bug#14399: octave-mode bug

bug#14399: octave-mode bug

[Tom Olin]

[-- Attachment #1: Type: text/plain, Size: 790 bytes --]

The following code does not indent correctly (IMHO) with octave-mode:

vec = [...
one;...
two;...
three];

What I get is this:

vec = [...
        one;...
           two;...
              three];



Emacs  : GNU Emacs 24.3.1 (x86_64-apple-darwin12.2.0, X toolkit, Xaw scroll bars)
 of 2013-03-21 on alum.local
Package: Emacs version 24.3.1

current state:
==============
(setq
 octave-blink-matching-block t
 octave-block-offset 2
 octave-comment-char 35
 octave-continuation-offset 4
 octave-continuation-string "..."
 octave-send-echo-input t
 octave-send-line-auto-forward t
 octave-send-show-buffer t
 )


--
Tom Olin
Senior Systems Engineer | PAR Government Systems Corporation
315.356.2142 | tom_olin@partech.com | http://pargovernment.com




[-- Attachment #2: smime.p7s --]
[-- Type: application/pkcs7-signature, Size: 5134 bytes --]

bug#14399: octave-mode bug

[Leo Liu]

I have written to octave's maintainers' list:
http://article.gmane.org/gmane.comp.gnu.octave.maintainers/32441

Leo

bug#14399: octave-mode bug

[Leo Liu]

Hi Stefan,

On 2013-05-15 20:00 +0800, Leo Liu wrote:
> http://article.gmane.org/gmane.comp.gnu.octave.maintainers/32441

It seems the octave community's consensus is: ... should not introduce
extra indentation in (), [] or {}. Could you help fix this bug?

Leo

bug#14399: octave-mode bug

[Leo Liu]

On 2013-05-18 07:13 +0800, Leo Liu wrote:
> It seems the octave community's consensus is: ... should not introduce
> extra indentation in (), [] or {}. Could you help fix this bug?

Hi Stefan,

I am unfamiliar with smie. I have been meaning to understand it but
haven't found time to. Could you help fix this indentation bug?

Leo

bug#14399: octave-mode bug

[Stefan Monnier]

> I am unfamiliar with smie. I have been meaning to understand it but
> haven't found time to. Could you help fix this indentation bug?

I can't remember much of the syntax of Octave.  What does ";..." mean?


        Stefan

bug#14399: octave-mode bug

[Rüdiger Sonderfeld]

On Friday 22 November 2013 09:27:57 Stefan Monnier wrote:
> > I am unfamiliar with smie. I have been meaning to understand it but
> > haven't found time to. Could you help fix this indentation bug?
> 
> I can't remember much of the syntax of Octave.  What does ";..." mean?

... introduces a continuation line.  Similar to \ in other languages.

a = [ foo;...
bar ]

is the same as a = [ foo; bar ].

See (info "(octave) Continuation Lines")

Regards,
Rüdiger

bug#14399: octave-mode bug

[Stefan Monnier]

>> > I am unfamiliar with smie. I have been meaning to understand it but
>> > haven't found time to. Could you help fix this indentation bug?
>> I can't remember much of the syntax of Octave.  What does ";..." mean?
> ... introduces a continuation line.  Similar to \ in other languages.

> a = [ foo;...
> bar ]

> is the same as a = [ foo; bar ].

That's the part I knew already ;-)

But in which way is 

   a = [ foo;...
         bar ]

different from

   a = [ foo
         bar ]
or
   a = [ foo;
         bar ]

-- Stefan

bug#14399: octave-mode bug

[Stefan Monnier]

Ping?

>>>>> "Stefan" == Stefan Monnier <monnier@iro.umontreal.ca> writes:

>>> > I am unfamiliar with smie. I have been meaning to understand it but
>>> > haven't found time to. Could you help fix this indentation bug?
>>> I can't remember much of the syntax of Octave.  What does ";..." mean?
>> ... introduces a continuation line.  Similar to \ in other languages.

>> a = [ foo;...
>> bar ]

>> is the same as a = [ foo; bar ].

> That's the part I knew already ;-)

> But in which way is 

>    a = [ foo;...
>          bar ]

> different from

>    a = [ foo
>          bar ]
> or
>    a = [ foo;
>          bar ]

> -- Stefan

bug#14399: octave-mode bug

[Achim Gratz]

Stefan Monnier writes:
> That's the part I knew already ;-)
>
> But in which way is 
>
>    a = [ foo;...
>          bar ]
>
> different from
>
>    a = [ foo
>          bar ]
> or
>    a = [ foo;
>          bar ]

For these three examples there is no difference because the newlines and
whitespace the continuation marker gobbles up wouldn't terminate or
alter the statement.  On the other hand

a = [ foo ...
      bar ]

_is_ different (introducing another column rather than a row).


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Factory and User Sound Singles for Waldorf rackAttack:
http://Synth.Stromeko.net/Downloads.html#WaldorfSounds

bug#14399: octave-mode bug

[Tom Olin]

[-- Attachment #1: Type: text/plain, Size: 1073 bytes --]

A semicolon separates rows; a comma or space separates columns. Thus [ foo bar ] or [ foo, bar ] is a 1x2 matrix. [ foo; bar ] is a 2x1 matrix.

--
Tom Olin
Senior Systems Engineer | PAR Government Systems Corporation
315.356.2142 | tom_olin@partech.com | http://pargovernment.com



On Nov 27, 2013, at 11:40 PM, Stefan Monnier <monnier@iro.umontreal.ca> wrote:

> Ping?
> 
>>>>>> "Stefan" == Stefan Monnier <monnier@iro.umontreal.ca> writes:
> 
>>>>> I am unfamiliar with smie. I have been meaning to understand it but
>>>>> haven't found time to. Could you help fix this indentation bug?
>>>> I can't remember much of the syntax of Octave.  What does ";..." mean?
>>> ... introduces a continuation line.  Similar to \ in other languages.
> 
>>> a = [ foo;...
>>> bar ]
> 
>>> is the same as a = [ foo; bar ].
> 
>> That's the part I knew already ;-)
> 
>> But in which way is 
> 
>>   a = [ foo;...
>>         bar ]
> 
>> different from
> 
>>   a = [ foo
>>         bar ]
>> or
>>   a = [ foo;
>>         bar ]
> 
>> -- Stefan


[-- Attachment #2: smime.p7s --]
[-- Type: application/pkcs7-signature, Size: 5134 bytes --]

bug#14399: octave-mode bug

[Stefan Monnier]

>> That's the part I knew already ;-)
>> 
>> But in which way is 
>> 
>> a = [ foo;...
>> bar ]
>> 
>> different from
>> 
>> a = [ foo
>> bar ]
>> or
>> a = [ foo;
>> bar ]

> For these three examples there is no difference because the newlines and
> whitespace the continuation marker gobbles up wouldn't terminate or
> alter the statement.

Thanks.  So it begs the question: why would someone write

 a = [ foo;...
       bar ]

instead of one of the other two?  I'm trying to understand what's the
intention behind it, to better understand how the code should behave.


        Stefan

bug#14399: octave-mode bug

[Rüdiger Sonderfeld]

On Saturday 30 November 2013 12:19:30 Stefan Monnier wrote:
> Thanks.  So it begs the question: why would someone write
> 
>  a = [ foo;...
>        bar ]
> 
> instead of one of the other two?  I'm trying to understand what's the
> intention behind it, to better understand how the code should behave.

I think it's just a matter of style.  I thought it had something to do with 
Matlab compatibility but a quick test showed that Matlab accepts both forms as 
well.

Regards,
Rüdiger

bug#14399: octave-mode bug

[Stefan Monnier]

>> It seems the octave community's consensus is: ... should not introduce
>> extra indentation in (), [] or {}. Could you help fix this bug?
> I am unfamiliar with smie.  I have been meaning to understand it but
> haven't found time to.  Could you help fix this indentation bug?

Looks like we've all been pretty dumb: not sure about the OP's real code
but at least his test case gets indented like this with Emacs's trunk:

  vec = [...
          one;...
          two;...
          three];

So it looks like the bug is already fixed.


        Stefan

bug#14399: octave-mode bug

[Leo Liu]

On 2013-12-02 23:15 +0800, Stefan Monnier wrote:
> Looks like we've all been pretty dumb: not sure about the OP's real code
> but at least his test case gets indented like this with Emacs's trunk:
>
>   vec = [...
>           one;...
>           two;...
>           three];
>
> So it looks like the bug is already fixed.

I was able to reproduce the bug some time ago but indeed this seems
fixed in trunk. but

x = ...
12

doesn't indent the second line correctly.

Leo

bug#14399: octave-mode bug

[Stefan Monnier]

> x = ...
> 12

> doesn't indent the second line correctly.

That's simply for lack of a corresponding rule in octave-smie-rules.
Try M-x smie-edebug RET


        Stefan

bug#14399: octave-mode bug

[Tom Olin]

[-- Attachment #1: Type: text/plain, Size: 987 bytes --]

Just FWIW, I'm still seeing the problem. I get:

vec = [...
	one;...
	   two;...
	      three];

I'm using:

GNU Emacs 24.3.1 (x86_64-apple-darwin12.2.0, X toolkit, Xaw scroll bars) of 2013-03-21

from MacPorts. How far from HEAD is that?

--
Tom Olin
Senior Systems Engineer | PAR Government Systems Corporation
315.356.2142 | tom_olin@partech.com | http://pargovernment.com



On Dec 2, 2013, at 10:58 AM, Leo Liu <sdl.web@gmail.com> wrote:

> On 2013-12-02 23:15 +0800, Stefan Monnier wrote:
>> Looks like we've all been pretty dumb: not sure about the OP's real code
>> but at least his test case gets indented like this with Emacs's trunk:
>> 
>>  vec = [...
>>          one;...
>>          two;...
>>          three];
>> 
>> So it looks like the bug is already fixed.
> 
> I was able to reproduce the bug some time ago but indeed this seems
> fixed in trunk. but
> 
> x = ...
> 12
> 
> doesn't indent the second line correctly.
> 
> Leo


[-- Attachment #2: smime.p7s --]
[-- Type: application/pkcs7-signature, Size: 5134 bytes --]

bug#14399: octave-mode bug

[Rüdiger Sonderfeld]

[-- Attachment #1: Type: text/plain, Size: 585 bytes --]

On Tuesday 03 December 2013 09:56:39 Tom Olin wrote:
> Just FWIW, I'm still seeing the problem. I get:
> 
> vec = [...
> 	one;...
> 	   two;...
> 	      three];
> 
> I'm using:
> 
> GNU Emacs 24.3.1 (x86_64-apple-darwin12.2.0, X toolkit, Xaw scroll bars) of
> 2013-03-21
> 
> from MacPorts. How far from HEAD is that?

Very far.  Thanks to Leo Octave mode was improved quite significantly since 
the release of 24.3.  It should be backward compatible to 24.3.  If you want 
to try it you could copy octave.el from bzr and load it in your Emacs.

Regards,
Rüdiger

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 230 bytes --]

bug#14399: octave-mode bug

[Tom Olin]

[-- Attachment #1: Type: text/plain, Size: 878 bytes --]

I will give that a try. Thanks.

--
Tom Olin
Senior Systems Engineer | PAR Government Systems Corporation
315.356.2142 | tom_olin@partech.com | http://pargovernment.com



On Dec 3, 2013, at 1:32 PM, Rüdiger Sonderfeld <ruediger@c-plusplus.de> wrote:

> On Tuesday 03 December 2013 09:56:39 Tom Olin wrote:
>> Just FWIW, I'm still seeing the problem. I get:
>> 
>> vec = [...
>> 	one;...
>> 	   two;...
>> 	      three];
>> 
>> I'm using:
>> 
>> GNU Emacs 24.3.1 (x86_64-apple-darwin12.2.0, X toolkit, Xaw scroll bars) of
>> 2013-03-21
>> 
>> from MacPorts. How far from HEAD is that?
> 
> Very far.  Thanks to Leo Octave mode was improved quite significantly since 
> the release of 24.3.  It should be backward compatible to 24.3.  If you want 
> to try it you could copy octave.el from bzr and load it in your Emacs.
> 
> Regards,
> Rüdiger


[-- Attachment #2: smime.p7s --]
[-- Type: application/pkcs7-signature, Size: 5134 bytes --]

bug#14399: octave-mode bug

[Tom Olin]

[-- Attachment #1: Type: text/plain, Size: 923 bytes --]

I've confirmed that the latest version of octave.el works correctly. Thanks!

--
Tom Olin
Senior Systems Engineer | PAR Government Systems Corporation
315.356.2142 | tom_olin@partech.com | http://pargovernment.com



On Dec 3, 2013, at 1:32 PM, Rüdiger Sonderfeld <ruediger@c-plusplus.de> wrote:

> On Tuesday 03 December 2013 09:56:39 Tom Olin wrote:
>> Just FWIW, I'm still seeing the problem. I get:
>> 
>> vec = [...
>> 	one;...
>> 	   two;...
>> 	      three];
>> 
>> I'm using:
>> 
>> GNU Emacs 24.3.1 (x86_64-apple-darwin12.2.0, X toolkit, Xaw scroll bars) of
>> 2013-03-21
>> 
>> from MacPorts. How far from HEAD is that?
> 
> Very far.  Thanks to Leo Octave mode was improved quite significantly since 
> the release of 24.3.  It should be backward compatible to 24.3.  If you want 
> to try it you could copy octave.el from bzr and load it in your Emacs.
> 
> Regards,
> Rüdiger


[-- Attachment #2: smime.p7s --]
[-- Type: application/pkcs7-signature, Size: 5134 bytes --]