Previous (cached?) version loads in Chrome

52 views
Skip to first unread message

Mat

unread,
Apr 26, 2016, 1:09:54 AM4/26/16
to tiddly...@googlegroups.com
Darn annoying; quite often I load a tiddlyspot only to realize it is not the latest save but the previous one. My workflow for developing in TW is to use a tiddlyspot, say TS1, and another copy, TS2, where I fiddle and when happy I save it do TS1. Now, when I reload TS1... it does not (always) show the new content. I need to reload it once again to have the latest kick in. I strongly suspect it is cached data showing up the first time.

This is very devilish as there is no indication that anything is wrong and if I do some little edit and save on TS1 it then overwrites what was really the latest version. I've almost lost a lot of work like this many times and probably also actually, but unknowingly, lost work at other times.

I find this info on disabling caching in Chrome but it only works as long as the DevTools are open. And here are some alternatives, including using Incognito mode. Best solution I find is a Chrome extension.

But what is really needed is permanently disabling cache for specific website (or, better, a domain) so I'm asking you if anyone has any suggestion?

...or if I misunderstand something? Might it really be a TW problem? or a TS problem? I believe it is primarily (only?) a problem in Chrome and not FF but not sure.

Does this happen to anyone else?

Thanks


<:-)

CodaCoder

unread,
Apr 28, 2016, 10:46:20 AM4/28/16
to TiddlyWikiDev
In this situation, I've used the time-honoured cache-busting technique of appending a unique parameter to the URL

http://blah.com/page.html?_=<unique-thing>

But of course, you'll need to avoid mangling any tiddler links you may pass in the URL.

Caveat-emptor: I don't recall ever having tried this using the file protocol.

HTH.

Mat

unread,
May 9, 2016, 6:58:04 PM5/9/16
to TiddlyWikiDev
Just reporting that the Chrome extension seems to work well for this. It turns off cashing all together which I guess should be less-than-optimal but I am not noticing anything other than the desired effect.


@CodaCoder - thank you also for your tip.

<:-)
Reply all
Reply to author
Forward
0 new messages