This is the mail archive of the cygwin mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

RE: cygwin setup window size, why no maximize button enabled ?


On Fri, 16 Apr 2004, Dave Korn wrote:

> > -----Original Message-----
> > From: cygwin-owner On Behalf Of Christopher Faylor
> > Sent: 16 April 2004 16:24
>
> > On Fri, Apr 16, 2004 at 04:10:55PM +0100, Dave Korn wrote:
> > >I've had a play with that myself, and I think the real
> > reason it hasn't
> > >been released so far is because it's not-yet-ready-for-primetime.
> >
> > Can you elaborate?  Is the implementation so bad that it would cause
> > more confusion that having a tiny setup screen seems to engender?
>
>   Nothing showstopping, just niggles:
>
>  - There is no resizing of the left-hand margin, and no rewrapping of the
> text or repositioning of items within the main client area of the dialog on
> any page except the "Select Packages..." page.  Everything just stays with
> the positioning it has when the window opens at its default initial size.

That's a question of specifying proper resizing controls for these
screens.

>  - There should be minimum size restrictions to prevent gadgets being shoved
> out of the visible area or half-drawn and half-undrawn when reducing the
> window's size from the default.

This is likely a deficiency in the mechanism, but one that can be fixed
without too much pain.

>  - Resizing the column headers at the top of the package display widget is
> buggy; at any rate, I've found some in the "Package" column header, possibly
> because it's the last column header at the right-hand side.  It's easy to
> end up with headers that don't fill the width of the widget, to get text
> displays under the header that exceed the apparent width of the header, and
> to get useless dead whitespace at the right hand side of the chooser that
> you can't get rid of by expanding the package column header because the dead
> space resizes in proportion.  Furthermore there's some bad interaction with
> the scroll bar, which doesn't resize or reproportion itself when the
> effective width of the chooser is increased or decreased; and both interact
> badly with changing the main window size.  I've just managed to get it into
> a state where the content of the chooser is scrolled partially off to the
> left of the screen (you can't see the Category column any more) but no
> scroll bar has appeared, not even after resizing the window a few times, so
> I don't know how I'm going to get it back.  The interactions between window
> size, column headers and scrollbars don't work properly at all.

FWIW, I've tried to fix some of this, but had little luck.

> > I've been advocating just releasing the current snapshot as
> > is,
>
>   It's a judgement call, so AFAICS it's largely just a matter of taste.  A
> buggy GUI is not the end of the world and if a user gets it into an
> unworkable state all they need do is cancel and start again.  OTOH it looks
> kind of flaky for a release.  It's about image and presentation and
> neat-and-tidy-ness.  I suppose there's no reason not to release it and list
> the gui problems as known bugs, PTC.

IMO, the snapshot can be no worse than the current version, in that if the
resizing is not used, the snapshot will behave identically to the current
official version (at least as far as the GUI is concerned -- there were
some logic bug fixes there too).

> I've just been doing some MFC coding for the first time, I'll take a
> look at fixing some bugs when[*] I get some spare time.

AFAIK, setup doesn't use MFC.

> > although it would also be nice to fix the pcre0 dependency problem.
> >
> > cgf
>
>   Can't speak to the dependency problem, as I haven't run up against it, but
> I remember seeing a lot of posts on the list from people who've not had
> libpcre0 properly installed until they re-run setup a second time; do you
> mean that it's a bug in setup's algorithm that it doesn't properly calculate
> the dependency of various packages that use it on the library, rather than
> some issue to do with the packaging/setup hint files?
>
>     cheers,
>       DaveK
>
> [*] If!

A bit of debugging (with judicious printfs) will go a long way here.  All
you need is setup built from source with debugging, and a blank system on
which to install.  I suspect a dual-boot/VMWare system, where you can wipe
one installation clean of Cygwin and reinstall from scratch, would be
ideal.  Anyone have the above, the skills, the time, and the inclination? ;-)
	Igor
-- 
				http://cs.nyu.edu/~pechtcha/
      |\      _,,,---,,_		pechtcha@cs.nyu.edu
ZZZzz /,`.-'`'    -.  ;-;;,_		igor@watson.ibm.com
     |,4-  ) )-,_. ,\ (  `'-'		Igor Pechtchanski, Ph.D.
    '---''(_/--'  `-'\_) fL	a.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

"I have since come to realize that being between your mentor and his route
to the bathroom is a major career booster."  -- Patrick Naughton

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]