[plt-scheme] Project management

On 7/1/07, Robby Findler <robby at cs.uchicago.edu> wrote:
> No, not from the interactions window. Programs running in drscheme
> cannot tell that they are running in drscheme (by design!). Otherwise,
> it would be too easy to write a program that would work in the
> development environment, but wouldn't work in the deployment
> environment.

oh!  Is that why it is?  I thought it was a "security through lack of
functionality" type reason :).  I guess that makes some sence.  But it
does make it much harder to develop developer tools then since you
can't "probe" anymore.

I've heard about tools before but that's about it.  I really know very
little about them.  My issue with tools is that I see myself putting
my project on halt, spending a bunch of time learning about "tools",
then usually finding it won't do what I actually want anyway.  {that'
s my typical development cycle for those types of things}

I'm not sure how easy this would be, but if it's something quick,
would you be willing to write me a "skeleton tool"?  something that
hooks to the drscheme GUI interface somewhere, that when activated,
opens a new window, and then tries to open a new file "file1.ss" in
it?

If have examples of the primatives, I can start trying to play around.

Corey

P.S. Happy Birthday

> You can, however, make a tool that would do that.
>
> Open file and open new window already have keybindings, of course. See
> the menu shortcuts. Also, control-x control-f opens a new file. (I
> don't think that there is anoter way to open a new window, but you
> coudl add one with a keybindings file).
>
> Finally, just as a side-note, adding sleeps to avoid timing issues is
> a sure way to write buggy code. But, even if you do go to the tool
> route, you wouldn't need them. There's only one thread involved.

I agree, race conditions can be bad.

> Robby
>
> On 7/1/07, Corey Sweeney <corey.sweeney at gmail.com> wrote:
> > I see.  Is there a "open new window" command, and a "load file"
> > command, so that I could do something like:
> >
> >
> > > (drscheme:new-window)
> > > (drscheme:load-file "file-for-win-1-tab-1")
> > > (drscheme:load-file "file-for-win-1-tab-2")
> > > (drscheme:new-window)
> > > (drscheme:load-file "file-for-win-2-tab-1")
> > > (drscheme:load-file "file-for-win-2-tab-2")
> >
> > from a "interactions window"?
> >
> > (with some possible "sleeps" to avoid timing issues)
> >
> >
> > Or perhaps some keybindings (that don't interfere with the emacs
> > keybindings) could be written for "open new window" and "open file",
> > and I could write a project manager from the macro language of the
> > window manager....
> >
> > Corey
> >
> >
> >
> > On 7/1/07, Robby Findler <robby at cs.uchicago.edu> wrote:
> > > Maybe it helps to know that, when you open a new file, it goes into
> > > the frontmost drscheme window as a tab (when you have the open-in-tabs
> > > preference set)?
> > >
> > > Robby
> > >
> > > On 7/1/07, Corey Sweeney <corey.sweeney at gmail.com> wrote:
> > > > I seem toI have managed to exceed the 2 Gigs of memory in my new
> > > > machine during development, so I'm interested in looking at more
> > > > efficent ways of doing project management.
> > > >
> > > > Right now, I have a linux (bash) script launch 4 copies of drscheme,
> > > > each copy with about 10 tabs.
> > > >
> > > > Is there a way to open a second window from the command prompt, and
> > > > load files into tabs in the second window?  My idea is that by only
> > > > loading drscheme once, I'm hoping it will save memory.
> > > >
> > > >
> > > > Or, alternatively, is there a way of opening windows, and loading
> > > > files into tabs from the interaction prompt?  Something like:
> > > >
> > > > > (define window1 (drscheme:new-window))
> > > > > (drscheme:load-file-in-tab "file1.scm" window1)
> > > > > (drscheme:load-file -in-tab "file2.ss" window1)
> > > > > (define window2 (drscheme:new-window))
> > > > > (drscheme:load-file-in-tab  "file3.ss" window2)
> > > > > (drscheme:load-file-in-tab  "file4.scm" window2)
> > > >
> > > > Due to some conversations I've had in the past, I'm guessing that
> > > > would have to be nested,  I.E. every time I wanted to edit a file,
> > > > load drscheme, and have it open a file, which has (require
> > > > "drscheme.ss") or something in it, and auto-run it.  (by the way, how
> > > > do you "auto-run" a program in drscheme?), then type something like
> > > > that at the interaction prompt, which would open the tabs in a nested
> > > > copy of drscheme?  Which would give me 2 running copies of drscheme,
> > > > which while non-ideal, would be better then 4.
> > > >
> > > > Or would the "load-file-in-tab" functions be protected behind a module
> > > > barrier, and thefrore inaccessable?
> > > >
> > > > I'm not really sure about anything here, so this is kind of a general
> > > > probe.  I'm just looking for a solution that I can interactively
> > > > improve from the "interactions area" as i'm using it.
> > > >
> > > >
> > > > Corey
> > > >
> > > >
> > > > --
> > > > ((lambda (y) (y y)) (lambda (y) (y y)))
> > > > _________________________________________________
> > > >   For list-related administrative tasks:
> > > >   http://list.cs.brown.edu/mailman/listinfo/plt-scheme
> > > >
> > >
> >
> >
> > --
> > ((lambda (y) (y y)) (lambda (y) (y y)))
> >
>


-- 
((lambda (y) (y y)) (lambda (y) (y y)))