q-lang-users — general discussion, questions, suggestions, etc.

Re: [q-lang-users] [qlist] SDL anyone?

[Andrew B. <and...@ya...>]

On 27 Dec 2006, at 08:39, Albert Graef wrote:

> I really have to write an introduction to Q-SWIG sometime, but I hope
> that with the SWIG docs and the Q examples you'll be able to find your
> way. ;-) Of course you could also use Q's "raw" C interface, which is
> documented fairly well in the manual, but SWIG is really the only
> practical option for wrapping bigger libs.

I just got the example from the SWIG website's tutorial to work with  
q.  Took about a half hour, most of which was spent fooling around  
building SWIG itself.  (One of the tests failed, and being the sucker  
that I am I got sucked into trying to figure out why.)

I don't know what the copyright is on their tutorial (http:// 
www.swig.org/tutorial.html) but just copying that and replacing all  
their 'Building a XXX Module' sections with a single 'Building a q  
module' would fit reasonably well in qdoc.  At least it would not  
take a lot of time to write.  ;^)

Well, my kids'll be up in about 7 hours, so it's bedtime for me.   
Tomorrow I'll try to rewrite SDL's 'testver' into q, adding SDL  
definitions as I go...  Ta, ta!

-andrew

Re: [q-lang-users] [qlist] Re: Feature request (or maybe bug report) for qcc

[Andrew B. <and...@ya...>]

On 27 Dec 2006, at 08:54, Albert Graef wrote:

> Andrew Berg wrote:
>> I don't know if this is hard, easy, or other.  What I would like  
>> is for
>> qcc to pick up the options passed to ./configure when q was built.
>
> It would be possible but not in general desirable. E.g, under Windows
> qcc provides different backends to gcc and msvc so it's not a good  
> idea
> to tie qcc to the set of options it was compiled with.

I'm not really looking for them to pick up the -I and -L flags, but  
rather to notice that I gave './configure' a '--prefix=' directive,  
so that qcc knows _at_least_ where its own 'qint.h', 'libqint.*', &c.  
are located.

I was rather hoping that in some sense qcc's job was congruent to  
that of apache's 'apxs'.  Am I wrong to expect qcc to be able to find  
its own components?

-andrew

Re: [q-lang-users] [qlist] SDL anyone?

[Andrew B. <and...@ya...>]

On 27 Dec 2006, at 17:02, Albert Graef wrote:

> Andrew Berg wrote:
>> It's really a shame about the ":" character not being
>> allowed in identifiers...  I'll have to campaign for that at some  
>> point.  :)
>
> Unfortunately that's not possible; ':' is already used for type
> matching, and '::' for qualified identifiers. But there are two
> delimiters in the ASCII range, '%' and '?', which aren't currently  
> used
> and thus can be declared as operators.
>

The crazy part is that I don't even want it as an operator.  The  
syntax I'm looking for is their funny named parameters.  Example (in  
Obj-C (meaningless, but I'm just trying to show syntax)):

r = [SomeDictThing new];
for (i = 0; i < 10; i ++) {
   [r at: i put: j];
}

The part I want to co-opt is the 'r at: i put: j' where r would be my  
class instance, and 'at:' and 'put:' are q constructors which I would  
just use as place holders, and the 'i' and 'j' would be some kind of  
variables (or parenthesized expressions).  I could do it with the ':'  
dropped, but the readability suffers.  I could use some Unicode ':'  
replacement, but the type-ability suffers.  I could use a character  
that is allowed in identifiers, like '_', but that's just silly.

I do understand that things are as they are for some good reasons.   
My point was just that if the grammar could be mangled a bit to  
support it, I will probably make a sub-dialect to support my own evil  
designs.  And I would probably whine and cajole to get it added into  
the trunk.  :^)  But, don't worry yet.  It is low priority.  Before  
that, I have a SDL wrapper to make work.

-andrew

Re: [q-lang-users] [qlist] SDL anyone?

[Albert G. <Dr....@t-...>]

John Cowan wrote:
> We stole "%" for rational division, remember?

Oops, yes. But of course we still have all the non-ASCII delimiters
above 0x7f. :)

Albert

-- 
Dr. Albert Gr"af
Dept. of Music-Informatics, University of Mainz, Germany
Email:  Dr....@t-..., ag...@mu...
WWW:    http://www.musikinformatik.uni-mainz.de/ag

Re: [q-lang-users] [qlist] SDL anyone?

[John C. <co...@cc...>]

Albert Graef scripsit:

> Unfortunately that's not possible; ':' is already used for type
> matching, and '::' for qualified identifiers. But there are two
> delimiters in the ASCII range, '%' and '?', which aren't currently used
> and thus can be declared as operators.

We stole "%" for rational division, remember?

-- 
You let them out again, Old Man Willow!                 John Cowan
What you be a-thinking of?  You should not be waking!   co...@cc...
Eat earth!  Dig deep!  Drink water!  Go to sleep!
Bombadil is talking.                                    http://ccil.org/~cowan

Re: [q-lang-users] [qlist] SDL anyone?

[Albert G. <Dr....@t-...>]

Andrew Berg wrote:
> It's really a shame about the ":" character not being
> allowed in identifiers...  I'll have to campaign for that at some point.  :)

Unfortunately that's not possible; ':' is already used for type
matching, and '::' for qualified identifiers. But there are two
delimiters in the ASCII range, '%' and '?', which aren't currently used
and thus can be declared as operators.

Albert

-- 
Dr. Albert Gr"af
Dept. of Music-Informatics, University of Mainz, Germany
Email:  Dr....@t-..., ag...@mu...
WWW:    http://www.musikinformatik.uni-mainz.de/ag

Re: [q-lang-users] [qlist] SDL anyone?

[Andrew B. <and...@ya...>]

On 27 Dec 2006, at 08:39, Albert Graef wrote:

> That would be great! :) Meanwhile I'm tempted to take another shot  
> at a
> decent GUI interface. A direct Qt wrapper for Q would be quite useful.

Being a Mac bigot--oh, I mean enthusiast--I have (waay down on my  
list of priorities) the idea to do a Obj-C bridge to q.  I think this  
would be groovy, because it would give all the Cocoa UI classes, and  
q is on the short list of languages where they syntax could be  
mangled to the point that the Obj-C/Smalltalk syntax for message  
sends could be preserved.  It's really a shame about the ":"  
character not being allowed in identifiers...  I'll have to campaign  
for that at some point.  :)

-a

Re: [q-lang-users] [qlist] SDL anyone?

[Andrew B. <and...@ya...>]

On 27 Dec 2006, at 08:39, Albert Graef wrote:

>
>> Drat, the sdl site seems to be down.  Or else my network here is
>> messed up.
>
> http://www.libsdl.org seems to work ok over here.

Ahh, good.  Somebody restarted their httpd.  I determined yesterday  
that I could get a ping back from them, but no http.  Also the cached  
version at Google showed a PHP error accessing their db, so I'm gonna  
stick with my "their site was down" story.  :)

-andrew

Re: [q-lang-users] [qlist] Re: Download page on Opera

[Andrew B. <and...@ya...>]

Yep, it's happy now.  Thanks,

-a

On 27 Dec 2006, at 08:04, Albert Graef wrote:

> Fixed (hopefully). Thanks a lot for the fix. Could you please check
> whether it renders ok in your Opera version now?
>

Re: [q-lang-users] Feature request (or maybe bug report) for qcc

[Albert G. <Dr....@t-...>]

Andrew Berg wrote:
> I don't know if this is hard, easy, or other.  What I would like is for
> qcc to pick up the options passed to ./configure when q was built.

It would be possible but not in general desirable. E.g, under Windows
qcc provides different backends to gcc and msvc so it's not a good idea
to tie qcc to the set of options it was compiled with.

> Anyhow, when I later use qcc, for example to build q-opengl, I get a bunch of
> errors:

Yes, the Makefile of q-opengl (and probably some other q-swig based
projects) lacks support for at least CFLAGS and LDFLAGS. I put this on
my TODO list. For the time being, you could set gcc's environment
(LIBRARY_PATH, C_INCLUDE_PATH) to work around this.

Albert

-- 
Dr. Albert Gr"af
Dept. of Music-Informatics, University of Mainz, Germany
Email:  Dr....@t-..., ag...@mu...
WWW:    http://www.musikinformatik.uni-mainz.de/ag

Re: [q-lang-users] [qlist] SDL anyone?

[<bri...@co...>]

I second the recommendation for SQLITE. It is a minor wonder :)

-------------- Original message -------------- 
From: Albert Graef <Dr....@t-...> 

> Andrew Berg wrote: 
> > That sounds like fun. Probably a good warm-up exercise for a wrapper 
> > that I'm actually a bit more interested in, with the GOODS database. 
> 
> That would be great! :) Meanwhile I'm tempted to take another shot at a 
> decent GUI interface. A direct Qt wrapper for Q would be quite useful. 
> 
> BTW, if you're interested in database interfaces, then you should maybe 
> take a look at the code of q-sqlite3. I know that OODBs are different, 
> but q-sqlite3 might give you a couple of ideas on how to handle result 
> sets and callbacks into the Q interpreter. 
> 
> > Drat, the sdl site seems to be down. Or else my network here is 
> > messed up. 
> 
> http://www.libsdl.org seems to work ok over here. 
> 
> > Anyhow, where is the best place to look for examples? I guess I 
> > should start by getting q-opengl to work, then look at how that 
> > works, eh? 
> 
> q-opengl is rather complex, with all the callback and special 
> typemapping stuff, but you might wish to take a look at the Makefile to 
> get an idea how to set up compilation of a q-swig module. I'd suggest 
> that you first take a look at the basic q-swig example in 
> /usr/share/q/examples/swig. And then there's a bunch of additional 
> introductory examples in the swig-1.3.22-q-20060225.tar.gz tarball at 
> SWIG-1.3.22/Examples/q. It should also be useful to have a look at the 
> html documenation for SWIG 1.3.22 which is also in that tarball. 
> 
> I really have to write an introduction to Q-SWIG sometime, but I hope 
> that with the SWIG docs and the Q examples you'll be able to find your 
> way. ;-) Of course you could also use Q's "raw" C interface, which is 
> documented fairly well in the manual, but SWIG is really the only 
> practical option for wrapping bigger libs. 
> 
> Cheers, 
> Albert 
> 
> -- 
> Dr. Albert Gr"af 
> Dept. of Music-Informatics, University of Mainz, Germany 
> Email: Dr....@t-..., ag...@mu... 
> WWW: http://www.musikinformatik.uni-mainz.de/ag 
> 
> ------------------------------------------------------------------------- 
> Take Surveys. Earn Cash. Influence the Future of IT 
> Join SourceForge.net's Techsay panel and you'll get the chance to share your 
> opinions on IT & business topics through brief surveys - and earn cash 
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV 
> _______________________________________________ 
> q-lang-users mailing list 
> q-l...@li... 
> https://lists.sourceforge.net/lists/listinfo/q-lang-users 

Re: [q-lang-users] [qlist] SDL anyone?

[Albert G. <Dr....@t-...>]

Andrew Berg wrote:
> That sounds like fun.  Probably a good warm-up exercise for a wrapper  
> that I'm actually a bit more interested in, with the GOODS database.

That would be great! :) Meanwhile I'm tempted to take another shot at a
decent GUI interface. A direct Qt wrapper for Q would be quite useful.

BTW, if you're interested in database interfaces, then you should maybe
take a look at the code of q-sqlite3. I know that OODBs are different,
but q-sqlite3 might give you a couple of ideas on how to handle result
sets and callbacks into the Q interpreter.

> Drat, the sdl site seems to be down.  Or else my network here is  
> messed up.

http://www.libsdl.org seems to work ok over here.

> Anyhow, where is the best place to look for examples?  I guess I  
> should start by getting q-opengl to work, then look at how that  
> works, eh?

q-opengl is rather complex, with all the callback and special
typemapping stuff, but you might wish to take a look at the Makefile to
get an idea how to set up compilation of a q-swig module. I'd suggest
that you first take a look at the basic q-swig example in
/usr/share/q/examples/swig. And then there's a bunch of additional
introductory examples in the swig-1.3.22-q-20060225.tar.gz tarball at
SWIG-1.3.22/Examples/q. It should also be useful to have a look at the
html documenation for SWIG 1.3.22 which is also in that tarball.

I really have to write an introduction to Q-SWIG sometime, but I hope
that with the SWIG docs and the Q examples you'll be able to find your
way. ;-) Of course you could also use Q's "raw" C interface, which is
documented fairly well in the manual, but SWIG is really the only
practical option for wrapping bigger libs.

Cheers,
Albert

-- 
Dr. Albert Gr"af
Dept. of Music-Informatics, University of Mainz, Germany
Email:  Dr....@t-..., ag...@mu...
WWW:    http://www.musikinformatik.uni-mainz.de/ag

Re: [q-lang-users] Download page on Opera

[Albert G. <Dr....@t-...>]

Andrew Berg wrote:
> I don't know if you care about supporting "alternative" web browsers,
> but when I look at it with the latest version of Opera, it is pretty
> funked up.  I can send a little screen grab if it helps, but Opera is a
> free download, and it might be easier to test it yourself.  Oh, I've
> also seen this on both Windows and the Mac.

Fixed (hopefully). Thanks a lot for the fix. Could you please check
whether it renders ok in your Opera version now?

NB: While at it, I also added links to John's Chicken interface to the
examples and links pages.

Albert

-- 
Dr. Albert Gr"af
Dept. of Music-Informatics, University of Mainz, Germany
Email:  Dr....@t-..., ag...@mu...
WWW:    http://www.musikinformatik.uni-mainz.de/ag

[q-lang-users] Feature request (or maybe bug report) for qcc

[Andrew B. <and...@ya...>]

I don't know if this is hard, easy, or other.  What I would like is  
for qcc to pick up the options passed to ./configure when q was built.

Here's my example.  I like to keep everything I build in $HOME/local  
so that it does not interfere (so much) with upgrading the O/S.  When  
I install q, I use "./configure --prefix=$HOME/local CFLAGS=-I$HOME/ 
local/include LDFLAGS=-L$HOME/local/lib".  This works fine, although  
it would be nice if prefix/include and prefix/lib were added to the  
CFLAGS and LDFLAGS by default.  It seems like a lot of projects do  
that, although I know that a lot of them do not.  Anyhow, when I  
later use qcc, for example to build q-opengl, I get a bunch of errors:

(cd src && make all)
qcc -o opengl.la opengl_wrap.c -- --link -lglut -lGL -lGLU
opengl_wrap.c:13:18: error: libq.h: No such file or directory
In file included from /usr/include/machine/_types.h:28,
                  from /usr/include/sys/_types.h:27,
                  from /usr/include/_types.h:27,
                  from /usr/include/string.h:60,
                  from opengl_wrap.c:58:/usr/include/i386/_types.h:  
In function 'MODULE':
/usr/include/i386/_types.h:31: error: storage class specified for  
parameter '__int8_t'
/usr/include/i386/_types.h:35: error: storage class specified for  
parameter '__uint8_t'
... &c.

It is clear that I need -I$HOME/include/include to get passed to gcc,  
so first try to build q-opengl with "make CFLAGS=-I$HOME/local/ 
include", but that does not work, and end up having to edit the  
Makefile.  This is rather tedious, and feels unnecessary.

-andrew

[q-lang-users] =?iso-2022-jp?b?GyRCIighIUxARnwkXiRHGyhC?=

[<pan...@ya...>]

q-l...@li...様へ、招待期限になりましたのでご確認下さい。
至急、下記の招待カードからアクセスしていただき、
準備したおります専用スペースへ入室お願いします。

 >>>  招待カード  :::: 明日期限 ::::

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
【No.20024】瑞穂様から
┌──┐
│＼／│　　ご招待カードを開封する
　　　　　http://www.now-on-site.com/acfree/20024

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

ずべて無料の自由恋愛コミュニティは、健全セフレの募集板です。

[q-lang-users] =?iso-2022-jp?b?GyRCIighIUxARnwkXiRHGyhC?=

[<pan...@ya...>]

q-l...@li...様へ、招待期限になりましたのでご確認下さい。
至急、下記の招待カードからアクセスしていただき、
準備したおります専用スペースへ入室お願いします。

 >>>  招待カード  :::: 明日期限 ::::

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
【No.20024】瑞穂様から
┌──┐
│＼／│　　ご招待カードを開封する
　　　　　http://www.now-on-site.com/acfree/20024

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

ずべて無料の自由恋愛コミュニティは、健全セフレの募集板です。

Re: [q-lang-users] [qlist] SDL anyone?

[Andrew B. <and...@ya...>]

That sounds like fun.  Probably a good warm-up exercise for a wrapper  
that I'm actually a bit more interested in, with the GOODS database.   
(see: http://www.ispras.ru/~knizhnik/goods.html)  I suspect that  
doing a decent integration with an OO database would be a bit more  
difficult than SDL, so I should start out easy.

Drat, the sdl site seems to be down.  Or else my network here is  
messed up.

Anyhow, where is the best place to look for examples?  I guess I  
should start by getting q-opengl to work, then look at how that  
works, eh?

-andrew

On 20 Dec 2006, at 01:29, Albert Graef wrote:

> Hi,
>
> does anyone here know SDL (http://www.libsdl.org) and could help  
> develop
> an SDL wrapper? With Q-SWIG that shouldn't be a big deal. I think we
> need a portable layer for basic multimedia and game programming stuff
> and SDL seems to be available on almost any device that has a graphics
> display. Most of the graphics and multimedia stuff available for Q  
> right
> now requires a system running X11 or Windows and so can't be ported  
> to,
> e.g., embedded systems with custom GUI toolkits like QtEmbedded.
>
> So, is there anyone here interested in game programming who might wish
> to help with this?

[q-lang-users] Download page on Opera

[Andrew B. <and...@ya...>]

I don't know if you care about supporting "alternative" web browsers,  
but when I look at it with the latest version of Opera, it is pretty  
funked up.  I can send a little screen grab if it helps, but Opera is  
a free download, and it might be easier to test it yourself.  Oh,  
I've also seen this on both Windows and the Mac.

Briefly looking at the source, then at the w3 standards, it looks  
like multiple "thead" elements are non-standard.  (See: http:// 
www.w3.org/TR/html401/struct/tables.html#h-11.2.3 "Tables rows may be  
grouped into a table head, table foot, and one or more table  
bodies.")  Looking again at the rendered HTML, it is grouping all of  
the "thead" elements before all of the "tbody" elements, which I  
guess is reasonable, given the definition of "thead".

Suggested fix:  Just change all instances of "thead" into "tbody"  
entries, since you are not really putting heading information in  
there anyhow.

-andrew

[q-lang-users] Q 7.6 released

[Albert G. <Dr....@t-...>]

Hi everybody,

(cue the music) jingle bells, jingle bells ... :)

Just in time for christmas there's a new bugfix release of the Q core
package. This release fixes a number of issues with temporary files
being created in the wrong directory on Windows, segfaults in custom
unparsing and bugs in the shebang processing. Thanks to Andrew Berg the
sources now also compile cleanly on Mac OSX Tiger. Moreover, the
OpenZaurus feed has finally been updated. There are no new features this
time, but some are in the queue for the 7.7 release; I hope that I can
start working on this after Christmas.

As usual, more information and the downloads can be found at
http://q-lang.sourceforge.net/download.html. The OpenZaurus feed is at
http://q-lang.sourceforge.net/feed/OpenZaurus. The README file there
gives some information on how to get started with Q on the Zaurus.

Taking a look at the news items on http://q-lang.sf.net, I realized that
Q has come quite a long way during this year. Unicode support, improved
special forms, Rob Hubbard's rational and polynomial libraries and the
new numeric tower, John Cowan's Chicken interface, the Pd interface, Q
Yacc and Lex... I have to thank you all for the discussions, bug
reports, patches and contributions. And I wish you nice holidays, a
peaceful christmas and a happy new year!

Cheers,
Albert

-- 
Dr. Albert Gr"af
Dept. of Music-Informatics, University of Mainz, Germany
Email:  Dr....@t-..., ag...@mu...
WWW:    http://www.musikinformatik.uni-mainz.de/ag

[q-lang-users] SDL anyone?

[Albert G. <Dr....@t-...>]

Hi,

does anyone here know SDL (http://www.libsdl.org) and could help develop
an SDL wrapper? With Q-SWIG that shouldn't be a big deal. I think we
need a portable layer for basic multimedia and game programming stuff
and SDL seems to be available on almost any device that has a graphics
display. Most of the graphics and multimedia stuff available for Q right
now requires a system running X11 or Windows and so can't be ported to,
e.g., embedded systems with custom GUI toolkits like QtEmbedded.

So, is there anyone here interested in game programming who might wish
to help with this?

Cheers,
Albert

-- 
Dr. Albert Gr"af
Dept. of Music-Informatics, University of Mainz, Germany
Email:  Dr....@t-..., ag...@mu...
WWW:    http://www.musikinformatik.uni-mainz.de/ag

Re: [q-lang-users] [qlist] Re: Suggestions on how to debug?

[Albert G. <Dr....@t-...>]

Andrew Berg wrote:
> I would like to see it fixed in the next version.

It's already on my TODO list for 7.7.

> Also, would you be receptive to the idea of
> switching to PCRE for your regular expression library?

That sounds like a good idea. There hasn't been a new GNU regex version
since 1993. PCRE has Unicode support and also provides a set of POSIXish
wrapper functions so it shouldn't be a big deal to just replace one with
the other without breaking the interface. And, as John pointed out it's
pretty standard these days. It's on the 7.7 TODO list now.

While I'm at it, I'd also like to remove the bundled readline from the
core package. The bundled version is fairly old, pre-Unicode, and
readline seems to be readily available on just about any platform these
days anyway.

> I'll admit that my eyes glazed over a bit while reading the long lists
> of supplied functions in the documentation.

Yes, last time I counted there were already well over 700 builtin and
standard library functions, and that doesn't even include the extension
modules not in the prelude. "Batteries included" indeed. ;-)

Albert

-- 
Dr. Albert Gr"af
Dept. of Music-Informatics, University of Mainz, Germany
Email:  Dr....@t-..., ag...@mu...
WWW:    http://www.musikinformatik.uni-mainz.de/ag

Re: [q-lang-users] [qlist] Q 7.6RC1 available

[Albert G. <Dr....@t-...>]

Andrew Berg wrote:
> It's all good now.  Thanks for getting this in.

Great. The new release is now up at
http://sourceforge.net/project/showfiles.php?group_id=96881&package_id=103965&release_id=472606

I'll update the website and make the formal announcements later today.

Albert

-- 
Dr. Albert Gr"af
Dept. of Music-Informatics, University of Mainz, Germany
Email:  Dr....@t-..., ag...@mu...
WWW:    http://www.musikinformatik.uni-mainz.de/ag

[q-lang-users] PCRE (was: Suggestions on how to debug?)

[John C. <co...@cc...>]

Andrew Berg scripsit:

> Also, would you be receptive to the idea of switching to PCRE for your
> regular expression library?  I know that it is yet another dependancy,
> but it is maintained, and quite a bit more powerful than the regular
> GNU one.

I strongly urge this also.  Perl-compatible regular expressions are
pretty standard these days, and as for being a dependency, it's easy to
package PCRE (which has a BSD license) with Q -- you'd want to do that
under Windows anyhow.  There is good UTF-8 support, and both native and
Posix-compatible APIs are available.

See http://www.pcre.org for more details.

-- 
How they ever reached any  conclusion at all    <co...@cc...>
is starkly unknowable to the human mind.        http://www.ccil.org/~cowan
        --"Backstage Lensman", Randall Garrett

Re: [q-lang-users] [qlist] Re: Suggestions on how to debug?

[Andrew B. <and...@ya...>]

On 19 Dec 2006, at 15:02, Albert Graef wrote:

>> No, the current version is not. And yes, I guess you're right with  
>> your
>> suspicion that regex_next is the culprit here. It shouldn't be a big
>> deal to rewrite regex_next so that it's tail-recursive, though.
>

Yeah, I realized this in the car this morning on my way to work.  The  
function call hidden in the [] syntax threw me off.

> Ok, here is a tail-recursive implementation of regex. Just put this at
> the beginning of your script and fetch_history "AA" goes through  
> without
> a hitch (takes a while, though). (I might also use this definition in
> the next version of clib, but not right now as I haven't tested it
> thoroughly yet and I don't want to hold back the 7.6 release any  
> longer.)
>

I would like to see it fixed in the next version.  If it bit me once,  
it'll bite me again.  Also, would you be receptive to the idea of  
switching to PCRE for your regular expression library?  I know that  
it is yet another dependancy, but it is maintained, and quite a bit  
more powerful than the regular GNU one.

> Also note that using regex to just split a string at a given delimiter
> is really overkill. There's the split function which does this much
> faster (clib provides a C implementation of that function). To use  
> this,
> just s/regsplit "g"/split/ in your script and fetch_history "AA" runs
> much faster.

I'll admit that my eyes glazed over a bit while reading the long  
lists of supplied functions in the documentation.  I should go back  
and re-read that part once I get a bit more comfortable with the  
language itself.

-andrew

Re: [q-lang-users] [qlist] Q 7.6RC1 available

[Andrew B. <and...@ya...>]

On 19 Dec 2006, at 13:51, Albert Graef wrote:

>
> It would be nice if you could check once more whether this now  
> compiles
> ok. If it's ok then I can release 7.6.

It's all good now.  Thanks for getting this in.

-andrew