sage-4.5.alpha1 -- GLPK test edition
Robert Miller
I've packaged up an alpha1 in the sage-4.5 series:
http://sage.math.washington.edu/home/rlmill/release/sage-4.5.alpha1.tar
I'm having fun managing this release, so in fact expect several more alphas.
This includes the following changes:
* removed doc/en/faq/Makefile from revision control
* 9152's sage library patches merged
* 9346 fixed fortran spkg for 32-bit Solaris
* 9368 sage_fortran ignores SAGE64 except on Solaris/OpenSolaris
* 9360 Sage 4.4.4 numerical noise in mwrank.pyx on 32-bit cicero
* 8284 IntervalGraph generator and a bug in is_chordal
* 7289 Multiway cut
* 7378 Subdivide edges in a graph
* 7476 Edge-disjoint spanning trees
* 7563 Interval Graphs : recognition
* 8748 Linear Arboricity, Acyclic edge coloring
* 8781 Overfull graph (and a bug in edge_coloring)
* 9269 clean up #optional tags in sage/graphs
* 9230 Broken docstrings in Travelling Salesman Problem
* 8870 Multiflow
* 9059 some shortcuts for "is_H-free" tests
* 9249 Wrong answer in is_hamiltonian if no LP solver available
* 7904 is_gallai_tree
* 8893 disjoint routed paths
* 8923 CompleteMultipartiteGraph generator
* 8927 is_triangle_free
* 9036 Graph.canonical_label(certify=True, edge_labels=True)
returns a name error
* 9073 Handle multigraphs better in planarity
* 9195 Recognition of split graphs (5 lines)
* 8403 Steiner Tree
I am now announcing an spkg feature freeze. I will still be merging
updates to the main sage library, but changes to scripts and spkgs
will be limited to mission critical bug fixes only, with one possible
exception (see below). I am paying particular attention to the
following issues:
9307 - remove GHMM
9365 - problems building R on OS X 10.6
9312 - update and fix GLPK
I would have done #9307, but it wasn't clear to me whether this ticket
required a reviewer, or what... I suppose there should be a patch to
deps there. I've been seeing some issues with mpmath that nobody else
has seen, so I'm waiting to see if those pop up in a fresh alpha1
build for anyone. (I'm hoping this will be the last "very alpha"
release in the series. Since I am expecting a few errors anyway, I am
posting this before I build and test myself, to give more time for the
testing mentioned below.) I am also wondering whether #9102 is really
a blocker, since we've done some releases without it. If it is, we
need to get it reviewed and merged.
=== PLEASE TEST GLPK! ===
You may have noticed a heavy emphasis on graph theory tickets. There
is a new spkg by David Kirkby for GLPK:
http://boxen.math.washington.edu/home/kirkby/patches/glpk-4.44.spkg
I am asking that all the usual build testers also do the following:
After building Sage, please download and install this new spkg, and
run both the usual set of tests, as well as
sage -t -long -only-optional=cbc,glpk $SAGE_ROOT/devel/sage-main
and report the results of this as well (the cbc,glpk thing will not
require CBC to be installed, but without that none of the tests for
GLPK will run - this is due to the way only-optional is implemented
right now).
There is a chance that GLPK will be included in sage-4.5 final. But
this is far from certain.
--
Robert L. Miller
http://www.rlmiller.org/
Nathann Cohen
Thank you very much !!!
Nathann
Robert Miller
> Need I mention that I quite *LOVE* this alpha release ? :-)
>
> Thank you very much !!!
>
Sage: increasing happiness around the world
> Nathann
Dr. David Kirkby
> === PLEASE TEST GLPK! ===
>
> You may have noticed a heavy emphasis on graph theory tickets. There
> is a new spkg by David Kirkby for GLPK:
>
> http://boxen.math.washington.edu/home/kirkby/patches/glpk-4.44.spkg
I did not really write it - I just attempted to sort out the quite numerous
issues with it. The title of that ticket
http://trac.sagemath.org/sage_trac/ticket/9312
does not convey the many problems that package had. (Error codes from
'configure', 'make', 'make install' were not checked. The test suite was not
run, etc etc etc)
> I am asking that all the usual build testers also do the following:
>
> After building Sage, please download and install this new spkg, and
> run both the usual set of tests, as well as
>
> sage -t -long -only-optional=cbc,glpk $SAGE_ROOT/devel/sage-main
>
> and report the results of this as well (the cbc,glpk thing will not
> require CBC to be installed, but without that none of the tests for
> GLPK will run - this is due to the way only-optional is implemented
> right now).
>
> There is a chance that GLPK will be included in sage-4.5 final. But
> this is far from certain.
I would have thought the earlier suggestion of Robert Bradshaw was a good one -
to make it optional for some time.
I believe the package in my directory still has some outstanding issues, as I
stated on the ticket.
Dave
Franco Saliola
> Hello,
>
> I've packaged up an alpha1 in the sage-4.5 series:
>
> http://sage.math.washington.edu/home/rlmill/release/sage-4.5.alpha1.tar
Build failed while trying to build opencdk:
****************************************************
Host system
uname -a:
Linux massena 2.6.23-gentoo-r6 #1 SMP Wed Feb 6 21:49:58 CET 2008
x86_64 Intel(R) Xeon(R) CPU X5365 @ 3.00GHz GenuineIntel
GNU/Linux
****************************************************
I've attached the relevant bit of install.log.
Franco
--
Robert Miller
<david....@onetel.net> wrote:
> I would have thought the earlier suggestion of Robert Bradshaw was a good
> one - to make it optional for some time.
The GLPK package has been optional for over a year.
> I believe the package in my directory still has some outstanding issues, as
> I stated on the ticket.
I am particularly interested in resolving these issues and getting it
into Sage. That is why I'm asking testers to check this new one.
Justin C. Walker
I've packaged up an alpha1 in the sage-4.5 series:
http://sage.math.washington.edu/home/rlmill/release/sage-4.5.alpha1.tar
--
Justin C. Walker, Curmudgeon at Large
Institute for the Absorption of Federal Funds
-----------
My wife 'n kids 'n dogs are gone,
I can't get Jesus on the phone,
But Ol' Milwaukee's Best is my best friend.
-----------
Nathann Cohen
The install log is at ~ncohen/install.log
I run it again without any threads, just to check.
Nathann
> --
> You received this message because you are subscribed to the Google Groups
> "sage-release" group.
> To post to this group, send email to sage-r...@googlegroups.com.
> To unsubscribe from this group, send email to
> sage-release...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/sage-release?hl=en.
>
Justin C. Walker
On 29 Jun, 2010, at 10:15 AM, Robert Miller wrote:I've packaged up an alpha1 in the sage-4.5 series:
http://sage.math.washington.edu/home/rlmill/release/sage-4.5.alpha1.tarStill having 64-bit problems on Mac OS X. This time, the problem is in the building of the rubik's cube package:
--
Justin C. Walker, Curmudgeon-At-Large
Director
Institute for the Enhancement of the Director's Income
--------
"Weaseling out of things is what separates us from the animals.
Well, except the weasel."
- Homer J Simpson
--------
leif
> Hello,
>
> I've packaged up an alpha1 in the sage-4.5 series:
>
> http://sage.math.washington.edu/home/rlmill/release/sage-4.5.alpha1.tar
>
> I'm having fun managing this release, so in fact expect several more alphas.
I'm fine with testing one alpha per week... ;-)
Ubuntu 9.04 x86_64 (Core2, sequential 64-bit build), gcc 4.5.0:
make build: OK
make doc: OK
make ptestlong: OK (all tests passed)
Haven't tested GLPK 4.44 yet.
[Won't test x86/32-bit before Sunday.]
> [...]
> (I'm hoping this will be the last "very alpha"
> release in the series. Since I am expecting a few errors anyway, I am
> posting this before I build and test myself, to give more time for the
> testing mentioned below.)
Ok for me - alphas are expected to have deficiencies...
> I am also wondering whether #9102 is really
> a blocker, since we've done some releases without it. If it is, we
> need to get it reviewed and merged.
I think not building the PDF documentation should be a blocker for a final.
> [...]
-Leif
Nathann Cohen
MAKE="make -j4" :-)
Nathann
Dr. David Kirkby
Was you previously trying to build packages in parallel with
SAGE_PARALLEL_SPKG_BUILD=yes ?
I think building packages in parallel should be considered experimental, but if
simply running 'make -j4' causes problems, then they need to be solved before
releasing since lots of people do that naturally. (Even on uni-processor
machines, it usually speeds the build process a bit).
The parallel building of packages can seem to create a few issues if there are
build failures. I think at the very least I need to understand it better, but
the ability to build Sage in 15 minutes or so is quite attractive to me.
Dave
John H Palmieri
> I am asking that all the usual build testers also do the following:
>
> After building Sage, please download and install this new spkg, and
> run both the usual set of tests, as well as
>
> sage -t -long -only-optional=cbc,glpk $SAGE_ROOT/devel/sage-main
"only-optional" doctesting does for me. By the way, with the -only-
optional tag, I can't stop doctesting with ctrl-c -- a bug?
----------------------------------------------------------------------
The following tests failed:
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/build/lib.macosx-10.6-i386-2.6/sage/graphs/digraph.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/build/lib.macosx-10.6-i386-2.6/sage/graphs/generic_graph.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/build/lib.macosx-10.6-i386-2.6/sage/graphs/graph.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/build/lib.macosx-10.6-i386-2.6/sage/graphs/
graph_coloring.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/build/lib.macosx-10.6-i386-2.6/sage/numerical/optimize.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/build/sage/graphs/digraph.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/build/sage/graphs/generic_graph.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/build/sage/graphs/graph.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/build/sage/graphs/graph_coloring.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/build/sage/numerical/mip.pyx"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/build/sage/numerical/optimize.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/sage/graphs/digraph.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/sage/graphs/generic_graph.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/sage/graphs/graph.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/sage/graphs/graph_coloring.py"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/sage/numerical/mip.pyx"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/sage/numerical/mip_coin.pyx"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/sage/numerical/mip_glpk.pyx"
sage -t -long -only-optional=cbc,glpk "builds/sage-4.5.alpha1/devel/
sage-main/sage/numerical/optimize.py"
Total time for all tests: 923.8 seconds
--
John
Nathann Cohen
So it seems all the files using LP failed. Could I ask you whether you
ran "sage -b" between the installation of the GLPK package and the
tests ? If you did not, could you give it a try ? :-)
The GLPK package currently needs such a command to be typed after its
installation. It prints a message asking the user to do it when
installed, though David's version does not as it is meant to become a
standard glpk package, in which case there is no reason to ask such a
thing from the user....
Thankssssssss ! :-)
Nathann
Robert Miller
> usage: mktemp [-d] [-q] [-t prefix] [-u] template ...
> mktemp [-d] [-q] [-u] -t prefix
This was caused by Dave Kirkby's patch at #9030. As Justin and I have
discovered here at MSRI in person, the way mktemp works does not
support
$ mktemp
on OS X. You need something more in that case, namely the usage
pattern mentioned above. The following syntax will work on Solaris,
Linux and OS X:
$ mktemp /tmp/temp.XXXXXXX
So if you use mktemp, beware! New spkg up at #9388 shortly.
leif
Exactly. (Without ./sage -b, "all" tests fail...)
Note that the tests are redundant and only 5 Python and 3 Cython files
had doctest failures (without rebuilding the Sage library), so I ran:
#!/bin/bash
./sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/mip_coin.pyx"
./sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/mip_glpk.pyx"
./sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/mip.pyx"
./sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/optimize.py"
./sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/graph.py"
./sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/graph_coloring.py"
./sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/generic_graph.py"
./sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/digraph.py"
With GLPK 4.44 (but not CBC) installed, I got the following:
============================================================
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/mip_coin.pyx"
**********************************************************************
File
"/home/leif/sage-4.5.alpha1/devel/sage-main/sage/numerical/mip_coin.pyx",
line 39:
sage: from sage.numerical.mip_coin import solve_coin # optional
- CBC
Exception raised:
Traceback (most recent call last):
File "/home/leif/sage-4.5.alpha1/local/bin/ncadoctest.py", line
1231, in run_one_test
self.run_one_example(test, example, filename, compileflags)
File "/home/leif/sage-4.5.alpha1/local/bin/sagedoctest.py", line
38, in run_one_example
OrigDocTestRunner.run_one_example(self, test, example, filename,
compileflags)
File "/home/leif/sage-4.5.alpha1/local/bin/ncadoctest.py", line
1172, in run_one_example
compileflags, 1) in test.globs
File "<doctest __main__.example_0[2]>", line 1, in <module>
from sage.numerical.mip_coin import solve_coin # optional -
CBC###line 39:
sage: from sage.numerical.mip_coin import solve_coin # optional
- CBC
ImportError: No module named mip_coin
**********************************************************************
File
"/home/leif/sage-4.5.alpha1/devel/sage-main/sage/numerical/mip_coin.pyx",
line 47:
sage: solve_coin(p,objective_only=True) # optional - CBC
Exception raised:
Traceback (most recent call last):
File "/home/leif/sage-4.5.alpha1/local/bin/ncadoctest.py", line
1231, in run_one_test
self.run_one_example(test, example, filename, compileflags)
File "/home/leif/sage-4.5.alpha1/local/bin/sagedoctest.py", line
38, in run_one_example
OrigDocTestRunner.run_one_example(self, test, example, filename,
compileflags)
File "/home/leif/sage-4.5.alpha1/local/bin/ncadoctest.py", line
1172, in run_one_example
compileflags, 1) in test.globs
File "<doctest __main__.example_0[9]>", line 1, in <module>
solve_coin(p,objective_only=True) # optional - CBC###line 47:
sage: solve_coin(p,objective_only=True) # optional - CBC
NameError: name 'solve_coin' is not defined
**********************************************************************
1 items had failures:
2 of 10 in __main__.example_0
***Test Failed*** 2 failures.
For whitespace errors, see the file
/home/leif/.sage//tmp/.doctest_mip_coin.py
[1.1 s]
----------------------------------------------------------------------
The following tests failed:
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/mip_coin.pyx"
Total time for all tests: 1.1 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/mip_glpk.pyx"
[1.1 s]
----------------------------------------------------------------------
All tests passed!
Total time for all tests: 1.1 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/mip.pyx"
[1.3 s]
----------------------------------------------------------------------
All tests passed!
Total time for all tests: 1.3 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/optimize.py"
**********************************************************************
File
"/home/leif/sage-4.5.alpha1/devel/sage-main/sage/numerical/optimize.py",
line 700:
sage: print sorted(binpacking([1/5,1/3,2/3,3/4, 5/7])) # optional -
GLPK, CBC
Expected:
[[1/5, 3/4], [1/3, 2/3], [5/7]]
Got:
[[1/5, 5/7], [1/3, 2/3], [3/4]]
**********************************************************************
1 items had failures:
1 of 6 in __main__.example_0
***Test Failed*** 1 failures.
For whitespace errors, see the file
/home/leif/.sage//tmp/.doctest_optimize.py
[1.1 s]
----------------------------------------------------------------------
The following tests failed:
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/optimize.py"
Total time for all tests: 1.1 seconds
sage -t -long -only-optional=cbc,glpk "devel/sage-main/sage/graphs/graph.py"
[6.1 s]
----------------------------------------------------------------------
All tests passed!
Total time for all tests: 6.1 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/graph_coloring.py"
[4.4 s]
----------------------------------------------------------------------
All tests passed!
Total time for all tests: 4.4 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/generic_graph.py"
[6.5 s]
----------------------------------------------------------------------
All tests passed!
Total time for all tests: 6.5 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/digraph.py"
[1.4 s]
----------------------------------------------------------------------
All tests passed!
Total time for all tests: 1.4 seconds
*******************************************************************
(The first error obviously occured just because CBC wasn't installed.)
With CBC 2.3.p2 installed (again after rerunning ./sage -b), I got:
===================================================================
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/mip_coin.pyx"
[1.4 s]
----------------------------------------------------------------------
All tests passed!
Total time for all tests: 1.4 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/mip_glpk.pyx"
[1.2 s]
----------------------------------------------------------------------
All tests passed!
Total time for all tests: 1.2 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/mip.pyx"
Writing problem data to
`/home/leif/.sage//temp/quadriga/22741//lp_problem.mps'...
17 records were written
Writing problem data to
`/home/leif/.sage//temp/quadriga/22741//lp_problem.lp'...
9 lines were written
**********************************************************************
File
"/home/leif/sage-4.5.alpha1/devel/sage-main/sage/numerical/mip.pyx",
line 740:
sage: p.solve() # optional - GLPK, CBC
Expected:
6.6666666666666661
Got:
6.6666665077209473
**********************************************************************
1 items had failures:
1 of 20 in __main__.example_5
***Test Failed*** 1 failures.
For whitespace errors, see the file /home/leif/.sage//tmp/.doctest_mip.py
[1.3 s]
----------------------------------------------------------------------
The following tests failed:
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/mip.pyx"
Total time for all tests: 1.3 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/optimize.py"
**********************************************************************
File
"/home/leif/sage-4.5.alpha1/devel/sage-main/sage/numerical/optimize.py",
line 700:
sage: print sorted(binpacking([1/5,1/3,2/3,3/4, 5/7])) # optional -
GLPK, CBC
Expected:
[[1/5, 3/4], [1/3, 2/3], [5/7]]
Got:
[[1/5, 5/7], [1/3, 2/3], [3/4]]
**********************************************************************
1 items had failures:
1 of 6 in __main__.example_0
***Test Failed*** 1 failures.
For whitespace errors, see the file
/home/leif/.sage//tmp/.doctest_optimize.py
[1.2 s]
----------------------------------------------------------------------
The following tests failed:
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/numerical/optimize.py"
Total time for all tests: 1.2 seconds
sage -t -long -only-optional=cbc,glpk "devel/sage-main/sage/graphs/graph.py"
python: ClpSimplexDual.cpp:5617: int
ClpSimplexDual::strongBranching(int, const int*, double*, double*,
double**, int*, int*, bool, bool, int): Assertion `inverseColumnScale_'
failed.
Aborted
[2.5 s]
----------------------------------------------------------------------
The following tests failed:
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/graph.py"
Total time for all tests: 2.5 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/graph_coloring.py"
python: ClpSimplexDual.cpp:5617: int
ClpSimplexDual::strongBranching(int, const int*, double*, double*,
double**, int*, int*, bool, bool, int): Assertion `inverseColumnScale_'
failed.
Aborted
[1.7 s]
----------------------------------------------------------------------
The following tests failed:
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/graph_coloring.py"
Total time for all tests: 1.7 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/generic_graph.py"
python: ClpSimplexDual.cpp:5617: int
ClpSimplexDual::strongBranching(int, const int*, double*, double*,
double**, int*, int*, bool, bool, int): Assertion `inverseColumnScale_'
failed.
Aborted
[2.0 s]
----------------------------------------------------------------------
The following tests failed:
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/generic_graph.py"
Total time for all tests: 2.0 seconds
sage -t -long -only-optional=cbc,glpk
"devel/sage-main/sage/graphs/digraph.py"
[2.9 s]
----------------------------------------------------------------------
All tests passed!
Total time for all tests: 2.9 seconds
So with CBC installed, I get numerical noise and failed assertions that
don't occur with only GLPK installed.
-Leif
John H Palmieri
On Jun 29, 5:21 pm, Nathann Cohen <nathann.co...@gmail.com> wrote:
> > The following tests failed:
>
> So it seems all the files using LP failed. Could I ask you whether you
> ran "sage -b" between the installation of the GLPK package and the
> tests ? If you did not, could you give it a try ? :-)
----------------------------------------------------------------------
The following tests failed:
lib.macosx-10.6-i386-2.6/sage/numerical/optimize.py"
sage -t -long -only-optional=cbc,glpk "devel/sage-main/build/sage/
numerical/optimize.py"
sage -t -long -only-optional=cbc,glpk "devel/sage-main/sage/numerical/
mip_coin.pyx"
sage -t -long -only-optional=cbc,glpk "devel/sage-main/sage/numerical/
optimize.py"
--
John
leif
> Robert Miller wrote:
>> Hello,
>>
>> I've packaged up an alpha1 in the sage-4.5 series:
>>
>> http://sage.math.washington.edu/home/rlmill/release/sage-4.5.alpha1.tar
>>
>> I'm having fun managing this release, so in fact expect several more alphas.
>
> I'm fine with testing one alpha per week... ;-)
>
>
> Ubuntu 9.04 x86_64 (Core2, sequential 64-bit build), gcc 4.5.0:
>
> make build: OK
> make doc: OK
> make ptestlong: OK (all tests passed)
>
>
> Haven't tested GLPK 4.44 yet.
(See also my previous post w.r.t. GLPK and CBC/optional tests.)
With GLPK and CBC installed, rerunning "make ptestlong" gives:
[...]
----------------------------------------------------------------------
The following tests failed:
sage -t -long
/leif/sage-4.5.alpha1/devel/sage/sage/graphs/generic_graph.py # 0
doctests failed
----------------------------------------------------------------------
Total time for all tests: 2134.2 seconds
[sic - filename left-truncated]
leif@quadriga:~/sage-4.5.alpha1$ ./sage -t -long -verbose
/home/leif/sage-4.5.alpha1/devel/sage/sage/graphs/generic_graph.py
[...]
Trying:
R = graphs.RandomGNP(Integer(2000), RealNumber('.01'))###line
13209:_sage_ >>> R = graphs.RandomGNP(2000, .01)
Expecting nothing
ok
Trying:
R.is_vertex_transitive()###line 13210:_sage_ >>> R.python:
ClpSimplexDual.cpp:5617: int ClpSimplexDual::strongBranching(int, const
int*, double*, double*, double**, int*, int*, bool, bool, int):
Assertion `inverseColumnScale_' failed.
is_vertex_transitive()
Expecting:
False
ok
Trying:
set_random_seed(0L)
Expecting nothing
ok
Trying:
change_warning_output(sys.stdout)
Expecting nothing
ok
Trying:
g = graphs.HeawoodGraph()###line 13259:_sage_ >>> g =
graphs.HeawoodGraph()
Expecting nothing
ok
Trying:
g = graphs.PetersenGraph()###line 13265:_sage_ >>> g =
graphs.PetersenGraph()
Expecting nothing
ok
Trying:
from sage.misc.exceptions import OptionalPackageNotFoundError###line
13274:_sage_ >>> from sage.misc.exceptions import
OptionalPackageNotFoundError
Expecting nothing
ok
Trying:
try:###line 13275:_sage_ >>> try:
g = graphs.ChvatalGraph()
if not g.is_hamiltonian():
print "There is something wrong here !"
except OptionalPackageNotFoundError:
pass
Expecting nothing
Aborted
[33.7 s]
----------------------------------------------------------------------
The following tests failed:
sage -t -long -verbose "devel/sage/sage/graphs/generic_graph.py"
Total time for all tests: 33.7 seconds
leif@quadriga:~/sage-4.5.alpha1$
-Leif
Robert Miller
On Tue, Jun 29, 2010 at 6:25 PM, leif <not.r...@online.de> wrote:
> Note that the tests are redundant and only 5 Python and 3 Cython files
> had doctest failures (without rebuilding the Sage library), so I ran:
>
> #!/bin/bash
> ./sage -t -long -only-optional=cbc,glpk
> "devel/sage-main/sage/numerical/mip_coin.pyx"
> ./sage -t -long -only-optional=cbc,glpk
> "devel/sage-main/sage/numerical/mip_glpk.pyx"
> ./sage -t -long -only-optional=cbc,glpk
> "devel/sage-main/sage/numerical/mip.pyx"
> ./sage -t -long -only-optional=cbc,glpk
> "devel/sage-main/sage/numerical/optimize.py"
> ./sage -t -long -only-optional=cbc,glpk
> "devel/sage-main/sage/graphs/graph.py"
> ./sage -t -long -only-optional=cbc,glpk
> "devel/sage-main/sage/graphs/graph_coloring.py"
> ./sage -t -long -only-optional=cbc,glpk
> "devel/sage-main/sage/graphs/generic_graph.py"
> ./sage -t -long -only-optional=cbc,glpk
> "devel/sage-main/sage/graphs/digraph.py"
>
In the future, can you use pastebin.com or something for logs this
big? Thank you!
Robert Miller
Do you still have the particular failures? Can you post them/link to them?
On Tue, Jun 29, 2010 at 6:27 PM, John H Palmieri <jhpalm...@gmail.com> wrote:
> ----------------------------------------------------------------------
> The following tests failed:
>
>
> sage -t -long -only-optional=cbc,glpk "devel/sage-main/build/
> lib.macosx-10.6-i386-2.6/sage/numerical/optimize.py"
> sage -t -long -only-optional=cbc,glpk "devel/sage-main/build/sage/
> numerical/optimize.py"
> sage -t -long -only-optional=cbc,glpk "devel/sage-main/sage/numerical/
> mip_coin.pyx"
> sage -t -long -only-optional=cbc,glpk "devel/sage-main/sage/numerical/
> optimize.py"
>
> --
> John
>
Justin C. Walker
On 29 Jun, 2010, at 10:15 AM, Robert Miller wrote:I've packaged up an alpha1 in the sage-4.5 series:
http://sage.math.washington.edu/home/rlmill/release/sage-4.5.alpha1.tarStill having 64-bit problems on Mac OS X. This time, the problem is in the building of the rubik's cube package:
--
Justin C. Walker, Curmudgeon-at-Large
() The ASCII Ribbon Campaign
/\ Help Cure HTML Email
John H Palmieri
On Jun 29, 8:42 pm, Robert Miller <r...@rlmiller.org> wrote:
> John,
>
> Do you still have the particular failures? Can you post them/link to them?
>
directory. I guess the errors in the first file are unrelated to GLPK
(I don't have CBC installed).
sage -t --long -only-optional=glpk,cbc "devel/sage/sage/numerical/
mip_coin.pyx"
**********************************************************************
File "/Applications/sage_builds/sage-4.5.alpha1/devel/sage/sage/
sage: from sage.numerical.mip_coin import solve_coin #
optional - CBC
Exception raised:
Traceback (most recent call last):
self.run_one_example(test, example, filename, compileflags)
OrigDocTestRunner.run_one_example(self, test, example,
filename, compileflags)
compileflags, 1) in test.globs
File "<doctest __main__.example_0[2]>", line 1, in <module>
from sage.numerical.mip_coin import solve_coin # optional
- CBC###line 39:
sage: from sage.numerical.mip_coin import solve_coin #
optional - CBC
ImportError: No module named mip_coin
**********************************************************************
sage: solve_coin(p,objective_only=True) # optional - CBC
Exception raised:
Traceback (most recent call last):
self.run_one_example(test, example, filename, compileflags)
OrigDocTestRunner.run_one_example(self, test, example,
filename, compileflags)
compileflags, 1) in test.globs
File "<doctest __main__.example_0[9]>", line 1, in <module>
solve_coin(p,objective_only=True) # optional - CBC###line
47:
sage: solve_coin(p,objective_only=True) # optional - CBC
NameError: name 'solve_coin' is not defined
**********************************************************************
1 items had failures:
2 of 10 in __main__.example_0
***Test Failed*** 2 failures.
tmp/.doctest_mip_coin.py
[7.9 s]
----------------------------------------------------------------------
sage -t --long -only-optional=glpk,cbc "devel/sage/sage/numerical/
optimize.py"
**********************************************************************
File "/Applications/sage_builds/sage-4.5.alpha1/devel/sage/sage/
sage: print sorted(binpacking([1/5,1/3,2/3,3/4, 5/7])) # optional
- GLPK, CBC
Expected:
[[1/5, 3/4], [1/3, 2/3], [5/7]]
Got:
[[1/5, 5/7], [1/3, 2/3], [3/4]]
**********************************************************************
1 items had failures:
1 of 6 in __main__.example_0
***Test Failed*** 1 failures.
tmp/.doctest_optimize.py
[2.4 s]
Rob Beezer
passes all tests on
(1) 64-bit Kubuntu 9.10 on Intel Core Duo
(2) 64-bit Ubuntu 10.04 on Intel Core i7
I'll test GLPK tomorrow.
Rob
Alex Ghitza
Built and ran make ptestlong with no failures on
[aghitza@soleil sage-4.5.alpha1]$ uname -a
Linux soleil.ms.unimelb.edu.au 2.6.18-194.3.1.el5 #1 SMP Sun May 2 04:17:42 EDT 2010 x86_64 x86_64 x86_64 GNU/Linux
[aghitza@soleil sage-4.5.alpha1]$ cat /etc/issue
Red Hat Enterprise Linux Server release 5.5 (Tikanga)
I'll give GLPK a try when I get a chance.
Best,
Alex
--
Alex Ghitza -- http://aghitza.org/
Lecturer in Mathematics -- The University of Melbourne -- Australia
Nathann Cohen
mip_coin.pyx, which is perfectly normal as Cbc is not available.
The error coming from optimize.py is the usual problem of having
different optimal solutions for the same LP.... And is fixed there :
http://trac.sagemath.org/sage_trac/ticket/9392
Nathann
davidloeffler
On 29 June, 18:15, Robert Miller <r...@rlmiller.org> wrote:
> === PLEASE TEST GLPK! ===
>
> You may have noticed a heavy emphasis on graph theory tickets. There
> is a new spkg by David Kirkby for GLPK:
>
> http://boxen.math.washington.edu/home/kirkby/patches/glpk-4.44.spkg
>
> I am asking that all the usual build testers also do the following:
>
> After building Sage, please download and install this new spkg, and
> run both the usual set of tests, as well as
>
> sage -t -long -only-optional=cbc,glpk $SAGE_ROOT/devel/sage-main
>
> and report the results of this as well
and long doctests pass. With the optional GLPK spkg, I get failures in
2 files (output below) which look like exactly the same failures also
reported by John Palmieri.
sage -t -long -only-optional=cbc,glpk "devel/sage-main/sage/numerical/
optimize.py"
File "/storage/masiao/sage-4.5.alpha1/devel/sage-main/sage/numerical/
sage: print sorted(binpacking([1/5,1/3,2/3,3/4, 5/7])) # optional
- GLPK, CBC
Expected:
[[1/5, 3/4], [1/3, 2/3], [5/7]]
Got:
[[1/5, 5/7], [1/3, 2/3], [3/4]]
**********************************************************************
mip_coin.pyx"
File "/storage/masiao/sage-4.5.alpha1/devel/sage-main/sage/numerical/
sage: from sage.numerical.mip_coin import solve_coin #
optional - CBC
Exception raised:
Traceback (most recent call last):
self.run_one_example(test, example, filename, compileflags)
OrigDocTestRunner.run_one_example(self, test, example,
filename, compileflags)
compileflags, 1) in test.globs
File "<doctest __main__.example_0[2]>", line 1, in <module>
from sage.numerical.mip_coin import solve_coin # optional
- CBC###line 39:
sage: from sage.numerical.mip_coin import solve_coin #
optional - CBC
ImportError: No module named mip_coin
**********************************************************************
sage: solve_coin(p,objective_only=True) # optional - CBC
Exception raised:
Traceback (most recent call last):
self.run_one_example(test, example, filename, compileflags)
OrigDocTestRunner.run_one_example(self, test, example,
filename, compileflags)
Marshall Hampton
On Jun 29, 12:15 pm, Robert Miller <r...@rlmiller.org> wrote:
>
> I am now announcing an spkg feature freeze. I will still be merging
> updates to the main sage library, but changes to scripts and spkgs
> will be limited to mission critical bug fixes only, with one possible
> exception (see below).
I would like to put in a plea for #9010, its an optional package
update that would make my life a bit easier.
-Marshall
Rob Beezer
After installing Kirby's package and running sage -b I get what
appear to be the same 4 failures John P gets:
sage -t -long -only-optional=cbc,glpk "devel/sage-main/sage/numerical/
optimize.py"
mip_coin.pyx"
numerical/optimize.py"
x86_64-2.6/sage/numerical/optimize.py"
Same result on two machines, both 64-bit (K)Ubuntu, different
processors (Core Duo, Core i7).
I could post logs with details if somebody reminds me just where to
find them!
Thanks to Robert Miller, Nathann Cohen and Dave Kirby (and anybody I'm
missing) for their coordinated work to make this happen. It'll be
great to have this functionality easily available.
Rob
Robert Miller
> I could post logs with details if somebody reminds me just where to
> find them!
Rob,
That's okay, it's most likely the same as what we've seen above.
Thanks for testing, as always!
----------------
My plan is to try to resolve these issues and merge GLPK in
sage-4.5.alpha2. This means most likely several days before alpha2,
which will focus just on getting GLPK in (plus fixing the build fails
mentioned earlier in the thread). I'm confident, based on what I've
seen here, that we can get this accomplished before the end of the
week. If there are still issues and we can't resolve them, then I will
base alpha3 on alpha1, and not include GLPK (but will include the
fixes for build fails).
Dave Kirkby will probably be pleased to hear that this will involve a
substantial stabilizing period, after which there will be one big push
to get sage library patches in, and then a full feature freeze
(followed by another stabilizing period before final release). Also
note that this is a two digit release which is not a bugfix only
release. ;)
leif
> My plan is to try to resolve these issues and merge GLPK in
> sage-4.5.alpha2. This means most likely several days before alpha2,
> which will focus just on getting GLPK in (plus fixing the build fails
> mentioned earlier in the thread). I'm confident, based on what I've
> seen here, that we can get this accomplished before the end of the
> week. If there are still issues and we can't resolve them, then I will
> base alpha3 on alpha1, and not include GLPK (but will include the
> fixes for build fails).
Please also consider #8680.
http://trac.sagemath.org/sage_trac/ticket/8680
> Dave Kirkby will probably be pleased to hear that this will involve a
> substantial stabilizing period, after which there will be one big push
> to get sage library patches in, and then a full feature freeze
> (followed by another stabilizing period before final release). Also
> note that this is a two digit release which is not a bugfix only
> release. ;)
Well, probably not just Dave (and me)... ;-)
-Leif
P.S.: 4.5.0 would compete with gcc..., but that's three digits?
leif
Note that this was meant to only refer to the (tab-removing) Sage
library patch there, not the patch to "sage-doctest" (which IMHO needs
work - more to come on the ticket, but not immediately); the former
could be merged independently.
(The latter patch is 3 month old, so I totally forgot about it - sorry.)
-Leif
John H Palmieri
> leif wrote:
> > Please also consider #8680.
> >http://trac.sagemath.org/sage_trac/ticket/8680
>
> Note that this was meant to only refer to the (tab-removing) Sage
> library patch there, not the patch to "sage-doctest" (which IMHO needs
> work - more to come on the ticket, but not immediately)
it, please clarify right away. I haven't seen any discussion on the
ticket about issues with the patch to sage-doctest.
--
John
davidloeffler
On 1 July, 22:54, John H Palmieri <jhpalmier...@gmail.com> wrote:
> The ticket has a positive review, so if there are issues with any of
> it, please clarify right away. I haven't seen any discussion on the
> ticket about issues with the patch to sage-doctest.
advocate the sage-doctest patch being merged ASAP. Anyone who doesn't
like the formatting of the warning messages it produces or stuff like
that it welcome to open a follow-up ticket, but we should definitely
get it merged now before this infernal plague of tab characters starts
to gain ground again! ;-)
Regards, David
Robert Miller
SAGE_ROOT/local/bin$ hg status
? fsdump
? fsoids
? fsrefs
? fstail
? mkzeoinst
? repozo
? runzeo
? zdaemon
? zeoctl
? zeopack
? zeopasswd
They all seem ZODB related, which is puzzling since I didn't update
it. But I'm sure there's a reason...
If anyone else knows what is up, please input. Otherwise I can do some digging.
leif
> Does anyone else have this on their alpha1?
>
> SAGE_ROOT/local/bin$ hg status
> ? fsdump
> ? fsoids
> ? fsrefs
> ? fstail
> ? mkzeoinst
> ? repozo
> ? runzeo
> ? zdaemon
> ? zeoctl
> ? zeopack
> ? zeopasswd
>
Funny, I just wanted to post this, too. ;-)
(I do have the same, but didn't try to track this down. The files are
also present in alpha0.)
-Leif
leif
> Does anyone else have this on their alpha1?
>
> SAGE_ROOT/local/bin$ hg status
> ? fsdump
> ? fsoids
> ? fsrefs
> ? fstail
> ? mkzeoinst
> ? repozo
> ? runzeo
> ? zdaemon
> ? zeoctl
> ? zeopack
> ? zeopasswd
In 4.4.4, we have:
-rwxr-xr-x 1 leif64 leif64 165 Jun 30 16:02 fsdump.py
-rwxr-xr-x 1 leif64 leif64 2352 Jun 30 16:02 fsoids.py
-rwxr-xr-x 1 leif64 leif64 5938 Jun 30 16:02 fsrefs.py
-rwxr-xr-x 1 leif64 leif64 1551 Jun 30 16:02 fstail.py
-rwxr-xr-x 1 leif64 leif64 6699 Jun 30 16:02 fstest.py
-rwxr-xr-x 1 leif64 leif64 744 Jun 30 16:02 mkzeoinst.py
-rwxr-xr-x 1 leif64 leif64 17030 Jun 30 16:02 repozo.py
-rwxr-xr-x 1 leif64 leif64 696 Jun 30 16:02 runzeo.py
-rwxr-xr-x 1 leif64 leif64 767 Jun 30 16:02 zeoctl.py
-rwxr-xr-x 1 leif64 leif64 3182 Jun 30 16:02 zeopack.py
-rwxr-xr-x 1 leif64 leif64 723 Jun 30 16:02 zeopasswd.py
leif64@portland:~/Sage/sage-4.4.4/local/bin$ hg status
leif64@portland:~/Sage/sage-4.4.4/local/bin$
(There's no zdaemon*, but in addition fstest.py.)
zodb3's install log (on 4.5.alpha0) contains:
[...]
Processing ZODB3-3.7.0-py2.6-linux-i686.egg
removing
'/home/leif/sage-4.5.alpha0/local/lib/python2.6/site-packages/ZODB3-3.7.0-py2.6-linux-i686.egg'
(and everything under it)
creating
/home/leif/sage-4.5.alpha0/local/lib/python2.6/site-packages/ZODB3-3.7.0-py2.6-linux-i686.egg
Extracting ZODB3-3.7.0-py2.6-linux-i686.egg to
/home/leif/sage-4.5.alpha0/local/lib/python2.6/site-packages
ZODB3 3.7.0 is already the active version in easy-install.pth
Installing mkzeoinst script to /home/leif/sage-4.5.alpha0/local/bin
Installing fstail script to /home/leif/sage-4.5.alpha0/local/bin
Installing zeopack script to /home/leif/sage-4.5.alpha0/local/bin
Installing runzeo script to /home/leif/sage-4.5.alpha0/local/bin
Installing zeopasswd script to /home/leif/sage-4.5.alpha0/local/bin
Installing zeoctl script to /home/leif/sage-4.5.alpha0/local/bin
Installing fsdump script to /home/leif/sage-4.5.alpha0/local/bin
Installing fsrefs script to /home/leif/sage-4.5.alpha0/local/bin
Installing repozo script to /home/leif/sage-4.5.alpha0/local/bin
Installing fsoids script to /home/leif/sage-4.5.alpha0/local/bin
Installed
/home/leif/sage-4.5.alpha0/local/lib/python2.6/site-packages/ZODB3-3.7.0-py2.6-linux-i686.egg
Processing dependencies for ZODB3==3.7.0
Searching for zdaemon==2.0.0
Best match: zdaemon 2.0.0
Processing zdaemon-2.0.0-py2.6.egg
zdaemon 2.0.0 is already the active version in easy-install.pth
Installing zdaemon script to /home/leif/sage-4.5.alpha0/local/bin
Using
/home/leif/sage-4.5.alpha0/local/lib/python2.6/site-packages/zdaemon-2.0.0-py2.6.egg
Searching for ZConfig==2.5
Best match: ZConfig 2.5
Processing ZConfig-2.5-py2.6.egg
ZConfig 2.5 is already the active version in easy-install.pth
Installing zconfig_schema2html script to
/home/leif/sage-4.5.alpha0/local/bin
Installing zconfig script to /home/leif/sage-4.5.alpha0/local/bin
[...]
-Leif
John H Palmieri
setuptools has not been installed, then the .py files are installed,
but if setuptools is present, then the other scripts are installed
instead. Since setuptools is not a dependency of zodb, it's luck, to
some extent, whether it's installed before or after zodb.
Were you building in parallel? That can have a large effect on the
order in which spkg's are installed, in my experience, thus revealing
potential bugs in the file spkg/standard/deps.
--
John
Robert Miller
>
> Here's a guess from looking at src/setup.py in the zodb spkg: if
> setuptools has not been installed, then the .py files are installed,
> but if setuptools is present, then the other scripts are installed
> instead. Since setuptools is not a dependency of zodb, it's luck, to
> some extent, whether it's installed before or after zodb.
>
> Were you building in parallel? That can have a large effect on the
> order in which spkg's are installed, in my experience, thus revealing
> potential bugs in the file spkg/standard/deps.
>
> --
> John
>
Thanks for the intel, John. I wasn't building in parallel, since I
never do that when I'm making releases. However I'm pretty sure Justin
was, and what he got was different.
Should I add a dependency to deps to include setuptools in the
dependencies for zodb, and add these files to hgignore?
I've made a blocker ticket for this, at
http://trac.sagemath.org/sage_trac/ticket/9412
leif
> Here's a guess from looking at src/setup.py in the zodb spkg: if
> setuptools has not been installed, then the .py files are installed,
> but if setuptools is present, then the other scripts are installed
> instead. Since setuptools is not a dependency of zodb, it's luck, to
> some extent, whether it's installed before or after zodb.
Well, I actually *reinstalled* zodb3 on 4.5.alpha0 ("sage -i -s zodb3"),
so setuptools were present at least the second time.
Note that there seems to be a bug in sage-spkg because it downloaded the
spkg: "sage: After decompressing the directory zodb3 does not exist..."
(After installation, the zodb3 scripts were "fresh".)
> Were you building in parallel? That can have a large effect on the
> order in which spkg's are installed, in my experience, thus revealing
> potential bugs in the file spkg/standard/deps.
In both cases no.
-Leif
Justin C. Walker
On Fri, Jul 2, 2010 at 1:09 PM, John H Palmieri <jhpalm...@gmail.com> wrote:Here's a guess from looking at src/setup.py in the zodb spkg: ifsetuptools has not been installed, then the .py files are installed,but if setuptools is present, then the other scripts are installedinstead. Since setuptools is not a dependency of zodb, it's luck, tosome extent, whether it's installed before or after zodb.Were you building in parallel? That can have a large effect on theorder in which spkg's are installed, in my experience, thus revealingpotential bugs in the file spkg/standard/deps.--John
Thanks for the intel, John. I wasn't building in parallel, since I
never do that when I'm making releases. However I'm pretty sure Justin
was, and what he got was different.
Justin C. Walker
On 2 Jul, 2010, at 2:26 PM, Robert Miller wrote:On Fri, Jul 2, 2010 at 1:09 PM, John H Palmieri <jhpalm...@gmail.com> wrote:Here's a guess from looking at src/setup.py in the zodb spkg: ifsetuptools has not been installed, then the .py files are installed,but if setuptools is present, then the other scripts are installedinstead. Since setuptools is not a dependency of zodb, it's luck, tosome extent, whether it's installed before or after zodb.Were you building in parallel? That can have a large effect on theorder in which spkg's are installed, in my experience, thus revealingpotential bugs in the file spkg/standard/deps.--John
Thanks for the intel, John. I wasn't building in parallel, since I
never do that when I'm making releases. However I'm pretty sure Justin
was, and what he got was different.My result, as best I can recall, was that the tree was clean (with the possible exception of some Makefile in the doc tree being gone but not forgotten).
--
Justin C. Walker, Curmudgeon-At-Large
--------
Men are from Earth.
Women are from Earth.
Deal with it.
--------
John H Palmieri
>
>
>
> > Here's a guess from looking at src/setup.py in the zodb spkg: if
> > setuptools has not been installed, then the .py files are installed,
> > but if setuptools is present, then the other scripts are installed
> > instead. Since setuptools is not a dependency of zodb, it's luck, to
> > some extent, whether it's installed before or after zodb.
>
> > Were you building in parallel? That can have a large effect on the
> > order in which spkg's are installed, in my experience, thus revealing
> > potential bugs in the file spkg/standard/deps.
>
> > --
> > John
>
> Thanks for the intel, John. I wasn't building in parallel, since I
> never do that when I'm making releases. However I'm pretty sure Justin
> was, and what he got was different.
>
> Should I add a dependency to deps to include setuptools in the
> dependencies for zodb, and add these files to hgignore?
of the installation for zodb, so I don't see why not. But I don't
actually know anything about this package or these scripts...
--
John
leif
Same on Ubuntu 9.04 x86 (P4 Prescott, gcc 4.3.3, sequ. build):
make build: OK
make doc: OK
make ptestlong: OK (All tests passed.)
After installing glpk-4.44 (and of course "sage -b"):
make ptestlong: OK (All tests passed.)
./sage -t -long -only-optional=cbc,glpk devel/sage-main:
Only sage/numerical/mip_coin.pyx failed (just because of CBC not being
installed), i.e. no "unexpected" solution in sage/numerical/optimize.py
on the 32-bit system (in contrast to x86_64/Core2/gcc 4.5.0).
-Leif
leif
> Same on Ubuntu 9.04 x86 (P4 Prescott, gcc 4.3.3, sequ. build):
>
> make build: OK
> make doc: OK
Not really:
sphinx-build -b html -d
/home/leif/sage-4.5.alpha1/devel/sage/doc/output/doctrees/en/thematic_tutorials
/home/leif/sage-4.5.alpha1/devel/sage/doc/en/thematic_tutorials
/home/leif/sage-4.5.alpha1/devel/sage/doc/output/html/en/thematic_tutorials
Error: Source directory doesn't contain conf.py file.
Build finished. The built documents can be found in
/home/leif/sage-4.5.alpha1/devel/sage/doc/output/html/en/thematic_tutorials
(See $SAGE_ROOT/dochtml.log; this error is present in Sage 4.4.4, too.)
I think the last line also indicates a bug in the documentation builder.
> make ptestlong: OK (All tests passed.)
-Leif
leif
> sphinx-build -b html -d
> /home/leif/sage-4.5.alpha1/devel/sage/doc/output/doctrees/en/thematic_tutorials
> /home/leif/sage-4.5.alpha1/devel/sage/doc/en/thematic_tutorials
> /home/leif/sage-4.5.alpha1/devel/sage/doc/output/html/en/thematic_tutorials
> Error: Source directory doesn't contain conf.py file.
> Build finished. The built documents can be found in
> /home/leif/sage-4.5.alpha1/devel/sage/doc/output/html/en/thematic_tutorials
>
> (See $SAGE_ROOT/dochtml.log; this error is present in Sage 4.4.4, too.)
>
> I think the last line also indicates a bug in the documentation builder.
This (the false message) is now ticket #9426.
http://trac.sagemath.org/sage_trac/ticket/9426
-Leif
(Robert: Please cc yourself if you think this should block 4.5.)
(Minh: Perhaps you can open a ticket for the missing conf.py - and
immediately fix it... ;-) )