With the latest build, or maybe including the builds prior to beta 7.5 I just never noticed, I'm getting annoyed with a little autofill feature Apple implemented for substitute tokens and their objects when creating log messages. I haven't really noticed this with class methods or instance methods or autocompleting anything else really, so far.
Before, in Xcode 6, this would display:
Ahhhhh, nice & crisp. As you can see, everything suffixing the user input is opportunity for autofill. All I had to do was press tab[enter] once to autofill 'error' and include the closing parenthesis, then it would automatically place my cursor after the closing parenthesis, hence ALL I had to do was type a ; for that statement to be complete and I was happy :)
Now, this displays :
Ehhhhh, troublesome & redundant. As you can see in this example, it includes an additional quotation mark. Now when I press tab to autofill 'error', it just autofills 'error' and places the cursor before the closing parenthesis. So it looks like this:
NSLog(#"%#", error|)
And if I were to press tab to go to the next sub word that needs autocomplete it will just indent the closing parenthesis:
NSLog(#"%#", error )
So essentially, I have to either move it with my mouse or press → and then and only then, I'll be able to enter a ;. Side note - pressing ↵ to autofill results in the same behavior as described.
I am a fast typer, don't want to say my speed because I don't want it to rub off the wrong way but it's 70+wpm, so this irritates me, especially since I've been used to a way for so long. Any way to correct the behavior?
I have tried turning off Enable type-over completions in the Xcode preferences pane, but it still places the cursor before the closing parenthesis.
printf resorts to the same behavior with Xcode 7.5
Related
BigDecimalField has a nice feature: it allows only digits, "+" and "-" signs, and a decimal separator defined by the current locale ("." or ","). It works fine until someone tries to enter a combination that is not a number or cannot be directly converted to the java BigDecimal type (e.g "123...45.6+++7-89--").
For sach combinations, the validation passes without any warning, and the background business object gets a null value.
Field is in the dialog and every time before opening the dialog there is a call to Binder.readBean() method (in order to set the appropriate data for editing), on "save" button in dialog I call Binder.writeBean(). Problem is that the next time I open the dialog, the problematic BigDecimalField still contains an invalid number entered,
actually Binder.readBean() stops working for that field.
There is a similar bug with IntegerField which I noticed: IntegerField validation bug.
How can I validate BigDecimalField and avoid this bug? How do i get the BigDecimalField to work again and not keep showing invalid data? Is there a elegant way to catch the NumberFormatException that probably occurs somewhere inside the vaadin api, and proces it i.e. warn the user that he has to enter a valid decimal number. Is there a way to do it via Binder?
I am using Vaadin 23.3.0
Checking format error is not yet by default on. You need to enable it by setting enforceFieldValidation=true in feature flags in src/main/resources/vaadin-featureflags.properties file.
com.vaadin.experimental.enforceFieldValidation=true
See more at: https://github.com/vaadin/platform/issues/3066
Regarding the spell checking in iOS, it's possible to tell the checker to ignore a word (or learn a word),
https://developer.apple.com/documentation/uikit/uitextchecker
func ignoreWord(String)
Tells the receiver to ignore the specified word when spell-checking.
- Apple doco
Say I have a UITextView which opens. I want spell checking On.
I know the user may type "fattie" which would get the red underline.
How do I tell that text view in that instance, to, ignore "fattie" ?
An obvious use case ...
User is typing in "#tag" type friends; in our data of course we know what all the tags are, it's absurd they get marked as spelling errors.
It seems incredible one can't just say "don't underline these words - - list".
Code example ....
So we have
var t: UITextView
and then, there must be "some way" to:
yourTextView.something->something.textChecker.ignoreWord("fattie"
.. some way to get to the text view's textChecker instance! How ?!?!
Partial answer: I just stumbled on to that, bizarrely, you can just call
UITextChecker.learnWord("fattie")
UITextChecker.learnWord("blahdee")
from, apparently, just anywhere in an app.
However this raises many issues,
• How to call the 'ignore' one, which seems better
• That one still makes the user tap the annoying, stupid, "in quotes" OK box in the suggestions bar - it seems to have not really "learned" anything
• Disturbingly, I think this goes for the "WHOLE PHONE". I only want it in that instance of the user using that text view.
A mystery!
How do I add use the bell character with Xterm.js? I see a couple of entries in the sources 'options' defaults, but dont get anything whether i send it from the shell or write it directly to the terminal, even with those set to true. I tried overloading the term.bell function with one containing a window.navigator.vibrate, and that works if I call it directly, but doesnt fire in response to a bell character.
Ok. So testing determined nothing is blocking the bell character ("\x07" or "\u0007") from getting through to xterm.js. It just doesnt respond, even if you enable the flags in the options. The code in the source looks like it should work, but something internal is not connected. Since I am avoiding actual changes to the source to make upgrades to xterm.js straightforward and out of the box, I worked a little magic. Here's a hack to get the bell character working:
Make a function that fires on the "message" event from your websocket. In it, check for a match for BEL. Then have it do your voodoo if it detects it. Something like:
sock.addEventListener('message',function(v){ if(v.match(/\x07/)){ /*voodoo here*/ } });
Where v is event, \x07 is BEL, and voodoo is code or an invocation to blip the screen white briefly, make a chirp, vibrate if your getting around on "mobile", or open a portal to the single-sock dimension.
Annoyingly, now I need to look up more ansi codes and find a pattern, as some end in a BEL character. So this will work by itself, but will be set off by some codes not intended to act as BEL, because they contain that character. More on this later....
FYI, not sure if this works with term.attach(). I have my own thing that does some preprocessing anyway, so basically all I had to do was splice in the if/match, but cut that out, the above code is the minimal cut-paste version.
Hope this helps someone else. Bug out.
I have a ListGrid with ListGridRecords where I am trying to display the straight xml output from a call I'm doing on the backend. Here's an example:
lgr.setAttribute("XML", "<xml><response>Bob</response></xml>");
However, when it goes to display the contents of lgr in the ListGrid, it doesn't display it. (i.e. it is blank). If I make the field editable, I can double click on the cell and it will then display it. Also, if I get rid of the '<' symbols it displays as well. However, that kind of defeats the purpose since I want to display the exact xml used.
Do I need to escape the '<' somehow in the ListGridRecord? What is the best way to do this in GWT?
Ok, just found a thread that seems to solve my problem at least halfway. I would still like to enable the user to copy it without the encoded values, but at least now they can see it. I believe I'll enable some sort of click handler on the fields to to enable the correct copying.
http://www.gwt-ext.com/forum/viewtopic.php?f=7&t=2693
I dont know what exactly i have to type in title for this ,i tried my best
anyway coming to topic
I am making one acc checker for that purpose ,i am sending user and pass from my bsskinedit1 and bsskinedit2
here is my code
s:='http:\\site.com..premlogin='+bsskinedit.text+'&password='+bsskinedit2.text
but it giving some error ,then i used showmessage whats wrong with it then i came with strange result
see below
observer after 4 & and p combining together and appearing as a some new symbol :(
can any one tell me why its coming like this ?
Your code (where you build the URL) is most likely correct (I guess the above has some typos?!), but when you display the URL in a label for instance, the & character is treated as indicator for an accelerator key.
By Windows design, accelerator keys
enable the user to access a menu
command from the keyboard by pressing
Alt along the appropriate letter,
indicated in your code by the
preceding ampersand (&). The character
after the and sign (&) appears
underlined in the menu.
If you want to display the & character itself, you have to set your string variable to &&.
By placing two ampesands together - you state that the character following the first one is not used as the accelerator - rather you actually want to get it (only one) displayed.
Just use your debugger if you want to see the real value that your string variables have, don't output them to a message box or the like... It may have side effects, as you can see.
Regarding the URL you build: I can't possibly know how it has to be correctly, but at least you should use the right slashes!
s := 'http://site.com...'
(All quotes from delphi.about.com)
In addition to what Mef said, you can use OutputDebugString to add your string to the event log in its raw form, so you don't need to modify it before displaying it. Delphi should capture those strings automatically if you're running from the debugger. If you aren't running it from Delphi you can use DebugView instead, which captures the messages from any running applications.