I am trying to set initial focus to the first input field in each page of my app.
I have implemented the code in the following post and it is working great:
How to Set Initial Focus in a View?
But I found that it is not working for first view/page in app when it opens from Fiori Launchpad. I found the following code which is getting executed after my onAfterShow:
From sap/ushell/renderers/fiori2/Shell-dbg.controller.js
Could you please help to solve this?
I am using SAP UI5 version 1.56.7.
Author of that answer here. After browsing through the source code and documentation quite a while, I have to admit I couldn't find any acceptable solutions to this question either. My impression is that FLP developers want to make sure that the app doesn't interfere with setting the initial focus.
The lack of APIs and documentation thereof strongly suggests that there are compelling reasons behind this prevention (probably a11y related).
My advice is not to rely on timeouts but maintaining consistent UX by keeping the focus on the app title on its launch - as designed by SAP.
Solved the issue by passing around 200ms into setTimeout function for first page and it is working fine
Related
Any help/advice on this would be greatly appreciated.
I'm looking for what I would consider a standard setup for modern day web design/development. Basically I have a dual monitor setup and I would like to code on one screen and have the changes displayed in real-time on the other screen.
Up until yesterday I was using jsbin in this way and it was working great. I had one browser setup with the coding stuff (HTML/CSS/jQuery) and I then had another separate browser open on the second monitor which updated instantly as I typed. So if I changed a CSS rule for example it was shown in real-time on the second monitor without me having to do ANYTHING. No saving, no refreshing, no switching tabs - NOTHING.
However, for whatever reason jsbin now refuses to update in real-time and it will only show the code changes if I manually refresh the browser. I've emailed jsbin about this but they can't diagnose the issue.
So what I'm looking for is either an online alternative, or a local alternative. However, everything (and I mean EVERYTHING) that I've tried so far can't do what jsbin did.
dabblet.com, jsFiddle.net, liveweave.com, codepen.io, cssdeck.com etc
But all of the above don't offer a second tab/browser that I can move to the second monitor to show the code updates in real-time which is the most important thing.
Surely there must be an easy solution to this? Is it not common to code in this way on 2 monitors whilst developing a website? I'd really appreciate any help on this.
Many thanks
Ben
For local option, Brackets has Live Preview mode.
After you open your project or file for editing, select File -> Live Preview
It launches a browser window, then your changes are reflected as you type, no save or browser refresh needed.
I use Liveweave with my dual monitor setup. I keep the HTML/CSS in my left window/monitor and the JS/Preview in my right monitor. Works great!
I installed the Rest Console extension onto Chrome. I can't use this Rest Console for anything, because the only fields that accept text are 'Request URL' and 'Request Method'.
I ask here as there is no resources or community accessible for even simple instructions on how this extension works, and there are likely members of this community who have used this extension.
Use the Postman - REST Client it works perfectly!
your_url_to_post: Fill the URL
your_json_param: The name of the param that you'll pass you json
your_json_data: The json content
Don't forget to select x-www-form-urlencoded if you'll do a post
http://i.stack.imgur.com/TISW1.png
You should be able to enter text in the other field if you tick the checkbox associated to the field first.
I started having a similar issue myself (Forms wouldnt let me input data I could click them but thats as far as it went) across a lot of web pages to include my own php game and I know the form code hasnt changed at all before this issue started. I came across a article of someone else having a similar issue themselves that I was having and someone said that they needed position: relative on their forms that are not given specific instructions on positioning when dealing with google chrome. I have since made this change myself at www.immoralattack.com and the issue has stopped on that site but still continues on many others so its pretty safe to say that is the issue.
Its only happening on the dev build of chrome for me atleast and I hope this isnt something they plan on implementing on live or this will mess up alot of websites.
Also try turning off auto fill from chrome if its on, while I was looking for this bug I found alot of similar bugs but they delt with auto fill being on.
Please forgive me if this answer is somewhere else on this site or online. If it is, I sure haven't found it in the past several days of searching.
What I am hoping to find is an "accurate" method of detecting a browser and redirecting to a simple, static page if not a recent browser.
The samples I have found until now often have not provided an accurate representation of the actual browser being used. For instance:
When testing with Navigator 9, I'll get a message that I'm using Firefox 2
When testing with Maxthon 3, it reports I'm using IE 9.
My site displays correctly in all the current browsers I've been testing it with. But I wish I could have a basic static page for those .01% who still are using an old browser for whatever reason. They could still get some basic information from my site, as well as encouraged to update to a more current browser.
If anyone has any useful suggestions, I'd greatly appreciate them.
Thanks so much.
Cheers,
David
Browser detection is never perfect, for a variery of reasons. If you are using jQuery, you should look into jQuery.browser.
I'd try to detect the browser on the server side and do an HTTP redirect if the browser is something non-standard. Most decent frameworks have functionality to detect the browser from the user agent string. Again, this is not perfect, mainly because of the data browsers report. Also, if Maxthon reports it's IE, that's because it is based on IE and therefore the layout engine should be the same.
So you either
support a small number of browsers and cater for their quirks, sending all other browsers to a basic page (this sucks for future versions of browsers because they might be standards-compliant but they will still display your very basic page), or
you have a standards-compliant page for all browsers and then you define alternatives for the ones that give you problems.
I'd go for the second option. It usually all boils down to one version for all browsers, and a number of hacks for various versions of IE. Also, remember to avoid padding in your CSS and use margins instead.
In the end, you probably shouldn't be testing for browsers and version numbers, but supported features. Try using Modernizr.
The $.browser property is deprecated in jQuery 1.3. On jQuery support site, they strongly recommend to use the detection feature (JQuery.support) instead of the jQuery.browser property.
Actually, this has been answered already in another question, please check here How can you detect the version of a browser?
I want to develop a GWT application. The application contains 8 modules and all the modules run in a single page.
I have links for all the 7 options on the top.
When the page loads I want only the home content to be loaded and displayed.
Only when the user clicks the menu options, the menu content should load.
Can anyone suggest the best way to achieve this.
This has been addressed in GWT 2.0... First you'll remove the separate entry points you have for each module - since they're all on the same page, you only need one entry point per page.
Then you can use the GWT.runAsync() method at each point that you feel can be a split... it automatically cuts up the code into chunks that are downloaded as and when necessary.
Look here for the docs.
AFAIK, lazily loaded (or pluggable) modules can't be done in GWT. Partly this is due to the way the GWT compiler works - it likes to import all the code that it is ever going to see and then optimises and prunes it as viciously as possible (to make the resulting JS files as small and lead as possible). If it doesn't have access to all the source code up front, it might make optimisations that will break the pluggable modules (especially since nothing in the "core" application may reference the classes that the lazily-loaded modules need to work properly - the GWT compiler would prune those "unused" classes in the core module).
This stackoverflow question from May 09 asks the same thing and has the same answer - it can't be done.
I searched high and low about six months ago for an answer to this problem, because I really wanted to do what I believe you're asking for. Never found a solution.
(I haven't used GWT 2.0 - it might have addressed the issue)
Thank you for your answer. You are right that the whole Javascript code gets downloaded.
But there must be a way like say on of module is View Profile. Only when the user clicks the view Profile link, the widgets related to displaying my profile needs to be created. is this not possible.
We we wondering what are some ways developers have added a help function to their apps. What are some techniques people have used?
One way we were thinking of is to us UIWebView to display a HTML file with help instructions.
Thoughts appreciated.
I'm using UIWebView right now which pretty much contains all the help in a single page, along with some JQuery things to display popups, etc. But I like the way iCab Mobile (et al.) are doing things which is a sectioned UITableView with each row a separate topic or section within their overall help information (complete with icons...) then in their bundle they have each section in its own html file, organized by localization.
Another thing in my queue for the next release is to provide a dynamic "News" view. The rough idea is as follows... I have on my server a file or CGI where I can place small bits of news I'd like to push out to users. On startup, my app checks for network availability and if present, start a thread to see if anything has changed on the server since last updating the News data. If changes present, post an alert letting user know, and asking if they'd like to read it now. At that point, the latest news is already downloaded and cached, so they can simply read it later if they want, and I won't post anymore alerts until the server file changes again. (And one could add a preference/setting to disable these alerts.)
I'm thinking this would be a good way to let people know that some nasty bug is known and fixed and an update is sitting in the queue, solicit beta testers, promote upcoming features or other apps, etc. I can see where constant alerts everytime I've got something new to promote would get annoying, so having a setting to disable them means the user never has to read them unless they want to. Although some kind of override to warn of recently discovered/fixed bugs seems sensible.
FWIW, the author of Mover+/Mover has just started doing a similar thing, though I think Emanuele is perhaps only showing one Notelet at a time, whereas I envision a bit more of a history (shown in UIWebView) until I decide to age stuff off the bottom of the stack.
I'm using a scroll/page view to show several images containing small notes. Each image then tells the user about the more advanced functions on a specific part of the app.
In my opinion the help should only contain information that isn't a 100% relevant for the use of the application. It should be things the advanced user should use to make more use of the app. It should contain gold for the power users. The "basics" should be so obvious that no help would ever be needed. If that's not the case, I think, you've failed as a developer on the iPhone platform.
(Here's a screen shot from my demo app)
I'm currently creating a fairly complicated app. I'm thinking of doing help as a semi-transparent overlay - help in text form is hard to swallow for users; it's much more helpful to just point at stuff and say "this does that".