My team is having a strange issue with the dojo combobox. When we moved the code to the integration environment, the dojo combobox started to render improperly. We verified that the code is the same. Strangely other dojo comboboxes in the same application render normally.
The integration server is on a older version: 9.0.1 HF32 64 bit
The unit server where it works is at: 9.0.1 FP3 HF632
UNIT SERVER rendering on left, INTG Server rendering on right.
My question is has anybody seen this and do they know how to fix it?
Notes:
this happens in IE and Chrome. It is coming wrong from the server.
We are using CIAO to promote between environments (which I hate).
Upgrading the server is not an option for me
I can post code if requested, but I suspect that it isn't a code issue.
UPDATE: For some reason the following two CSS files are not loaded on that particular page. They are loaded on every other page in the application. I can't find any property or setting that would effect this. The admin verified that these files are on the server and the dates are identical. These are system generated resources.
<link rel="stylesheet" type="text/css" href="/oneuiv2.1/base/dojo.css">
<link rel="stylesheet" type="text/css" href="/oneuiv2.1/defaultTheme/dojoTheme.css">
Does anyone know why these two stylesheets are not being loaded??
I've seen some strange positioning issues depending on where Dojo comboboxes are relative to the browser screen, i.e. if there's not enough space left in the screen to position relative to the field they're bound to. That may have been in XPiNC and was in older versions. The thing to check is whether the rendering issue is related to where the field is relative to the page in the browser.
Related
I am trying to use HTTP/2 server push, powered by Cloudflare. I am trying to push two assets using the following HTTP Header...
Link: </img/whysoslow-hero.jpg>; rel=preload; as=image, </lib/bootstrap.min.css>; rel=preload; as=style
The image is reference as a background image in a stylesheet and this is being pushed and then used correctly (downloaded once).
However, I am referencing the stylesheet in the HTML of my page like this...
<link rel="stylesheet" href="/lib/bootstrap.min.css">
...and it's being pushed but then also being fetched as well (downloaded twice).
My website is... https://whysoslow.co.uk
Here's a screenshot of the Network tab in Chrome showing that the assets are pushed, but then the stylesheet is downloaded again...
Can anyone explain to me why this is happening and how I can resolve it?
EDIT:
I've since noticed that the console is saying...
A preload for 'https://whysoslow.co.uk/lib/bootstrap.min.css' is found, but is not used due to an integrity mismatch.
So it's the SRI check that's failing. But I don't understand why, it's fetching the same file, so it should be identical. Why would they not be identical?
Now that I've figured that out, I've also managed to find the answer...
How to preload script using integrity and crossorigin
Short answer: You can't preload and use SRI, it's not supported yet.
I am out of wit as how to solve this problem in Javascript or HTML. I have customers currently using web applications built for ie9 and under. These legacy applications do not work well on IE11. IT solutions was to enable enterprises mode. Enterprises mode was designed to avoid "common compatibility problems associated with web apps written and tested on older versions of Internet Explorer".
See: http://www.eightforums.com/tutorials/43972-ie11-enterprise-mode-enable-disable-users.html
Enabling enterprises mode appear to be problematic on web application written using Bootstrap and AngularJS. Ie, responsive does not work at all unless enterprises mode is disabled. Not just AngularJS and Bootstrap but other libraries as well.
The solution that I am looking for is a way to check the status of enterprises mode via javascript, then tell the users to either enable / disable the mode. Better, if it can be turn off / on automatically via JS or HTML attributes.
Snooping in the document.x and window.x objects, I do not see any properties that we would give me an indication that enterprises mode is enable. Any suggestion?
Repro(s):
IE11 > Developer Tool > Console > Type window
IE11 > Developer Tool > Console > Type document
There is no DOM property that indicates that you're running EMIE. The whole idea of EMIE is to emulate IE8 behavior better than the IE8 document mode emulates IE8 behavior. EMIE should only be used in specific cases where it's needed; it should not be used wholesale.
It is possible to detect EMIE in certain cases. If you look carefully at the list of user-agent strings over the last couple of releases, there's a noticeable difference between EMIE on IE11 and the user agent string for IE11 RTM.
However, before you take that as your magic bullet, there are two caveats:
You cannot disable EMIE programmatically. It's a local configuration change only.
The user agent for IE11 is completely different today than it was when IE11 was released. Based on reports from the IE team, the UA string is going to be even more complicated, especially once "IE Spartan" (or whatever they choose to call it") hits the wire.
My recommendation? Create a small launcher page that does a simple feature detection for the web app in question. If you detect features consistent with what's needed for the app, then display a link to launch the app. If feature detection fails to detect IE8, IE11, or whatever version you've targeted, display a warning with a link to more troubleshooting information. Be sure to include a launch link anyway, just in case.
This way, the user has the information they need and you have a lightweight way of handling the issue, one that doesn't require too many updates to the app in question.
Hope this helps...
-- Lance
To my mind the reason of the issue is that IE 11 Enterprise mode emulates IE 8. But bootstrap doesn't support IE 8. To overcome it just use HTML5 shim and Respond.js as described here.
<script src="js/respond.min.js"></script>
<script src="js/html5shiv.min.js"></script>
But without check like <!--[if lt IE 9]> - because it seems that it doesn't work in the enterprise mode.
The better solution would be not just include the mentioned scripts without conditions but find out the appropriate condition instead of < IE 9.
To fix the problem with angularjs just use the following meta tag:
<meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1">
For those who also struggled with this problem. I have submitted a feature request to MS IE Team.
See:
https://connect.microsoft.com/IE/feedback/details/1159543/need-a-way-for-client-side-codes-to-detect-enterprises-mode
My solution is a workaround that involve checking the width of the container div. Enterprises mode do not support responsive.
I'm using phantomjs version 1.9.7 which I believe is supposed to support web fonts. I have inserted the font with google web fonts however it is just displaying my fallback font when I automate a screengrab. The webfont is displaying fine in all my browsers. Are there any workarounds for this?
I have struggled with this issue for several hours. Well, there is a simple reason for this controversy: user-agent!
Some services such as Google Fonts are returning different CSS content based on user-agent. When you call a webpage which includes Google Fonts with default PhantomJs user-agent, Google would return TTF version which is supported in PhantomJs.
However, if you set a custom user agent (Chrome, FF, etc.), Google Fonts would return .woff2 version. .woff2 is not supported in PhantomJs 2.x. Obviously, fonts are not going to be loaded.
So, for users who are testing PhantomJs without setting a generic user-agent, Google Fonts is working. If they set for example Google Chrome user-agent, it does not work.
So, you have two options:
Avoid setting a custom user-agent if it is possible.
Avoid using dumb-smart font providers like Google Fonts which does not outputting all font versions in CSS and let the browser decide what it needs.
I believe your belief is wrong :-)
The 1.9.x series of Phantom is still based on the old WebKit code, which people have been reporting web font problems with all along.
The good news is that a technical preview of Phantom 2.0 is apparently about to be released; https://github.com/ariya/phantomjs/wiki/PhantomJS-2 is intended to be the best place to follow status. And, from memory, someone reported success with web fonts using it (but I may be wrong on that as a quick search of the mailing list archives didn't turn up a definitive message saying web fonts work... but they definitely should).
BTW, SlimerJS is an almost drop-in replacement, based on Firefox's Gecko engine, and does support web fonts (though some problems in corner cases, IIRC).
I just had the same issue with PhantomJS 2.1.1.
In my case I was working under a proxy that was blocking PhantomJS from loading the font from Google. After connecting to an open network it rendered correctly.
I tried all fixes listed here to no avail, but here is a work around. Either option resolved this for me.
First Option:
Install the missing web fonts on the local computer that is running PhantomJS. Most web font providers including Google allow you to easily download the fonts for local install. No change needed at the target URL.
Second:
I sites I have control over, I was able to resolve by splitting up the link tags.
Change This from:
<link href='//fonts.googleapis.com/css?family=Roboto:400,500,700|Open+Sans:400italic,700italic,400,700' rel='stylesheet' type='text/css'>
To:
<link href='//fonts.googleapis.com/css?family=Roboto:400,500,700' rel='stylesheet' type='text/css'>
<link href='//fonts.googleapis.com/css?family=Open+Sans:400italic,700italic,400,700' rel='stylesheet' type='text/css'>
I know this is not an ideal solution but either one works, It depends on if you can modify the target URL.
I have tried this on numerous pc's and are receiving the same error. My webbrowser control isnt displaying javascript or jquery.
Here is a example:
when i visit this url in my browser: https://binarykings.co.uk/traderoom-aspx, it looks great with the charts and prices etc.
example: http://i.imgur.com/1Lmq4em.png
same in internet explorer: http://i.imgur.com/3KP864X.png
however in my vb.net webbrowser control it looks like this: http://i.imgur.com/qjme2Yg.jpg
What is causing this ? Its not just this site but any other site that has a similar way of displaying things via jquery/javascript.
I do have script errors suppressed however when i disabled this and clicked yes to the errors manually it still displayed the same empty page.
Thanks
The WebBrowser control uses the Internet Explorer rendering engine. But by default, for compatibility reasons, it is stuck using the IE 7 rendering engine, even if you have a later version of the browser installed on your computer.
There are a couple of ways to convince it to use a newer version of IE to render the page. One involves editing the registry, while the other involves modifying the web page.
In general, it is best to avoid attempts to solve local problems with global solutions like registry edits. There might be another app running on the same machine that uses the WebBrowser control and relies on the behavior of the IE 7 rendering engine. If you go changing global settings, you'll mess up that other app.
Therefore, the best solution, if possible, is to edit the web page. This is quite simple, and probably a good idea anyway. All you need to do is add the following code inside of the <head> section:
<meta http-equiv="X-UA-Compatible" content="IE=9" >
If you absolutely must use the alternative approach of editing the registry, you'll find the information you need here in the documentation.
I am part of the developer team for a quite a large online system using ASP.NET(4).
Asp.net Ajax completely breaks down for Webkit browsers and we are getting full page postbacks when we should be getting partial only for the UpdatePanels.
I am starting to believe it has something to do with my Application Configuration, mainly for the following reasons.
If I move the ajax enabled controls to a new project they will work as expected for all browsers, including Webkit.
I created a static .aspx file with nothing but an UpdatePanel,ScriptManager and a button making a literal visible on click.
I get no Javascript errors from any browser, and i see an http request for the asp.net-ajax (ScriptResource.axd) in both Firebug and Chrome Developer tools
I tried ye'old safari fix from this highly referenced thread
Edit: After a bit more testing and http sniffing i noticed a major difference between the test application and the actual application. The test application generates 2 additional .axd files which are not generated from the actual application. These WebResource.axd, seem to contain data related to the async postback. However this is only the case for Webkit browsers. The WebResource.axd files are generated for Firefox as i can see them in firebug
What i am asking from the community, is any ideas or suggestions as to what could be the cause of this problem and if i am correct to assume that the problem is probably on the server side
Thanks for any help
The problem was due to a deprecated config file that's used to limit the content that bots/spiders/crawlers receive, which was loading by mistake thanks to our lovely inhouse CMS
In short if u get behavior similar to my case, check your or configs
I was having a similar issue however my problem was with all browsers and not just webkit. I ended up going through and tearing up the web.config file and found out that a line: <xhtmlConformance mode="Legacy"> was preventing webresource.axd from working properly. The fix was to simply remove that line from my web.config file.
For a little more information on xhtmlConformance, visit http://technet.microsoft.com/en-us/librarY/ms228268(v=vs.85).aspx.
If you scroll all the way to the bottom you'll notice it explicitly states that it causes issues with webresource.axd and scriptresource.axd.