Re: option description - always give default value

2002-07-31 Thread Sven Guckes

* Vincent Lefevre <[EMAIL PROTECTED]> [2002-07-31 15:14]:
> If the default value given by the manual is not the same as
> the real default value used by Mutt, it is very confusing.

errr... that's self-evident, isn't it?

> > the default value should *always* be documented.
> What do you mean by "documented"?

the manual to "muttrc" should show it.

somehow i think i've been pointing out the bleeding obvious.
or is there something i'm missing here?

are there any options without a default?  (i hope not)

are there any options with different defaults
according to configure options being used?
if so then this should be documented, too.

Sven



Re: option description - always give default value

2002-07-31 Thread Vincent Lefevre

On Wed, Jul 31, 2002 at 18:26:40 +0200, Sven Guckes wrote:
> * Vincent Lefevre <[EMAIL PROTECTED]> [2002-07-31 15:14]:
> > > the default value should *always* be documented.
> > What do you mean by "documented"?
> 
> the manual to "muttrc" should show it.

OK, so what happens if two Mutt binaries use different default values?

> are there any options with different defaults
> according to configure options being used?

The default value of dotlock_program depends on the configuration
options.

> if so then this should be documented, too.

Yes, and my point was: for these options, the default value shouldn't
be written in the manual, because it may be incorrect. The user should
get the actual value with :set xxx=[TAB]. And of course, this should
be documented. Moreover, if mutt -v could give a hint, this would be
useful.

-- 
Vincent Lefèvre <[EMAIL PROTECTED]> - Web:  - 100%
validated (X)HTML - Acorn Risc PC, Yellow Pig 17, Championnat International
des Jeux Mathématiques et Logiques, TETRHEX, etc.
Work: CR INRIA - computer arithmetic / SPACES project at LORIA



Re: option description - always give default value

2002-08-01 Thread Cameron Simpson

On 19:15 31 Jul 2002, Vincent Lefevre <[EMAIL PROTECTED]> wrote:
| On Wed, Jul 31, 2002 at 18:26:40 +0200, Sven Guckes wrote:
| > * Vincent Lefevre <[EMAIL PROTECTED]> [2002-07-31 15:14]:
| > > > the default value should *always* be documented.
| > > What do you mean by "documented"?
| > the manual to "muttrc" should show it.
| OK, so what happens if two Mutt binaries use different default values?

The installed manual should be preprocessed during the build to have the
correct defaults.
-- 
Cameron Simpson, DoD#743[EMAIL PROTECTED]http://www.zip.com.au/~cs/

Life is like arriving late for a movie, having to figure out what was going
on without bothering everybody with a lot of questions, and then being
unexpectedly called away before you find out how it ends.



Re: option description - always give default value

2002-08-02 Thread Vincent Lefevre

On Fri, Aug 02, 2002 at 16:39:57 +1000, Cameron Simpson wrote:
> On 19:15 31 Jul 2002, Vincent Lefevre <[EMAIL PROTECTED]> wrote:
> | On Wed, Jul 31, 2002 at 18:26:40 +0200, Sven Guckes wrote:
> | > * Vincent Lefevre <[EMAIL PROTECTED]> [2002-07-31 15:14]:
> | > > > the default value should *always* be documented.
> | > > What do you mean by "documented"?
> | > the manual to "muttrc" should show it.
> | OK, so what happens if two Mutt binaries use different default values?
> 
> The installed manual should be preprocessed during the build to have the
> correct defaults.

But how can it have the correct defaults since there is only one
manual for several binaries?

-- 
Vincent Lefèvre <[EMAIL PROTECTED]> - Web:  - 100%
validated (X)HTML - Acorn Risc PC, Yellow Pig 17, Championnat International
des Jeux Mathématiques et Logiques, TETRHEX, etc.
Work: CR INRIA - computer arithmetic / SPACES project at LORIA



Re: option description - always give default value

2002-08-02 Thread Chris Palmer

Vincent Lefevre writes:

> > The installed manual should be preprocessed during the build to have
> > the correct defaults.
> 
> But how can it have the correct defaults since there is only one
> manual for several binaries?

What several binaries? My system has only one /usr/local/bin/mutt.

Cameron means that, at compile/configure time, the source of the man
page would be modified programmatically to agree with the ./configure
options.

(Which is, btw, the best solution.)


-- 
Chris music is what numbers feel like San Francisco, CA




Re: option description - always give default value

2002-08-02 Thread Vincent Lefevre

On Fri, Aug 02, 2002 at 09:14:27 -0700, Chris Palmer wrote:
> What several binaries? My system has only one /usr/local/bin/mutt.

There are systems with multiple binaries. Please read the thread.

-- 
Vincent Lefèvre <[EMAIL PROTECTED]> - Web:  - 100%
validated (X)HTML - Acorn Risc PC, Yellow Pig 17, Championnat International
des Jeux Mathématiques et Logiques, TETRHEX, etc.
Work: CR INRIA - computer arithmetic / SPACES project at LORIA



Re: option description - always give default value

2002-08-02 Thread Chris Palmer

Vincent Lefevre writes:

> There are systems with multiple binaries.

It's up to the sysadmin to keep the man pages in the same directory
prefix as the binaries. /usr/foo/man/man1 should correspond to
/usr/foo/bin, et c.


-- 
Chris music is what numbers feel like San Francisco, CA




Re: option description - always give default value

2002-08-02 Thread Vincent Lefevre

On Fri, Aug 02, 2002 at 10:01:18 -0700, Chris Palmer wrote:
> Vincent Lefevre writes:
> 
> > There are systems with multiple binaries.
> 
> It's up to the sysadmin to keep the man pages in the same directory
> prefix as the binaries. /usr/foo/man/man1 should correspond to
> /usr/foo/bin, et c.

We agreed that there should be only one manual. Again, read the thread.

-- 
Vincent Lefèvre <[EMAIL PROTECTED]> - Web:  - 100%
validated (X)HTML - Acorn Risc PC, Yellow Pig 17, Championnat International
des Jeux Mathématiques et Logiques, TETRHEX, etc.
Work: CR INRIA - computer arithmetic / SPACES project at LORIA



Re: option description - always give default value

2002-08-03 Thread Cameron Simpson

On 09:14 02 Aug 2002, Chris Palmer <[EMAIL PROTECTED]> wrote:
| Vincent Lefevre writes:
| > > The installed manual should be preprocessed during the build to have
| > > the correct defaults.
| > But how can it have the correct defaults since there is only one
| > manual for several binaries?

Only on sloppy installs.

| What several binaries? My system has only one /usr/local/bin/mutt.
| 
| Cameron means that, at compile/configure time, the source of the man
| page would be modified programmatically to agree with the ./configure
| options.
| (Which is, btw, the best solution.)

Or that mutt has a -dump-defaults option, which is used post build,
during install, to mangle the muttrc man page. Hell, with the embedded
help text in the binary, much of the manual entry could be _generated_
from the binary!

On my systems I frequently have several mutts installed.

Each has its own manual page because each has its own install tree.
The /usr/local/bin/mutt is a symlink to the appropriate mutt binary
in its respective tree, and so is the manual entry.
Users wanting the nondefault mutt put
/opt/mutt-version/bin
in their PATH and
/opt/mutt-version/man
in their MANPATH. And lo, they get matching binaries and manuals.
This is not hard. It's almost trivially easy.

Ask yourselves: _why_ do you have multiple mutt binaries? Because they
have _different_ behaviours! And so the should have different manual
entries.

Cheers,
-- 
Cameron Simpson, DoD#743[EMAIL PROTECTED]http://www.zip.com.au/~cs/

The betas never stop at Microsoft.  - news.com



Re: option description - always give default value

2002-08-03 Thread Sven Guckes

* Cameron Simpson <[EMAIL PROTECTED]> [2002-08-04 01:02]:
> On my systems I frequently have several mutts installed.
> Each has its own manual page because each has its own install tree.
> The /usr/local/bin/mutt is a symlink to the appropriate mutt
> binary in its respective tree, and so is the manual entry.

> Users wanting the nondefault mutt put /opt/mutt-version/bin
> in their PATH and /opt/mutt-version/man in their MANPATH.

Users like history majors and non-cs minors never edit
their PATH because they no freakin' clue about this.

> And lo, they get matching binaries and manuals.
> This is not hard. It's almost trivially easy.

come over here - there are 30,000 users on this system whom you
can teach about shells, PATHs, editors and editing, and generally
shooting yourself in the foot.  (no, sorry, no payment for that.)

> Ask yourselves: _why_ do you have multiple mutt binaries?
> Because they have _different_ behaviours!
> And so the should have different manual entries.

let's see - i need a different shell setup for each
of my mutt binaries, and for each of the systems.
now, if i follow this for every program i have..
that would be.. *punch* *punch* *punch*
something above a thousand different versions.
just *why* do i *still* think this is a silly idea?

Sven  [wondering what package maintainers think about this idea.
   can you say "dependecy"? well - i thought you could!]



Re: option description - always give default value

2002-08-03 Thread Cameron Simpson

On 04:22 04 Aug 2002, Sven Guckes <[EMAIL PROTECTED]> wrote:
| let's see - i need a different shell setup for each
| of my mutt binaries,

No, JUST ONE that peers at $0. If you're doing the shell wrapper thing.
Just make links.

| and for each of the systems.
| now, if i follow this for every program i have..

Only the ones where you _really_ maintain several production versions.

| that would be.. *punch* *punch* *punch*
| something above a thousand different versions.
| just *why* do i *still* think this is a silly idea?

Suit yourself. Works for me. The workload is VERY SMALL.

| Sven  [wondering what package maintainers think about this idea.
|can you say "dependecy"? well - i thought you could!]

Most package maintainers don't need to consider multiple versions.  That's
what --prefix is for - to save them the trouble of second guessing the system
maintainer.

I do this "--prefix for each version" thing for several packages.  None of
the package authors know anything about it - provided they honour --prefix
or some similar install prefix string, IT JUST WORKS.

I'm basicly saying here that it can be made to work, with the current
stuff, without extra effort on theauthor end. Whether you _want_ to make
that effort for your users is your call, but saying it's silly merely
means you don't want to make the effort and will live with inaccurate
man pages.

Some of us don't like that idea very much. We would _like_ to be able
to install multiple, _accurate_ man pages. _We_ will handle keeping
them separate/accessible for our users. We ask only that _you_ make it
possibly to _generate_ an accurate man page. As long as it goes in the
_current_ place w.r.t the install prefix, _we_ can do the rest.
-- 
Cameron Simpson, DoD#743[EMAIL PROTECTED]http://www.zip.com.au/~cs/

As long as you're willing to distract people with odd host names, the British
Indian Ocean Territory appears to be a safe bet; unlike .cv, .cm, and .vg,
there is still no name service for .io, making it ripe for fictional hosts.
According to my almanac, it contains the Chagos Archipelago, with a surface
area of 23 square miles and no civilian population whatsoever, although both
the UK and the US "maintain a military presence". Presumably any computers it
contains will be in .mil or .uk. - Elizabeth Zwicky <[EMAIL PROTECTED]>



Re: option description - always give default value

2002-08-04 Thread Thomas Dickey

On Sun, Aug 04, 2002 at 04:22:03AM +0200, Sven Guckes wrote:
> * Cameron Simpson <[EMAIL PROTECTED]> [2002-08-04 01:02]:
> > On my systems I frequently have several mutts installed.
> > Each has its own manual page because each has its own install tree.
> > The /usr/local/bin/mutt is a symlink to the appropriate mutt
> > binary in its respective tree, and so is the manual entry.
> 
> > Users wanting the nondefault mutt put /opt/mutt-version/bin
> > in their PATH and /opt/mutt-version/man in their MANPATH.
> 
> Users like history majors and non-cs minors never edit
> their PATH because they no freakin' clue about this.

that's too broad a generalization.  Last week I was having a nice chat with
a fellow whose field of expertise is economics.  He's also a well-known
contributor to XEmacs...

-- 
Thomas E. Dickey <[EMAIL PROTECTED]>
http://invisible-island.net
ftp://invisible-island.net