Re: `with` as a list.

[Kyle Meyer <kyle@kyleam.com>]

Mario Frasca writes:

> btw. are you on irc.freenode.net?  I'm `mariotomo' there.  and I just 
> joined `#org-mode'.  I don't think that my terminal will ring a bell if 
> I'm mentioned there.

No, sorry, I'm not.

> On 30/05/2020 01:22, Kyle Meyer wrote:
>> Mario Frasca writes:
>
>> Two meta-comments:
>>
>>    * Please provide a patch with a commit message.  See
>>      <https://orgmode.org/worg/org-contribute.html> for more information.
>>
>>    * The link above also explains the copyright assignment requirement
>>      for contributions.  Please consider assigning copyright to the FSF.
>
> I'm editing in my cloned repository, and committing often, so I do not 
> have a single commit, consequently also not a single commit message.

There are various ways to get those fixups into a single commit.  Here's
one that assumes that you're working from the command line and on a
topic branch rather than directly on master.  Do an interactive rebase
(`git rebase -i master').  When the editor pops up, replace "pick" with
"fixup" in all of the lines but the first.  (If any of the "fixup"
commits have message bits that you want to keep in the final message,
you can use "squash" instead.)

For later fixups, you can do the same process.  You could also use
git-commit's --amend.  (Again, there are lots of possible workflows, and
of course once you're working on a series of commits rather than a
single one, the process changes a bit.)

> I just sent my form request to assign@gnu.org.

Thank you!  Please let us know when the process is complete.

I see you already sent an updated patch, so I'll just respond to one
more point from this message.

>> I haven't looked at this bit too closely (and I know the handling around
>> col-labels changed a bit in the last message you sent), but I did try
>> out a few org-plot invocations and things seemed to work as I expected.
>> I'll plan to take a closer when you send a reroll.
>
> the whole org-plot.el has (had) no unit tests, so if you share a couple 
> of your org-plot invocations, I can convert them to regression tests.

I don't use org-plot.el, so to try this out, I just copied an example
from the Worg tutorial, stripped the table down a bit, and tried a few
`deps' and `with' values.  So there's nothing too useful, but here's is
the last one I had:

--8<---------------cut here---------------start------------->8---
#+PLOT: title:"example table" ind:1 deps:(2 3) type:2d with:(lines points)
#+PLOT: labels:("xlab" "ylab" "zlab")
#+TBLNAME:org-plot-example-1
|   x |    y |    z |
|-----+------+------|
| 0.1 | 0.42 | 0.37 |
| 0.2 | 0.31 | 0.33 |
| 0.3 | 0.24 | 0.28 |
| 0.4 | 0.27 | 0.23 |
--8<---------------cut here---------------end--------------->8---

Thanks for your interest in adding tests.  Of course it'd be good for
org-plot.el to have some, and there are existing spots where I think it
should be easy enough to hook into and test (in particular generating
the script file and data file).  However, that doesn't necessarily have
to be tacked on to this series.