Neil Hodgson | 1 Mar 2003 07:29
Picon

Re: RFE: Close SciTE with Ctrl-W

Jens Hatlak:

> I'd like to voice a feature request: I find myself using SciTE's
> keyboard shortcuts very often, esp. things like Ctrl-O or Ctrl-W. One
> thing I really miss compared to Mozilla+MultiZilla is the ability to
> close the application using Ctrl-W.

   Alt+F X will close the application. Have you experimented with
user.shortcuts?

> There should, however, be an option
> for an alert to be displayed if the last tab is to be closed; on by
> default. For users who dislike this behaviour there should also be an
> option to switch it off completely.

   So what you are after is [Closing the last buffer closes the
application]? I'm not going to implement this but if a well written patch
with documentation arrives, I'll include it.

   Neil
Neil Hodgson | 1 Mar 2003 08:27
Picon

Re: Patches: Minor corrections and lexer addons

   Hi Pedro,

> (1) fortran_corrections.diff: Fixes a couple of problems with the
> fortran.properties file,

   OK, although I expect you want to remove the comment character '#' before
"braces.fortran.style".

> and adds the fortran extensions to source.files
> field of SciTEGlobal.properties.

   Not OK. As the comment says, this setting is a limited resource so I have
to choose which subset of extensions go in. It would be very useful if this
limit could be fixed.

> (2) fortran_lexer.diff: Adds a couple of error tags to Scintilla.iface, to
> allow scintilla to recognize the Fortran compilers that I use: Essential
> Lahey Fortran 90 (ELF90) (www.lahey.com) in Windows, and Intel Fortran
> Compiler (IFC) in Linux
> (http://www.intel.com/software/products/compilers/flin/index.htm).
> Implementing the IFC check in LexOthers.cxx has to be done *before* the
> Borland's error check, because both of them check for the "Error " string.
If
> the Borland check is first, it swallows the error, and later is not
capable
> of finding the error line.

   These use strstr on "Error " and "Line " which finds them anywhere on the
line although the examples show them at the beginning. It is best for
detection to be as specific as possible to avoid interfering with other
(Continue reading)

Neil Hodgson | 1 Mar 2003 12:02
Picon

Re: Bug with japanese characters

Tyler Stratton:

> - If i set the code.page to 0 (not sure what it actually is),

   That assumes 8 bit coding so no need to call DBCS functions.

> Syntax higlighting comes out correctly, but character movement
> works incorrectly over japanese characters.

   As expected.

> - If i set the code.page to 932 (microsoft sjis), character
> movement comes out correctly, but the syntax higlighting
> sometimes messes up when it encounters japanese
> characters. for example strings arent recognzied as being
> closed, and more annoyingly, multiline comments arent
> closed. please see the attached image.

   Thanks for the report. The bug is reproducible on my machine and I think
I have fixed it.
Fix in CVS and available from
http://www.scintilla.org/scite.zip Source
http://www.scintilla.org/wscite.zip Windows executable

> I am including an image of the incorrect syntax hilighting
> and the text from the sample code.

    Extra merit points for sending an illustrative sample and screen shot
evidence.

(Continue reading)

Jens Hatlak | 2 Mar 2003 20:08
Picon
Favicon

Re: RFE: Close SciTE with Ctrl-W

Neil wrote:
>>I'd like to voice a feature request: I find myself using SciTE's
>>keyboard shortcuts very often, esp. things like Ctrl-O or Ctrl-W. One
>>thing I really miss compared to Mozilla+MultiZilla is the ability to
>>close the application using Ctrl-W.
> 
>    Alt+F X will close the application. Have you experimented with
> user.shortcuts?

This is a) dependant on the menu language and b) not in the least 
comfortable enough for me. Then there is Alt+F4, independent from the 
menu language, but too complicated as well: I'd press it with thumb and 
much-spanned pointing finger, whereas Ctrl+W is as easy as small and 
middle finger (no spanning). Simple ergonomics.

>>There should, however, be an option
>>for an alert to be displayed if the last tab is to be closed; on by
>>default. For users who dislike this behaviour there should also be an
>>option to switch it off completely.
> 
>    So what you are after is [Closing the last buffer closes the
> application]? I'm not going to implement this but if a well written patch
> with documentation arrives, I'll include it.

Very sad to hear that, I thought it would be easy to implement. Anyone 
else? I know I'd get lost in source if I tried myself, and the result 
would be anything but an acceptable patch. You'd have to spend even more 
time on making it applicable than me to try and make it work at all...

--

-- 
(Continue reading)

Bruce Dodson | 3 Mar 2003 03:22
Picon
Favicon

Re: RFE: Close SciTE with Ctrl-W

Alt-F4 is the standard keybinding for "close this app".  However, as Neil
mentioned, there is user.shortcuts.  If all you're looking for is a simpler
keystroke that does the same as Alt-F4, you could put something like this in
SciTEGlobals:

user.shortcuts=F4|IDM_QUIT|

Then you wouldn't need to hold down the Alt.

----- Original Message -----
From: "Jens Hatlak" <jh <at> junetz.de>
To: <scite-interest <at> lyra.org>
Sent: Sunday, March 02, 2003 3:08 PM
Subject: Re: [scite] RFE: Close SciTE with Ctrl-W

> Neil wrote:
> >>I'd like to voice a feature request: I find myself using SciTE's
> >>keyboard shortcuts very often, esp. things like Ctrl-O or Ctrl-W. One
> >>thing I really miss compared to Mozilla+MultiZilla is the ability to
> >>close the application using Ctrl-W.
> >
> >    Alt+F X will close the application. Have you experimented with
> > user.shortcuts?
>
> This is a) dependant on the menu language and b) not in the least
> comfortable enough for me. Then there is Alt+F4, independent from the
> menu language, but too complicated as well: I'd press it with thumb and
> much-spanned pointing finger, whereas Ctrl+W is as easy as small and
> middle finger (no spanning). Simple ergonomics.
>
(Continue reading)

Brian Sullivan | 4 Mar 2003 06:13
Picon

browse button-rollover textbox editor

 

 Google found you while  I was looking for a browse button to put on my web page so people could upload images to my web.

I could also use a rollover text box editor...

Any suggestions appreciated...

Thanks in advance,

 

   Thanks,

   Brian Sullivan

   brian <at> ohkc.com
   913-940-0338
   www.ohkc.com

 

 
Neil Hodgson | 4 Mar 2003 13:10
Picon

Calltip, POSIX regex, CSS lexer and user.shortcuts patches

   Jakub Vrana has contributed some patches to improve calltips, add some
optional POSIX compatibility to regular expressions and allow quoted strings
in the CSS lexer. I have fixed a problem with user.shortcuts on GTK+ where
the key was also inserted in the documented if it was a visible character.
The POSIX change allows use of bare ( and ) for tagged sections rather than
\( and \).

   Committed to CVS and available from
http://www.scintilla.org/scite.zip Source
http://www.scintilla.org/wscite.zip Windows executable

   Neil

   More information from Jakub:

1. If text 'foo(bar(' is written, calltip for bar() appears. In old
version, if foo() calltip was visible, nothing happens.

2. If text 'foo(bar(a, b),' is written, calltip for foo() appears with
second parameter highlighted. In old version, it was third parameter.

3. If user moves caret to 'foo(bar(a, |b),' afterwards and say Show
Calltip, calltip for bar() is displayed. In old version, nothing
happens if other calltip was displayed.

Here is it. There are no |, {, } in Scintilla RexExp implementation,
so I implemented POSIX syntax just for ( and ).

If you don't want RE support in Scintilla and want to move it to
SciTE, it will be probably best done by you.

There is new SciTE property $(find.replace.regexp.posix) and new value
SCFIND_POSIX=0x00400000 in Scintilla.iface.

I have also corrected Find Previous with RexExp. In old version, if
string matched somewhere in line, Find Previous stays on it. It was
working only if string has started at 1st column.
Philip Mateescu | 4 Mar 2003 15:36

Keyword missing from ruby.properties

Hi,

I believe the "require" keyword is missing from the ruby.properties
file.

--

-- 
Best regards,
 Philip                          mailto:pmateescu <at> novosadhayes.com
Neil Hodgson | 5 Mar 2003 11:10
Picon

Re: browse button-rollover textbox editor

Brian Sullivan:

> I could also use a rollover text box editor...

   What do you mean by "rollover text box editor"?

   Neil
Neil Hodgson | 5 Mar 2003 11:18
Picon

Re: Keyword missing from ruby.properties

Philip Mateescu:

> I believe the "require" keyword is missing from the ruby.properties
> file.

   OK, committed.

   Neil

Gmane