make doc fails on current head

make doc fails on current head

[Jonathan Leech-Pepin]

Hello,

Under the current git head (4144c55) I get the following error when
trying to run =make doc=.

#+begin_src sh
  ~/build/org-mode $ make doc
  /usr/bin/make -C doc info
  make[1]: Entering directory `/cygdrive/d/Users/jleechpe/build/org-mode/doc'
  makeinfo --no-split org.texi -o org
  org.texi:2450: Unknown command `#$2)'.
  makeinfo: Removing output file `org' due to errors; use --force to preserve.
  Makefile:53: recipe for target `org' failed
  make[1]: *** [org] Error 1
  make[1]: Leaving directory `/cygdrive/d/Users/jleechpe/build/org-mode/doc'
  targets.mk:76: recipe for target `info' failed
  make: *** [info] Error 2
  ~/build/org-mode $
#+end_src

If I revert to =git checkout HEAD~1= make doc succeeds as it had previously.

Regards,

Jonathan

Re: make doc fails on current head

[Michael Brand]

Hi Bastien

On Thu, May 31, 2012 at 5:21 PM, Jonathan Leech-Pepin
<jonathan.leechpepin@gmail.com> wrote:
> If I revert to =git checkout HEAD~1= make doc succeeds as it had previously.

Funny that just out of curiosity and without any suspicion I came
across this regression of one of the org.texi lines written by me. The
intention of your change

-$3 = remote(FOO, @@@@#$2)      @r{copy column 2 from table FOO into}
+$3 = remote(FOO, @@@#$2)       @r{copy column 2 from table FOO into}

maybe was

+$3 = remote(FOO, @@#$2)        @r{copy column 2 from table FOO into}

but I can confirm that it should really compile to @@#$2 and not to
@#$2 or something else.

I'll think about if for this Org TBLFM expression I can find a less
confusing alternative for the texi domain. :-)

Michael

Re: make doc fails on current head

[Nick Dokos]

Michael Brand <michael.ch.brand@gmail.com> wrote:

> Hi Bastien
> 
> On Thu, May 31, 2012 at 5:21 PM, Jonathan Leech-Pepin
> <jonathan.leechpepin@gmail.com> wrote:
> > If I revert to =git checkout HEAD~1= make doc succeeds as it had previously.
> 
> Funny that just out of curiosity and without any suspicion I came
> across this regression of one of the org.texi lines written by me. The
> intention of your change
> 
> -$3 = remote(FOO, @@@@#$2)      @r{copy column 2 from table FOO into}
> +$3 = remote(FOO, @@@#$2)       @r{copy column 2 from table FOO into}
> 
> maybe was
> 
> +$3 = remote(FOO, @@#$2)        @r{copy column 2 from table FOO into}
> 
> but I can confirm that it should really compile to @@#$2 and not to
> @#$2 or something else.
> 
> I'll think about if for this Org TBLFM expression I can find a less
> confusing alternative for the texi domain. :-)
> 

Just to make sure, you are saying that commit
4144c55ec78a4fdf246c64a4130f807eec50a913 should be reverted - the four @
signs in org.texi produce two @ signs in the produced info file, and
that's the way the info file should be. Do I undertand correctly?

Thanks,
Nick

Re: make doc fails on current head

[Michael Brand]

Hi Nick

On Thu, May 31, 2012 at 11:58 PM, Nick Dokos <nicholas.dokos@hp.com> wrote:
> Just to make sure, you are saying that commit
> 4144c55ec78a4fdf246c64a4130f807eec50a913 should be reverted - the four @
> signs in org.texi produce two @ signs in the produced info file, and
> that's the way the info file should be. Do I undertand correctly?

Yes.

Michael

Re: make doc fails on current head

[Bastien]

Hi Michael,

I just reverted my commit, thanks.

Michael Brand <michael.ch.brand@gmail.com> writes:

> but I can confirm that it should really compile to @@#$2 and not to
> @#$2 or something else.

So what does @@#$2 really means?  Does the first "@" stand for "This is
a field coordinate" and the rest for the coordinates range itself?

-- 
 Bastien

Re: make doc fails on current head

[Michael Brand]

Hi Bastien

On Fri, Jun 1, 2012 at 9:48 AM, Bastien <bzg@gnu.org> wrote:
> So what does @@#$2 really means?  Does the first "@" stand for "This is
> a field coordinate"

yes

> and the rest for the coordinates range itself?

it is not a range, but as "@# and $# can be used to get the row or
column number of the field where the formula result goes" it will
evaluate to @1$2, @2$2 and so on. I tried to be brief in the manual
but there are more examples on Worg:

"Field coordinates in formulas (@#  and $#)"
http://orgmode.org/worg/org-hacks.html#field-coordinates-in-formulas

Michael

Re: make doc fails on current head

[Nick Dokos]

Bastien <bzg@gnu.org> wrote:


> So what does @@#$2 really means?  Does the first "@" stand for "This is
> a field coordinate" and the rest for the coordinates range itself?
> 

@# is the current row number, so @@#$2 is a reference to the current row,
second column. Michael has a couple of nontrivial examples (e.g. transposing
a table) using this facility on worg:

  http://orgmode.org/worg/org-hacks.html#field-coordinates-in-formulas-transpose-table

where he is using the current row and current col to form a reference to the transposed
location:

	@$#$@#

"The row whose number is the number of the current column and the column
whose number is the number of the current row".

Nick

Re: make doc fails on current head

[Bastien]

Nick Dokos <nicholas.dokos@hp.com> writes:

> Bastien <bzg@gnu.org> wrote:
>
>
>> So what does @@#$2 really means?  Does the first "@" stand for "This is
>> a field coordinate" and the rest for the coordinates range itself?
>
> @# is the current row number, so @@#$2 is a reference to the current row,
> second column. 

Got it, thanks to you and Michael for the detailed answers.

-- 
 Bastien

Re: make doc fails on current head

[Jonathan Leech-Pepin]

I can confirm it's fixed

And thanks for the answer, hadn't realized you could use @# and $# for
references.

On Fri, Jun 1, 2012 at 8:56 AM, Bastien <bzg@gnu.org> wrote:
> Nick Dokos <nicholas.dokos@hp.com> writes:
>
>> Bastien <bzg@gnu.org> wrote:
>>
>>
>>> So what does @@#$2 really means?  Does the first "@" stand for "This is
>>> a field coordinate" and the rest for the coordinates range itself?
>>
>> @# is the current row number, so @@#$2 is a reference to the current row,
>> second column.
>
> Got it, thanks to you and Michael for the detailed answers.
>
> --
>  Bastien
>

Re: make doc fails on current head

[Bastien]

Hi Jonathan,

Jonathan Leech-Pepin <jonathan.leechpepin@gmail.com> writes:

> Under the current git head (4144c55) I get the following error when
> trying to run =make doc=.

Fixed, thanks for reporting this.

-- 
 Bastien