info-find-source

info-find-source

[Emanuel Berg]

Did anyone do

(defun info-find-source ()
  (interactive)
  (let ((file (concat Info-current-file ".info")))
    (if (file-exists-p file)
        (find-file-read-only file)
      (message
       "No file: %s (Did you gunzip the info files?)" file) )))

?

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Marcin Borkowski]

On 2018-01-10, at 06:51, Emanuel Berg <moasen@zoho.com> wrote:

> Did anyone do
>
> (defun info-find-source ()
>   (interactive)
>   (let ((file (concat Info-current-file ".info")))
>     (if (file-exists-p file)
>         (find-file-read-only file)
>       (message
>        "No file: %s (Did you gunzip the info files?)" file) )))
>
> ?

No, why would I want to?

-- 
Marcin Borkowski

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

>> Did anyone do
>>
>> (defun info-find-source ()
>>   (interactive)
>>   (let ((file (concat Info-current-file ".info")))
>>     (if (file-exists-p file)
>>         (find-file-read-only file)
>>       (message
>>        "No file: %s (Did you gunzip the info files?)" file) )))
>>
>> ?
>
> No, why would I want to?

I think my desire to do it is a consequence of
everything that has ever happened since
Big Bang, ~13.8 billion years ago, when the
array of discontinued proto-algorithms from the
religious-mythical era finally ended their
cycle of complete disintegration?

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Marcin Borkowski]

On 2018-01-11, at 06:25, Emanuel Berg <moasen@zoho.com> wrote:

> Marcin Borkowski wrote:
>
>>> Did anyone do
>>>
>>> (defun info-find-source ()
>>>   (interactive)
>>>   (let ((file (concat Info-current-file ".info")))
>>>     (if (file-exists-p file)
>>>         (find-file-read-only file)
>>>       (message
>>>        "No file: %s (Did you gunzip the info files?)" file) )))
>>>
>>> ?
>>
>> No, why would I want to?
>
> I think my desire to do it is a consequence of
> everything that has ever happened since
> Big Bang, ~13.8 billion years ago, when the
> array of discontinued proto-algorithms from the
> religious-mythical era finally ended their
> cycle of complete disintegration?

Joking aside, I have never had a need to look at raw info files (outside
info).  And I haven't gunzipped them (why would I?).

Best,

-- 
Marcin Borkowski

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

>>> No, why would I want to?
>>
>> I think my desire to do it is a consequence
>> of everything that has ever happened since
>> Big Bang, ~13.8 billion years ago, when the
>> array of discontinued proto-algorithms from
>> the religious-mythical era finally ended
>> their cycle of complete disintegration?
>
> Joking aside

... joking?

> I have never had a need to look at raw info
> files (outside info).

Info, with all those tiny nodes hanging
everywhere like ornaments from
a paleo-Christmas tree on Terra Prima, can be
painfully slow to navigate.

On the other hand, if you get the "raw" file,
which is actually just a bunch of marked-up
text, you can make a search for a term using
your everyday search-mechanism, and find every
occurence of that term in the entire body
of documentation.

> And I haven't gunzipped them (why would I?).

Because otherwise that Elisp won't work :)

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Marcin Borkowski]

On 2018-01-11, at 22:55, Emanuel Berg <moasen@zoho.com> wrote:

> Marcin Borkowski wrote:
>
>> I have never had a need to look at raw info
>> files (outside info).
>
> Info, with all those tiny nodes hanging
> everywhere like ornaments from
> a paleo-Christmas tree on Terra Prima, can be
> painfully slow to navigate.

On the contrary, I find Info extremely fast.

> On the other hand, if you get the "raw" file,
> which is actually just a bunch of marked-up
> text, you can make a search for a term using
> your everyday search-mechanism, and find every
> occurence of that term in the entire body
> of documentation.

Why not just use `C-s'?  (Or, even better, `i'?)  I use both all the
time, and they let me search through the whole manual (not only the
current node).

Best,

-- 
Marcin Borkowski

RE: info-find-source

[Drew Adams]

Not too on-topic, but FWIW `info+.el' provides command
`Info-merge-subnodes', bound to `+' in Info, which
gives you a plain-text, prettified version of the
current node and all its child nodes (or all of its
descendant nodes, au choix), creating kind of a
little book-in-a-buffer.

(Not necessarily so little.  You could merge
_everything_ in a manual into a single such book.)

You can also create virtual Info manuals, collecting
arbitrary nodes from any manuals.  Unlike what `+'
gives you, such a virtual book is in Info mode, with
all the usual Info features.

Command `Info-virtual-book' is bound to `v' in Info.
You "save" a node using `.' (`Info-save-current-node').
Using `v' creates a book from the currently "saved"
nodes.

https://www.emacswiki.org/emacs/InfoPlus

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

>> Info, with all those tiny nodes hanging
>> everywhere like ornaments from
>> a paleo-Christmas tree on Terra Prima, can
>> be painfully slow to navigate.
>
> On the contrary, I find Info extremely fast.

Not so fast. Because searching a plain text
file is faster still.

Perhaps your perception of info as fast is
a function of your own velocity?

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Bob Proulx]

Emanuel Berg wrote:
> Marcin Borkowski wrote:
> >> Info, with all those tiny nodes hanging
> >> everywhere like ornaments from
> >> a paleo-Christmas tree on Terra Prima, can
> >> be painfully slow to navigate.
> >
> > On the contrary, I find Info extremely fast.
> 
> Not so fast. Because searching a plain text
> file is faster still.
> 
> Perhaps your perception of info as fast is
> a function of your own velocity?

I also find info to be very fast.  I am not using fast computers.
Mostly I use older Thinkpads that most people think ancient.  I am
typing this message on a laptop built in 2008 and it is my newest.

Unfortunately info searching being interactive I don't know of a way
to benchmark it.  I can only say that the time between invoking a
search and the display of the match is so fast that I find it
imperceptible.  It "feels" like 0.05 seconds or faster.  (I played
with "time sleep 0.05" and numbers in that range felt about the same
as searching for random things in the emacs info manual.)  At that
speed even if search in a raw text file was faster in absolute time I
just wouldn't care because it is well below my interactive threshold
of noticing any delay at all.

Bob

Re: info-find-source

[Emanuel Berg]

Bob Proulx wrote:

> Unfortunately info searching being
> interactive I don't know of a way to
> benchmark it. I can only say that the time
> between invoking a search and the display of
> the match is so fast that I find it
> imperceptible. It "feels" like 0.05 seconds
> or faster. (I played with "time sleep 0.05"
> and numbers in that range felt about the same
> as searching for random things in the emacs
> info manual.) At that speed even if search in
> a raw text file was faster in absolute time
> I just wouldn't care because it is well below
> my interactive threshold of noticing any
> delay at all.

What I can see, info, i.e., the framework, is
just a bunch of sparsely annotated and
highlighted hypertext.

Because I don't spend a lot of time with info,
and isn't active with it myself, I do not
care about the markup.

As for the interconnectivity, I was never fond
of moving back and forth by hitting any other
buttons than those on my keyboard, which
I don't think need re-wiring for the purpose of
browsing documentation.

So to me, I can just as well, or better
actually, access the entire manual of a piece
of software by means of a huge text file.

It is more honest. No gleaming links that lures
you into doing something else just because you
can. No new keystrokes to learn and no new
hooks to fill with Elisp. Same old - great -
stuff to use.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Bob Proulx]

Emanuel Berg wrote:
> What I can see, info, i.e., the framework, is
> just a bunch of sparsely annotated and
> highlighted hypertext.

What I see looks the same as the web version:

  https://www.gnu.org/software/emacs/manual/html_node/emacs/index.html

Except mine is local and therefore always available to me even when I
am not connected to a network.  And the local version always matches
the version of emacs I am using with it.  And my local version is
inside emacs so uses my preferred fonts and colors.

For people who don't use emacs but use vim I recommend using the
'pinfo' browser.  It would match their keystroke memory better.

> Because I don't spend a lot of time with info,
> and isn't active with it myself, I do not
> care about the markup.

Markup?  What markup?  All I see is text.  It is formated into
paragraphs.  The most annoying markup is the quoting using characters
I don't prefer with the stylized curly quotes.  But it's a compromise
there for certain because everyone prefers something different there.

> As for the interconnectivity, I was never fond
> of moving back and forth by hitting any other
> buttons than those on my keyboard, which
> I don't think need re-wiring for the purpose of
> browsing documentation.

WAT?  I can only hit the buttons on my keyboard.  I can't hit buttons
that are not on my keyboard.

Personally I prefer the emacs movement keys and so I use emacs
movement keys when browsing info documenation.  And since this is the
help-gnu-emacs mailing list I can say that proudly without vim users
being unhappy with me. :-)

> So to me, I can just as well, or better
> actually, access the entire manual of a piece
> of software by means of a huge text file.

Since it is a multi-function document that is also designed to be
printed and viewed as a book then I can't find any complaint with you
wanting to browse it all printed out like it was a book.

Go for it!

> It is more honest. No gleaming links that lures
> you into doing something else just because you
> can. No new keystrokes to learn and no new
> hooks to fill with Elisp. Same old - great -
> stuff to use.

Oh, so the links to related documenation is like ... *squirrel*
:-)

Bob

Re: info-find-source

[Emanuel Berg]

Another example where info is slower is this
very common situation.

1. You look something up with info and use it.

2. You look something else up and use that.

3, or I mean 1b - because the first thing
didn't quite work! Now, with info, you have to
navigate back to that node to see what it
really said. And maybe between steps 1 and 2,
you looked up something else still, which you
didn't use, which might even belong to some
different program, etc. etc.!

With a text file, you bring that file up -
which is always the same for a given piece of
software - and make a search up or down in
the document.

this is faster and easier, again using only the
everyday commands and finger-habits which has
nothing to do with info or any other major or
minor mode.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Kaushal Modi]

On Fri, Jan 12, 2018, 7:55 PM Emanuel Berg <moasen@zoho.com> wrote:

> Another example where info is slower is this
> very common situation.
>
> 1. You look something up with info and use it.
>
> 2. You look something else up and use that.
>
> 3, or I mean 1b - because the first thing
> didn't quite work! Now, with info, you have to
> navigate back to that node to see what it
> really said. And maybe between steps 1 and 2,
> you looked up something else still, which you
> didn't use, which might even belong to some
> different program, etc. etc.!
>
> With a text file, you bring that file up -
> which is always the same for a given piece of
> software - and make a search up or down in
> the document.
>
> this is faster and easier, again using only the
> everyday commands and finger-habits which has
> nothing to do with info or any other major or
> minor mode.
>

All of what you said is a non-issue. You basically need to learn to use
Info (C-h i h). If you don't want to learn it and search just plain text,
that's fine too.

> --

Kaushal Modi

Re: info-find-source

[Emanuel Berg]

Kaushal Modi wrote:

> All of what you said is a non-issue.

This is what the thread is about. If it is
beyond the horizon of a little grind like you,
why don't you start a thread about how you
always did the right thing the right way but it
never quite clicked anyway, and head down ask
the elders to explain this
unfathomable discrepancy?

> You basically need

I don't need to do anything and especially not
because someone tells me to do it from the
comfort of his computer.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Kaushal Modi]

On Fri, Jan 12, 2018, 11:01 PM Emanuel Berg <moasen@zoho.com> wrote:

> Kaushal Modi wrote:
>
> > All of what you said is a non-issue.
>
> This is what the thread is about. If it is
> beyond the horizon of a little grind like you,
> why don't you start a thread about how you
> always did the right thing the right way but it
> never quite clicked anyway, and head down ask
> the elders to explain this
> unfathomable discrepancy?
>
> > You basically need
>
> I don't need to do anything and especially not
> because someone tells me to do it from the
> comfort of his computer.
>

Yep, I didn't expect a better reply than that from you. I've been following
these threads for years and many times I see you drown the thread in a lot
of non-sense. I have seen this pattern where you ask/suggest something,
someone suggests a better solution, and you simply demean/disrespect that
person.

As an example, in your reply to Marcin earlier:

> I think my desire to do it is a consequence of
everything that has ever happened since
Big Bang, ~13.8 billion years ago, when the
array of discontinued proto-algorithms from the
religious-mythical era finally ended their
cycle of complete disintegration?

> Perhaps your perception of info as fast is
a function of your own velocity?

There is so much noise in your replies!.. that has nothing to do with the
technical topic of discussion.

Back to the topic..

Just try learning Info, C-h i h, and ask questions here if you face any
trouble.

(You do see that everyone else in this thread has had a positive experience
searching using Info.)

> --

Kaushal Modi

Re: info-find-source

[Emanuel Berg]

Kaushal Modi wrote:

> Yep, I didn't expect a better reply than that
> from you. I've been following these threads
> for years and many times I see you drown the
> thread in a lot of non-sense. I have seen
> this pattern where you ask/suggest something,
> someone suggests a better solution, and you
> simply demean/disrespect that person.

Well, I don't know about that, but if anyone
ever perceives that to happen, at least Kaushal
Modi has got that going for him he didn't let
it make him hesitant to participate further in
the discussion...

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[tomas]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sat, Jan 13, 2018 at 01:54:06AM +0100, Emanuel Berg wrote:
> Another example where info is slower is this
> very common situation.
> 
> 1. You look something up with info and use it.
> 
> 2. You look something else up and use that.
> 
> 3, or I mean 1b - because the first thing
> didn't quite work! Now, with info, you have to
> navigate back to that node to see what it
> really said. And maybe between steps 1 and 2,
> you looked up something else still, which you
> didn't use, which might even belong to some
> different program, etc. etc.!

Ahh... the "lost in cyberspace" syndrome. It was widespread
in the Internets around 1990. These days less, because all
there is is Facebook ;-)

(a bit tongue-in-cheek, as you might guess).

Perhaps some kind of visual orientation help
(akin to some breadcrumbs) might help there?

Cheers
- -- t
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlpZuVwACgkQBcgs9XrR2kYo6gCeJNMw4OwS03zY6UqGZIyAHKDI
V/0Anj+r7GLw+idvYx5M/FDQ4RlqhD55
=Zoq5
-----END PGP SIGNATURE-----

Re: info-find-source

[Marcin Borkowski]

On 2018-01-13, at 08:46, tomas@tuxteam.de wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Sat, Jan 13, 2018 at 01:54:06AM +0100, Emanuel Berg wrote:
>> Another example where info is slower is this
>> very common situation.
>> 
>> 1. You look something up with info and use it.
>> 
>> 2. You look something else up and use that.
>> 
>> 3, or I mean 1b - because the first thing
>> didn't quite work! Now, with info, you have to
>> navigate back to that node to see what it
>> really said. And maybe between steps 1 and 2,
>> you looked up something else still, which you
>> didn't use, which might even belong to some
>> different program, etc. etc.!
>
> Ahh... the "lost in cyberspace" syndrome. It was widespread
> in the Internets around 1990. These days less, because all
> there is is Facebook ;-)
>
> (a bit tongue-in-cheek, as you might guess).
>
> Perhaps some kind of visual orientation help
> (akin to some breadcrumbs) might help there?

Info-history-back (bound to `l' by default) anyone?

Best,

-- 
Marcin Borkowski

Re: info-find-source

[Eli Zaretskii]

> From: Marcin Borkowski <mbork@mbork.pl>
> Date: Sat, 13 Jan 2018 09:10:12 +0100
> Cc: help-gnu-emacs@gnu.org
> 
> > Perhaps some kind of visual orientation help
> > (akin to some breadcrumbs) might help there?
> 
> Info-history-back (bound to `l' by default) anyone?

And, of course, breadcrumbs are already there.

Re: info-find-source

[tomas]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sat, Jan 13, 2018 at 10:22:22AM +0200, Eli Zaretskii wrote:
> > From: Marcin Borkowski <mbork@mbork.pl>
> > Date: Sat, 13 Jan 2018 09:10:12 +0100
> > Cc: help-gnu-emacs@gnu.org
> > 
> > > Perhaps some kind of visual orientation help
> > > (akin to some breadcrumbs) might help there?
> > 
> > Info-history-back (bound to `l' by default) anyone?
> 
> And, of course, breadcrumbs are already there.

Yes to both. I do use `l' regularly (I'm not the OP, I'm more or
less happy with it: for me it beats browser+html most of the time),
but I can connect wich Emanuel's sense of "lost in cyberspace".

Some additional creative (and non-intrusive) navigation help might
make Info more accessible to people.

Yes, Eli, "breadcrumbs" are there, as in "where am I, relative to
root", but not so much as in "how on Earth I arrived here". The
`l' key answers that, but then, you gotta find that first :)

Sorry I can't offer a more concrete idea...

Cheers
- -- tomás
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlpZ70EACgkQBcgs9XrR2kbUwwCfXggXXcayx9t4hR4FHob9wLSO
KNMAnicuLvSNajTCxokJG0OVIald8jPc
=JP05
-----END PGP SIGNATURE-----

Re: info-find-source

[Eli Zaretskii]

> Date: Sat, 13 Jan 2018 12:36:33 +0100
> From: <tomas@tuxteam.de>
> 
> Yes, Eli, "breadcrumbs" are there, as in "where am I, relative to
> root", but not so much as in "how on Earth I arrived here".

Does 'L' do what you want?  (Note: upper-case L.)

And breadcrumbs are not only a way of telling where you are in the
tree, it is also a means of jumping several levels up in one go, by
clicking on them.

Re: info-find-source

[tomas]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sat, Jan 13, 2018 at 01:59:57PM +0200, Eli Zaretskii wrote:
> > Date: Sat, 13 Jan 2018 12:36:33 +0100
> > From: <tomas@tuxteam.de>
> > 
> > Yes, Eli, "breadcrumbs" are there, as in "where am I, relative to
> > root", but not so much as in "how on Earth I arrived here".
> 
> Does 'L' do what you want?  (Note: upper-case L.)

This one is nice, thank you!

(it's like receiving an unexpected Christmas present)

> And breadcrumbs are not only a way of telling where you are in the
> tree, it is also a means of jumping several levels up in one go, by
> clicking on them.

Yes, I guessed that (but the keybindings on Info are so quick that
for us keyboard junkies, a quick thrice-u does it).

I was rather trying to understand and interpret Emanuel's unease,
and to think more philosophically about the Emacs whereabouts.

And yes, in some ways, Emacs and the landscape around is sometimes
lacking in discoverability. This is somewhat of a paradox, since
Emacs is at the same time the most discoverable software I know
of (and has ever been), but there you go.

To be fair, it has done impressive strides the last ~10 years in
this department.

Cheers
- -- tomás
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlpZ+C8ACgkQBcgs9XrR2ka6CwCeO0I1YsvnXfual8fmwDTzU5EF
74QAniyM1SSTN/auuDfcEy1nwJLIEEcw
=d+u7
-----END PGP SIGNATURE-----

Re: info-find-source

[Marcin Borkowski]

On 2018-01-13, at 13:14, tomas@tuxteam.de wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Sat, Jan 13, 2018 at 01:59:57PM +0200, Eli Zaretskii wrote:
>> > Date: Sat, 13 Jan 2018 12:36:33 +0100
>> > From: <tomas@tuxteam.de>
>> > 
>> > Yes, Eli, "breadcrumbs" are there, as in "where am I, relative to
>> > root", but not so much as in "how on Earth I arrived here".
>> 
>> Does 'L' do what you want?  (Note: upper-case L.)
>
> This one is nice, thank you!
>
> (it's like receiving an unexpected Christmas present)

Same here.  I had no idea that exists!  Thank you!

> To be fair, it has done impressive strides the last ~10 years in
> this department.

For me, periodically doing C-h m in various modes is one of the nice
pastimes.  It gives you more Christmas presents like that:-).

Best,

-- 
Marcin Borkowski

Re: info-find-source

[Kaushal Modi]

On Sat, Jan 13, 2018, 3:11 AM Marcin Borkowski <mbork@mbork.pl> wrote:

>
> Info-history-back (bound to `l' by default) anyone?
>

The `l'/`r' are one of my favorite bindings in Emacs special modes. It
works as expected in Info mode, .. and also in eww, and in Help buffers
too! (Have you ever done C-h v foo and then C-h v bar, and wanted to look
at the earlier "foo" help buffer.. just hit `l' to go to "foo", and then
`r' to go back to "bar" :) ).

> --

Kaushal Modi

Re: info-find-source

[Marcin Borkowski]

On 2018-01-13, at 15:20, Kaushal Modi <kaushal.modi@gmail.com> wrote:

> On Sat, Jan 13, 2018, 3:11 AM Marcin Borkowski <mbork@mbork.pl> wrote:
>
>>
>> Info-history-back (bound to `l' by default) anyone?
>>
>
> The `l'/`r' are one of my favorite bindings in Emacs special modes. It
> works as expected in Info mode, .. and also in eww, and in Help buffers
> too! (Have you ever done C-h v foo and then C-h v bar, and wanted to look
> at the earlier "foo" help buffer.. just hit `l' to go to "foo", and then
> `r' to go back to "bar" :) ).

Been there, done that.  You're right, l/r is great!

Best,

-- 
Marcin Borkowski

RE: info-find-source

[Drew Adams]

> Ahh... the "lost in cyberspace" syndrome...
> 
> Perhaps some kind of visual orientation help
> (akin to some breadcrumbs) might help there?

I added breadcrumbs to `info+.el' in 2008.
I added it to vanilla Emacs in Emacs 23.

In Info+ you can have breadcrumbs near the top
of the buffer (as in vanilla Emacs) or on the
mode-line (or both).  The former is via option
`Info-breadcrumbs-in-header-flag'.  The latter is
via minor mode `Info-breadcrumbs-in-mode-line-mode'.

In both cases, you can click any segment of the
breadcrumbs to go directly to that node.

Breadcrumbs give you a structural orientation.
And `L' gives you a chronological orientation,
with direct access to any part of that history
(just as breadcrumbs give you direct access to
any part of the structure above you.)

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

> Info-history-back (bound to `l' by
> default) anyone?

All this climbing up and down the tree and
travelling back and forth in time is just
confusing no matter what dedicated keys there
are, because it is unnecessary when one can
just navigate up and down a text file
Emacs buffer.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Marcin Borkowski]

On 2018-01-14, at 08:17, Emanuel Berg <moasen@zoho.com> wrote:

> Marcin Borkowski wrote:
>
>> Info-history-back (bound to `l' by
>> default) anyone?
>
> All this climbing up and down the tree and
> travelling back and forth in time is just
> confusing no matter what dedicated keys there
> are, because it is unnecessary when one can
> just navigate up and down a text file
> Emacs buffer.

If you only navigate up/down, yes.  If you navigate using links, how do
you know whether you went up or down?  (You can look at the percentage
in the buffer, of course.  But it's way easier to hit `l' for going
back.)

Best,

--
Marcin Borkowski

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

> If you only navigate up/down, yes. If you
> navigate using links, how do you know whether
> you went up or down?

Indeed, and this is the/a problem with the tree
structure/representation of the material.
Instead of by and by getting a sense where you
are, you are forever hanging from some branch,
somewhere, without getting a sense how it all
fits together.

Now, in all fairness, if you navigate the tree
long enough, you should get the spacial
awareness there just as well as in a file.
Because awareness of space should be a function
of time and movement, right? It is just
a matter or MORE time compared with
a simpler structure.

-- 
underground experts united
http://user.it.uu.se/~embe8573

RE: info-find-source

[Drew Adams]

> Another example where info is slower is this
> very common situation.
> 1. You look something up with info and use it.
> 2. You look something else up and use that.
> 3, or I mean 1b - because the first thing
> didn't quite work! Now, with info, you have to
> navigate back to that node to see what it
> really said. And maybe between steps 1 and 2,
> you looked up something else still, which you
> didn't use, which might even belong to some
> different program, etc. etc.!

`M-n' (`clone-buffer') is your friend.

Re: info-find-source

[Marcin Borkowski]

On 2018-01-13, at 16:47, Drew Adams <drew.adams@oracle.com> wrote:

>> Another example where info is slower is this
>> very common situation.
>> 1. You look something up with info and use it.
>> 2. You look something else up and use that.
>> 3, or I mean 1b - because the first thing
>> didn't quite work! Now, with info, you have to
>> navigate back to that node to see what it
>> really said. And maybe between steps 1 and 2,
>> you looked up something else still, which you
>> didn't use, which might even belong to some
>> different program, etc. etc.!
>
> `M-n' (`clone-buffer') is your friend.

Wow, yet another gift from Info!  Thanks again!

Sometimes when using Emacs I feel like a child in a toy store.  With the
parents' credit card. ;-P

Best,

--
Marcin Borkowski

Re: info-find-source

[Bob Proulx]

Marcin Borkowski wrote:
> Drew Adams wrote:
> > Emanuel Berg wrote:
> >> Another example where info is slower is this
> >> very common situation.
> >> 1. You look something up with info and use it.
> >> 2. You look something else up and use that.
> >> 3, or I mean 1b - because the first thing
> >> didn't quite work! Now, with info, you have to
> >> navigate back to that node to see what it
> >> really said. And maybe between steps 1 and 2,
> >> you looked up something else still, which you
> >> didn't use, which might even belong to some
> >> different program, etc. etc.!

There are some documentation pages that I refer to often.  This is
typically when I am trying to learn a new feature.  I haven't
committed the details into long term memory yet.  I need to get back
to those pages often.  If I do that twice in a row then I set an emacs
bookmark on the info page I am wanting to get back to often.  Let's
say that is (intentionally recursive) emacs book marks.

Afer navigating to the page I want to refer to again:

  C-x r m  (bookmark-set)
  q        (Info-exit)

So then I am going along and doing whatever.  Then I want to refer to
that documentation again.  I've bookmarked it.  I bring up the
bookmark menu.

  C-x r l  (bookmark-bmenu-list)
  % Bookmark                    File
  (emacs) Bookmarks	        /usr/share/info/emacs-25/emacs

I use C-n C-p (or n and p), or C-s to search down to it, to move the
selection to the bookmark I want.  Then Enter to select it.  With the
selection on the Bookmarks line above Enter takes me immediately back
to that node and the documentation I am wanting.

> > `M-n' (`clone-buffer') is your friend.
> 
> Wow, yet another gift from Info!  Thanks again!

Cool!  That is also a new feature I did not know about either.  I can
think of several cases where this is going to be very useful to me.
Thank you for this very interesting hint!

> Sometimes when using Emacs I feel like a child in a toy store.  With the
> parents' credit card. ;-P

Well said!  Me too!  :-)

Bob

RE: info-find-source

[Drew Adams]

> > > `M-n' (`clone-buffer') is your friend.
> > Wow, yet another gift from Info!  Thanks again!
> 
> Cool!  That is also a new feature I did not know about either.  I can
> think of several cases where this is going to be very useful to me.
> Thank you for this very interesting hint!

Yes; it is very useful.  No, it is not a new feature.
It's been around at least since Emacs 22.

> > Sometimes when using Emacs I feel like a child in a
> > toy store.  With the parents' credit card. ;-P
> 
> Well said!  Me too!  :-)

Me too.  And like everyone else, some things that are
new to me are not new to Emacs.

One moral here is that we have our habitual ways of
using Emacs but there are probably, for each of us,
other ways to use it that we could benefit from
learning.  And sometimes that new acquaintance is
just a doc perusal away.

Re: info-find-source

[Bob Proulx]

Drew Adams wrote:
> > > > `M-n' (`clone-buffer') is your friend.
> > > Wow, yet another gift from Info!  Thanks again!
> > 
> > Cool!  That is also a new feature I did not know about either.  I can
> > think of several cases where this is going to be very useful to me.
> > Thank you for this very interesting hint!
> 
> Yes; it is very useful.  No, it is not a new feature.
> It's been around at least since Emacs 22.

Sorry.  I meant feature new to me.  I didn't mean new to Emacs.

> One moral here is that we have our habitual ways of
> using Emacs but there are probably, for each of us,
> other ways to use it that we could benefit from
> learning.  And sometimes that new acquaintance is
> just a doc perusal away.

Actually that is one of the reasons I have started to post usage
questions here.  To challenge the way I have been doing things for so
long that I don't think about them anymore and to breath some fresh
air into them.

Bob

Re: info-find-source

[Marcin Borkowski]

On 2018-01-13, at 22:33, Bob Proulx <bob@proulx.com> wrote:

>> > `M-n' (`clone-buffer') is your friend.
>> 
>> Wow, yet another gift from Info!  Thanks again!
>
> Cool!  That is also a new feature I did not know about either.  I can
> think of several cases where this is going to be very useful to me.
> Thank you for this very interesting hint!

BTW, did you see this?  http://mbork.pl/2014-12-27_Info_dispatch

Best,

-- 
Marcin Borkowski

Re: info-find-source

[tomas]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sun, Jan 14, 2018 at 07:55:59AM +0100, Marcin Borkowski wrote:

[...]
> BTW, did you see this?  http://mbork.pl/2014-12-27_Info_dispatch

Nice :-)

thanks
- -- t
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlpbHxcACgkQBcgs9XrR2ka1/gCeKpOXpJqA25Z/x8FgDzVD5b4m
KJcAn1ICUNn1kKoLN6iMnBdBt4Bj2QfO
=xOYP
-----END PGP SIGNATURE-----

Re: info-find-source

[Emanuel Berg]

Drew Adams wrote:

> One moral here is that we have our habitual
> ways of using Emacs but there are probably,
> for each of us, other ways to use it that we
> could benefit from learning. And sometimes
> that new acquaintance is just a doc
> perusal away.

In the Colosseum, the champion will choose the
sword (the same type of sword) that he has used
and mastered all his life. But in his spare
time, he might practice and play with the spear
and bow just as well. And you know what?
This will make him an even better
sword fighter!

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

> Sometimes when using Emacs I feel like
> a child in a toy store. With the parents'
> credit card. ;-P

Well, yeah, if you never found new stuff to
learn anymore, you would stop taking an active
interest in Emacs itself and start to actually
*do* things with it. Now what a world that
would be...

PS. Were there really credit cards when you
were a kid? I remember cheques and I've heard
"old people" (?) still use them to a large
extent...

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Emanuel Berg]

Emanuel Berg wrote:

>> Sometimes when using Emacs I feel like
>> a child in a toy store. With the parents'
>> credit card. ;-P
>
> Well, yeah, if you never found new stuff to
> learn anymore, you would stop taking an
> active interest in Emacs itself and start to
> actually *do* things with it. Now what
> a world that would be...

N.B. here "you" = YOU and ME, or ONE.

Of course, I did other stuff with Emacs, that
wasn't about Emacs. But a lot sure was :)

It is like the radio amateurs (ham radio guys).
They could communicate all over the known world
at least from the mid-50s! After telling their
names and locations, which was mandatory, the
topic of discussion was for the most part the
radio equipment itself...

And this is exactly what happens right now.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Marcin Borkowski]

On 2018-01-15, at 01:57, Emanuel Berg <moasen@zoho.com> wrote:

> Emanuel Berg wrote:
>
>>> Sometimes when using Emacs I feel like
>>> a child in a toy store. With the parents'
>>> credit card. ;-P
>>
>> Well, yeah, if you never found new stuff to
>> learn anymore, you would stop taking an
>> active interest in Emacs itself and start to
>> actually *do* things with it. Now what
>> a world that would be...
>
> N.B. here "you" = YOU and ME, or ONE.
>
> Of course, I did other stuff with Emacs, that
> wasn't about Emacs. But a lot sure was :)
>
> It is like the radio amateurs (ham radio guys).
> They could communicate all over the known world
> at least from the mid-50s! After telling their
> names and locations, which was mandatory, the
> topic of discussion was for the most part the
> radio equipment itself...
>
> And this is exactly what happens right now.

Well, this is expected on this list, no? ;-) Emacs is a hobby for many
of us, right?

Best,

--
Marcin Borkowski

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

> Well, this is expected on this list, no? ;-)
> Emacs is a hobby for many of us, right?

That's true. Something can be a hobby *and*
a tool to do amazing things.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Marcin Borkowski]

On 2018-01-15, at 01:47, Emanuel Berg <moasen@zoho.com> wrote:

> Marcin Borkowski wrote:
>
>> Sometimes when using Emacs I feel like
>> a child in a toy store. With the parents'
>> credit card. ;-P
>
> Well, yeah, if you never found new stuff to
> learn anymore, you would stop taking an active
> interest in Emacs itself and start to actually
> *do* things with it. Now what a world that
> would be...

Well, I do stuff in Emacs all the time...

> PS. Were there really credit cards when you
> were a kid? I remember cheques and I've heard
> "old people" (?) still use them to a large
> extent...

No, and neither were cheques.  Cheques were never very popular in
Poland, I think.  (Maybe before WW2...?)  I remember using them to get
cash from my bank account before using plastic cards, but that's pretty
much it.

I hear cheques are pretty popular in the US, which is yet another proof
that it's one of the weirdest places on Earth.  (Not necessarily in
a bad sense in this case - although I consider cash superior to cheques,
they may be superior to credit cards, which are among the worst of
humanity's inventions.  I do not own one, and I do not plan to own one
ever in my life.)

Best,

--
Marcin Borkowski

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

> Well, I do stuff in Emacs all the time...

You don't say...

> No, and neither were cheques. Cheques were
> never very popular in Poland, I think. (Maybe
> before WW2...?) I remember using them to get
> cash from my bank account before using
> plastic cards, but that's pretty much it.
>
> I hear cheques are pretty popular in the US,
> which is yet another proof that it's one of
> the weirdest places on Earth. (Not
> necessarily in a bad sense in this case -
> although I consider cash superior to cheques,
> they may be superior to credit cards, which
> are among the worst of humanity's inventions.
> I do not own one, and I do not plan to own
> one ever in my life.)

I don't know how cheques work but people use
them so they must serve some purpose. Here,
credit cards are now legio and people pay
everything with them.

A couple of years ago there were a bunch of
spectacular armed robberies in Sweden, one
involving a helicopter dropping people on the
roof of a cash deposit for example, and more
such stuff, involving people both from the
Swedish "disorganized crime" and from Serbian
gangsters who snapped from/during all their
wars. Anyway since then, and partly because of
that, almost the entire society has abandoned
cash, and that kind of crime doesn't happen
anymore. So that is one benefit, at least.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Stefan Monnier]

> wars. Anyway since then, and partly because of
> that, almost the entire society has abandoned
> cash, and that kind of crime doesn't happen
> anymore. So that is one benefit, at least.

Does that mean that there are no more anonymous monetary transactions?
That sounds scary!


        Stefan

Re: info-find-source

[tomas]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Tue, Jan 16, 2018 at 08:45:11AM -0500, Stefan Monnier wrote:
> > wars. Anyway since then, and partly because of
> > that, almost the entire society has abandoned
> > cash, and that kind of crime doesn't happen
> > anymore. So that is one benefit, at least.
> 
> Does that mean that there are no more anonymous monetary transactions?

Hard cash. Until they implement routine DNA scanning on coins and
notes, that is. Then... hard cash with gloves.

Or they do away with cash altogether. Because tarrists and mafia.

> That sounds scary!

It does.

Cheers (nonetheless)
- -- tomás
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlpeGgIACgkQBcgs9XrR2kbPMwCfXus6JrPowzktbXczzn3fzmbK
cn4AniFUNb+kejDHkdewYVOg5EULGC0r
=9bnm
-----END PGP SIGNATURE-----

Re: info-find-source

[Marcin Borkowski]

On 2018-01-16, at 14:45, Stefan Monnier <monnier@iro.umontreal.ca> wrote:

>> wars. Anyway since then, and partly because of
>> that, almost the entire society has abandoned
>> cash, and that kind of crime doesn't happen
>> anymore. So that is one benefit, at least.
>
> Does that mean that there are no more anonymous monetary transactions?
> That sounds scary!

Exactly my thought.  I'm not a tinfoil-hat level paranoid, but I almost
never use my debit card (I don't have a credit one) for anything but
taking cash out of ATMs.  In principle, I could share the knowledge of
my transactions with a bank, but I'd expect a solid financial reward for
that.

And to think that we went so far down the madness route that we consider
paper money to be genuine and old-school, whereas it's in fact only
virtual and more or less forged version of the _real_ thing...

Best,

--
Marcin Borkowski

Re: info-find-source

[Robert Thorpe]

Emanuel Berg <moasen@zoho.com> writes:
>
> I don't know how cheques work but people use
> them so they must serve some purpose. Here,
> credit cards are now legio and people pay
> everything with them.

They're simple for the person recieving the payments and good for
irregular or one-off payments.

Take my accountant for example.  I pay him by cheque.

He doesn't have a card payment machine, his business isn't large so he
probably can't justify getting a machine.  I could pay cash but it would
be more than I usually carry.  So, I'd have to make a special trip to an
ATM.  Then there's the risk of losing cash or of crime.  He has the same
risks, so he wouldn't like cash either.

I could use the electronic payment feature of my bank's website.  The
problem there is that he would need to give me his bank details.  Those
may change in the future, so I'd have to ask him every time.  Then, I
only have 14 characters to identify the transaction.  On my cheque book
stub I have as much as I can write in, which is a lot more.

I use cheques much less often than I did.  I only write maybe 10-20 per
year.  I used to write maybe three times as many.

This is getting really off-topic.  So, I'm not going to say anything
more about it.

BR,
Robert Thorpe

Re: info-find-source

[Emanuel Berg]

Stefan Monnier wrote:

>> because of that, almost the entire society
>> has abandoned cash, and that kind of crime
>> doesn't happen anymore. So that is one
>> benefit, at least.
>
> Does that mean that there are no more
> anonymous monetary transactions?
> That sounds scary!

There is cash but its use and significance has
plummeted in the last 10-15 years and what
I can determine this is a process that
only continues.

When we first heard one could pay directly with
a credit card in the store, what we thought was
OK, this is what happens when some dude from
Virgin Records gets on a jet from London and
wants to buy a snowmobile, only he is afraid he
will be robbed on the way from the ATM to the
vendor... But today, kids use credit cards when
they buy one "Lion" candy bar at 7.95 SEK!
(7.95 SEK ~= $0.99 | £0.72 | €0.81)

As for privacy, without being an aluminium
foil hat (or "aluminum" as you say in NA), yes
it would be a problem if one could never do
small favors to people for small sums in cash.
But we are not exactly there quite yet...

In the long run, remember that it is not only
the plebeians that would suffer from this.
Also all the dealings of shady BB could
potentially come to the public's eye.
While I don't think that will ever happen (they
will just come up with yet another layer to
hide it) it could, in principle.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Emanuel Berg]

Drew Adams wrote:

>> Another example where info is slower is this
>> very common situation. 1. You look something
>> up with info and use it. 2. You look
>> something else up and use that. 3, or I mean
>> 1b - because the first thing didn't quite
>> work! Now, with info, you have to navigate
>> back to that node to see what it really
>> said. And maybe between steps 1 and 2, you
>> looked up something else still, which you
>> didn't use, which might even belong to some
>> different program, etc. etc.!
>
> `M-n' (`clone-buffer') is your friend.

I'm aware of `clone-buffer'. But actually it
was you who once told me about it so I can give
you the credit anyway.

However all these examples only emphasize my
argument! I know that you can do a lot of stuff
with the info browser. But with a text file,
you either don't have to do it, *or* you can do
it, only not with whatever unknown interface
the Info browser provides, but with what you
use every day to edit, search, and view text
and code, i.e. stuff that you are already
embarrassingly familiar with!

So let's put it another way, what *advantages*
are there to the info browser? N.B. here we are
only talking about the browser, not the style
or organization of the info documentation
itself, which, as is evident by this
discussion, thru its uniformity can be accessed
in many different ways.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Robert Thorpe]

Emanuel Berg <moasen@zoho.com> writes:

> Marcin Borkowski wrote:
>
>>> Info, with all those tiny nodes hanging
>>> everywhere like ornaments from
>>> a paleo-Christmas tree on Terra Prima, can
>>> be painfully slow to navigate.
>>
>> On the contrary, I find Info extremely fast.
>
> Not so fast. Because searching a plain text
> file is faster still.
>
> Perhaps your perception of info as fast is
> a function of your own velocity?

Searching info is very fast on the computers I use.

Have you tried it with emacs -Q?  I wonder if you have customized
something that's made it slower.

I doubt that using fundamental mode would be much faster.  Internally, Info
search uses normal isearch regexp functions.

BR,
Robert Thorpe

Re: info-find-source

[Emanuel Berg]

OK, there seems to be a bit of confusion here
with respect to what I mean. So I'd like to
clarify a few thing. That said, I'm not saying
any of this to try to sway anyone. It is just
my POV.

First, I don't think info is bad in any way.
Actually, I think it is very good! To have
a uniform interface to documentation and to
have people add new pieces to it, in a uniform
way, which will then fit seamlessly, is great.
I encourage everyone who has written a larger
piece of software to do it, no doubt.

Markup and interconnectivity, if that is
a word, is also good. The man pages are both as
well and I never felt the need to browse the
groff source.

That the documentation comes with Emacs, or is
on-line (i.e., not on paper) - remember the
terrible Sierra On-Line adventure games? - is
also a good thing, even tho a web version is
also good. And because of the uniformity one
can easily use or write a tool that will
translate info material into HTML or whatever
format is desired.

The issue I have with info is that it is easy
to get lost when navigating all those node back
and forth in the tree structure, back and forth
in history, up to the parent and down to the
child until you are stuck at a leaf and you
still haven't found what you are looking for.
And you do all this with keys that you do not
use every day for editing.

Compare this to the man pages where this never
happens (because of less complexity), *or*
a plain text files, where by definition it
cannot ever happen.

But doesn't this mean the files will be very
long? Yes, and I don't have a problem with that
as this volume is linear, not broken down into
a complicated tree structure one has
to traverse to get to the rainbow's end.

The speed I've mentioned isn't the speed it
takes to execute a command, it the the general
speed of access, the human-computer interface
if you will, which again per definition (unless
your cognitive "humanity" differs from mine),
this will be much, much faster with text
because I edit text and code every day, using
the same functions and finger-habits, and no
matter how fluent an info user I'll ever be, it
could never, ever match that.

Also, how does info look to you guys? To me, it
looks like this:

    http://user.it.uu.se/~embe8573/pics/info.png

The problems getting an overview what's going
on may be related to that, as well.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Robert Thorpe]

Emanuel Berg <moasen@zoho.com> writes:

> The speed I've mentioned isn't the speed it
> takes to execute a command, it the the general
> speed of access, the human-computer interface
> if you will, which again per definition (unless
> your cognitive "humanity" differs from mine),
> this will be much, much faster with text
> because I edit text and code every day, using
> the same functions and finger-habits, and no
> matter how fluent an info user I'll ever be, it
> could never, ever match that.

I don't understand what you mean.  I usually don't understand what you
mean in these type of situations.

Info has it's own keybindings.  That can be a little tricky, since "s"
is the normal way to search rather than "C-s" or "M-s".  You can rebind
these functions though.

> Also, how does info look to you guys? To me, it
> looks like this:
>
>     http://user.it.uu.se/~embe8573/pics/info.png

If you use Emacs in a GUI environment then there is more markup.  Some
things are in bold and some are in larger fonts.  I generally think this
is useful, but opinions differ.

BR,
Robert Thorpe

Re: info-find-source

[Marcin Borkowski]

On 2018-01-13, at 04:43, Robert Thorpe <rt@robertthorpeconsulting.com> wrote:

> Info has it's own keybindings.  That can be a little tricky, since "s"
> is the normal way to search rather than "C-s" or "M-s".  You can rebind
> these functions though.

I didn't know that!  I used to use C-s/C-r in Info all the time.  They
_do_ work there!

Then, maybe a year or two ago, I learned about the index and the `i'
keystroke, and now I can't understand how I used Info before.

Also, `C-h S', by the way.

I would really like it if someone set up a contest of "how fast can
someone use Info to search for things and come back to other things and
not get distracted by links and/or writing one's own functions to do
what you can do with stock Emacs better", and made Emanuel and me the
contestants.

;-)

Best,

--
Marcin Borkowski

RE: info-find-source

[Drew Adams]

> I would really like it if someone set up a contest of "how fast can
> someone use Info to search for things and come back to other things and
> not get distracted by links and/or writing one's own functions to do
> what you can do with stock Emacs better", and made Emanuel and me the
> contestants.

What I find most useful for finding stuff: `i', combined
with better pattern-matching for the index-entry
completion candidates.

I use Icicles.  That means that index entries, which
are what `i' completes your minibuffer contents to,
can be matched with regexps, including just substrings.

In Icicle mode, `i' is bound to command `icicle-Info-index'.
Besides giving you better pattern-matching, you can
optionally have it highlight index-entry candidates
in *Completions* that correspond to already-visited
nodes.  That way, you don't end up trying multiple
index entries in hopes of getting to some nodes you
haven't already checked out.

(Wrt seeing which nodes you've visited by link color,
face `info-xref-visited' helps, but if you want that
indication to persist across Emacs sessions you can
get that (togglable anytime) with `info+.el' minor
mode `Info-persist-history-mode'.) 

Even just substring matching makes `i' much, much more
useful, IMO.  (And yes, you can set vanilla Emacs to use
substring matching for `i'.  If you don't use Icicles
or similar then this is a good workaround/substitute.)

With Icicles you can incrementally match any number of
simple patterns (progressive completion), which is much
simpler and quicker than trying to come up with a single
regexp to match what you need.  (Not to mention more
powerful, as a single regexp is limited in terms of what
it matches.)  You can also match the complements of
patterns.

Beyond pattern-matching and indicated previously visited
nodes, `icicle-Info-index' is a multi-command, which
means that you can, with a single invocation, visit
any number of nodes, matching any patterns, in any
order.

Icicles also enhances other Info commands, in particular
`g'.  When you use it you can optionally match node names
or node _content_, or both at once.

https://www.emacswiki.org/emacs/Icicles_-_Info_Enhancements

https://www.emacswiki.org/emacs/Icicles_-_Nutshell_View#ProgressiveCompletion

Re: info-find-source

[Marcin Borkowski]

On 2018-01-13, at 17:31, Drew Adams <drew.adams@oracle.com> wrote:

>> I would really like it if someone set up a contest of "how fast can
>> someone use Info to search for things and come back to other things and
>> not get distracted by links and/or writing one's own functions to do
>> what you can do with stock Emacs better", and made Emanuel and me the
>> contestants.
>
> What I find most useful for finding stuff: `i', combined
> with better pattern-matching for the index-entry
> completion candidates.
>
> I use Icicles.  That means that index entries, which
> are what `i' completes your minibuffer contents to,
> can be matched with regexps, including just substrings.

I stopped using Icicles, since I did not use it often enough to memorize
all the cool stuff in there; also, it was not as fast as I wanted.
I switched to Ivy, which is definitely less powerful, but good enough
for me, and much faster than Icicles.

In Ivy, if you search for "abc xyz", it basically transforms it to
"abc.*xyz" under the hood.  Very useful, and covers 99% of my use cases.

Still, I do appreciate Icicles - I just don't really need its power (or
at least I haven't yet discovered that I do;-)).

Best,

--
Marcin Borkowski

RE: info-find-source

[Drew Adams]

[-- Attachment #1: Type: text/plain, Size: 5211 bytes --]

>> What I find most useful for finding stuff: `i',
>> combined with better pattern-matching for the
>> index-entry completion candidates.

Further in that post I said "Icicles or similar"
to characterize such "better pattern-matching".

My point was that key `i' is typically the best
way (IMO) to find stuff in Info.  And its power is
greatly increased by libraries that provide better
pattern-matching than what is offered by vanilla
Emacs.  Icicles is one such library.

>> I use Icicles.  That means that index entries,
>> which are what `i' completes your minibuffer
>> contents to, can be matched with regexps,
>> including just substrings.
> 
> I stopped using Icicles...  I switched to Ivy, which
> is definitely less powerful, but good enough...

If you prefer this or that or you don't use/need
this or that, that's fine.

Wrt Icicles and other pattern-matching libraries, FWIW:

When Icicles started exploring completion and what
could be done with it there was essentially nothing
besides vanilla-Emacs completion, which was itself
coded only in C (no `minibuffer.el' library yet, no
`completion-styles' - just basic prefix completion).

And completion wasn't used much in Emacs - mainly
just for file-finding, buffer-switching, and `M-x'.

IswitchB was the only completion-related thing that
did something interesting before Icicles.  Well,
there was also `icomplete.el', which incrementally
showed you some input completions, but you couldn't
do anything with them except use them as a guide
for what to continue typing.

Over time many Icicles features have been introduced
into new packages (Ido, Helm/Anything, Ivy) - years
later.  Ivy apparently introduced `ivy-occur'/`swiper'
in 2015.  Icicles introduced it (as `icicle-occur')
in 2006 (along with `icicle-search': same, but with
regexp-defined search contexts, not just lines).

That's all good, not bad.  "Imitation...flattery."

Icicles has introduced original ideas/features,
including:

 incremental completion (matches updated as you
 type), help on individual completion candidates,
 multi-commands (multiple actions on multiple
 candidates), progressive completion (narrowing,
 successive search patterns), match complementing,
 multi-completions (matching multiple things
 together - e.g. file names & contents), cycling
 candidates, sorting candidates on the fly, saving
 completion matches & combining them using set
 operations, key completion (which also shows the
 keys currently available), fuzzy completion,
 using completion for search...

Some Icicles ideas might be hare-brained or
half-baked.  Some that I originally thought were
probably crazy have turned out to be among the most
useful.  Others I thought might be more useful were
not so.

Any or all of them could be implemented in different,
some better, ways.  And different UIs could be used
to present them to users.  And there are bugs to be
fixed...

If another package picks up this or that Icicles
idea and implements it faster or in an easier-to-use
way than what Icicles provides that's a good thing,
not a bad thing.  Improvement is good.

One of the explicit purposes of Icicles, from the
outset, has been to serve as food for thought and
experiment (for me, in particular). The existence
of Helm (formerly Anything) and Ivy is, among
other things, a testament to the usefulness of
Icicles ideas - at least some of them ;-).

Other Icicles ideas have found their way to vanilla
Emacs and to other of my libraries: Isearch+, Info+,
Bookmark+, Dired+, LaCarte, highlight.el, mouse3.el,
palette.el, synonyms.el, ucs-cmds.el.

> since I did not use it often enough to memorize all
> the cool stuff in there;

There's really nothing to memorize.  But perhaps the
first thing is to know how to ask it.

* `S-TAB', to see all currently available keys and
  their commands (navigate the key hierarchy,
  including menus).

* `M-?' during minibuffer input for general help,
  with links to the complete help - in local files
  and on the Web - and with links to customizing
  the Icicles options & faces.

The `M-?' help also gives you the current status
of options, and (linked) key-sequences to change
status on the fly.  See attached, and imagine that
the commands and keys shown there are links that
perform their actions. 

If someone can't remember `M-?' then s?he can find
it in menu-bar `Icicles > Icicles Help' anytime.

> In Ivy, if you search for "abc xyz", it basically
> transforms it to "abc.*xyz" under the hood.  Very
> useful, and covers 99% of my use cases.

Same with Icicles, FWIW.  That's one of the 7
"fuzzy" completion methods it supports, besides
regexp matching and vanilla `completion-styles'.
You can make it your default method or choose it
or another using `M-('.

> Still, I do appreciate Icicles - I just don't
> really need its power (or at least I haven't yet
> discovered that I do;-)).

You probably don't "need" most of what Emacs or
Lisp has to offer either.  Few (none?) of us do.
It's available on demand, for when you do.  It
doesn't bother you when you don't.  Same for
Icicles, I hope.

[-- Attachment #2: general-help.txt --]
[-- Type: text/plain, Size: 50349 bytes --]

[Icicles Help on the Web]                        [Icicles Doc, Part 1]
[Icicles Options & Faces]                        [Icicles Doc, Part 2]

You are completing input for an Icicles multi-command.
To show help on individual candidates:

     Current candidate                       C-M-RET, C-M-mouse-2
     Next, previous candidate                C-M-down, C-M-up,
                                              C-M- plus mouse wheel
                    prefix-match candidate   C-M-end, C-M-home
                    apropos-match candidate  C-M-next, C-M-prior

To act on individual candidates:
     Current candidate                       C-RET, C-mouse-2
     Next, previous candidate                C-down, C-up,
                                              C- plus mouse wheel
                    prefix-match candidate   C-end, C-home
                    apropos-match candidate  C-next, C-prior
     All candidates at once                  C-! (each) or M-! (list)
     Delete object named by candidate        S-delete
     Object-action: apply a fn to candidate  M-RET

For alt action, use `C-S-' instead of `C-', but use `C-|' or `M-|',
     instead of `C-!' or `M-!', to act on all.



                    Icicles Minibuffer Completion
                    -----------------------------

Completion indicators:

  Mode line `Icy' lighter (additive):
    red = Completion available      (use `TAB' or `S-TAB' to complete)
    +   = Multi-command completion   (use `C-RET' to act on candidate)
    ||  = Multi-completion candidates  (use `C-M-j' to separate parts)
    ... = `icicle-max-candidates' shown        (use `C-x #' to change)

  Prompt prefix (exclusive):
    . = Simple completion
    + = Multi-command completion

You can complete your minibuffer input in several ways.
These are the main Icicles actions and their minibuffer key bindings:

 * Show Icicles minibuffer help (this).      M-?
     For help on individual completion candidates, see "Show help on
     individual completion candidates", below.

 * Abandon or commit your input.
     Abandon input                           C-g
     Commit input to Emacs                   RET
       Complete partial input, then commit   S-return

 * Toggle/cycle Icicles options on the fly.  Key:   	Currently:
     Highlighting of past inputs             C-pause	yes
     Highlighting of saved candidates        M-i s	yes
     Showing candidates with WYSIWYG         M-i w	yes
     Removal of duplicate candidates         M-i $	no
     Sort order                              C-,	alphabetical
     Alternative sort order                  M-,	by previous use alphabetically
     Swap alternative/normal sort            M-i M-,
     Case sensitivity                        M-i A	yes
     `.' matching newlines too (any char)    M-i M-.	no
     Escaping of special regexp chars        C-`	no
     Incremental completion                  M-i #	yes, if *Completions* showing
     Input expansion to common match (toggle)M-i "	yes
     Input expansion to common match (cycle) M-i M-"	always
     Hiding common match in `*Completions*'  C-x .	no
     Hiding no-match lines in `*Completions*' C-u C-x .	no
     Horizontal/vertical candidate layout    C-M-^	horizontal
     Completion-mode keys                    M-i TAB	unchanged
     S-TAB completion method                 M-(	apropos
     TAB completion method                   C-(	vanilla
     Vanilla completion style set (E23+)     C-M-(	nil
     Showing image-file thumbnails (E22+)    C-x t	image and name
     Showing candidate annotations           C-x C-a	yes
     Inclusion of proxy candidates           C-M-_	no
     Ignoring certain file extensions        M-i .	yes
     Expansion of directory candidates       C-x /	no
     Checking for remote file names          C-^	yes
     Considering network drives as remote    C-x :	yes
     Ignoring space prefix for buffer names  M-_	yes
     Using `C-' for multi-command actions    M-g	yes
     Using `~' for your home directory       M-~	yes
     `icicle-search' all-current highlights  C-^	no
     Whole-word searching                    M-q	no
     Removal of `icicle-search' highlighting M-i .	yes
     Replacement of whole search hit         M-_	yes
     Replacement of expanded common match    M-;	yes
     Searching complements of contexts       M-i ~	no

 * Regexp-quote input, then apropos-complete M-%

 * Change the set of completion candidates.  Modify your input.
     Edit your input                         (just edit in minibuffer)
     Erase your input (clear minibuffer)     M-k
     Goto/kill non-matching portion of input C-M-l
     Retrieve previous completion inputs     C-l, C-S-l
     Match another regexp (chaining)         M-*
     Satisfy another predicate (chaining)    M-&
     Remove a candidate from set of matches  delete, S-mouse-2
     Yank text at cursor into minibuffer     M-.
     Insert text (string) from a variable    C-=
     Insert `icicle-list-join-string'        C-M-j
     Insert previously entered input(s)      M-o
     Insert completion candidates(s)         M-r
     Insert key description (key completion) M-q

 * Complete your current input in the minibuffer.
     Apropos (regexp) completion             backtab
       Without displaying candidates         C-M-S-TAB
       Complete and match another regexp     S-SPC
     Prefix completion
       As much as possible                   TAB
         Without displaying candidates       C-M-tab
       A word at a time                      M-SPC
     Complete and commit S-return
     Complete search string using past input backtab

 * Display/navigate completions for current input (in `*Completions*').
     Show completion candidates
       Prefix completion                     TAB (repeat)
       Apropos completion                    backtab
     Move between minibuffer and list        C-insert
     Cycle among completion candidates       right, left, TAB, backtab
       Within a `*Completions*' column       down, up
     Choose a completion candidate           RET, M-x mouse-choose-completion

 * Cycle among input candidates.
     Completion candidates
       Current mode                          down, up, mouse wheel
       Prefix completion                     end, home
       Apropos completion                    next, prior
     Minibuffer history items                M-n, M-p
     Completion history items                C-l, C-S-l

 * Show help on individual completion candidates.
     Current candidate                       C-M-RET, C-M-mouse-2
     Next, previous candidate                C-M-down, C-M-up,
                                              C-M- plus mouse wheel
                    prefix-match candidate   C-M-end, C-M-home
                    apropos-match candidate  C-M-next, C-M-prior

 * Choose a previous input from the minibuffer history.
     Complete to insert a previous input     M-o
     Complete against history items          M-h, M-pause
     Restrict candidates to history items    M-pause
     Change to another history               C-M-pause
     List history items first in Completions M-i M-,
     Cycle among minibuffer history items    M-n, M-p

 * Delete history entries
     Delete current entry (cycling)          M-k
     Delete any or all entries               M-K

 * Multi-commands: Act on completion candidates.
   For alternative action, use `C-S-' instead of `C-', but
   `C-|' and `M-|' are alternative action versions of `C-!' and `M-!'.
     Current candidate                       C-RET, C-mouse-2
     Next, previous candidate                C-down, C-up,
                                             C- with mouse wheel
                    prefix-match candidate   C-end, C-home
                    apropos-match candidate  C-next, C-prior
     Act on each matching candidate, in turn C-!
     Act on the list of matching candidates  M-!
     Delete object named by candidate        S-delete
     Remove candidate from set of matches    delete, S-mouse-2
     Save candidate (add to those saved)     insert, M-S-mouse-2
     Object-action: apply a fn to candidate  M-RET

 * Act on multiple minibuffer inputs         M-R

 * Search and replace (e.g. `C-c `').  See also `icicle-search'.
     Use action keys (prefix `C-') to navigate.
     Use alternative action keys (prefix `C-S-') to replace matches.
     Toggle input highlighting at all hits   C-^
     Toggle whole-word searching             M-q
     Toggle `.' matching newlines too        M-i M-.
     Toggle escaping of special regexp chars C-`
     Toggle removal of search highlighting   M-i .

     Replace all                             M-|
     Redefine the replacement string         M-,
     Toggle literal replacement              M-i `
     Toggle replacement of whole search hit  M-_
     Toggle replacement of common match      M-;

 * Perform set operations on candidate sets.
     Remove candidate from current set       delete, S-mouse-2
     Add current candidate to saved set      insert, M-S-mouse-2
     Retrieve saved candidates from...
       `icicle-saved-completion-candidates'  C-M-<
       another variable                      C-M-{
       a cache file                          C-{
     Retrieve more saved candidates          C-<
     Save candidates in current set to...
       `icicle-saved-completion-candidates'  C-M->
       another variable                      C-M-}
       a cache file                          C-}
     Save more candidates to current set     C->
     Save, save more selected candidates     C-M-), C-)  with region
     Save multiple minibuffer inputs         M-S
     Clear all saved candidates              C-M-) with empty region
     Add new or update existing saved set
       M-x icicle-add/update-saved-completion-set
     Remove a saved completion set
       M-x icicle-remove-saved-completion-set
     Swap current and saved sets             C-%
     Define current set by evaluating sexp   C-:
     Restrict candidates to history items    M-pause
     Set complement                          C-~
     Set difference                          C--
     Set union                               C-+
     Set intersection                        C-*
     Set intersection using regexp           M-*
     Set intersection using predicate        M-&
       Save current predicate to a variable  C-M-&
       Insert string variable as input       C-=

 * Adjust Icicles options incrementally on the fly (uses Do Re Mi).
     `icicle-candidate-width-factor'        C-x w
     `icicle-max-candidates'                C-x #
     `icicle-swank-timeout'                 C-x 1
     `icicle-swank-prefix-length'           C-x 2
     `icicle-inter-candidates-min-spaces'   C-x |
     Zoom `*Completions*' (not an option)   C-x -   (Emacs 23+)

Remember: You can always input any character (e.g. TAB) that is bound
          to a command by preceding it with C-q.

Though it has no direct connection with completion, you can use `M-:'
in the minibuffer at any time to evaluate an Emacs-Lisp expression.
This calls `icicle-pp-eval-expression-in-minibuffer', which displays
the result in the echo area or a popup buffer, *Pp Eval Output*.
It also provides some of the Emacs-Lisp key bindings during expression
editing.
\f

Customize Icicles: `M-x icicle-customize-icicles-group'.
Summary of customizable options and faces (alphabetical order).

Some of the options can be toggled or cycled - the keys for this are
noted in parentheses.

* `case-fold-search', `completion-ignore-case',
  (`C-u') `read-file-name-completion-ignore-case'
                                         - Case sensitivity? (`C-A')
* `completion-ignored-extensions'        - Ignored filenames (`C-.')
* `icicle-act-before-cycle-flag'         - Act then cycle or reverse?
* `icicle-add-proxy-candidates-flag'     - Include proxies? (`C-M-_')
* `icicle-alternative-actions-alist'     - Overriding alt actions
* `icicle-alternative-sort-comparer'     - Other sort (`M-,', `C-M-,')
* `icicle-apropos-complete-keys*'        - Keys to apropos-complete
* `icicle-apropos-cycle-*-keys'          - Keys to apropos-cycle
* `icicle-bookmark-name-length-max'      - Max length of bookmark name
* `icicle-bookmark-refresh-cache-flag'   - Refresh bookmarks cache?
* `icicle-top-level-key-bindings'        - Bind top-level commands
* `icicle-buffer-*'                      - `icicle-buffer' options
* `icicle-candidate-width-factor'        - Width %%, candidate columns
* `icicle-change-region-background-flag' - Change region color?
* `icicle-change-sort-order-completion'  - Control `C-,' behavior
* `icicle-C-l-uses-completion-flag'      - `C-l' uses completion?
* `icicle-color-themes'                  - For `icicle-color-theme'
* `icicle-comint-dynamic-complete-replacements' - Comint complete fns
* `icicle-command-abbrev*'               - Command abbrev behavior
* `icicle-complete-key-anyway-flag'      - `S-TAB' must complete keys
* `icicle-complete-keys-self-insert-ranges'- `S-TAB' for self-insert?
* `icicle-completing-read+insert-keys'   - Keys for complete-on-demand
* `icicle-completion-history-max-length' - Completion history length
* `icicle-completion-key-bindings'       - minibuffer completion keys
* `icicle-completion-list-key-bindings'  - `*Completions*' bindings
* `icicle-Completions-display-min-input-chars'- Remove `*Completions*'
                                           if fewer chars input
* `icicle-completions-format'            - `*Completions*' layout
* `icicle-move-Completions-frame'        - `*Completions*' at edge?
* `icicle-Completions-text-scale-decrease'- `*Completions*' shrink
* `icicle-Completions-window-max-height' - Max lines, `*Completions*'
* `icicle-customize-save-flag'           - Save some options on quit?
* `icicle-default-cycling-mode'          - Default completion mode for
                                           per-mode cycling
* `icicle-default-thing-insertion'       - Control behavior of M-.
* `icicle-default-value'                 - How to treat default value
* `icicle-define-alias-commands-flag'    - Define top-level aliases?
* `icicle-deletion-action-flag'          - `S-delete' deletes?
* `icicle-dot-show-regexp-flag'          - Show regexp for `.'?
* `icicle-dot-string'                    - String that `.' inserts
* `icicle-expand-input-to-common-match'  - Expand your input? (`C-M-"')
* `icicle-expand-input-to-common-match-alt' - Expand your input? (`C-"')
* `icicle-file-*'                        - `icicle-file' options
* `icicle-filesets-as-saved-completion-sets-flag'- Use filesets?
* `icicle-guess-commands-in-path'        - Shell commands to complete
* `icicle-help-in-mode-line-delay'       - Secs to show candidate help
* `icicle-hide-common-match-in-Completions-flag'- Show common match?
* `icicle-hide-non-matching-lines-flag'  - Hide non-match lines?
* `icicle-highlight-historical-candidates-flag'
                                         - Highlight past input?
* `icicle-highlight-input-completion-failure*'- Input non-match sign
* `icicle-highlight-input-initial-whitespace-flag'
                                         - Highlight input whitespace?
* `icicle-highlight-lighter-flag'        - Highlight mode-line `Icy'
* `icicle-incremental-completion'        - Icompletion? (`C-#')
* `icicle-incremental-completion-delay'  - Delay before update cands
* `icicle-incremental-completion-threshold'- # of candidates for delay
* `icicle-inhibit-advice-functions'      - Advice-inhibited functions
* `icicle-inhibit-ding-flag'             - Suppress audible bell
* `icicle-input-string'                  - String inserted by `C-='
* `icicle-inter-candidates-min-spaces'   - Min spaces among candidates
* `icicle-isearch-complete-keys'         - Keys to complete search
* `icicle-key-complete-keys'             - Keys to complete keys
* `icicle-key-descriptions-use-<>-flag'  - Show key names with "<>"?
* `icicle-keymaps-for-key-completion'    - `S-TAB' = key-complete maps
* `icicle-kmacro-ring-max'               - Icicles `kmacro-ring-max'
* `icicle-levenshtein-distance'          - Levenshtein match distance
* `icicle-list-join-string'              - Multi-completion join
* `icicle-list-nth-parts-join-string'    - Join split-candidate parts
* `icicle-mark-position-in-candidate'    - Mark position in cycling
* `icicle-menu-items-to-history-flag'    - Add menus to history?
* `icicle-minibuffer-key-bindings'       - general minibuffer keys
* `icicle-minibuffer-setup-hook'         - Functions run after setup
* `icicle-modal-cycle-*-keys'            - Keys for modal cycling
* `icicle-option-type-prefix-arg-list'   - Prefix-args for `C-h C-o'
* `icicle-point-position-in-candidate'   - Cursor position in cycling
* `icicle-populate-interactive-history-flag'- Track interactive use?
* `icicle-pp-eval-expression-print-*'    - Print control for `pp-*'
* `icicle-prefix-complete-keys*'         - Keys to prefix-complete
* `icicle-prefix-cycle-*-keys'           - Keys to prefix-cycle
* `icicle-quote-shell-file-name-flag'    - Quote file name in shell?
* `icicle-read+insert-file-name-keys'    - Keys for on-demand file
* `icicle-regexp-quote-flag'             - Escape chars? (`C-`')
* `icicle-regexp-search-ring-max'        - `regexp-search-ring-max'
* `icicle-region-background'             - Background for region
* `icicle-require-match-flag'            - Override REQUIRE-MATCH?
* `icicle-saved-completion-sets'         - Completion sets for `C-M-<'
* `icicle-search-cleanup-flag'           - Remove search highlighting?
                                           (`C-.')
* `icicle-search-from-isearch-keys'      - Isearch-to-Icicles keys
* `icicle-search-highlight-all-current-flag'- In each hit (`C-^')
* `icicle-search-highlight-context-levels-flag' -
                                           Highlight match subgroups?
* `icicle-search-highlight-threshold'    - # hits to highlight at once
* `icicle-search-hook'                   - Functions run by `C-c `'
* `icicle-search-replace-common-match-flag' - Replace ECM? (`M-;')
* `icicle-search-replace-literally-flag' - Replace text literally?
* `icicle-search-replace-whole-candidate-flag' - Replace input match
                                           or whole search hit?(`M-_')
* `icicle-search-ring-max'               - Icicles `search-ring-max'
* `icicle-search-whole-word-flag'        - Find whole words? (`M-q')
* `icicle-show-Completions-help-flag'    - Show `*Completions*' help?
* `icicle-show-Completions-initially-flag'- Show `*Completions*' 1st?
* `icicle-show-multi-completion-flag'    - Show extra candidate info?
* `icicle-sort-comparer'                 - Sort candidates (`C-,')
* `icicle-sort-orders-alist'             - Predicates for sorting
* `icicle-special-candidate-regexp'      - To highlight special cands
* `icicle-S-TAB-completion-methods-alist'- `S-TAB' methods (`M-(')
* `icicle-swank-*'                       - Swank completion control
* `icicle-TAB-completion-methods'        - `TAB' methods (`C-(')
* `icicle-TAB-shows-candidates-flag'     - 1st `TAB' shows candidates?
* `icicle-test-for-remote-files-flag'    - Check remote files? (`C-^')
* `icicle-thing-at-point-functions'      - Functions to yank things
* `icicle-top-level-key-bindings'        - Top-level key bindings
* `icicle-top-level-when-sole-completion-*'- Exiting if one completion
* `icicle-touche-pas-aux-menus-flag'     - Add to standard menus?
* `icicle-transform-function'            - Remove duplicates (`C-$')
* `icicle-type-actions-alist'            - Objects and their types
* `icicle-unpropertize-completion-result-flag'- Properties in result?
* `icicle-update-input-hook'             - Fns run when input changes
* `icicle-use-~-for-home-dir-flag'       - Use `~' for $HOME? (`M-~')
* `icicle-use-C-for-actions-flag'        - `C-' for actions? (`M-g')
* `icicle-use-candidates-only-once-flag' - Remove used candidate?
* `icicle-word-completion-keys'          - Keys for word completion
* `icicle-WYSIWYG-Completions-flag'      - WYSIWYG `*Completions*'?
* `icicle-yank-function'                 - Yank function to use

Faces that highlight input in minibuffer.

* `icicle-complete-input'               - Input when it is complete
* `icicle-completion'                   - Completing?
* `icicle-input-completion-fail*'       - Non-match part of input
* `icicle-match-highlight-minibuffer'   - Matched part of input
* `icicle-multi-command-completion'     - Multi-command completion?
* `icicle-mustmatch-completion'         - Strict completion?
* `icicle-whitespace-highlight'         - Initial whitespace in input

Faces that highlight candidates in buffer `*Completions*'.

* `icicle-candidate-part'               - Part of candidate
* `icicle-common-match-highlight-Completions' - Max common substring
* `icicle-current-candidate-highlight'  - Current candidate (cycling)
* `icicle-extra-candidate'              - Extra candidate
* `icicle-historical-candidate'         - Highlight candidates used
* `icicle-match-highlight-Completions'  - Matched part of input
* `icicle-proxy-candidate'              - Proxy candidate
* `icicle-saved-candidate'              - Saved candidate
* `icicle-special-candidate'            - Special candidate

Faces that highlight information in the mode line.

* `icicle-completion'                   - Completing?
* `icicle-mode-line-help'               - Candidate help
* `icicle-multi-command-completion'     - Multi-command completion?
* `icicle-mustmatch-completion'         - Strict completion?

Faces that highlight for command `icicle-search'.

* `icicle-search-context-level-*'       - Regexp subgroup highlighting
* `icicle-search-current-input'         - What input matches
* `icicle-search-main-regexp-current'   - Current match of 1st regexp
* `icicle-search-main-regexp-others'    - Other matches of 1st regexp

Icicle mode defines many top-level commands.  For a list, see the
Commentary headers of files `icicles-cmd1.el' and `icicles-cmd2.el'.
\f

These are all of the top-level bindings in Icicle mode:

key             binding
---             -------

C-c             Prefix Command
C-h             Prefix Command
C-x             Prefix Command
ESC             Prefix Command
<S-f10>         icicle-complete-menu-bar
<S-f4>          icicle-kmacro
<f10>           lacarte-execute-menu-command
<pause>         icicle-switch-to/from-minibuffer
<remap>         Prefix Command

<remap> <abort-recursive-edit>  icicle-abort-recursive-edit
<remap> <apropos>               icicle-apropos
<remap> <apropos-command>       icicle-apropos-command
<remap> <apropos-user-option>   icicle-apropos-option
<remap> <apropos-value>         icicle-apropos-value
<remap> <apropos-zippy>         icicle-apropos-zippy
<remap> <bmkp-all-tags-jump>    icicle-bookmark-all-tags
<remap> <bmkp-all-tags-jump-other-window>
                                icicle-bookmark-all-tags-other-window
<remap> <bmkp-all-tags-regexp-jump>
                                icicle-bookmark-all-tags-regexp
<remap> <bmkp-all-tags-regexp-jump-other-window>
                                icicle-bookmark-all-tags-regexp-other-window
<remap> <bmkp-autofile-all-tags-jump>
                                icicle-bookmark-autofile-all-tags
<remap> <bmkp-autofile-all-tags-jump-other-window>
                                icicle-bookmark-autofile-all-tags-other-window
<remap> <bmkp-autofile-all-tags-regexp-jump>
                                icicle-bookmark-autofile-all-tags-regexp
<remap> <bmkp-autofile-all-tags-regexp-jump-other-window>
                                icicle-bookmark-autofile-all-tags-regexp-other-window
<remap> <bmkp-autofile-jump>    icicle-bookmark-autofile
<remap> <bmkp-autofile-jump-other-window>
                                icicle-bookmark-autofile-other-window
<remap> <bmkp-autofile-set>     icicle-bookmark-a-file
<remap> <bmkp-autofile-some-tags-jump>
                                icicle-bookmark-autofile-some-tags
<remap> <bmkp-autofile-some-tags-jump-other-window>
                                icicle-bookmark-autofile-some-tags-other-window
<remap> <bmkp-autofile-some-tags-regexp-jump>
                                icicle-bookmark-autofile-some-tags-regexp
<remap> <bmkp-autofile-some-tags-regexp-jump-other-window>
                                icicle-bookmark-autofile-some-tags-regexp-other-window
<remap> <bmkp-autonamed-jump>   icicle-bookmark-autonamed
<remap> <bmkp-autonamed-jump-other-window>
                                icicle-bookmark-autonamed-other-window
<remap> <bmkp-autonamed-this-buffer-jump>
                                icicle-bookmark-autonamed-this-buffer
<remap> <bmkp-bookmark-file-jump>
                                icicle-bookmark-bookmark-file
<remap> <bmkp-bookmark-list-jump>
                                icicle-bookmark-bookmark-list
<remap> <bmkp-bookmark-set-confirm-overwrite>
                                icicle-bookmark-cmd
<remap> <bmkp-desktop-jump>     icicle-bookmark-desktop
<remap> <bmkp-dired-jump>       icicle-bookmark-dired
<remap> <bmkp-dired-jump-other-window>
                                icicle-bookmark-dired-other-window
<remap> <bmkp-file-all-tags-jump>
                                icicle-bookmark-file-all-tags
<remap> <bmkp-file-all-tags-jump-other-window>
                                icicle-bookmark-file-all-tags-other-window
<remap> <bmkp-file-all-tags-regexp-jump>
                                icicle-bookmark-file-all-tags-regexp
<remap> <bmkp-file-all-tags-regexp-jump-other-window>
                                icicle-bookmark-file-all-tags-regexp-other-window
<remap> <bmkp-file-jump>        icicle-bookmark-file
<remap> <bmkp-file-jump-other-window>
                                icicle-bookmark-file-other-window
<remap> <bmkp-file-some-tags-jump>
                                icicle-bookmark-file-some-tags
<remap> <bmkp-file-some-tags-jump-other-window>
                                icicle-bookmark-file-some-tags-other-window
<remap> <bmkp-file-some-tags-regexp-jump>
                                icicle-bookmark-file-some-tags-regexp
<remap> <bmkp-file-some-tags-regexp-jump-other-window>
                                icicle-bookmark-file-some-tags-regexp-other-window
<remap> <bmkp-file-this-dir-all-tags-jump>
                                icicle-bookmark-file-this-dir-all-tags
<remap> <bmkp-file-this-dir-all-tags-jump-other-window>
                                icicle-bookmark-file-this-dir-all-tags-other-window
<remap> <bmkp-file-this-dir-all-tags-regexp-jump>
                                icicle-bookmark-file-this-dir-all-tags-regexp
<remap> <bmkp-file-this-dir-all-tags-regexp-jump-other-window>
                                icicle-bookmark-file-this-dir-all-tags-regexp-other-window
<remap> <bmkp-file-this-dir-jump>
                                icicle-bookmark-file-this-dir
<remap> <bmkp-file-this-dir-jump-other-window>
                                icicle-bookmark-file-this-dir-other-window
<remap> <bmkp-file-this-dir-some-tags-jump>
                                icicle-bookmark-file-this-dir-some-tags
<remap> <bmkp-file-this-dir-some-tags-jump-other-window>
                                icicle-bookmark-file-this-dir-some-tags-other-window
<remap> <bmkp-file-this-dir-some-tags-regexp-jump>
                                icicle-bookmark-file-this-dir-some-tags-regexp
<remap> <bmkp-file-this-dir-some-tags-regexp-jump-other-window>
                                icicle-bookmark-file-this-dir-some-tags-regexp-other-window
<remap> <bmkp-find-file>        icicle-find-file-handle-bookmark
<remap> <bmkp-find-file-all-tags>
                                icicle-find-file-all-tags
<remap> <bmkp-find-file-all-tags-other-window>
                                icicle-find-file-all-tags-other-window
<remap> <bmkp-find-file-all-tags-regexp>
                                icicle-find-file-all-tags-regexp
<remap> <bmkp-find-file-all-tags-regexp-other-window>
                                icicle-find-file-all-tags-regexp-other-window
<remap> <bmkp-find-file-other-window>
                                icicle-find-file-handle-bookmark-other-window
<remap> <bmkp-find-file-some-tags>
                                icicle-find-file-some-tags
<remap> <bmkp-find-file-some-tags-other-window>
                                icicle-find-file-some-tags-other-window
<remap> <bmkp-find-file-some-tags-regexp>
                                icicle-find-file-some-tags-regexp
<remap> <bmkp-find-file-some-tags-regexp-other-window>
                                icicle-find-file-some-tags-regexp-other-window
<remap> <bmkp-gnus-jump>        icicle-bookmark-gnus
<remap> <bmkp-gnus-jump-other-window>
                                icicle-bookmark-gnus-other-window
<remap> <bmkp-image-jump>       icicle-bookmark-image
<remap> <bmkp-image-jump-other-window>
                                icicle-bookmark-image-other-window
<remap> <bmkp-info-jump>        icicle-bookmark-info
<remap> <bmkp-info-jump-other-window>
                                icicle-bookmark-info-other-window
<remap> <bmkp-local-file-jump>  icicle-bookmark-local-file
<remap> <bmkp-local-file-jump-other-window>
                                icicle-bookmark-local-file-other-window
<remap> <bmkp-man-jump>         icicle-bookmark-man
<remap> <bmkp-man-jump-other-window>
                                icicle-bookmark-man-other-window
<remap> <bmkp-non-file-jump>    icicle-bookmark-non-file
<remap> <bmkp-non-file-jump-other-window>
                                icicle-bookmark-non-file-other-window
<remap> <bmkp-region-jump>      icicle-bookmark-region
<remap> <bmkp-region-jump-other-window>
                                icicle-bookmark-region-other-window
<remap> <bmkp-remote-file-jump>
                                icicle-bookmark-remote-file
<remap> <bmkp-remote-file-jump-other-window>
                                icicle-bookmark-remote-file-other-window
<remap> <bmkp-some-tags-jump>   icicle-bookmark-some-tags
<remap> <bmkp-some-tags-jump-other-window>
                                icicle-bookmark-some-tags-other-window
<remap> <bmkp-some-tags-regexp-jump>
                                icicle-bookmark-some-tags-regexp
<remap> <bmkp-some-tags-regexp-jump-other-window>
                                icicle-bookmark-some-tags-regexp-other-window
<remap> <bmkp-specific-buffers-jump>
                                icicle-bookmark-specific-buffers
<remap> <bmkp-specific-buffers-jump-other-window>
                                icicle-bookmark-specific-buffers-other-window
<remap> <bmkp-specific-files-jump>
                                icicle-bookmark-specific-files
<remap> <bmkp-specific-files-jump-other-window>
                                icicle-bookmark-specific-files-other-window
<remap> <bmkp-tag-a-file>       icicle-tag-a-file
<remap> <bmkp-temporary-jump>   icicle-bookmark-temporary
<remap> <bmkp-temporary-jump-other-window>
                                icicle-bookmark-temporary-other-window
<remap> <bmkp-this-buffer-jump>
                                icicle-bookmark-this-buffer
<remap> <bmkp-this-buffer-jump-other-window>
                                icicle-bookmark-this-buffer-other-window
<remap> <bmkp-untag-a-file>     icicle-untag-a-file
<remap> <bmkp-url-jump>         icicle-bookmark-url
<remap> <bmkp-url-jump-other-window>
                                icicle-bookmark-url-other-window
<remap> <bmkp-w3m-jump>         icicle-bookmark-w3m
<remap> <bmkp-w3m-jump-other-window>
                                icicle-bookmark-w3m-other-window
<remap> <bookmark-jump>         icicle-bookmark
<remap> <bookmark-jump-other-window>
                                icicle-bookmark-other-window
<remap> <bookmark-set>          icicle-bookmark-cmd
<remap> <customize-apropos>     icicle-customize-apropos
<remap> <customize-apropos-faces>
                                icicle-customize-apropos-faces
<remap> <customize-apropos-groups>
                                icicle-customize-apropos-groups
<remap> <customize-apropos-options>
                                icicle-customize-apropos-options
<remap> <customize-face>        icicle-customize-face
<remap> <customize-face-other-window>
                                icicle-customize-face-other-window
<remap> <dabbrev-completion>    icicle-dabbrev-completion
<remap> <delete-window>         icicle-delete-window
<remap> <delete-windows-for>    icicle-delete-window
<remap> <describe-package>      icicle-describe-package
<remap> <eval-expression>       icicle-pp-eval-expression
<remap> <exchange-point-and-mark>
                                icicle-exchange-point-and-mark
<remap> <execute-extended-command>
                                icicle-execute-extended-command
<remap> <find-file>             icicle-file
<remap> <find-file-other-window>
                                icicle-file-other-window
<remap> <find-file-read-only>   icicle-find-file-read-only
<remap> <find-file-read-only-other-window>
                                icicle-find-file-read-only-other-window
<remap> <find-tag>              icicle-find-tag
<remap> <find-tag-other-window>
                                icicle-find-first-tag-other-window
<remap> <insert-buffer>         icicle-insert-buffer
<remap> <kill-buffer>           icicle-kill-buffer
<remap> <kill-buffer-and-its-windows>
                                icicle-kill-buffer
<remap> <load-library>          icicle-load-library
<remap> <minibuffer-keyboard-quit>
                                icicle-abort-recursive-edit
<remap> <other-window>          icicle-other-window-or-frame
<remap> <other-window-or-frame>
                                icicle-other-window-or-frame
<remap> <pop-global-mark>       icicle-goto-global-marker-or-pop-global-mark
<remap> <pop-tag-mark>          icicle-pop-tag-mark
<remap> <pp-eval-expression>    icicle-pp-eval-expression
<remap> <repeat-complex-command>
                                icicle-repeat-complex-command
<remap> <set-mark-command>      icicle-goto-marker-or-set-mark-command
<remap> <switch-to-buffer>      icicle-buffer
<remap> <switch-to-buffer-other-window>
                                icicle-buffer-other-window
<remap> <where-is>              icicle-where-is
<remap> <yank>                  icicle-yank-maybe-completing
<remap> <yank-pop>              icicle-yank-pop-commands
<remap> <yank-pop-commands>     icicle-yank-pop-commands
<remap> <zap-to-char>           icicle-zap-to-char

C-h C-o         icicle-describe-option-of-type

ESC ESC         Prefix Command
M-`             lacarte-execute-menu-command
M-s             Prefix Command
C-M-/           icicle-dispatch-C-M-/

M-ESC C-x       icicle-command-abbrev
M-ESC x         lacarte-execute-command

M-s ESC         Prefix Command

C-x ESC         Prefix Command
C-x 4           Prefix Command
C-x 5           Prefix Command
C-x j           Prefix Command

C-c "           icicle-search-text-property
C-c $           icicle-search-word
C-c '           icicle-occur
C-c /           icicle-complete-thesaurus-entry
C-c =           icicle-imenu
C-c ^           icicle-search-keywords
C-c `           icicle-search-generic

M-s M-s         Prefix Command

M-s M-s C-l     icicle-search-pages
M-s M-s ESC     Prefix Command
M-s M-s ,       icicle-tags-search
M-s M-s D       icicle-search-defs-full
M-s M-s I       icicle-imenu-full
M-s M-s J       icicle-search-bookmarks-together
M-s M-s O       icicle-search-overlay-property
M-s M-s T       icicle-search-text-property
M-s M-s X       icicle-search-xml-element-text-node
M-s M-s b       icicle-search-buffer
M-s M-s c       icicle-search-char-property
M-s M-s d       icicle-search-defs
M-s M-s f       icicle-search-file
M-s M-s g       icicle-grep-saved-file-candidates
M-s M-s i       icicle-imenu
M-s M-s j       icicle-search-bookmark
M-s M-s k       icicle-search-keywords
M-s M-s l       icicle-search-lines
M-s M-s o       icicle-occur
M-s M-s p       icicle-search-paragraphs
M-s M-s s       icicle-search-sentences
M-s M-s t       icicle-search-thing
M-s M-s w       icicle-search-word
M-s M-s x       icicle-search-xml-element

C-x 4 j         Prefix Command

C-x j t         Prefix Command

C-x 5 o         icicle-select-frame

C-x M-e         icicle-execute-named-keyboard-macro

M-s M-s M-s     icicle-search-generic

C-x 4 j t       Prefix Command

C-x j t C-f     Prefix Command
C-x j t j       icicle-bookmark-tagged

C-x 4 j t C-f   Prefix Command
C-x 4 j t j     icicle-bookmark-tagged-other-window

C-x j t C-f C-f                 icicle-find-file-tagged

C-x 4 j t C-f C-f               icicle-find-file-tagged-other-window



These are all of the minibuffer bindings during completion:

key             binding
---             -------

C-a             icicle-beginning-of-line+
C-e             icicle-end-of-line+
C-g             icicle-abort-recursive-edit
TAB             icicle-prefix-complete
C-j             icicle-insert-newline-in-minibuffer
C-l             icicle-retrieve-previous-input
RET             exit-minibuffer
C-v             icicle-scroll-Completions-forward
C-w             icicle-kill-region
C-x             Prefix Command
ESC             Prefix Command
C-^             icicle-dispatch-C-^
SPC             icicle-self-insert
.               icicle-insert-dot-command
?               icicle-self-insert
C-S-a           icicle-toggle-case-sensitivity
C-S-l           icicle-retrieve-next-input
S-SPC           icicle-apropos-complete-and-narrow
C-!             icicle-all-candidates-action
C-"             icicle-toggle-expand-to-common-match
C-#             icicle-cycle-incremental-completion
C-$             icicle-toggle-transforming
C-%             icicle-candidate-set-swap
C-(             icicle-next-TAB-completion-method
C-)             icicle-candidate-set-save-more-selected
C-*             icicle-candidate-set-intersection
C-+             icicle-candidate-set-union
C-,             icicle-change-sort-order
C--             icicle-candidate-set-difference
C-.             icicle-dispatch-C-.
C-:             icicle-candidate-set-define
C-<             icicle-candidate-set-retrieve-more
C-=             icicle-insert-string-from-variable
C->             icicle-candidate-set-save-more
C-`             icicle-toggle-regexp-quote
C-{             icicle-candidate-set-retrieve-persistent
C-|             icicle-all-candidates-alt-action
C-}             icicle-candidate-set-save-persistently
C-~             icicle-candidate-set-complement
<C-M-S-TAB>             icicle-apropos-complete-no-display
<C-M-S-tab>                     icicle-apropos-complete-no-display
<C-M-down>      icicle-next-candidate-per-mode-help
<C-M-end>       icicle-help-on-next-prefix-candidate
<C-M-f1>        icicle-help-on-candidate
<C-M-help>      icicle-help-on-candidate
<C-M-home>      icicle-help-on-previous-prefix-candidate
<C-M-next>      icicle-help-on-next-apropos-candidate
<C-M-pause>     icicle-other-history
<C-M-prior>     icicle-help-on-previous-apropos-candidate
<C-M-return>    icicle-help-on-candidate
<C-M-tab>       icicle-prefix-complete-no-display
<C-M-up>        icicle-previous-candidate-per-mode-help
<C-M-wheel-down>                icicle-next-candidate-per-mode-help
<C-M-wheel-up>                  icicle-previous-candidate-per-mode-help
<C-S-down>      icicle-next-candidate-per-mode-alt-action
<C-S-end>       icicle-next-prefix-candidate-alt-action
<C-S-home>      icicle-previous-prefix-candidate-alt-action
<C-S-next>      icicle-next-apropos-candidate-alt-action
<C-S-pause>     icicle-toggle-WYSIWYG-Completions
<C-S-prior>     icicle-previous-apropos-candidate-alt-action
<C-S-return>    icicle-candidate-alt-action
<C-S-tab>       icicle-toggle-completion-mode-keys
<C-S-up>        icicle-previous-candidate-per-mode-alt-action
<C-S-wheel-down>                icicle-next-candidate-per-mode-alt-action
<C-S-wheel-up>                  icicle-previous-candidate-per-mode-alt-action
<C-down>        icicle-next-candidate-per-mode-action
<C-end>         icicle-next-prefix-candidate-action
<C-f1>          icicle-help-on-candidate
<C-help>        icicle-help-on-candidate
<C-home>        icicle-previous-prefix-candidate-action
<C-insert>      icicle-switch-to-Completions-buf
<C-next>        icicle-next-apropos-candidate-action
<C-pause>       icicle-toggle-highlight-historical-candidates
<C-prior>       icicle-previous-apropos-candidate-action
<C-return>      icicle-candidate-action
<C-tab>         file-cache-minibuffer-complete
<C-up>          icicle-previous-candidate-per-mode-action
<C-wheel-down>  icicle-next-candidate-per-mode-action
<C-wheel-up>    icicle-previous-candidate-per-mode-action
<M-S-backspace>                 icicle-erase-minibuffer
<M-S-delete>                    icicle-erase-minibuffer
<M-backtab>                     icicle-complete-keys
<M-pause>       icicle-keep-only-past-inputs
<M-return>      icicle-candidate-read-fn-invoke
<M-up>          1on1-fit-minibuffer-frame
<S-backspace>   icicle-apropos-complete-and-widen
<S-delete>      icicle-delete-candidate-object
<S-pause>       icicle-toggle-highlight-saved-candidates
<S-return>      icicle-apropos-complete-and-exit
<XF86Back>      previous-history-element
<XF86Forward>   next-history-element
<backtab>       icicle-apropos-complete
<delete>        icicle-remove-candidate
<down>          icicle-next-candidate-per-mode
<end>           icicle-next-prefix-candidate
<home>          icicle-previous-prefix-candidate
<icicle-is-completion-map>      ignore
<insert>        icicle-save/unsave-candidate
<next>          icicle-next-apropos-candidate
<nil>           Prefix Command
<prior>         icicle-previous-apropos-candidate
<remap>         Prefix Command
<tab>           icicle-prefix-complete
<up>            icicle-previous-candidate-per-mode
<wheel-down>    icicle-next-candidate-per-mode
<wheel-up>      icicle-previous-candidate-per-mode

C-x C-a         icicle-toggle-annotation
C-x ESC         Prefix Command
C-x #           icicle-doremi-increment-max-candidates+
C-x -           icicle-doremi-zoom-Completions+
C-x .           icicle-dispatch-C-x.
C-x /           icicle-toggle-expand-directory
C-x :           icicle-toggle-network-drives-as-remote
C-x t           icicle-cycle-image-file-thumbnail
C-x w           icicle-doremi-candidate-width-factor+
C-x |           icicle-doremi-inter-candidates-min-spaces+
C-x C-0         icicle-recomplete-from-original-domain
C-x C-<         bmkp-retrieve-more-icicle-search-hits

<nil> <C-M-wheel-down>          icicle-next-candidate-per-mode-help
<nil> <C-M-wheel-up>            icicle-previous-candidate-per-mode-help
<nil> <C-S-wheel-down>          icicle-next-candidate-per-mode-alt-action
<nil> <C-S-wheel-up>            icicle-previous-candidate-per-mode-alt-action
<nil> <C-wheel-down>            icicle-next-candidate-per-mode-action
<nil> <C-wheel-up>              icicle-previous-candidate-per-mode-action
<nil> <wheel-down>              icicle-next-candidate-per-mode
<nil> <wheel-up>                icicle-previous-candidate-per-mode

<remap> <backward-delete-char-untabify>
                                icicle-backward-delete-char-untabify
<remap> <backward-kill-paragraph>
                                icicle-backward-kill-paragraph
<remap> <backward-kill-sentence>
                                icicle-backward-kill-sentence
<remap> <backward-kill-sexp>    icicle-backward-kill-sexp
<remap> <backward-kill-word>    icicle-backward-kill-word
<remap> <delete-backward-char>  icicle-delete-backward-char
<remap> <delete-char>           icicle-delete-char
<remap> <digit-argument>        icicle-digit-argument
<remap> <kill-line>             icicle-kill-line
<remap> <kill-paragraph>        icicle-kill-paragraph
<remap> <kill-sexp>             icicle-kill-sexp
<remap> <kill-word>             icicle-kill-word
<remap> <mouse-yank-secondary>  icicle-mouse-yank-secondary
<remap> <negative-argument>     icicle-negative-argument
<remap> <reposition-window>     icicle-goto/kill-failed-input
<remap> <self-insert-command>   icicle-self-insert
<remap> <transpose-chars>       icicle-transpose-chars
<remap> <transpose-sexps>       icicle-transpose-sexps
<remap> <transpose-words>       icicle-transpose-words
<remap> <universal-argument>    icicle-universal-argument
<remap> <yank-pop>              icicle-yank-pop

C-M-j           icicle-insert-list-join-string
M-RET           icicle-candidate-read-fn-invoke
C-M-^           icicle-toggle-completions-format
C-M-_           icicle-toggle-proxy-candidates
M-SPC           icicle-prefix-word-complete
M-!             icicle-all-candidates-list-action
M-$             icicle-candidate-set-truncate
M-%             icicle-regexp-quote-input
M-&             icicle-narrow-candidates-with-predicate
M-(             icicle-next-S-TAB-completion-method
M-*             icicle-narrow-candidates
M-+             icicle-widen-candidates
M-,             icicle-dispatch-M-comma
M-;             icicle-toggle-search-replace-common-match
M-_             icicle-dispatch-M-_
M-g             icicle-toggle-C-for-actions
M-h             icicle-history
M-i             icicle-toggle-map
M-m             icicle-toggle-show-multi-completion
M-q             icicle-dispatch-M-q
M-r             icicle-roundup
M-v             icicle-scroll-Completions-backward
M-|             icicle-all-candidates-list-alt-action
M-~             icicle-toggle-~-for-home-dir
C-M-"           icicle-cycle-expand-to-common-match
C-M-#           icicle-toggle-icomplete-mode
C-M-&           icicle-save-predicate-to-variable
C-M-(           icicle-next-completion-style-set
C-M-)           icicle-candidate-set-save-selected
C-M-+           icicle-plus-saved-sort
C-M-,           icicle-toggle-alternative-sorting
C-M-.           icicle-toggle-dot
C-M-;           icicle-toggle-ignoring-comments
C-M-<           icicle-candidate-set-retrieve
C-M->           icicle-candidate-set-save
C-M-`           icicle-toggle-literal-replacement
C-M-{           icicle-candidate-set-retrieve-from-variable
C-M-}           icicle-candidate-set-save-to-variable
C-M-~           icicle-toggle-search-complementing-domain

M-i TAB         icicle-toggle-completion-mode-keys
M-i ESC         Prefix Command
M-i "           icicle-toggle-expand-to-common-match
M-i #           icicle-cycle-incremental-completion
M-i $           icicle-toggle-transforming
M-i ,           icicle-toggle-sorting
M-i .           icicle-dispatch-C-.
M-i /           icicle-toggle-expand-directory
M-i :           icicle-toggle-network-drives-as-remote
M-i ;           icicle-toggle-ignoring-comments
M-i <           icicle-toggle-angle-brackets
M-i A           icicle-toggle-case-sensitivity
M-i F           icicle-toggle-include-cached-files
M-i ^           icicle-dispatch-C-^
M-i _           icicle-dispatch-M-_
M-i `           icicle-toggle-literal-replacement
M-i a           icicle-toggle-annotation
M-i g           icicle-toggle-C-for-actions
M-i h           icicle-dispatch-C-x.
M-i m           icicle-toggle-show-multi-completion
M-i p           icicle-toggle-proxy-candidates
M-i q           icicle-dispatch-M-q
M-i r           icicle-toggle-include-recent-files
M-i s           icicle-toggle-highlight-saved-candidates
M-i t           icicle-cycle-image-file-thumbnail
M-i w           icicle-toggle-WYSIWYG-Completions
M-i ~           icicle-toggle-search-complementing-domain
M-i C-`         icicle-toggle-regexp-quote
M-i <backtab>   icicle-complete-keys
M-i <pause>     icicle-toggle-highlight-historical-candidates

C-x C-f         icicle-resolve-file-name

C-M-v           icicle-scroll-forward
C-M-y           icicle-yank-secondary
M-.             icicle-insert-string-at-point
M-:             icicle-pp-eval-expression-in-minibuffer
M-?             icicle-minibuffer-help
M-K             icicle-clear-current-history
M-R             icicle-multi-inputs-act
M-S             icicle-multi-inputs-save
M-k             icicle-erase-minibuffer-or-history-element
M-n             next-history-element
M-o             icicle-insert-history-element
M-p             previous-history-element
M-r             previous-matching-history-element
  (that binding is currently shadowed by another mode)
M-s             next-matching-history-element
C-M-S-c         icicle-completing-read+insert
C-M-S-f         icicle-read+insert-file-name
C-M-S-t         icicle-top-level
C-M-S-v         icicle-scroll-backward
ESC <M-backtab>                 icicle-complete-keys

C-x C-M-l       icicle-display-candidates-in-Completions
C-x C-M-<       bmkp-retrieve-icicle-search-hits
C-x C-M->       bmkp-set-icicle-search-hits-bookmark

M-i M-"         icicle-cycle-expand-to-common-match
M-i M-#         icicle-toggle-icomplete-mode
M-i M-,         icicle-toggle-alternative-sorting
M-i M-.         icicle-toggle-dot
M-i M-;         icicle-toggle-search-replace-common-match
M-i M-^         icicle-toggle-completions-format
M-i M-i         icicle-toggle-option
M-i M-~         icicle-toggle-~-for-home-dir
M-i ESC <backtab>               icicle-complete-keys
______________________________________________________________________

Send an Icicles bug report: `M-x icicle-send-bug-report'.

[Icicles Help on the Web]                        [Icicles Doc, Part 1]
[Icicles Options & Faces]                        [Icicles Doc, Part 2]

RE: info-find-source

[Drew Adams]

> Info has it's own keybindings.  That can be a little tricky,
> since "s" is the normal way to search rather than "C-s" or "M-s".

`C-s' and `C-M-s' are normal ways of searching in Info.

In the beginning there was only `s': `C-s' searched only
the current node, but `s' searched across nodes.  As of
several releases ago `C-s' and `C-M-s' work fine across nodes.

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

>> Info has it's own keybindings. That can be
>> a little tricky, since "s" is the normal way
>> to search rather than "C-s" or "M-s".
>> You can rebind these functions though.
>
> I didn't know that!

To quote the late Cipher, dreaming of
re-entering the Matrix without remembering
anything, "ignorance is a bliss".

> Then, maybe a year or two ago, I learned
> about the index and the `i' keystroke, and
> now I can't understand how I used
> Info before.

Unnecessary in text-mode, of course.

> I would really like it if someone set up
> a contest of "how fast can someone use Info
> to search for things and come back to other
> things and not get distracted by links and/or
> writing one's own functions to do what you
> can do with stock Emacs better", and made
> Emanuel and me the contestants.

If so, I will not only defeat you. I will
punish you :)

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Marcin Borkowski]

On 2018-01-14, at 03:57, Emanuel Berg <moasen@zoho.com> wrote:

> Marcin Borkowski wrote:
>
>>> Info has it's own keybindings. That can be
>>> a little tricky, since "s" is the normal way
>>> to search rather than "C-s" or "M-s".
>>> You can rebind these functions though.
>>
>> I didn't know that!
>
> To quote the late Cipher, dreaming of
> re-entering the Matrix without remembering
> anything, "ignorance is a bliss".
>
>> Then, maybe a year or two ago, I learned
>> about the index and the `i' keystroke, and
>> now I can't understand how I used
>> Info before.
>
> Unnecessary in text-mode, of course.

Seriously?  Try C-s'ing the string "interactive" your way to find out
about the (interactive) codes.

In Emacs Lisp Reference:

C-x n w
M-x how-many RET
interactive RET

result: 515 hits.  The "right" one is about 1/3 down the list.

>> I would really like it if someone set up
>> a contest of "how fast can someone use Info
>> to search for things and come back to other
>> things and not get distracted by links and/or
>> writing one's own functions to do what you
>> can do with stock Emacs better", and made
>> Emanuel and me the contestants.
>
> If so, I will not only defeat you. I will
> punish you :)

Wouldn't be so sure - see above. ;-)

Best,

--
Marcin Borkowski

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

> Seriously? Try C-s'ing the string
> "interactive" your way to find out about the
> (interactive) codes.
>
> In Emacs Lisp Reference:
>
> C-x n w M-x how-many RET interactive RET
>
> result: 515 hits. The "right" one is about
> 1/3 down the list.

This is the nature of this kind of search.
If you are searching for an item that appears
several times, and an instance somewhere in the
middle is the desired one, then yes, search
will have to visit each preceding instance if
it starts at the beginning of the document.

This situation is a part of, but not restricted
to, the described "one file" method.

However in time you will be more sophisticated
with your searches. It'll also give you
a spacial awareness (spacial as in "space", not
special) of the document and what it contains,
which will benefit searching - your use of it -
even more.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Marcin Borkowski]

On 2018-01-15, at 05:17, Emanuel Berg <moasen@zoho.com> wrote:

> Marcin Borkowski wrote:
>
>> Seriously? Try C-s'ing the string
>> "interactive" your way to find out about the
>> (interactive) codes.
>>
>> In Emacs Lisp Reference:
>>
>> C-x n w M-x how-many RET interactive RET
>>
>> result: 515 hits. The "right" one is about
>> 1/3 down the list.
>
> This is the nature of this kind of search.
> If you are searching for an item that appears
> several times, and an instance somewhere in the
> middle is the desired one, then yes, search
> will have to visit each preceding instance if
> it starts at the beginning of the document.
>
> This situation is a part of, but not restricted
> to, the described "one file" method.
>
> However in time you will be more sophisticated
> with your searches. It'll also give you
> a spacial awareness (spacial as in "space", not
> special) of the document and what it contains,
> which will benefit searching - your use of it -
> even more.

Yes, I could train myself to do that.

But...  I have `i' in Info!  So why train myself to do something my
computer (using the information Emacs developers put in) can do better?

Best,

--
Marcin Borkowski

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

>> This is the nature of this kind of search.
>> If you are searching for an item that
>> appears several times, and an instance
>> somewhere in the middle is the desired one,
>> then yes, search will have to visit each
>> preceding instance if it starts at the
>> beginning of the document. This situation is
>> a part of, but not restricted to, the
>> described "one file" method. However in time
>> you will be more sophisticated with your
>> searches. It'll also give you a spacial
>> awareness (spacial as in "space", not
>> special) of the document and what it
>> contains, which will benefit searching -
>> your use of it - even more.
>
> Yes, I could train myself to do that.
>
> But... I have `i' in Info! So why train
> myself to do something my computer (using the
> information Emacs developers put in) can
> do better?

Because you have already trained yourself to
edit text and code and that happens every day
from now on as well. No matter how much you
train with Info, you will never get to the
text/code level. And when you "train" with text
and code, you do amazing stuff - well,
hopefully, but almost certainly something more
interesting than how to navigate a browser...

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Robert Thorpe]

Emanuel Berg <moasen@zoho.com> writes:

> Marcin Borkowski wrote:
>
....
>> Yes, I could train myself to do that.
>>
>> But... I have `i' in Info! So why train
>> myself to do something my computer (using the
>> information Emacs developers put in) can
>> do better?
>
> Because you have already trained yourself to
> edit text and code and that happens every day
> from now on as well. No matter how much you
> train with Info, you will never get to the
> text/code level. And when you "train" with text
> and code, you do amazing stuff - well,
> hopefully, but almost certainly something more
> interesting than how to navigate a browser...

I agree with Marcin Borkowski.

I think it's a mistake to over-emphasise the editing modes, and their
keybindings.  The viewing modes and the special modes are just as
important.

I spend a great deal of time in Dired, Info, Help and reading mail.  I
also spend a fair amount of time in View, Occur, Grep, Find, Compile and
Shell.

In my experience it's worth becoming reasonably familiar with those
modes and their keybindings.  It's true that doing that means less
practice with the normal editing keybindings.  However, as you get
better at using those, each increment of extra skill comes more slowly.
You reach a point of diminishing returns.  At that stage it's worthwhile
to pay more attention to the special modes.

I don't expect you to necessarily take my advice.  I'm just giving my
opinion.

BR,
Robert Thorpe

Re: info-find-source

[Marcin Borkowski]

On 2018-01-15, at 20:55, Emanuel Berg <moasen@zoho.com> wrote:

> [...]  No matter how much you
> train with Info, you will never get to the
> text/code level.  [...]

Wrong.  Info is a special mode, so uses keys without modifiers like...
"escape-meta-alt-control-shift";-).  That's faster.  (And operable with
one hand!)

Best,

-- 
Marcin Borkowski

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

>> No matter how much you train with Info, you
>> will never get to the text/code level.
>
> Wrong. Info is a special mode, so uses keys
> without modifiers like...
> "escape-meta-alt-control-shift";-).
> That's faster. (And operable with one hand!)

Perhaps you should look into your editing
mode keybindings?

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Marcin Borkowski]

On 2018-01-19, at 08:12, Emanuel Berg <moasen@zoho.com> wrote:

> Marcin Borkowski wrote:
>
>>> No matter how much you train with Info, you
>>> will never get to the text/code level.
>>
>> Wrong. Info is a special mode, so uses keys
>> without modifiers like...
>> "escape-meta-alt-control-shift";-).
>> That's faster. (And operable with one hand!)
>
> Perhaps you should look into your editing
> mode keybindings?

In editing mode, `a' to `z' run self-insrt-command.  That's enough.
(Unless you use some kind of Vi(m) emulation, that's another story.)

Best,

-- 
Marcin Borkowski

RE: info-find-source

[Drew Adams]

BTW/FWIW, about using Info in "editing mode" or not,
here's some trivia from the past...

Info mode used to bind command `Info-edit' to `e'.

That command lets you edit Info nodes, so you can
easily update or annotate your version of a given
manual, e.g., to correct or explain something to
yourself or add more information.

The command was declared "obsolete" in Emacs 24.4.
(Personally, I see no point in that.)  The reason
given is told by `C-h f':

   This function is obsolete since 24.4;
   editing Info nodes by hand is not recommended.

   Edit the contents of this Info node.

And `e' now just takes you to the end of the `*info*'
buffer.  (On n'arrete pas le progres.)

Re: info-find-source

[Emanuel Berg]

Drew Adams wrote:

> This function is obsolete since 24.4; editing
> Info nodes by hand is not recommended.

I've seen this as well and wondered what it
meant... "By hand" - is the recommended way
having a bunch of text files and have that
compiled, or should one generate the
documentation straight from Elisp, including
docstrings and/or comments?

Also, when you could edit Info nodes that way,
where did those edits go? Using the useful and
sensible program that was the initial topic of
this thread, we find out that for example Gnus'
is here

    /usr/share/info/emacs-24/gnus.info

while SLIME is here

    ~/.emacs.d/elpa/slime-20170921.1000/slime.info

so while the SLIME's manual could be annotated
without any hazard, how will this be done with
Gnus' without sudo right and without the
changes/annotations being threatened
by updates?

-- 
underground experts united
http://user.it.uu.se/~embe8573

RE: info-find-source

[Drew Adams]

> > This function is obsolete since 24.4; editing
> > Info nodes by hand is not recommended.
> 
> I've seen this as well and wondered what it
> meant... "By hand" - is the recommended way
> having a bunch of text files and have that
> compiled, or should one generate the
> documentation straight from Elisp, including
> docstrings and/or comments?

AFAIK, the "recommended" way is to build Info
files using `makeinfo' with TexInfo source files.
But I'm no expert on that.

> Also, when you could edit Info nodes that way,
> where did those edits go?

You can still edit Info nodes that way.  As I
said, command `Info-edit' still exists, and it
still works.

Try it.  You are asked to confirm whether you
really want to do it.  After you make your
changes, by editing normally, you use `C-c C-c'
and enter the name of the Info file you want
to write.

If you give the name of the current file (see
variable `Info-current-file', for instance)
then you are asked to confirm overwriting it.
If you instead give the name of a new file then
you create a new Info file.

In either case you provide an absolute file name,
so that's "where" your edits go.

> Using the useful and
> sensible program that was the initial topic of
> this thread, we find out that for example Gnus'
> is here /usr/share/info/emacs-24/gnus.info
> while SLIME is here:
> ~/.emacs.d/elpa/slime-20170921.1000/slime.info
> so while the SLIME's manual could be annotated
> without any hazard, how will this be done with
> Gnus' without sudo right and without the
> changes/annotations being threatened
> by updates?

If you change that slime.info file and then you
download the SLIME again to the same disk location
then that will overwrite your edits.

And yes, if you want to change that gnus.info
file then you need the necessary permissions
write to that location.  Nothing new here.

But you can have Info files wherever you want.
See option `Info-additional-directory-list',
and variables `Info-directory-list', and
`Info-default-directory-list'.  And see env
var `INFOPATH'.

To be clear, I'm not encouraging anyone to edit
Info files.  I'm just mentioning that you can.

Re: info-find-source

[Emanuel Berg]

Drew Adams wrote:

>>> This function is obsolete since 24.4;
>>> editing Info nodes by hand is
>>> not recommended.
>>
>> I've seen this as well and wondered what it
>> meant... "By hand" - is the recommended way
>> having a bunch of text files and have that
>> compiled, or should one generate the
>> documentation straight from Elisp, including
>> docstrings and/or comments?
>
> AFAIK, the "recommended" way is to build Info
> files using `makeinfo' with TexInfo source
> files.

Yes of course, it even says so first thing:

    This is slime.info, produced by makeinfo
    version 5.2 from slime.texi.

OK, so it is the *.info files* that one is
disencouraged from editing? Well yeah, why
would you do that?!

The "by hand" phrasing is where the confusion
begins because it seems to imply there is
a better way to "edit" them. But to me it seems
totally backward to edit the result of
compilation and I don't think I ever did that
because wouldn't not only update but also
recompilation overwrite the edits?

In general one should edit the source! and here
one could simply keep a text file with any
extra material.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Eli Zaretskii]

> From: Emanuel Berg <moasen@zoho.com>
> Date: Sat, 20 Jan 2018 20:49:22 +0100
> 
> The "by hand" phrasing is where the confusion
> begins because it seems to imply there is
> a better way to "edit" them. But to me it seems
> totally backward to edit the result of
> compilation and I don't think I ever did that
> because wouldn't not only update but also
> recompilation overwrite the edits?
> 
> In general one should edit the source! and here
> one could simply keep a text file with any
> extra material.

That command exists because Info files predate the makeinfo program;
the first Info files were made by hand, because makeinfo didn't yet
exist.

RE: info-find-source

[Drew Adams]

> > The "by hand" phrasing is where the confusion
> > begins because it seems to imply there is
> > a better way to "edit" them. But to me it seems
> > totally backward to edit the result of
> > compilation and I don't think I ever did that
> > because wouldn't not only update but also
> > recompilation overwrite the edits?
> >
> > In general one should edit the source! and here
> > one could simply keep a text file with any
> > extra material.
> 
> That command exists because Info files predate the makeinfo program;
> the first Info files were made by hand, because makeinfo didn't yet
> exist.

Maybe `Info-edit' existed before `makeinfo' and `Texinfo';
I don't recall.  But all three are very old.  I recall all
three back in the 80s, if I'm not mistaken.

And `Info-edit' was not deprecated until recently.  So
unless my memory is mistaken here, the lack of `makeinfo'
was not at all the reason that `Info-edit' remained
available (and bound to `e') all those years, even if it
was the case that it existed before `makeinfo'.

`Info-edit' can be useful for someone to simply modify
or add a bit of text, without needing `makeinfo' to be
available (installed).  IOW, it's use cases were never
limited to hand-creation of entire manuals.  That would
give a very false impression, IMHO, of what this command
was/is about.

For one thing, `Info-edit' is used from Info.  I'm not
sure it was ever intended to be used to write whole
manuals.  I'd guess that at least that was not the main
use case, especially all those years long.

RE: info-find-source

[Drew Adams]

> OK, so it is the *.info files* that one is
> disencouraged from editing? Well yeah, why
> would you do that?!

Quick correction/annotation/addition to one's
own copy of a manual.

Versus re-creating lots of stuff (after having
installed `makeinfo', if it's not installed
locally).

The existence of `makeinfo' does not obviate
the usefulness of `Info-edit'.

> The "by hand" phrasing is where the confusion
> begins because it seems to imply there is
> a better way to "edit" them. But to me it seems
> totally backward to edit the result of
> compilation

When that result is human-readable text it's
not a big deal to edit it.  Of course, if
you want the change to be reflected more
globally or to be shared etc. then you want
to only modify Texinfo source and generate
Info output.

You yourself argued for using *.info files in
plain editing mode (e.g. after `C-x n w').
Something like `Info-edit' is nowhere near as
extreme as that.  It's used for simple, quick
one-off changes or additions.

One doesn't have to argue _against_ generating
Info from Texinfo to see some utility in a 
command such as `Info-edit'.  Granted, that
utility is limited, and most people have never
even heard of it.  But that's not the same as
saying that it has no raison d'etre.
____

I mentioned that Info+ has a command,
`Info-merge-subnodes', for creating a
plain-text, prettified merge of Info nodes
(even a whole manual, but more typically a
section of a manual, however small).

Such a flat buffer can be useful sometimes
(e.g., plain-text printing, some kinds of
searching, sending excerpts), but I wouldn't
argue that `Info-merge-subnodes' is a _super_
useful command.  Some commands have limited
usefulness and use cases.  `Info-edit', like
`Info-merge-subnodes' is one such.

Re: info-find-source

[Emanuel Berg]

Drew Adams wrote:

> You yourself argued for using *.info files in
> plain editing mode (e.g. after `C-x n w').
> Something like `Info-edit' is nowhere near as
> extreme as that. It's used for simple, quick
> one-off changes or additions.

Well, it is a thought to read the documentation
that way, and have the whole program file in
a single buffer, which operates like a regular
file and doesn't disappear when you do something
else or bring something else up thru some
interface which you aren't that apt with...

But after doing `C-x n w', and even `text-mode'
on top of that, you still have to disable
`read-only-mode' in order to edit it. And no
one has suggested that. So you hear me loud and
clear out there? DON'T DO THAT! Ha ha ha :)

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

> In editing mode, `a' to `z' run
> self-insrt-command. That's enough. (Unless
> you use some kind of Vi(m) emulation, that's
> another story.)

The really short and close shortcuts, which
involve either a single C or ditto M, and then
a single key, which doesn't require hand
movement, e.g. M-i, M-k, etc. - these are not
slower to any practical extent, and whats more,
you know them much better than anything that
comes out of info, because without even
practising 'em, that is what happens every day
of writing/editing text and code.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Marcin Borkowski]

On 2018-01-13, at 01:43, Emanuel Berg <moasen@zoho.com> wrote:

> The issue I have with info is that it is easy
> to get lost when navigating all those node back
> and forth in the tree structure, back and forth
> in history, up to the parent and down to the
> child until you are stuck at a leaf and you
> still haven't found what you are looking for.
> And you do all this with keys that you do not
> use every day for editing.
>
> Compare this to the man pages where this never
> happens (because of less complexity), *or*
> a plain text files, where by definition it
> cannot ever happen.
>
> But doesn't this mean the files will be very
> long? Yes, and I don't have a problem with that
> as this volume is linear, not broken down into
> a complicated tree structure one has
> to traverse to get to the rainbow's end.

This is the point.  Underneath, Info is *still* linear.  You can search
it with C-s/C-r, if that's your way.  Also, you can bookmark Info pages
(using Emacs stock bookmarks or bookmark+).

I can't see *any* gain from your function.  And if I really wanted
something like that, I'd probably use "no new keystrokes to learn"
(using your own words!) and stick with plain old `C-x n w'.

Please try it out and tell me how your solution is superior.

Best,

-- 
Marcin Borkowski

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

> This is the point. Underneath, Info is
> *still* linear.

Of course, because otherwise it couldn't be put
into a single file where all the items appear
in order!

(Here "Info" means the data, or documentation,
not the browser or Emacs mode.)

> I can't see *any* gain from your function.

OK, the gain is it brings up the raw source
file associated with the info file you are
browsing. Or at least that is the purpose as
I haven't had time testing it a lot.
But because there exists a correlation between
the source file and what is displayed with the
info browser, the Holy Grail cannot be farther
away than the expected initial problems.

Actually, I cannot see "any gain" from
displaying the data as a tree! The only gain
I can see is hypertext. I see the point of
that, but I see that as something similar of,
you ask a person "how do you do X? what do you
need to do it?" and s/he says "Ask P about
that. Or go to the HW store and ask them" - if
you on the contrary know what data you desire,
and you know where it is, there is no need for
any form of hypertext, be it "On-Line" or on
the Internet. Instead of following links you go
directly to the Rainbow's End, acquire the
item, and return.

The drawbacks from using Info compared to text
are it's a mode from which you have much less
experience and thus much less fluency, and it
isn't always true that all your tweaks and
extensions fit seamlessly into info, while this
cannot be an issue for text, as that is the
original habitate anyway!

> And if I really wanted something like that,
> I'd probably use "no new keystrokes to learn"
> (using your own words!) and stick with plain
> old `C-x n w'.
>
> Please try it out and tell me how your
> solution is superior.

You don't have to do `M-x text-mode RET' after
you do `C-x n w'?

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Marcin Borkowski]

On 2018-01-14, at 03:54, Emanuel Berg <moasen@zoho.com> wrote:

> Marcin Borkowski wrote:
>
> Actually, I cannot see "any gain" from
> displaying the data as a tree! The only gain
> I can see is hypertext. I see the point of
> that, but I see that as something similar of,
> you ask a person "how do you do X? what do you
> need to do it?" and s/he says "Ask P about
> that. Or go to the HW store and ask them" - if
> you on the contrary know what data you desire,
> and you know where it is, there is no need for
> any form of hypertext, be it "On-Line" or on
> the Internet. Instead of following links you go
> directly to the Rainbow's End, acquire the
> item, and return.

Hypertext is an important gain.  Convenience is another: many keys in
Info do not need two keystrokes (`i' compared to `C-s'), so I can use
Info with one hand.  (And yes, I do it quite often.)

> The drawbacks from using Info compared to text
> are it's a mode from which you have much less
> experience and thus much less fluency, and it
> isn't always true that all your tweaks and
> extensions fit seamlessly into info, while this
> cannot be an issue for text, as that is the
> original habitate anyway!

But Info derives from special-mode, and shares many keys with other
modes derived from that, like eww, dired etc.

>> And if I really wanted something like that,
>> I'd probably use "no new keystrokes to learn"
>> (using your own words!) and stick with plain
>> old `C-x n w'.
>>
>> Please try it out and tell me how your
>> solution is superior.
>
> You don't have to do `M-x text-mode RET' after
> you do `C-x n w'?

Again: why would I?  I would lose all the benefits (e.g., keybindings)
from Info-mode!

Best,

-- 
Marcin Borkowski

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

> Hypertext is an important gain.

Hypertext is an important gain to text systems
just like inheritance is to OO systems, that
said both are very easy to over-use and it
shouldn't be relied upon, neither from the
producer nor from the consumer side of
a computer system.

> Again: why would I? I would lose all the
> benefits (e.g., keybindings) from Info-mode!

That's the benefit: with text only, source file
only, there is no need for Info-mode or
any key that comes with it.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Emanuel Berg]

Robert Thorpe wrote:

> I don't understand what you mean. I usually
> don't understand what you mean in these type
> of situations.

:D

OK, I'll explain it again tomorrow.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Emanuel Berg]

Robert Thorpe wrote:

>> Also, how does info look to you guys? To me,
>> it looks like this:
>>
>> http://user.it.uu.se/~embe8573/pics/info.png
>
> If you use Emacs in a GUI environment then
> there is more markup. Some things are in bold
> and some are in larger fonts. I generally
> think this is useful, but opinions differ.

I also think markup is useful. I prefer when it
is color-only but that is besides more than
just the point.

If you take a look at that screenshot [1]
again, or even use file(1) on it, you find out
that the resolution is only 576 x 416. This is
the resolution of my setup, which employs
a projector, and not the most
expensive/advanced projector at that.

Practically speaking, I can operate only *two*
windows! [2] This is perhaps yet another reason
I can't use info like some of you guys,
virtually having a dedicated window for it open
all day long and still have sufficient space to
do everything else. Here, everything needs to
be kept just as simple as possible.

[1] http://user.it.uu.se/~embe8573/pics/info.png
[2] http://user.it.uu.se/~embe8573/pics/two.png

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Bob Proulx]

Emanuel Berg wrote:
> If you take a look at that screenshot [1]
> again, or even use file(1) on it, you find out
> that the resolution is only 576 x 416. This is
> the resolution of my setup, which employs
> a projector, and not the most
> expensive/advanced projector at that.

I sympathize.  I am typing this on a Thinkpad X220 and the vertical
resolution is only a mere 768 pixels.  I realize that is still
significantly more than your 416 pixels but still in the same general
size neighborhood.  I can comfortably get only 47 vertical lines of
text.  And of course there is the minibuffer and modeline.  But of
course I worked for years with 80x24 terminals of similar size using
emacs, live by electric-buffer-mode which I can't do without, and flip
between buffers frequently.

> Practically speaking, I can operate only *two*
> windows!

That's really my limit too.  But when I split vertically I get two 24
line windows.  I know that would be luxury for you.  I worked for
years on that size of display when that was all that was available.
So while it feels cramped to me now I put up with it for the
lightweight carrying around laptop that weighs a fraction of my
heavier beast.  Coming back to my desktop now feels like infinite
living space.

> This is perhaps yet another reason
> I can't use info like some of you guys,
> virtually having a dedicated window for it open
> all day long and still have sufficient space to
> do everything else. Here, everything needs to
> be kept just as simple as possible.

I don't keep the window open in this dedicated way you imagine.  I am
always using 'C-h i' and 'q' to flip between buffers.  Do something.
Look something up.  Do something.  Look something up.  And also
Bookmarks with C-x r l to open a bookmark.

To a lessor extend I sometimes use frames.  I sometimes use frames
when I must take a high priority interrupt and do something else for a
while and want to save the window state of what I was doing and then
return exactly to it.  Yes this also works in the text terminal too.
First use 'C-x 5 2' to open a new frame.  Do whatever.  Then use 'C-x
5 o' to swap between frames.  Use 'C-x 5 1' to select the current
frame as the only one and close other frames collapsing back to the
one you are in only to clean up.  There is something that annoys me
about using frames though.  I'll ask that separately when I have time
to discuss it.

Bob

Re: info-find-source

[Marcin Borkowski]

On 2018-01-14, at 23:30, Bob Proulx <bob@proulx.com> wrote:

> To a lessor extend I sometimes use frames.  I sometimes use frames
> when I must take a high priority interrupt and do something else for a
> while and want to save the window state of what I was doing and then
> return exactly to it.  Yes this also works in the text terminal too.
> First use 'C-x 5 2' to open a new frame.  Do whatever.  Then use 'C-x
> 5 o' to swap between frames.  Use 'C-x 5 1' to select the current
> frame as the only one and close other frames collapsing back to the
> one you are in only to clean up.  There is something that annoys me
> about using frames though.  I'll ask that separately when I have time
> to discuss it.

That's what I do, too, ut keep in mind that you can also save window
configurations in registers.

(info "(emacs) Configuration Registers")

Best,

-- 
Marcin Borkowski

Re: info-find-source

[Bob Proulx]

Marcin Borkowski wrote:
> Bob Proulx wrote:
> > To a lessor extend I sometimes use frames.  I sometimes use frames
> > when I must take a high priority interrupt and do something else for a
> > while and want to save the window state of what I was doing and then
> > return exactly to it. ...
> 
> That's what I do, too, ut keep in mind that you can also save window
> configurations in registers.
> 
> (info "(emacs) Configuration Registers")

I have not previously used registers for more than text.  Playing
around with saving the widow config into registers seems to be exactly
what I was looking for.  I think I will be using this feature a lot
now that I have tried it.

Thank you Marcin for expanding my Emacs vocabulary! :-)

Bob

Re: info-find-source

[Marcin Borkowski]

On 2018-01-20, at 18:37, Bob Proulx <bob@proulx.com> wrote:

> Marcin Borkowski wrote:
>> Bob Proulx wrote:
>> > To a lessor extend I sometimes use frames.  I sometimes use frames
>> > when I must take a high priority interrupt and do something else for a
>> > while and want to save the window state of what I was doing and then
>> > return exactly to it. ...
>>
>> That's what I do, too, ut keep in mind that you can also save window
>> configurations in registers.
>>
>> (info "(emacs) Configuration Registers")
>
> I have not previously used registers for more than text.  Playing
> around with saving the widow config into registers seems to be exactly
> what I was looking for.  I think I will be using this feature a lot
> now that I have tried it.
>
> Thank you Marcin for expanding my Emacs vocabulary! :-)

You're welcome!

As M-x doctor says, my secretary will send you a bill;-).

Best,

--
Marcin Borkowski

Re: info-find-source

[HASM]

> Bob Proulx <bob@proulx.com> writes:

Somehow I think I remember this name from the old hp.* groups.

-- HASM

Re: info-find-source

[Bob Proulx]

Hello HASH,

HASM wrote:
> > Bob Proulx <bob@proulx.com> writes:
> 
> Somehow I think I remember this name from the old hp.* groups.

Yes.  I am alumni of HP and was frequently posting to the hp.*
internal newsgroups.  Ah...  The good old days! :-)

(This is off-topic for help-gnu-emacs but I had no other way to
respond to an @example.invalid address and therefore I hope the group
will forgive the intrusion.)

Bob

Re: info-find-source

[Emanuel Berg]

Marcin Borkowski wrote:

> (info "(emacs) Configuration Registers")

Here on the other hand, admittedly, is one good
use of info!

But one could do the same with text, only
imposing the restriction that all captions
(chapters, sections, and subsections) should be
unique. That is actually good practise anyway
as it makes sense. A book entitled
"The Universe" shouldn't have a chapter,
a section, and absolutely not a subsection with
the exact same phrasing!

But yes, it is one example of a good side of
info. I said a didn't like hypertext but it is
actually the reliance and excessive
"mindfullnessless" use I don't like. This is
more of a direct reference than
a hyperreference, altho I suppose it employs
the same technology.

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Stefan Monnier]

> I sympathize.  I am typing this on a Thinkpad X220 and the vertical
> resolution is only a mere 768 pixels.

I suggest you upgrade to a X201(s) where you can get 1440x900.


        Stefan "using 1400x1050 here on my T61"

Re: info-find-source

[Bob Proulx]

Stefan Monnier wrote:
> > I sympathize.  I am typing this on a Thinkpad X220 and the vertical
> > resolution is only a mere 768 pixels.
> 
> I suggest you upgrade to a X201(s) where you can get 1440x900.

Ha!  I have an X201 as well!  It is a very good machine too.  Actually
I have them sitting next to each other both up and running right now.
I have IRC and other always on things running on one that I want to
keep an eye on and am using the other for this email.  A "poor man's"
way to have dual monitors. :-)  [And yes I also use x2x and other
utilities for this as it happens.]

I like the X201 as well.  It is a very good machine too.  It has a few
more vertical pixels, which are the important pixels.

I have my two machines configured differently from each other having
tuned them for different things.  One is a very thin network client
that I used when traveling.  The other is a fully self contained
system for working offline that I use as my daily carrrying around
machine.  And also the keyboards, while both are very good, have a
slightly different feel from each other making me prefer to type on
one more than the other.

Note the X220 is compatible with Libreboot whereas the X201 is not.

>         Stefan "using 1400x1050 here on my T61"

For some reason my beloved T60p has started overheating.  It hits 100C
too often now.  Something has changed.  In spite of the fan running at
high speed all of the time.  It really produces a lot of heat!  It has
become my coffee table machine and sits in wait most of the time now
due to the new overheating problem.  I need to pull it apart and
double check everything.  Or replace the cpu in it.  Or something.  It
wasn't previously this way.  Making me worry the newer kernels are the
cause of the overheating problem.

Even before the overheating problem on my T60p began I had already
started to use the X201 as a travel machine.  The kit is a full pound
lighter in weight and much reduces my carry around weight.  I miss the
vertical pixels.  But it is an experiment for me in "small living". :-)

Bob

Re: info-find-source

[Stefan Monnier]

> Note the X220 is compatible with Libreboot whereas the X201 is not.

Important detail, indeed, but not sufficient to make me buy a new
machine and definitely not making up for the braindead 16/9 aspect ratio
which would bring me back to the crippling 768 vertical pixels of my
old X30.

>>         Stefan "using 1400x1050 here on my T61"
> For some reason my beloved T60p has started overheating.

Most likely the fan is too dirty and/or the thermal compound on top of
the CPU (or GPU) should be replaced.

My T60 (with ATI graphics) tends to get hot as well, but that's always
been the case.  It was one of the motivations (along with the need for
more RAM) for me to buy a second hand T61 with Intel graphics to replace
it (AFAIK that T61 model is still the "best" available machine with
a 14" 4/3 screen).

The T61's 14" 4/3 and the X201s's 12" 16/10 result in the same final
width (that of a normal fullsize keyboard).

> Making me worry the newer kernels are the cause of the
> overheating problem.

Seems very unlikely.

> Even before the overheating problem on my T60p began I had already
> started to use the X201 as a travel machine.  The kit is a full pound
> lighter in weight and much reduces my carry around weight.

Indeed, my X201s is lighter, smaller, faster and with a much better
autonomy than my T61 (and it doesn't have that many fewer vertical
pixels), so it's my machine of choice when traveling.


        Stefan

Re: info-find-source

[Emanuel Berg]

Bob Proulx wrote:

>> If you take a look at that screenshot [1]
>> again, or even use file(1) on it, you find
>> out that the resolution is only 576 x 416.
>> This is the resolution of my setup, which
>> employs a projector, and not the most
>> expensive/advanced projector at that.
>
> I sympathize. [...]

Interesting stories, but I'm not complaining if
it came across that way! I've since long grown
accustomed to the smaller resolution and made
many hacks to counteract/sharpen it. The setup
here, in a wooden house, resembles the
subterranean high-tech bases from which the
criminal masterminds operates, protected by
their disgusting lackeys. I always tell all the
young athletic females around to stop wearing
their lycra tights in magenta and cyan less
people might think is an X-Men comic book!

-- 
underground experts united
http://user.it.uu.se/~embe8573

Re: info-find-source

[Emanuel Berg]

Robert Thorpe wrote:

> I think it's a mistake to over-emphasise the
> editing modes, and their keybindings.
> The viewing modes and the special modes are
> just as important.

They are as important (or unimportant
I suppose) as the editing modes, however at
least I spend only a fraction of my time there
so the fluency compared to editing is - well,
it hasn't reached the point where it can even
be called "fluency", in all honesty.

At the end of the day, editing and writing is
where the creativity is. Everything else is
support for that. (Support is important,
of course.)

> I spend a great deal of time in Dired, Info,
> Help and reading mail.

I also use Dired a lot. The creativity-support
model doesn't really apply to that tho as Dired
is about mucking around the file system. I find
the original bindings to Dired generally long
and out of place, and I have written a bunch of
other Dired related Elisp as well [1]. However that
was one of the first things I did with Emacs so
it is possible some of it was unneccessary if
the "mere" functionality is concerned.

Info is what this thread is all about.

Help I just look up a function or keystroke
with without really doing anything else. But it
has served me well. I don't get lost in it.

For reading mails I use Gnus and it doesn't
involve a lot of keys. Writing mails I consider
creative, only perhaps the need/desire to do
it seems to be greater when there is no other
project that really holds a strong appeal, so
it is more of an in-between pastime for me
personally. Undisputably, the whole FOSS
development movement is based on mails.

> I also spend a fair amount of time in View,
> Occur, Grep, Find, Compile and Shell.

Of those I'm aware of "Compile" if you mean the
compilation buffer. I never do anything there
tho. Shell I consider creative or perhaps in
the "Dired" category.

> In my experience it's worth becoming
> reasonably familiar with those modes and
> their keybindings. It's true that doing that
> means less practice with the normal
> editing keybindings.

Well, you don't really need MORE time there :)

[1] http://user.it.uu.se/~embe8573/emacs-init/dired-my.el

-- 
underground experts united
http://user.it.uu.se/~embe8573