 
" + cacheEntryDescriptor.dataSize );

        if( cacheEntryDescriptor.isStreamBased() )
        {
          var inputStream = cacheEntryDescriptor.openInputStream(0);
          const scriptableStream = 
Components.classes["@mozilla.org/scriptableinputstream;1"].createInstance(Components.interfaces.nsIScriptableInputStream);
          scriptableStream.init( inputStream );

          var s = scriptableStream.read(scriptableStream.available());
          alert( s );
          scriptableStream.close();
          inputStream.close();
        }
      }
    }
    catch(ex)
    {
...
It works in some cases, when the page is ASCII and not compressed.
It does not work when:
- the HTTP header has: Content-Encoding: gzip
  -> I get the compressed file and not the HTML
- the protocol is "file://"
  -> it seems not to be cached
- for non ascii pages
In a way, it is a better than the 1rst test. Due that I am sure that the 
content is cached and that I never request the data again to the 
webserver. But I have the problems above.

Any idea or pointer on how to do in C or javascript ? There is maybe a 
simple way that I miss completely.

Thanks by advance,
Marc
_______________________________________________
Mozilla-netlib mailing list
Mozilla-netlib@mozilla.org
http://mail.mozilla.org/listinfo/mozilla-netlib