dmdz, take 2

[Graham Fawcett]

On Tue, 01 Jun 2010 18:59:23 -0500, Ellery Newcomer wrote:

> it doesn't really need itself to compile itself anymore (there's only 1
> or 2 files), but for posterity's sake:
> 
> http://personal.utulsa.edu/~ellery-newcomer/dmdz/dmdz.zip

How about adding a README? To a casual reader like me, it's not clear
what dmdz is supposed to do...

Graham


> 
> Rewrote split or whatever it was to be a bidirectional range.
> 
> Also wrote a nifty join implementation (of which I am very proud don't
> you dare criticize <g>).
> 
> Got rid of the modified basename and dirname, but the troubles are
> recorded in
> 
> http://d.puremagic.com/issues/show_bug.cgi?id=3937
> http://d.puremagic.com/issues/show_bug.cgi?id=4260
> 
> One side note: I couldn't think of an obvious way to make normalize a
> lazy range, unless you want it to only be able to iterate backwards.
> 
> std.process has been giving me trouble. I included some custom code for
> Posix, which I think Lars' efforts should obviate.
> 
>  From the windows shell, the output of the command passed to system()
> doesn't get printed to the screen. Ideas?
> 
> 
> Now, about dmdz:
> 
> I restricted the format of the zip files a bit for some reason or other
> which may have made sense at the time (it was a while ago). The idea is
> in the zip there are directories
> 
> bin
> lib
> obj
> resources
> src
> 
> with appropriate files going in each.
> 
> I also added support for [platform specific] extra compiler flags in
> dmdz.conf.
> 
> Haven't done much stress testing on it, but I have been using it quite a
> bit in the past few months for one project in particular. ranges are a
> bit new, though.
> 
> I meant to kick the tires on windows, but dmd is failing me, so screw
> that.
> 
> So yeah.