* Re: need editor
1999-09-10 0:00 ` David W. Glessner
@ 1999-09-10 0:00 ` David C. Hoos, Sr.
1999-09-14 0:00 ` ada-mode (was Re: need editor) David Kristola
1999-09-11 0:00 ` need editor Robert Dewar
1999-09-13 0:00 ` Emmanuel Briot
2 siblings, 1 reply; 23+ messages in thread
From: David C. Hoos, Sr. @ 1999-09-10 0:00 UTC (permalink / raw)
David W. Glessner <dwglessn@collins.rockwell.com> wrote in message
news:37D975A7.9BBA26DC@collins.rockwell.com...
> Robert Dewar wrote:
> >
> > In article <7r67hm$mi2@hobbes.crc.com>,
> > "David C. Hoos, Sr." <david.c.hoos.sr@ada95.com> wrote:
> > > But I find for editing, and general navigation, the ability of
> > > the emacs Ada-mode to format the code in a uniform style, to jump
> > > to declarations, to produce templates for Ada constructs, and
> > > in general to do things in an Ada-aware manner makes for much
> > > greater productivity.
> > >
> > > To be sure, to take advantage of these features does require
> > > some learning, but .. there's no free lunch.
> >
> > Note that although the Ada-mode of EMACS can be used with
> > any Ada compiler, the full power of this tools is only available
> > when used in an integrated development environment that includes
> > GNAT. For example, jumping to a declaration requires the cross-
> > reference information generated by GNAT.
>
> No, GNAT isn't required.
My original posting probably was not sufficiently couched in "legalese"
to make it clear to all that we were talking about the ada-mode that
is specifically tailored to work with the GNAT tools. Here's the
relevant sentence from that original post"
The latest Ada-mode for Emacs (platform-independent) is available
from http://ada.eu.org/ada-mode/
So, given that for context, Robert is absolutely correct -- to
take advantage of the features in the Ada-mode under discussion --
GNAT _is_ required.
> Rational's Apex provides an emacs/ada-mode combination with
> similar functionality. It looks like they modified emacs 19.30
> and an older ada-mode to work somewhat seemlessly with the rest
> of the Apex tools. There are a few quirks, but I find it much,
> much nicer than the standard Apex editor. I especially like
> the "Show Usage" function.
>
^ permalink raw reply [flat|nested] 23+ messages in thread
* ada-mode (was Re: need editor)
1999-09-10 0:00 ` David C. Hoos, Sr.
@ 1999-09-14 0:00 ` David Kristola
1999-09-14 0:00 ` Robert Dewar
` (2 more replies)
0 siblings, 3 replies; 23+ messages in thread
From: David Kristola @ 1999-09-14 0:00 UTC (permalink / raw)
In article 6ku@hobbes.crc.com, "David C. Hoos, Sr." <david.c.hoos.sr@ada95.com> () writes:
{snip}
>The latest Ada-mode for Emacs (platform-independent) is available
>from http://ada.eu.org/ada-mode/
>
>So, given that for context, Robert is absolutely correct -- to
>take advantage of the features in the Ada-mode under discussion --
>GNAT _is_ required.
>
>> Rational's Apex provides an emacs/ada-mode combination with
>> similar functionality. It looks like they modified emacs 19.30
>> and an older ada-mode to work somewhat seemlessly with the rest
>> of the Apex tools. There are a few quirks, but I find it much,
>> much nicer than the standard Apex editor. I especially like
>> the "Show Usage" function.
Would there be a way to make an ada-mode that could work in
either environment?
I have my own ada-mode which has evolved over the years and
periodically incorporated some of the features of the official
ada-mode. It uses the file names to figure out if it is working
with Verdix (*.a), Apex (*.[12].ada), or GNAT (*.ad[sb]) ada
files.
I really need to isolate a few features of my ada-mode and put
them in an enhancement package, then go with the official
version.
--djk, keeper of arcane lore & trivial fluff
Home: David95037 at aol dot com
Spam: goto.hades@welovespam.com
^ permalink raw reply [flat|nested] 23+ messages in thread* Re: ada-mode (was Re: need editor)
1999-09-14 0:00 ` ada-mode (was Re: need editor) David Kristola
@ 1999-09-14 0:00 ` Robert Dewar
[not found] ` <37ddf8d6.814751@news.geccs.gecm.com>
1999-09-14 0:00 ` Robert Dewar
2 siblings, 0 replies; 23+ messages in thread
From: Robert Dewar @ 1999-09-14 0:00 UTC (permalink / raw)
In article <7rk53d$iog1@svlss.lmms.lmco.com>,
dkristol@see-my.sig wrote:
> Would there be a way to make an ada-mode that could work in
> either environment?
Sure, there might be, but the GNU project has no particular
interest in making tools like this compatible with proprietary
software, and the official policy for EMACS and all similar
GNU tools is that they should basically consider only the needs
of other related free software components.
We discussed this with Richard Stallman, as part of his
appointment of Emmanuel Briot as the official maintainer
of GNU EMACS Ada mode, and he made the above policy quite
explicit, so this mode is definitely maintained primarily
with GNAT in mind (at such time as some other Ada compilers
become free software, the situation would change).
That does not mean we go out of our way to make it unusable
with other compilers, and indeed we try to keep as many features
as possible compiler neutral where there is no utility in doing
otherwise. But we do NOT go out of our way to accomodate other
proprietary compilers, and certainly trying to accomodate the
idiosyncrasies of the proprietary Rational Ada compiler would
fall in this category.
Of course, people are free to add any enhancements they want,
that's the whole point of free software. I am talking here
just about the policies as to what goes into the official
GNU versions from the FSF.
By the way, we are investing a lot of effort in improvements
in this environment, expect some interesting announcements
at the SIG Ada meeting :-)
Robert Dewar
>
> I have my own ada-mode which has evolved over the years and
> periodically incorporated some of the features of the official
> ada-mode. It uses the file names to figure out if it is
working
> with Verdix (*.a), Apex (*.[12].ada), or GNAT (*.ad[sb]) ada
> files.
>
> I really need to isolate a few features of my ada-mode and put
> them in an enhancement package, then go with the official
> version.
>
> --djk, keeper of arcane lore & trivial fluff
> Home: David95037 at aol dot com
> Spam: goto.hades@welovespam.com
>
>
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
^ permalink raw reply [flat|nested] 23+ messages in thread[parent not found: <37ddf8d6.814751@news.geccs.gecm.com>]
* Re: ada-mode (was Re: need editor)
[not found] ` <37ddf8d6.814751@news.geccs.gecm.com>
@ 1999-09-14 0:00 ` John McCabe
0 siblings, 0 replies; 23+ messages in thread
From: John McCabe @ 1999-09-14 0:00 UTC (permalink / raw)
abuse@borpin.demon.co.uk (Brian Orpin) wrote:
>There was some discussion about this either here or on the ada-mode list.
>There was a strong feeling that there should be an ada-mode and a
>gnat-mode with functions available only to users of Gnat in a separate
>package. IIRC Emmanuel is trying to move in this direction. We use the
>Tartan compiler and I have made some local modifications to ada-mode and
>created a tartan-mode. We currently do not have Gnat installed and
>available for use.
There was actually quite a lot of discussion on this at one point. I
was very keen for ada-mode to be compiler independent where possible
and was initially not very happy for maintenance to be taken over by
ACT. I was particularly disturbed about the amount of GNAT-specific
code that had gone into their first (?) attempt, and also that it was
based on what was then an 'old' version of ada-mode compared to what
those on the ada-mode mailing list had access to at the time.
During the discussions it was pointed out that as GNAT is a GNU
product, it seemed perfectly reasonable for emacs' ada-mode to depend
on it, but it was pretty much agreed between most of the people who
got into the discussion that the purely Ada based features of ada-mode
should be kept separate from anything that was compiler dependent and
Emmanuel has done a lot of work reorganising the files that comprise
ada-mode into something that fits the bill.
ada-xref and so on use features of GNAT, however most of these
features will be available in other compilers and it is really a
matter for either the users of those compilers, or the compiler
vendors themselves to implement this functionality for their products.
I would suggest that if you do use a compiler other than GNAT that
either:
a) Doesn't come with an editor
b) Has a really naff editor (e.g. ObjectAda)
then contact the vendor and suggest that they put some effort into
replicating the ACT developed parts of ada-mode for their own
products. Emmanuel has already done a lot of the work providing the
structure, all they need to do is mess around with it to provide the
interface to their stuff.
I know for example that Stephen Leake has developed some Emacs Lisp to
interface to the ObjectAda Command Line Debugger. There's nothing
stopping Aonix from doing that themselves.
IMO, I can't see why any compiler vendor produces an editor when Emacs
is available!
>IMHO it is wrong for ACT to hijack ada-mode and make it dependent on
>gnat. A gnat-mode would be more in keeping with the spirit of Emacs.
c.g. cc-mode - aren't the default settings based in GCC?
Best Regards
John McCabe <john@assen.demon.co.uk>
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: ada-mode (was Re: need editor)
1999-09-14 0:00 ` ada-mode (was Re: need editor) David Kristola
1999-09-14 0:00 ` Robert Dewar
[not found] ` <37ddf8d6.814751@news.geccs.gecm.com>
@ 1999-09-14 0:00 ` Robert Dewar
1999-09-15 0:00 ` David Kristola
2 siblings, 1 reply; 23+ messages in thread
From: Robert Dewar @ 1999-09-14 0:00 UTC (permalink / raw)
In article <7rk53d$iog1@svlss.lmms.lmco.com>,
dkristol@see-my.sig wrote:
> I really need to isolate a few features of my ada-mode and put
> them in an enhancement package, then go with the official
> version.
If you have general improvements that you would like to
contribute to the official version, that would be most
welcome, please send them to Emmanuel Briot (briot@gnat.com).
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
^ permalink raw reply [flat|nested] 23+ messages in thread* Re: ada-mode (was Re: need editor)
1999-09-14 0:00 ` Robert Dewar
@ 1999-09-15 0:00 ` David Kristola
0 siblings, 0 replies; 23+ messages in thread
From: David Kristola @ 1999-09-15 0:00 UTC (permalink / raw)
In article 1@nnrp1.deja.com, Robert Dewar <robert_dewar@my-deja.com> () writes:
>In article <7rk53d$iog1@svlss.lmms.lmco.com>,
> dkristol@see-my.sig wrote:
>> I really need to isolate a few features of my ada-mode and put
>> them in an enhancement package, then go with the official
>> version.
>
>If you have general improvements that you would like to
>contribute to the official version, that would be most
>welcome, please send them to Emmanuel Briot (briot@gnat.com).
I plan on it. I just need to figure out which parts
are real improvements or additions. ada-mode.el has
changed so much over the years. Many of the things
i have added in my own version were added into the
official version. Generally, mine are more of a
hack to get around a certain problem. :-(
A little code cleanup and some time with the latest
version of ada-mode, then i'll see if i have anything
worth contributing.
--djk, keeper of arcane lore & trivial fluff
Home: David95037 at aol dot com
Spam: goto.hades@welovespam.com
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: need editor
1999-09-10 0:00 ` David W. Glessner
1999-09-10 0:00 ` David C. Hoos, Sr.
@ 1999-09-11 0:00 ` Robert Dewar
1999-09-13 0:00 ` Emmanuel Briot
1999-09-13 0:00 ` Emmanuel Briot
2 siblings, 1 reply; 23+ messages in thread
From: Robert Dewar @ 1999-09-11 0:00 UTC (permalink / raw)
In article <37D975A7.9BBA26DC@collins.rockwell.com>,
"David W. Glessner" <dwglessn@collins.rockwell.com> wrote:
> > Note that although the Ada-mode of EMACS can be used with
> > any Ada compiler, the full power of this tools is only
available
> > when used in an integrated development environment that
includes
> > GNAT. For example, jumping to a declaration requires the
cross-
> > reference information generated by GNAT.
>
> No, GNAT isn't required.
>
> Rational's Apex provides an emacs/ada-mode combination with
> similar functionality. It looks like they modified emacs
> 19.30
Just to be clear, when I talk about "the Ada mode of EMACS",
I am talking about the official FSF distribution as maintained
by Emmanuel Briot of ACT for FSF, based on the latest version
of EMACS. And this indeed can only provide this kind of
functionality for GNAT.
Most certainly, since EMACS is programmable, there may be
non-FSF versions that do all kinds of things, and indeed I
believe that Rational technology does make use of some of the
GNU technology (including I think RCS), but this EMACS
version are not "the Ada mode of EMACS".
However, there is still a LOT of useful functionality in
this mode that can be used with any Ada 95 compiler, it is
only semantic issues, like finding declarations, that need
the input from the compiler. Our philosophy in maintaing
the Ada mode is to accomodate other Ada compilers where
this can be done without too much extra effort, but
naturally the FSF project favors GNU software over
proprietary software, so specific compiler-dependent
work in the official EMACS release will always generally be
keyed to GNAT.
We make a significant effort to make sure that the Ada mode
keeps up to date with the very latest Emacs releases. We also
try to be as compatible as we can with XEmacs, and will likely
fully support XEmacs in the future.
Robert Dewar
Ada Core Technologies
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: need editor
1999-09-11 0:00 ` need editor Robert Dewar
@ 1999-09-13 0:00 ` Emmanuel Briot
1999-09-13 0:00 ` Fraser Wilson
0 siblings, 1 reply; 23+ messages in thread
From: Emmanuel Briot @ 1999-09-13 0:00 UTC (permalink / raw)
Robert Dewar <dewar@gnat.com> wrote:
> We make a significant effort to make sure that the Ada mode
> keeps up to date with the very latest Emacs releases. We also
> try to be as compatible as we can with XEmacs, and will likely
> fully support XEmacs in the future.
Note that I recently improved quite a lot the support for XEmacs (I have
finally be able to get it working on my linux laptop :-). The menus
are now correct, there is no more know indentation problem, ... This
will be in the next release of the ada-mode.
People who find Emacs not to be visual enough should give XEmacs a try.
The menus have been thought much better, and are probably more
accessible to the beginners than Emacs is.
There might be a few stability problems still, but XEmacs tries hard not
to lose anything even when it crashes, so this should not be a real
problem.
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: need editor
1999-09-13 0:00 ` Emmanuel Briot
@ 1999-09-13 0:00 ` Fraser Wilson
0 siblings, 0 replies; 23+ messages in thread
From: Fraser Wilson @ 1999-09-13 0:00 UTC (permalink / raw)
paene lacrimavi postquam Emmanuel Briot <briot@gnat.com> scripsit:
>Note that I recently improved quite a lot the support for XEmacs (I have
>finally be able to get it working on my linux laptop :-). The menus
>are now correct, there is no more know indentation problem, ... This
>will be in the next release of the ada-mode.
I've been using the Ada mode with XEmacs for some time now ... what
sort of indendation problems have you had? Well, apart from the one
that manifests after a generic package instantiation. And the project
stuff is a little dodgy -- I generally edit the .adp file directly,
then associate it, which works just fine.
>People who find Emacs not to be visual enough should give XEmacs a try.
Seconded. I use XEmacs, and it's very pretty.
Fraser.
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: need editor
1999-09-10 0:00 ` David W. Glessner
1999-09-10 0:00 ` David C. Hoos, Sr.
1999-09-11 0:00 ` need editor Robert Dewar
@ 1999-09-13 0:00 ` Emmanuel Briot
1999-09-13 0:00 ` David W. Glessner
2 siblings, 1 reply; 23+ messages in thread
From: Emmanuel Briot @ 1999-09-13 0:00 UTC (permalink / raw)
"David W. Glessner" <dwglessn@collins.rockwell.com> wrote:
> Rational's Apex provides an emacs/ada-mode combination with
> similar functionality. It looks like they modified emacs 19.30
> and an older ada-mode to work somewhat seemlessly with the rest
> of the Apex tools. There are a few quirks, but I find it much,
> much nicer than the standard Apex editor. I especially like
> the "Show Usage" function.
Could you explain a bit more what the "show usage" function is about ?
It would be interesting to see if we can put it in the official ada-mode
distribution, if it is not already there.
What the FSF ada-mode currently has is a "goto references" function,
to show where an identifier/variable/subprogram is referenced and used
all other your project. Is the "Show usage" function the same kind
of things, or does it show how the function should be used ??
Emmanuel
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: need editor
1999-09-13 0:00 ` Emmanuel Briot
@ 1999-09-13 0:00 ` David W. Glessner
1999-09-14 0:00 ` John McCabe
0 siblings, 1 reply; 23+ messages in thread
From: David W. Glessner @ 1999-09-13 0:00 UTC (permalink / raw)
Emmanuel Briot wrote:
>
> "David W. Glessner" <dwglessn@collins.rockwell.com> wrote:
>
> > Rational's Apex provides an emacs/ada-mode combination with
> > similar functionality. It looks like they modified emacs 19.30
> > and an older ada-mode to work somewhat seemlessly with the rest
> > of the Apex tools. There are a few quirks, but I find it much,
> > much nicer than the standard Apex editor. I especially like
> > the "Show Usage" function.
>
> Could you explain a bit more what the "show usage" function is about ?
> It would be interesting to see if we can put it in the official ada-mode
> distribution, if it is not already there.
> What the FSF ada-mode currently has is a "goto references" function,
> to show where an identifier/variable/subprogram is referenced and used
> all other your project. Is the "Show usage" function the same kind
> of things, or does it show how the function should be used ??
It sounds like Apex's "show usage" is similar to your ada-mode's
"goto references", assuming that an identifier comprises at least
type names, package names, record components, and enumeration
literals. It also handles renaming. For example, in the
following snippet, "show usage" for package xyzzy highlights
three items:
with Xyzzy; -- Xyzzy
package Aaa is
package X renames Xyzzy; -- Xyzzy
A : X.Some_Type; -- X
end Aaa;
Based on the ada-mode web page, I was under the impression that
ada-mode only allowed jumping to the declaration or body of an
entity, which is quite limited when compared with a full "goto
references". Perhaps the web page needs to be updated?
I'll give the latest ada-mode a try.
--
David
^ permalink raw reply [flat|nested] 23+ messages in thread* Re: need editor
1999-09-13 0:00 ` David W. Glessner
@ 1999-09-14 0:00 ` John McCabe
0 siblings, 0 replies; 23+ messages in thread
From: John McCabe @ 1999-09-14 0:00 UTC (permalink / raw)
"David W. Glessner" <dwglessn@collins.rockwell.com> wrote:
>Emmanuel Briot wrote:
>It sounds like Apex's "show usage" is similar to your ada-mode's
>"goto references", assuming that an identifier comprises at least
>type names, package names, record components, and enumeration
>literals. It also handles renaming. For example, in the
>following snippet, "show usage" for package xyzzy highlights
>three items:
This sounds like a prime example of why some functions of ada-mode
need to be targetted to a specific compiler. Can you imagine the
complexity that would be required to implement the semantic processing
of Ada projects to produce this information in Emacs Lisp? Why
re-invent the wheel.
Best Regards
John McCabe <john@assen.demon.co.uk>
^ permalink raw reply [flat|nested] 23+ messages in thread