At ILC 2003, I learned that some people are, but were having trouble
getting started. If you're curious about trying CMUCL's built-in
editor/IDE, try the instructions at
<http://www.cliki.net/GettingStartedWithHemlock>. And in particular,
if there are problems with them, let me know.
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
>>>>> "Thomas" == Thomas F Burdick <···@famine.OCF.Berkeley.EDU> writes:
Thomas> At ILC 2003, I learned that some people are, but were
Thomas> having trouble getting started. If you're curious about
Thomas> trying CMUCL's built-in editor/IDE, try the instructions
Thomas> at <http://www.cliki.net/GettingStartedWithHemlock>. And
Thomas> in particular, if there are problems with them, let me
Thomas> know.
Would you happen to know if it currently works for people in a Debian
(unstable) setup? I get the following, although of course it could be
the "unstable" part. In any case, there is no longer a cmucl-hemlock
package.
Thanks,
Tim
* (ed)
; In: LAMBDA (&REST REST)
; (MULTIPLE-VALUE-BIND (RETURN ERRORP) (IGNORE-ERRORS #) (IF ERRORP # #))
; --> MULTIPLE-VALUE-CALL
; ==>
; #'(LAMBDA (&OPTIONAL RETURN ERRORP &REST #:G0) (DECLARE #) (IF ERRORP # #))
; Note: Variable RETURN defined but never used.
; Loading #p"/usr/share/common-lisp/repositories/cmucl/src/hemlock/cmucl-hemlock.asd".
; Loading #p"/usr/share/common-lisp/repositories/cmucl/src/hemlock/cmucl-hemlock-dict.asd".
; Loading #p"/usr/share/common-lisp/repositories/cmucl/src/hemlock/cmucl-hemlock-base.asd".
; Loading #p"/usr/share/common-lisp/repositories/cmucl/src/clx/cmucl-clx.asd".
; Loading #p"/usr/share/common-lisp/repositories/cmucl/src/pcl/cmucl-graystream.asd".
Bignoings.2..
;;; Please wait, recompiling library...
;;; Please wait, recompiling library...
;;; Please wait, recompiling library...
;;; Please wait, recompiling library...
;;; Please wait, recompiling library...
Error in function ED:
Sorry, cannot find hemlock, please install and use ilisp. Reason: #<SIMPLE-ERROR
{4878B59D}>
Restarts:
0: [ABORT] Return to Top-Level.
Debug (type H for help)
(ED)[:OPTIONAL]
Source: (ERROR
"Sorry, cannot find hemlock, please install and use ilisp. Reason: ~S"
ERRORP)
0] 0
Tim Lavoie <········@spamcop.net> writes:
> >>>>> "Thomas" == Thomas F Burdick <···@famine.OCF.Berkeley.EDU> writes:
>
> Thomas> At ILC 2003, I learned that some people are, but were
> Thomas> having trouble getting started. If you're curious about
> Thomas> trying CMUCL's built-in editor/IDE, try the instructions
> Thomas> at <http://www.cliki.net/GettingStartedWithHemlock>. And
> Thomas> in particular, if there are problems with them, let me
> Thomas> know.
>
> Would you happen to know if it currently works for people in a Debian
> (unstable) setup?
"Yes, but" -- I assume that you installed the standard distribution,
available from cons.org. On my Debian/x86, I use the standard CMUCL
distribution, not the Debian packages; you can certainly install both,
though.
> I get the following, although of course it could be the "unstable"
> part. In any case, there is no longer a cmucl-hemlock package.
That sucks -- I'd just use the cons.org tarballs, or complain to the
package maintainer. I'd bet he removed Hemlock because he thought no
one was using it; if people want it, he might bring it back.
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
>>>>> "Thomas" == Thomas F Burdick <···@famine.OCF.Berkeley.EDU> writes:
>> Would you happen to know if it currently works for people in a
>> Debian (unstable) setup?
Thomas> "Yes, but" -- I assume that you installed the standard
Thomas> distribution, available from cons.org. On my Debian/x86,
Thomas> I use the standard CMUCL distribution, not the Debian
Thomas> packages; you can certainly install both, though.
>> I get the following, although of course it could be the
>> "unstable" part. In any case, there is no longer a
>> cmucl-hemlock package.
Thomas> That sucks -- I'd just use the cons.org tarballs, or
Thomas> complain to the package maintainer. I'd bet he removed
Thomas> Hemlock because he thought no one was using it; if people
Thomas> want it, he might bring it back.
Actually, I think it's all in there, but no longer installed as a
separate package. The process complains when compiling the
cmucl-graystream stuff, which I assume Hemlock needs to run. Typing
(ed) gets all that going, it just doesn't quite finish.
In any case, I'll probably just install from the tarball to muck
around with Hemlock plus any other bits newer than 18e.
Thanks,
Tim
> Actually, I think it's all in there, but no longer installed as a
> separate package. The process complains when compiling the
> cmucl-graystream stuff, which I assume Hemlock needs to run. Typing
> (ed) gets all that going, it just doesn't quite finish.
Weird. Hemlock doesn't need CLOS to run. It doesn't even need the
compiler. Well, it needs some functions that are in the compiler code
to run slave lisps, but not to function as an editor. I've built
Hemlock runtimes without CLOS or the compiler.
If you create executables and use gzexe on them you can get amusingly
small binaries:
-rwxrwxr-x 1 root wheel 3909438 Aug 22 17:01 /usr/local/lib/cmucl/freebsd/lisp/hemlock
However, I stopped trying to force myself to use Hemlock because of
Slime. It's better integrated than Ilisp, lets me use keybindings I'm
familiar with, and, to be honest, I've just gotten used to coloration.
The code looks dead without it. :-)
--
Fred Gilham ······@csl.sri.com
...And then it got late, and I had to stop researching. But it appears
that every slab of concrete in West Virginia is named after Bob Byrd.
-- Ann Coulter
Fred Gilham <······@snapdragon.csl.sri.com> writes:
> However, I stopped trying to force myself to use Hemlock because of
> Slime. It's better integrated than Ilisp, lets me use keybindings I'm
> familiar with, and, to be honest, I've just gotten used to coloration.
> The code looks dead without it. :-)
The keybinding thing I've always found weird -- they're so similar
between Hemlock and Emacs! Then again, I mess with my Emacs
keybindings like crazy, so maybe other people mind rebinding things
more than I do. As for the coloration -- yuck -- after using mostly
Hemlock then recently MCL, font-locked Lisp looks like Las Vegas. I
prefer the calmer prettiness of Paris. :-)
(Not to slight Slime, it looks great. Particularly with font-lock
mode turned off.)
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
Thomas F. Burdick wrote:
> Fred Gilham <······@snapdragon.csl.sri.com> writes:
>
>
>>However, I stopped trying to force myself to use Hemlock because of
>>Slime. It's better integrated than Ilisp, lets me use keybindings I'm
>>familiar with, and, to be honest, I've just gotten used to coloration.
>>The code looks dead without it. :-)
>
>
> The keybinding thing I've always found weird -- they're so similar
> between Hemlock and Emacs! Then again, I mess with my Emacs
> keybindings like crazy, so maybe other people mind rebinding things
> more than I do. As for the coloration -- yuck -- after using mostly
> Hemlock then recently MCL, font-locked Lisp looks like Las Vegas. I
> prefer the calmer prettiness of Paris. :-)
>
> (Not to slight Slime, it looks great. Particularly with font-lock
> mode turned off.)
I forget who said it first, but I used to agree with the assessment that
syntax coloration looked like 'angry fruit salad'. However, I had to
work for a while in an environment that did it, and I got used to it.
Now that I'm used to it I much prefer it; it makes code navigation much
faster for me.
mikel <·····@evins.net> writes:
> I forget who said it first, but I used to agree with the assessment that
> syntax coloration looked like 'angry fruit salad'. However, I had to
> work for a while in an environment that did it, and I got used to it.
> Now that I'm used to it I much prefer it; it makes code navigation much
> faster for me.
Are we still talking Lisp here? In C, C++, Perl, etc., I need syntax
coloring. Which is why I used it at first in Lisp. Nowadays, I still
use it for most languages, but for Lisp, I find myself more productive
without it -- not a lot more, but I used to spend some amount of time
trying to figure out why Emacs' broken regexp had colored something
the way it did. Not no mo
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
···@famine.OCF.Berkeley.EDU (Thomas F. Burdick) writes:
> Are we still talking Lisp here? In C, C++, Perl, etc., I need syntax
> coloring. Which is why I used it at first in Lisp. Nowadays, I still
> use it for most languages, but for Lisp, I find myself more productive
> without it -- not a lot more, but I used to spend some amount of time
> trying to figure out why Emacs' broken regexp had colored something
> the way it did. Not no mo
There's degrees of colour. I like having string literals coloured,
and I like having comments coloured (especially if #| |# comments are
likewise coloured, as I often use them for commenting out large chunks
of code). I can live without having (error in bold red, though.
And if you think emacs is bad, Red Hat's vi is worse yet -
http://ww.telent.net/angry-fruit-salad.png
-dan
--
http://web.metacircles.com/cirCLe_CD - Free Software Lisp/Linux distro
Daniel Barlow wrote:
> ···@famine.OCF.Berkeley.EDU (Thomas F. Burdick) writes:
>
>
>>Are we still talking Lisp here? In C, C++, Perl, etc., I need syntax
>>coloring. Which is why I used it at first in Lisp. Nowadays, I still
>>use it for most languages, but for Lisp, I find myself more productive
>>without it -- not a lot more, but I used to spend some amount of time
>>trying to figure out why Emacs' broken regexp had colored something
>>the way it did. Not no mo
>
>
> There's degrees of colour. I like having string literals coloured,
> and I like having comments coloured (especially if #| |# comments are
> likewise coloured, as I often use them for commenting out large chunks
> of code).
That would be enough for me, though I also kind of like having the name
in a defining form colored.
I can live without having (error in bold red, though.
>
> And if you think emacs is bad, Red Hat's vi is worse yet -
>
> http://ww.telent.net/angry-fruit-salad.png
>
>
> -dan
>
Daniel Barlow <···@telent.net> writes:
> ···@famine.OCF.Berkeley.EDU (Thomas F. Burdick) writes:
>
> > Are we still talking Lisp here? In C, C++, Perl, etc., I need syntax
> > coloring. Which is why I used it at first in Lisp. Nowadays, I still
> > use it for most languages, but for Lisp, I find myself more productive
> > without it -- not a lot more, but I used to spend some amount of time
> > trying to figure out why Emacs' broken regexp had colored something
> > the way it did. Not no mo
>
> There's degrees of colour. I like having string literals coloured,
> and I like having comments coloured (especially if #| |# comments are
> likewise coloured, as I often use them for commenting out large chunks
> of code). I can live without having (error in bold red, though.
>
> And if you think emacs is bad, Red Hat's vi is worse yet -
>
> http://ww.telent.net/angry-fruit-salad.png
If you're using vim, remember that ":syntax off" is your friend. I've
never seen an editor so intent on making shell scripts hard to read
and hard to edit. I so wish that nvi was the default vi everywhere. :)
//Ingvar
--
Q: What do you call a Discworld admin?
A: Chelonius Monk
Thomas F. Burdick wrote:
> mikel <·····@evins.net> writes:
>
>
>>I forget who said it first, but I used to agree with the assessment that
>>syntax coloration looked like 'angry fruit salad'. However, I had to
>>work for a while in an environment that did it, and I got used to it.
>>Now that I'm used to it I much prefer it; it makes code navigation much
>>faster for me.
>
>
> Are we still talking Lisp here?
Yes.
> In C, C++, Perl, etc., I need syntax
> coloring. Which is why I used it at first in Lisp. Nowadays, I still
> use it for most languages, but for Lisp, I find myself more productive
> without it -- not a lot more, but I used to spend some amount of time
> trying to figure out why Emacs' broken regexp had colored something
> the way it did. Not no mo
Okay, but I prefer syntax-colored lisp because I can navigate gross
blocks of code much more quickly with it than without it. I notice the
advantage most often when I lose it, as when I edit a file of lisp code
in an editor that lacks syntax coloring, or when (as occasionally
happens) the Emacs syntax coloring breaks in a file I'm editing and
formerly-different swaths of code are recolored all the same.
mikel <·····@evins.net> writes:
> I forget who said it first, but I used to agree with the assessment
> that syntax coloration looked like 'angry fruit salad'. However, I had
> to work for a while in an environment that did it, and I got used to
> it. Now that I'm used to it I much prefer it; it makes code navigation
> much faster for me.
I have my Emacs set to rather *extreme* syntax coloring. It really does
seem to make navigation faster.
I also have the parens rendered in a slightly subdued color (but the
*matching* paren blinks brightly, and a *mismatching* paren is bold
RED)
···@famine.OCF.Berkeley.EDU (Thomas F. Burdick) writes:
> As for the coloration -- yuck -- after using mostly Hemlock then
> recently MCL, font-locked Lisp looks like Las Vegas.
Gee wiz, I can't even read printed listings without syntax
colouring. Let me know when someone ports `ps-print-buffer-with-faces'
to Hemlock. :-)
-Luke
···@famine.OCF.Berkeley.EDU (Thomas F. Burdick) writes:
[ no cmucl-hemlock pacage ]
> That sucks -- I'd just use the cons.org tarballs, or complain to the
> package maintainer. I'd bet he removed Hemlock because he thought no
> one was using it; if people want it, he might bring it back.
I think it's actually found in the cmucl-source package (which,
perhaps confusingly, is a binary package) these days. I'm not sure
what the reason for the move is.
(and I too run the cons.org tarballs instead - albeit mostly only for
testing and for compiling SBCL)
-dan
--
http://web.metacircles.com/cirCLe_CD - Free Software Lisp/Linux distro
On 28 Oct 2003, Tim Lavoie wrote:
>
> Would you happen to know if it currently works for people in a
> Debian (unstable) setup? I get the following, although of course it
> could be the "unstable" part. In any case, there is no longer a
> cmucl-hemlock package.
It's all in the cmucl-source package at this point.
I have installed the whole system from the Debian (unstable) packages
and it works quite nicely.
Dan
>>>>> "Dan" == Dan Katz <·····@profitlogic.com> writes:
Dan> It's all in the cmucl-source package at this point.
Dan> I have installed the whole system from the Debian (unstable)
Dan> packages and it works quite nicely.
OK, maybe something is just slightly messed up in my particular
setup. That is good news, in a perverse sort of way. :)
What you need to do under Debian (unstable) is to fire up `lisp' and
then evaluate:
(require :cmucl-hemlock)
Depending on how common-lisp-controller and/or cmucl is configured, it
may have to compile hemlock, so be patient, but when the prompt
returns, you may continue with:
(ed)
------------------------+-----------------------------------------------------
Christian Lynbech | christian ··@ defun #\. dk
------------------------+-----------------------------------------------------
Hit the philistines three times over the head with the Elisp reference manual.
- ·······@hal.com (Michael A. Petonic)
>>>>> "Christian" == Christian Lynbech <·················@ericsson.com> writes:
Christian> What you need to do under Debian (unstable) is to fire
Christian> up `lisp' and then evaluate:
Christian> (require :cmucl-hemlock)
Christian> Depending on how common-lisp-controller and/or cmucl is
Christian> configured, it may have to compile hemlock, so be
Christian> patient, but when the prompt returns, you may continue
Christian> with:
Christian> (ed)
For me, it starts the same process as (ed), but still chokes on
compiling cmucl-graystream. That may just be my setup though, so I'll
need to play with that.
Tim Lavoie <········@spamcop.net> writes:
> >>>>> "Christian" == Christian Lynbech <·················@ericsson.com> writes:
>
> Christian> What you need to do under Debian (unstable) is to fire
> Christian> up `lisp' and then evaluate:
>
> Christian> (require :cmucl-hemlock)
>
> Christian> Depending on how common-lisp-controller and/or cmucl is
> Christian> configured, it may have to compile hemlock, so be
> Christian> patient, but when the prompt returns, you may continue
> Christian> with:
>
> Christian> (ed)
>
> For me, it starts the same process as (ed), but still chokes on
> compiling cmucl-graystream. That may just be my setup though, so I'll
> need to play with that.
Yeah, that's a bad sign; Gray stream support should work in CMUCL,
*and* Hemlock doesn't use Gray streams anyway.
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
>>>>> "Thomas" == Thomas F Burdick <···@famine.OCF.Berkeley.EDU> writes:
Thomas> Yeah, that's a bad sign; Gray stream support should work in CMUCL,
It does.
Thomas> *and* Hemlock doesn't use Gray streams anyway.
Perhaps not, but doesn't CLX use graystreams? And hemlock does use
CLX, I think.
------------------------+-----------------------------------------------------
Christian Lynbech | christian ··@ defun #\. dk
------------------------+-----------------------------------------------------
Hit the philistines three times over the head with the Elisp reference manual.
- ·······@hal.com (Michael A. Petonic)
Christian Lynbech <·················@ericsson.com> writes:
> Perhaps not, but doesn't CLX use graystreams? And hemlock does use
> CLX, I think.
No and yes.
CLX needs a stream connected to a unix or tcp socket. There may be a
Lisp implementation out there that does this with Gray streams, but
CMUCL is not it.
-dan
--
http://web.metacircles.com/cirCLe_CD - Free Software Lisp/Linux distro
>>>>> "Daniel" == Daniel Barlow <···@telent.net> writes:
Daniel> Christian Lynbech <·················@ericsson.com> writes:
Daniel> CLX needs a stream connected to a unix or tcp socket. There may be a
Daniel> Lisp implementation out there that does this with Gray streams, but
Daniel> CMUCL is not it.
If CLX does not need graystreams, then why does the Debian version of
the cmucl-clx.asd add graystreams as a prerequisite, me wonders?
In /usr/share/common-lisp/systems/cmucl-clx.asd, I see
(defsystem :cmucl-clx
:depends-on (:cmucl-graystream)
:components
...)
This is coming from cmucl-source version 18e-7.
------------------------+-----------------------------------------------------
Christian Lynbech | christian ··@ defun #\. dk
------------------------+-----------------------------------------------------
Hit the philistines three times over the head with the Elisp reference manual.
- ·······@hal.com (Michael A. Petonic)
>>>>> "Christian" == Christian Lynbech <·················@ericsson.com> writes:
Christian> If CLX does not need graystreams, then why does the Debian
Christian> version of the cmucl-clx.asd add graystreams as a
Christian> prerequisite, me wonders?
I did an experiment of copying the clx sources from
/usr/share/common-lisp/source/cmucl-clx and copying thas asd file to
remove the :depends-on option, and started the compilation of CLX,
without graystreams.
The compilation went smooth and there does not seem to be any reports
of missing or unknown functions.
So it would look as if the graystream requirement in the Debian
packaging is indeed unnecessary (not that I have tried to run any code
with graystream-less lisp).
------------------------+-----------------------------------------------------
Christian Lynbech | christian ··@ defun #\. dk
------------------------+-----------------------------------------------------
Hit the philistines three times over the head with the Elisp reference manual.
- ·······@hal.com (Michael A. Petonic)
I tried Hemlock as the web page said. But I just can't get CMU CL
connected.
I'm used to Emacs.
What's the difference between Hemlock and Emacs?
And I can't connect to the ilisp site.
--
Yin Wang,
EDA Lab,
Deparment of Computer Science and Technology,
Tsinghua University,
100084
Beijing China
Wang Yin <··@wangyin.com> writes:
> I tried Hemlock as the web page said. But I just can't get CMU CL
> connected.
What do you mean "connected"? Do you mean, when you run (ed), it
won't connect to your X11 display? Be sure you apply the CLX patch
mentioned on the GettingStarted page. If you're still having
problems, post or email me a complete transcript of the session, and
I'll see if I can figure it out.
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
Wang Yin wrote:
> I tried Hemlock as the web page said. But I just can't get CMU CL
> connected.
>
> I'm used to Emacs.
> What's the difference between Hemlock and Emacs?
> And I can't connect to the ilisp site.
Which ILISP site? There may be a broken link somewhere. Try
http://ilisp.sf.net
Cheers
--
Marco
Marco Antoniotti wrote:
>
>
> Wang Yin wrote:
>
>> I tried Hemlock as the web page said. But I just can't get CMU CL
>> connected.
>> I'm used to Emacs.
>> What's the difference between Hemlock and Emacs?
>> And I can't connect to the ilisp site.
>
>
> Which ILISP site? There may be a broken link somewhere. Try
> http://ilisp.sf.net
I stand corrected. Try
http://sourceforge.net/projects/ilisp/
Cheers
--
Marco