currently i added this line to Mojolicious::Lite app to disable caching
app->renderer->cache->max_keys(0);
but sometimes webpage showing updated content and sometimes showing cached template
i do not know exactly what the problem is, there is no problem with morbo, this is happening only with hypnotoad. i want to try disabling the cache.
edit: i solved my problem. problem is not with the mojolicious caching, it is with the database caching. if you experience similar problem, try setting 'AutoCommit' to 1 in your DBI->connect() method.
Related
I am having a problem using the "Tabs" component: https://material-ui.com/components/tabs/ In fact, when developing locally, the rendering of the component is fine. But pushing to the server, it looks weird (with differences in the borders on each side of the bar).
Moreover: when refreshing the page in which it has been embedded or browsing manually to this page, the whole template is broken all of a sudden!
I tried generating a build folder locally and launching the page from that build, and fell back on the same problem.
Thank you in advance for your help
From your description, I saw that your "build" is not work even in local development. That's mean it should not work on the production, of course.
The thing is, what is your actual "build" action? Depends on what library/framework you use, but basically with Material-UI, most popular problem comes from not load some CSS before using it's components.
Please read here first: https://material-ui.com/guides/server-rendering/#server-rendering
Just in case you use Gatsby, read here: https://www.gatsbyjs.org/packages/gatsby-plugin-material-ui/.
I'm using Gatsby and use this following config to fix some CSS issues.
stylesProvider: {
injectFirst: true,
}
Hope this help.
Generally my application response times are good and pages load quickly. However login, using Zend_Auth with mysql db, takes an age compared to other tasks. I'm wondering has anyone else encountered this?
If not can anyone suggest a good way to debug and find the cause of the slow down?
FYI most pages are rendered under a second after their link is clicked whereas login takes around 7 seconds.
First you need to figure out what is causing slowdown. Add this to your application.ini:
resources.db.setDefaultTableAdapter = 1
resources.db.params.profiler.enabled = true
resources.db.params.profiler.class = Zend_Db_Profiler_Firebug
Then use Firefox+Firebug+FirePHP to see (in Firebug Console) list of all queries. Maybe one of them is really slow? Or maybe you make a lot of them?
Other thing to do is to check Net tab (Firebug) with 'keep changes' turned on to see if you don't make too many redirects or maybe some JS files are slowing down downloading page?
Last step is to install XDebug and run your login with XDEBUG_PROFILE parameter and then use KCacheGrind tool to see what is happening during login and what takes most time.
(Check manual)
I am still a beginner in GWT . Firstly I write a sample project and run it . And open with browser giving url by Development Mode of my Eclipse IDE. At this time , I noticed that my browser was stopped just a few seconds and I can't do anythings on my browser . If I have some other pages are loading to open , they also stop loading . I think in this time browsers will download JavaScript files these need to show my page. That is worse thing for me .
By using RPC , I retrieve many datas from my database and try to render to my view page. That may also happen stop for a moment to my browser. So , I had add a Gif image in my web-page and press some button to retrieve data and try to render my view page. Supprisingly , this Gif image also stop animation. Why has this situation happend with GWT project ?
I am testing it in localhost. Can I avoid it ? Any suggestions ? Thanks...
Development mode (localhost) is very different from the production mode. It is many times slower, and it requires a lot of memory. This is why your browser freezes. You can:
Try a different browser. You may have a better performance on your system.
Add memory to your computer.
Optimize your code. There are many ways to do it. As a first step you need to understand what causes a problem: processing data or rendering of a page.
Finally, deploy your code to the production environment and check if you experience the same problems.
I have applications which are displayed as tabs in Facebook pages which have been working fine. They suddenly started displaying this output in the tab:
/1336720089,176820405/
if (window.CavalryLogger) { CavalryLogger.start_js(["EgxV3"]); }
__d("UFIUpdate",
Any ideas what is going on???
Here is a link to one of them: http://www.facebook.com/TweakShoes/app_132692060112327
A temporary solution is to add https:// to facebook itself. This doesn't solve the problem, but it'll allow you to see your page on a per client basis.
The best fix in the longer term until facebook fixes this issue is to go to your account settings > Security > Enable secure browsing. This will enforce HTTPS wherever it can and should resolve the issue for a lot of pages you're trying to access.
Good Luck!
After encountering this issue yesterday, I tracked it down to an apparent conflict with the JS log wrapper included as part of HTML5 Boilerplate's script.js file. In particular, the "make it safe to use console.log always" snippet. After commenting it out, the FB lib error went away and my app displayed properly.
I didn't dig into what the conflict was, but here is the snippet. If you use something similar on your page, then it may be worth investigating.
/* make it safe to use console.log always */
(function(b){function c(){}for(var d="assert,count,debug,dir,dirxml,error,exception,group,groupCollapsed,groupEnd,info,log,timeStamp,profile,profileEnd,time,timeEnd,trace,warn".split(","),a;a=d.pop();){b[a]=b[a]||c}})((function(){try
{console.log();return window.console;}catch(err){return window.console={};}})());
It's probably not a coincidence that FB's own logger bugs out with this.
Facebook has opened a bug for this issue and recently triaged it to medium priority - no word on when it will be addressed.
I have a Webview that must load an image! When I upload this image I see every time the same image as before, and i must reboot my app to see the new image...
I think is a cache problem..How can I solve that??
One quick and easy method would be to append the current time stamp onto the url whenever you load it.
So instead of loading:
http://www.myhost.com/myimg.jpg
you'd load
http://www.myhost.com/myimg.jpg?12345689
Using a cache breaker like this is a very common method in web development to force reloading of content.
I did do some quick googling and it appears clearing out NSURL's cache won't do the trick. In 10.6 the api reloadFromOrigin: may do the trick, but I'm not aware if this has made it's way onto the iphone yet.
Edit:
I found this page in the docs. It looks like you can use the preferences system to say whether or not to use caching. Not tested, but that'd be something to look at.