Re: [sage-combinat-devel] WARNING: Docbuilder patch (#6495) going into combinat

[Nathann Cohen]

Helllo !!!

Now while we are at rebasing, the subdoc combinat is by far the biggest 
> one. So if we can break it further it will speedup the whole doc 
> compilation. I think we should do it right now otherwise there will be 
> another 
> big rebasing. Does anyone have some suggestion on how to break the doc in 
> smaller doc ? Should we have subdocs for subdirectories (eg: posets, 
> crystals, 
> root systems) ? 
>

Hmmm... Well, I've been thinking about it with my little brain, but it 
turns out I do not understand half of what is implemented in the combinat/ 
folder. I know what my technique would have been however : as I have been 
looking at this page [1] with very big eyes for a while (it's rather messy, 
don't you think ?) I would have first tried to reorder its elements, then 
try to regroup them into thematic subgroups. A bit like that [2] if 
possible. Then, hope that it can be done without splitting into different 
subgroups related documents (hey... Finding those groups IS a graph problem 
!) and try to regroup the Sphinx documents the same way :-)

At the very least, I think that Words should have a group of its own. But 
it's not a good idea to say so if I have no idea how I would regroup the 
rest of the documents ^^;

Oh, and also : these pages are usually easier to read when the .rst file 
hae :maxdepth: 1 instead of :maxdepth: 2.

 Hey ! I just noticed ! If you look at doc/en/reference/combinat you 
will find many "subgroups" of the combinat files already. Wouldn't that 
partiton work for you ? :-)

Have fuun !

Nathann

[1] http://www.sagemath.org/doc/reference/combinat/index.html
[2] http://www.sagemath.org/doc/reference/graphs.html

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[sage-combinat-devel] Re: WARNING: Docbuilder patch (#6495) going into combinat

[Travis Scrimshaw]

Hey Andrew,
There could likely be other dependencies that don't show up on trac 
because they were already merged into sage before the ticket came up for 
review. What I found was look for files which you get rejects on and see if 
there are any patches which modify said files between your version and 
5.7.rc0. I wish there was a better way to look through trac, but everytime 
I've done a search for a particular file, it hasn't worked for me... I'll 
do a more through search tomorrow (US Eastern time) if you haven't resolved 
it by then.

   There's another option (more of a hack solution), but create a dummy 
patch which just syncs up what the first #6495 patch expects to see and 
what is really there and guard it under #+5_7_beta2.

On Tuesday, February 19, 2013 10:11:57 PM UTC-5, Andrew Mathas wrote:
>
>
> You might have to add the patches from the dependency tickets listed at 
>> #6495.
>>
>> All of the patches listed on the ticket are already included BUT there is 
> also a dependency on 5.7.beta2, so I guess that in the series file the 
> guards should really be
>
> trac_6495-part1-moving-files-link.patch #+5_8_beta0 
> #-5_7_beta2
> trac_6495-part2-everything-else.patch  #+5_8_beta0 
> #-5_7_beta2
> ...
>
 
I'm fairly certain that would not be the correct guards because they would 
never apply.

Best,
Travis

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[sage-combinat-devel] Re: WARNING: Docbuilder patch (#6495) going into combinat

[Andrew Mathas]

> You might have to add the patches from the dependency tickets listed at 
> #6495.
>
> All of the patches listed on the ticket are already included BUT there is 
also a dependency on 5.7.beta2, so I guess that in the series file the 
guards should really be

trac_6495-part1-moving-files-link.patch #+5_8_beta0 
#-5_7_beta2
trac_6495-part2-everything-else.patch  #+5_8_beta0 
#-5_7_beta2
...

Unfortunately, the queue still won't apply to 5.6 even with these guards in 
place because later the patches which rely on #6495 now fail. 

As a result, I don't see an easy way to make the queue apply to both 5.6 
and 5.7.beta2+.

Andrew

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[sage-combinat-devel] Re: WARNING: Docbuilder patch (#6495) going into combinat

[John H Palmieri]

On Tuesday, February 19, 2013 6:23:08 PM UTC-8, Andrew Mathas wrote:
>
> Hi Travis,
>
> The queue isn't applying for in using version 5.6 even with this spkg 
> installed. Specifically, trac_6495-part1-moving-files-link.patch does not 
> apply. It's not a big deal as I can just move to 5.7.rc0 but it might 
> affect others.
>
> Andrew
>

You might have to add the patches from the dependency tickets listed at 
#6495.

-- 
John

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[sage-combinat-devel] Re: WARNING: Docbuilder patch (#6495) going into combinat

[Andrew Mathas]

Hi Travis,

The queue isn't applying for in using version 5.6 even with this spkg 
installed. Specifically, trac_6495-part1-moving-files-link.patch does not 
apply. It's not a big deal as I can just move to 5.7.rc0 but it might 
affect others.

Andrew

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[sage-combinat-devel] Re: Queue broken under 5.6 because of error in series file with #13862

[Travis Scrimshaw]

Hey Andrew,
   Whoops. Thank you for catching and fixing that.

Best,
Travis


On Tuesday, February 19, 2013 7:36:02 PM UTC-5, Andrew Mathas wrote:
>
> Hi,
>
> It seems that in the series file #13862 is incorrectly marked for merging 
> in 5.7.beta2 whereas trac says that it is merged in 5.6-beta2. As a 
> consequence, the queue does not currently apply to 5.6. 
>
> I will update the series file and push the change to the queue. Anyone who 
> objects is free to reply quickly or to reverse this change at their 
> leisure:)
>
> Andrew
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[sage-combinat-devel] Queue broken under 5.6 because of error in series file with #13862

[Andrew Mathas]

Hi,

It seems that in the series file #13862 is incorrectly marked for merging 
in 5.7.beta2 whereas trac says that it is merged in 5.6-beta2. As a 
consequence, the queue does not currently apply to 5.6. 

I will update the series file and push the change to the queue. Anyone who 
objects is free to reply quickly or to reverse this change at their 
leisure:)

Andrew

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sage-combinat-devel] WARNING: Docbuilder patch (#6495) going into combinat

[Florent Hivert]

  Hi Travis,

> More precisely, the new spkg in only needed here if you run sphinx for the 
> > first time (during the first run of sphinx it pickle Python's grammar 
> > leading 
> > to some race condition if it is done in parallel). So if you already 
> > compiled 
> > the doc (for example during Sage installation) and apply the new patch 
> > then 
> > you don't need the spkg. So I think no one will actually need it except 
> > for 
> > installing a brand new Sage. 
> >
> 
> Good to know. Thank you. Do you know if the sphinx spkg will be/is included 
> in the next beta/stable release?

Sure. Otherwise Sage doesn't compile ;-)

> > Now while we are at rebasing, the subdoc combinat is by far the biggest 
> > one. So if we can break it further it will speedup the whole doc 
> > compilation. I think we should do it right now otherwise there will be 
> > another 
> > big rebasing. Does anyone have some suggestion on how to break the doc in 
> > smaller doc ? Should we have subdocs for subdirectories (eg: posets, 
> > crystals, 
> > root systems) ? 

> My 2 cents is we could make a sucdoc subdirectory for anything that is 
> already in a subfolder of combinat. Also I think we could group the 
> crystals and rigged configurations grouped together. The #13605 crew were 
> thinking of grouping partitions and tableaux into two folders and I think 
> they could justifiably be put into a separate subdirectory as well (this 
> would likely be the most painful rebase which is why we didn't do it 
> there...). ...Maybe that was more like a nickel...

When creating a subdoc, there is no requirement on where the *source* (.py)
code should be. IE: there could be two source code in a different directory
and the same subdoc and on the contrary two source code in the same directory
could be in a different subdoc. Doc and source organizations are orthogonal.
Now, to create a subdoc you need to

 - create a new directory inside doc/en/reference/
 - copy the basic conf.py (eg: from doc/en/reference/misc/conf.py)
 - create an index.rst which include the files as usual
 - add the new created doc to multidocs_subdoc_list in doc/en/reference/conf.py

For example, I can fairly imagine having a
doc/en/reference/combinat-crystals. So the only change is that the index.rst
which include the doc is no longer at the same place. That's why I suggested
that while we are rebasing the patches because of the moving of the index.rst
file, we could as well split the index.rst.


Cheers,

Florent

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sage-combinat-devel] WARNING: Docbuilder patch (#6495) going into combinat

[Travis Scrimshaw]

Hey everyone,

More precisely, the new spkg in only needed here if you run sphinx for the 
> first time (during the first run of sphinx it pickle Python's grammar 
> leading 
> to some race condition if it is done in parallel). So if you already 
> compiled 
> the doc (for example during Sage installation) and apply the new patch 
> then 
> you don't need the spkg. So I think no one will actually need it except 
> for 
> installing a brand new Sage. 
>

Good to know. Thank you. Do you know if the sphinx spkg will be/is included 
in the next beta/stable release?
 

>
> When I worked on this one, I did it in a different branch. I therefore 
> didn't 
> anticipate that It will induce a huge rebasing of the queue. I'm sorry for 
> that though I don't really thing of a better solution than rebasing. I 
> nevertheless thing that it was indeed a needed feature. 
>

It actually wasn't as bad as I was fearing. I would have rebased the tree 
patch but I know it was getting some attention recently.


> Now while we are at rebasing, the subdoc combinat is by far the biggest 
> one. So if we can break it further it will speedup the whole doc 
> compilation. I think we should do it right now otherwise there will be 
> another 
> big rebasing. Does anyone have some suggestion on how to break the doc in 
> smaller doc ? Should we have subdocs for subdirectories (eg: posets, 
> crystals, 
> root systems) ? 
>
>  
My 2 cents is we could make a sucdoc subdirectory for anything that is 
already in a subfolder of combinat. Also I think we could group the 
crystals and rigged configurations grouped together. The #13605 crew were 
thinking of grouping partitions and tableaux into two folders and I think 
they could justifiably be put into a separate subdirectory as well (this 
would likely be the most painful rebase which is why we didn't do it 
there...). ...Maybe that was more like a nickel...

Best,
Travis

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sage-combinat-devel] WARNING: Docbuilder patch (#6495) going into combinat

[Florent Hivert]

  Hi there,

> > On Tue, Feb 19, 2013 at 05:20:20AM -0800, Travis Scrimshaw wrote:
> > >   I have (a few) patches which modifies the documentation, so I 
> > need to
> > >add ticket #6495 to the combinat queue. This will likely break a 
> > fair
> > >number of patches and you *will* need to download the spkg
> > >at 
> > http://sage.math.washington.edu/home/palmieri/SPKG/sphinx-1.1.2.p2.spkg 
> > .
> > 
> > Just a quick note: the spkg is only needed for compiling the
> > documentation. On the bright side, you will get parallel compilation
> > of the documentation! Finally!

More precisely, the new spkg in only needed here if you run sphinx for the
first time (during the first run of sphinx it pickle Python's grammar leading
to some race condition if it is done in parallel). So if you already compiled
the doc (for example during Sage installation) and apply the new patch then
you don't need the spkg. So I think no one will actually need it except for
installing a brand new Sage.

When I worked on this one, I did it in a different branch. I therefore didn't
anticipate that It will induce a huge rebasing of the queue. I'm sorry for
that though I don't really thing of a better solution than rebasing. I
nevertheless thing that it was indeed a needed feature.

Now while we are at rebasing, the subdoc combinat is by far the biggest
one. So if we can break it further it will speedup the whole doc
compilation. I think we should do it right now otherwise there will be another
big rebasing. Does anyone have some suggestion on how to break the doc in
smaller doc ? Should we have subdocs for subdirectories (eg: posets, crystals,
root systems) ?

cheers,

Florent

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sage-combinat-devel] WARNING: Docbuilder patch (#6495) going into combinat

[Anne Schilling]

It is pretty nice indeed! But now it does not tell you any longer which
file to look at. That was very useful information (at least for me)!

Best,

Anne

On 2/19/13 9:13 AM, Travis Scrimshaw wrote:
>Also, if you change a single file, don't run "sage -docbuild reference 
> html". Instead run "sage -docbuild reference/folder_name html", otherwise it 
> will go through the entire reference and check
> each sub-folder which takes some time.
> 
> Best,
> Travis
> 
> 
> On Tuesday, February 19, 2013 11:52:01 AM UTC-5, Nicolas M. Thiery wrote:
> 
> On Tue, Feb 19, 2013 at 05:20:20AM -0800, Travis Scrimshaw wrote:
> >   I have (a few) patches which modifies the documentation, so I 
> need to
> >add ticket #6495 to the combinat queue. This will likely break a fair
> >number of patches and you *will* need to download the spkg
> >at 
> http://sage.math.washington.edu/home/palmieri/SPKG/sphinx-1.1.2.p2.spkg 
> .
> 
> Just a quick note: the spkg is only needed for compiling the
> documentation. On the bright side, you will get parallel compilation
> of the documentation! Finally!
> 
> Cheers,
> Nicolas
> -- 
> Nicolas M. Thi�ry "Isil" >
> http://Nicolas.Thiery.name/
> 
> -- 
> You received this message because you are subscribed to the Google Groups 
> "sage-combinat-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-combinat-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-combinat-devel@googlegroups.com.
> Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
> For more options, visit https://groups.google.com/groups/opt_out.
>  
>  

-- 
Anne Schilling  E-mail: a...@math.ucdavis.edu
Professor   Web:http://www.math.ucdavis.edu/~anne
Department of Mathematics   Office: MSB 3222
University of CaliforniaPhone:  (530) 554 2326
One Shields Ave Fax:(530) 752 6635
Davis, CA 95616

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sage-combinat-devel] WARNING: Docbuilder patch (#6495) going into combinat

[Travis Scrimshaw]

   Also, if you change a single file, don't run "sage -docbuild reference 
html". Instead run "sage -docbuild reference/folder_name html", otherwise 
it will go through the entire reference and check each sub-folder which 
takes some time.

Best,
Travis


On Tuesday, February 19, 2013 11:52:01 AM UTC-5, Nicolas M. Thiery wrote:
>
> On Tue, Feb 19, 2013 at 05:20:20AM -0800, Travis Scrimshaw wrote: 
> >   I have (a few) patches which modifies the documentation, so I need 
> to 
> >add ticket #6495 to the combinat queue. This will likely break a fair 
> >number of patches and you *will* need to download the spkg 
> >at 
> http://sage.math.washington.edu/home/palmieri/SPKG/sphinx-1.1.2.p2.spkg. 
>
> Just a quick note: the spkg is only needed for compiling the 
> documentation. On the bright side, you will get parallel compilation 
> of the documentation! Finally! 
>
> Cheers, 
> Nicolas 
> -- 
> Nicolas M. Thi�ry "Isil" > 
> http://Nicolas.Thiery.name/ 
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sage-combinat-devel] WARNING: Docbuilder patch (#6495) going into combinat

[Nicolas M. Thiery]

On Tue, Feb 19, 2013 at 05:20:20AM -0800, Travis Scrimshaw wrote:
>   I have (a few) patches which modifies the documentation, so I need to
>add ticket #6495 to the combinat queue. This will likely break a fair
>number of patches and you *will* need to download the spkg
>at http://sage.math.washington.edu/home/palmieri/SPKG/sphinx-1.1.2.p2.spkg.

Just a quick note: the spkg is only needed for compiling the
documentation. On the bright side, you will get parallel compilation
of the documentation! Finally!

Cheers,
Nicolas
--
Nicolas M. Thiéry "Isil" 
http://Nicolas.Thiery.name/

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[sage-combinat-devel] Re: WARNING: Docbuilder patch (#6495) going into combinat

[Travis Scrimshaw]

Here's the list of patches I disabled:

- trac_10193-graded_enumerated_sets-vd_no_more_nt.patch
- trac_10193-review-nb.patch
- trac_10193-more-vd.patch

- trac_12940_affine_permutations-td.patch

- trac_8703-trees-fh.patch
- trac_13987_mary_trees-vp.patch
- trees_symmetry_factor-fh.patch
- trac_11529-rooted_trees-fh.patch
- shape_tree-fc.patch
- operads-fh.patch
- operads_more-fc.patch
- bintrees_leaf_paths-fh.patch
- q_tree_factorial-fc.patch
- algebras_over_operads-fc.patch
- shuffle-operads-fc.patch

- trac_14086-parking_functions-dm.patch

- sage-demos-and-tutorials-nt.patch

- trac_Kleshchev-partitions-am.patch

Here is the list of patches I rebased:

- trac_13580-map_reduce-old-fh.patch
- trac_10194-factories_policy-fh.patch
- trac_11407-list_clone_improve-fh.patch
- mutator-fh.patch
- trac_9280-graded-algebras-example-fs.patch
- trac_10963-more_functorial_constructions-nt.patch
- missing-doc-includes-nt.patch
- finite_semigroup-nt.patch

>From the rejects, I suspect most of the patches in the trees block will 
apply once #8703 is rebased.

(Miraculously) Sage starts with the full queue applied.

Best,
Travis

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[sage-combinat-devel] WARNING: Docbuilder patch (#6495) going into combinat

[Travis Scrimshaw]

Hey everyone,
   I have (a few) patches which modifies the documentation, so I need to 
add ticket #6495  to the 
combinat queue. This will likely break a fair number of patches and you 
*will* need to download the spkg at 
http://sage.math.washington.edu/home/palmieri/SPKG/sphinx-1.1.2.p2.spkg.

   It's going to take me a little bit to track down all of the dependencies 
and add them into the queue, but expect me to push and be disabling patches 
for rebase soon.

Best,
Travis

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.