Exporter interface question

Exporter interface question

[Carsten Dominik]

Hi,

I have a question about the exporter dispatcher.

I much appreciate that `C-u C-c C-e' repeats the
previous export action, that is extremely useful.

I often only export a subtree of a document, and I
was wondering if others would also find this proposal
useful:

When exporting a subtree, lets drop a mark at the
headline and then go to this mark when C-c C-e is
called with a prefix arguments.  This would allow
to keep the cursor somewhere deep in the subtree while
working on it, and still be able to quickly repeat the
export command.

If others would find this useful, I'd make a patch.

- Carsten

Re: Exporter interface question

[Nicolas Goaziou]

Hello,

Carsten Dominik <carsten.dominik@gmail.com> writes:

> I much appreciate that `C-u C-c C-e' repeats the
> previous export action, that is extremely useful.
>
> I often only export a subtree of a document, and I
> was wondering if others would also find this proposal
> useful:
>
> When exporting a subtree, lets drop a mark at the
> headline and then go to this mark when C-c C-e is
> called with a prefix arguments.  This would allow
> to keep the cursor somewhere deep in the subtree while
> working on it, and still be able to quickly repeat the
> export command.

It would be nice indeed.

Though, it may be useful to put some additional limitations to this
feature. For example, it would trigger only when on the same buffer as
before, perhaps even within the same subtree.

What do you think about it?

> If others would find this useful, I'd make a patch.

Be my guest. ;)


Regards,

-- 
Nicolas Goaziou

Re: Exporter interface question

[Carsten Dominik]

On 27.2.2013, at 09:52, Nicolas Goaziou <n.goaziou@gmail.com> wrote:

> Hello,
> 
> Carsten Dominik <carsten.dominik@gmail.com> writes:
> 
>> I much appreciate that `C-u C-c C-e' repeats the
>> previous export action, that is extremely useful.
>> 
>> I often only export a subtree of a document, and I
>> was wondering if others would also find this proposal
>> useful:
>> 
>> When exporting a subtree, lets drop a mark at the
>> headline and then go to this mark when C-c C-e is
>> called with a prefix arguments.  This would allow
>> to keep the cursor somewhere deep in the subtree while
>> working on it, and still be able to quickly repeat the
>> export command.
> 
> It would be nice indeed.
> 
> Though, it may be useful to put some additional limitations to this
> feature. For example, it would trigger only when on the same buffer as
> before, perhaps even within the same subtree.

Yes, this is possible.  I'll look into it.

- Carsten

> 
> What do you think about it?
> 
>> If others would find this useful, I'd make a patch.
> 
> Be my guest. ;)
> 
> 
> Regards,
> 
> -- 
> Nicolas Goaziou

Re: Exporter interface question

[Suvayu Ali]

On Wed, Feb 27, 2013 at 01:00:30PM +0100, Carsten Dominik wrote:
> 
> On 27.2.2013, at 09:52, Nicolas Goaziou <n.goaziou@gmail.com> wrote:
> 
> > Carsten Dominik <carsten.dominik@gmail.com> writes:
> > 
> >> When exporting a subtree, lets drop a mark at the
> >> headline and then go to this mark when C-c C-e is
> >> called with a prefix arguments.  This would allow
> >> to keep the cursor somewhere deep in the subtree while
> >> working on it, and still be able to quickly repeat the
> >> export command.
> > 
> > It would be nice indeed.
> > 
> > Though, it may be useful to put some additional limitations to this
> > feature. For example, it would trigger only when on the same buffer as
> > before, perhaps even within the same subtree.
> 
> Yes, this is possible.  I'll look into it.

I love the idea, including Nicolas' proposed restrictions.  Until now I
was defining keyboard macros for this.

Cheers,

-- 
Suvayu

Open source is the future. It sets us free.

Re: Exporter interface question

[Carsten Dominik]

On 27.2.2013, at 09:52, Nicolas Goaziou <n.goaziou@gmail.com> wrote:

> Hello,
> 
> Carsten Dominik <carsten.dominik@gmail.com> writes:
> 
>> I much appreciate that `C-u C-c C-e' repeats the
>> previous export action, that is extremely useful.
>> 
>> I often only export a subtree of a document, and I
>> was wondering if others would also find this proposal
>> useful:
>> 
>> When exporting a subtree, lets drop a mark at the
>> headline and then go to this mark when C-c C-e is
>> called with a prefix arguments.  This would allow
>> to keep the cursor somewhere deep in the subtree while
>> working on it, and still be able to quickly repeat the
>> export command.
> 
> It would be nice indeed.
> 
> Though, it may be useful to put some additional limitations to this
> feature. For example, it would trigger only when on the same buffer as
> before, perhaps even within the same subtree.
> 
> What do you think about it?
> 
>> If others would find this useful, I'd make a patch.
> 
> Be my guest. ;)

OK, I see that this is a bit less trivial then I thought, requires that I study the ui interface a bit more.  If you want to rush ahead and implement it let me know - otherwise this will be a few days before I find the time.

- Carsten

Re: Exporter interface question

[Nicolas Goaziou]

Carsten Dominik <carsten.dominik@gmail.com> writes:

> OK, I see that this is a bit less trivial then I thought, requires
> that I study the ui interface a bit more. 

I think the only pieces of the UI involved are `org-export-dispatch'
function and `org-export-dispatch-last-action' variable.

> If you want to rush ahead and implement it let me know - otherwise
> this will be a few days before I find the time.

There's no hurry. It can wait a few days.

Thank you for looking into it.


Regards,

-- 
Nicolas Goaziou

Re: Exporter interface question

[Carsten Dominik]

On 27.2.2013, at 16:22, Nicolas Goaziou <n.goaziou@gmail.com> wrote:

> Carsten Dominik <carsten.dominik@gmail.com> writes:
> 
>> OK, I see that this is a bit less trivial then I thought, requires
>> that I study the ui interface a bit more. 
> 
> I think the only pieces of the UI involved are `org-export-dispatch'
> function and `org-export-dispatch-last-action' variable.
> 
>> If you want to rush ahead and implement it let me know - otherwise
>> this will be a few days before I find the time.
> 
> There's no hurry. It can wait a few days.

OK, I pushed this change, maybe it is good if you have a brief look, Nicolas.
It was great fun and educational to study your very clear code.

Cheers

- Carsten

Re: Exporter interface question

[Carsten Dominik]

On 28.2.2013, at 09:39, Carsten Dominik <carsten.dominik@gmail.com> wrote:

> 
> On 27.2.2013, at 16:22, Nicolas Goaziou <n.goaziou@gmail.com> wrote:
> 
>> Carsten Dominik <carsten.dominik@gmail.com> writes:
>> 
>>> OK, I see that this is a bit less trivial then I thought, requires
>>> that I study the ui interface a bit more. 
>> 
>> I think the only pieces of the UI involved are `org-export-dispatch'
>> function and `org-export-dispatch-last-action' variable.
>> 
>>> If you want to rush ahead and implement it let me know - otherwise
>>> this will be a few days before I find the time.
>> 
>> There's no hurry. It can wait a few days.
> 
> OK, I pushed this change, maybe it is good if you have a brief look, Nicolas.
> It was great fun and educational to study your very clear code.

Sorry for following up on my own post.  I forgot to mention that
I did put in the restriction to make this work only if the export
command is called in the same buffer.  However, I am not checking
if the cursor is in the same subtree - it seems to me that the
behavior would become too unpredictable.

Also, I have not yet documented the change in the manual, of course.

- Carsten

Re: Exporter interface question

[Nicolas Goaziou]

Hello,

Carsten Dominik <carsten.dominik@gmail.com> writes:

> OK, I pushed this change, maybe it is good if you have a brief look,
> Nicolas.

Thanks for the patch. It looks good. I have one minor suggestion,
though. In the following snippet:

  (eq (marker-buffer org-export-dispatch-last-position)
  		   (current-buffer))

wouldn't it be better to check equality with:

  (org-base-buffer (current-buffer))

instead or simply (current-buffer)?

> It was great fun and educational to study your very clear code.

Thank you. 


Regards,

-- 
Nicolas Goaziou

Re: Exporter interface question

[Carsten Dominik]

On 28.2.2013, at 11:22, Nicolas Goaziou <n.goaziou@gmail.com> wrote:

> Hello,
> 
> Carsten Dominik <carsten.dominik@gmail.com> writes:
> 
>> OK, I pushed this change, maybe it is good if you have a brief look,
>> Nicolas.
> 
> Thanks for the patch. It looks good. I have one minor suggestion,
> though. In the following snippet:
> 
>  (eq (marker-buffer org-export-dispatch-last-position)
>  		   (current-buffer))
> 
> wouldn't it be better to check equality with:
> 
>  (org-base-buffer (current-buffer))
> 
> instead or simply (current-buffer)?

I agree, but I think then also setting the mark need to take care to set the mark in the base buffer. Will do this.

- Carsten

Re: Exporter interface question

[Nicolas Goaziou]

Carsten Dominik <carsten.dominik@gmail.com> writes:

> I agree, but I think then also setting the mark need to take care to
> set the mark in the base buffer. Will do this.

AFAICT, this is not necessary. Indirect buffers share markers with base
buffer.


Regards,

-- 
Nicolas Goaziou

Re: Exporter interface question

[Carsten Dominik]

On 28.2.2013, at 13:30, Nicolas Goaziou <n.goaziou@gmail.com> wrote:

> Carsten Dominik <carsten.dominik@gmail.com> writes:
> 
>> I agree, but I think then also setting the mark need to take care to
>> set the mark in the base buffer. Will do this.
> 
> AFAICT, this is not necessary. Indirect buffers share markers with base
> buffer.

I think it is necessary.  If I move a marker to a position
in a buffer and another marker to the same position in a connected
indirect buffer, both markers are different and point to different
buffers.  So if I start an export from an indirect buffer, the
dispatch marker will point to the indirect buffer.  If I later
compare the marker-buffer with (org-base-buffer (current-buffer)) I
do get a mismatch.  So if you want to compare to org-base-buffer,
I think I clearly need to set the marker to a position in the base
buffer.

Regards

- Carsten

Re: Exporter interface question

[Nicolas Goaziou]

Carsten Dominik <carsten.dominik@gmail.com> writes:

> On 28.2.2013, at 13:30, Nicolas Goaziou <n.goaziou@gmail.com> wrote:
>
>> Carsten Dominik <carsten.dominik@gmail.com> writes:
>> 
>>> I agree, but I think then also setting the mark need to take care to
>>> set the mark in the base buffer. Will do this.
>> 
>> AFAICT, this is not necessary. Indirect buffers share markers with base
>> buffer.
>
> I think it is necessary.  If I move a marker to a position
> in a buffer and another marker to the same position in a connected
> indirect buffer, both markers are different and point to different
> buffers.  So if I start an export from an indirect buffer, the
> dispatch marker will point to the indirect buffer.  If I later
> compare the marker-buffer with (org-base-buffer (current-buffer)) I
> do get a mismatch.  So if you want to compare to org-base-buffer,
> I think I clearly need to set the marker to a position in the base
> buffer.

Then what about comparing base buffer for both entities?


Regards,

-- 
Nicolas Goaziou

Re: Exporter interface question

[Carsten Dominik]

On 28.2.2013, at 22:10, Nicolas Goaziou <n.goaziou@gmail.com> wrote:

> Carsten Dominik <carsten.dominik@gmail.com> writes:
> 
>> On 28.2.2013, at 13:30, Nicolas Goaziou <n.goaziou@gmail.com> wrote:
>> 
>>> Carsten Dominik <carsten.dominik@gmail.com> writes:
>>> 
>>>> I agree, but I think then also setting the mark need to take care to
>>>> set the mark in the base buffer. Will do this.
>>> 
>>> AFAICT, this is not necessary. Indirect buffers share markers with base
>>> buffer.
>> 
>> I think it is necessary.  If I move a marker to a position
>> in a buffer and another marker to the same position in a connected
>> indirect buffer, both markers are different and point to different
>> buffers.  So if I start an export from an indirect buffer, the
>> dispatch marker will point to the indirect buffer.  If I later
>> compare the marker-buffer with (org-base-buffer (current-buffer)) I
>> do get a mismatch.  So if you want to compare to org-base-buffer,
>> I think I clearly need to set the marker to a position in the base
>> buffer.
> 
> Then what about comparing base buffer for both entities?

That would be the alternative, right.  Do you think this would be preferred to what I am doing right now?

- Carsten

> 
> 
> Regards,
> 
> -- 
> Nicolas Goaziou

Re: Exporter interface question

[Nicolas Goaziou]

Hello,

Carsten Dominik <carsten.dominik@gmail.com> writes:

>> Then what about comparing base buffer for both entities?
>
> That would be the alternative, right.  Do you think this would be
> preferred to what I am doing right now?

I didn't realize you had already modified the code. I think both methods
are equivalent.

Thank you again.


Regards,

-- 
Nicolas Goaziou

Re: Exporter interface question

[Carsten Dominik]

OK, great.

- Carsten

On 4.3.2013, at 15:45, Nicolas Goaziou <n.goaziou@gmail.com> wrote:

> Hello,
> 
> Carsten Dominik <carsten.dominik@gmail.com> writes:
> 
>>> Then what about comparing base buffer for both entities?
>> 
>> That would be the alternative, right.  Do you think this would be
>> preferred to what I am doing right now?
> 
> I didn't realize you had already modified the code. I think both methods
> are equivalent.
> 
> Thank you again.
> 
> 
> Regards,
> 
> -- 
> Nicolas Goaziou