[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Caching and JS - Bug?
I second your thoughts. I've been rewriting scripts daily and can count the
number of times this *caching* has happened to me on one hand (twice -- but
they were with beta Navigators). Clicking in the location box and hitting
[ENTER] seems to *always* load the resaved code. I can be assured of this
because many of the rewrites I've done have been with document.write()
calls, and the new page is redrawn accordingly every time.
Like I said before, if you've got some code that gets 'stuck' in the cache,
could you post it for the rest of us to see?
At 03:58 PM 3/2/96 -0800, Gordon McComb wrote:
>lots of revisions and reloads, and I think I've hit this bug maybe 3-4 times
>total. It's always fixed by leaving Netscape and restarting -- a five
>second job in Windows 95. (Often a complete reload of the page -- reopening
>it or retyping it in the location box -- also does the trick.)
>Are you sure you don't have something funky like a weird cache (especially
>with extra long delayed writes)? I'm astonished that if the problem is this
>big it should have generated a massive amount of debate on the several
>lot of people chiming in.
>At 11:50 AM 3/1/96 -0800, you wrote:
>>Ed Kloskowski wrote:
>>> I also do an INTRAnet site, with the same worrting about what software
>>> everyone is using. However, the caching problem is real. So real in
>>> because the old code kept "popping" up (its a huge site with many test
>>> deleting the netscape cache.
>>Unfortunatly deleting the caches doesn't even do the job 100% of
>>the time. I have modified code and deleted both caches four
>>times and sometimes I still get the first or second version of
>>the code being executed even though view source shows the
>>I personally consider this a giant bug in JS since it serious
>>impedes the development effort and greatly reduces my over
>>confidence in Netscape, Navigator, and JS.
For help about the list, please send a message to 'email@example.com'
with the message body 'help'. To unsubscribe, send a message to