Sage 6.2.beta5 released

[Volker Braun]

The "develop" branch on git has been updated to 6.2.beta5. Alternatively, you can download the source tarball:

http://boxen.math.washington.edu/home/release/sage-6.2.beta5/sage-6.2.beta5.tar.gz

$ git log --first-parent develop ^6.2.beta4 --oneline

9db8c5c Updated Sage version to 6.2.beta5

cca671d Trac #15975: FiniteStateMachine.determine_alphabets: corrected description of parameter reset

76d428a Trac #15286: Latin squares

e0eee52 Trac #15907: Improvements for affine factorization crystal

ebd2c00 Trac #15331: Do not try to create embedded number field morphisms for non-embedded number fields

5e64ce2 Trac #15925: Use Brzozowski' algorithm as default for minimizing non-deterministic automata

6fdfd13 Trac #15924: Allow non-hashable output labels in FiniteStateMachine.prepone_output

671feb5 Trac #15923: Incorrect TypeError in FiniteStateMachine.add_from_transition_function

3065fe6 Trac #15922: Allow list of transitions in FiniteStateMachine.add_transistions_from_function

25361b7 Trac #13774: Irreducible Cartan types created as reducibles

8bd99a7 Trac #13609: symbolic arithmetic errors

4f70b4d Trac #15973: Allow GAP to compile on OS X 10.4

25255ca Trac #15824: Remove deprecated code from matrix/

2092418 Trac #15940: Improve printing for rigged configurations

cde5f2c Trac #15952: import statements with both absolute and (implicit) relative imports

47403f6 Trac #15960: New function: FiniteStateMachine.set_coordinates

cb44fb9 Trac #15849: Default ordering of States and Transitions in FiniteStateMachine

0c33fd1 Trac #15848: Simplification for non-deterministic transducers via Moore's algorithm

20c4867 Trac #15847: Do deepcopy + reset label in FiniteStateMachine.quotient instead of creating a new state

6f65fec Trac #15966: Fix paralle_iter

d242c97 Trac #15951: Fix last string exception for python3

d59db5b Trac #12661: Include GMP 5.1.3 as an optional package

f105b3c Trac #13174: Wrong documentation for methods @cached_method

0c10a1b Trac #15937: Singular won't compile with latest Xcode (5.1, build 5B130a)

c8cf92e Trac #15979: Toric varieties can be tricked into accepting base rings that are not fields

de53f11 Trac #14993: Upgrade matplotlib to 1.3.1

21255d6 Trac #15918: Update FLINT to 2.4.2

9acfe5e Trac #9524: Nasty bug with polynomial arithmetic and NTL contexts

0fd9907 Trac #15815: rational preimages for projective morphisms returns incorrect points

33eb663 Trac #13048: __floordiv__ does not work when dividing polynomials over QQbar by a constant

82f6e16 Trac #15627: (disjoint) Union of graphs with multiple edges

7ebf80e Trac #14844: Implement cluster algebras with universal coefficients

07baf37 Trac #15910: Fix doctest in root_systems/coxeter_group.py when coxeter3 is install

426c91d Trac #15888: CC -> PARI: return t_REAL for real numbers

cd7dff2 Trac #15890: EllipticCurve_number_field.is_isogenous() raises error for non-minimal models

4d888e6 Trac #15895: Bug in Graph.subdivide_edge

a28a9b1 Trac #15908: guava must be explicitly loaded

c88356c Trac #15521: Deprecations: default LP variables will become real instead of nonnegative

7890f8a Trac #15333: remove the imports of is_* from modules/all.py and structure/all.py

807f60d Trac #15306: partial_fraction_decomposition over QQ[] wrong

d694634 Trac #8573: prod(primes_first_n(64)).divisors() crashes

955c1ab Trac #15599: Improve support for PARI series (t_SER)

3464296 Trac #15843: Add reduced Burau representation, Alexander polynomial to Braid

[Nathann Cohen]

Hellooooooooooo !

Updated and recompiled with "make".

Everything went fine, except that building the doc never terminates. It just does nothing after a while, probably some multithreading problem.

I typed "make doc-clean && make", same result.

...

Oh.

I just noticed this by scrolling up a bit :

Exception in thread Thread-7:
Traceback (most recent call last):
  File "/home/ncohen/.Sage/local/lib/python/threading.py", line 808, in __bootstrap_inner
    self.run()
  File "/home/ncohen/.Sage/local/lib/python/threading.py", line 761, in run
    self.__target(*self.__args, **self.__kwargs)
  File "/home/ncohen/.Sage/local/lib/python/multiprocessing/pool.py", line 325, in _handle_workers
    pool._maintain_pool()
  File "/home/ncohen/.Sage/local/lib/python/multiprocessing/pool.py", line 229, in _maintain_pool
    self._repopulate_pool()
  File "/home/ncohen/.Sage/local/lib/python/multiprocessing/pool.py", line 222, in _repopulate_pool
    w.start()
  File "/home/ncohen/.Sage/local/lib/python/multiprocessing/process.py", line 130, in start
    self._popen = Popen(self)
  File "/home/ncohen/.Sage/local/lib/python/multiprocessing/forking.py", line 121, in __init__
    self.pid = os.fork()
OSError: [Errno 12] Cannot allocate memory

And well.. I was not running anything else at that time ^^;

Nathann

--
You received this message because you are subscribed to the Google Groups "sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sage-release...@googlegroups.com.
To post to this group, send email to sage-r...@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

[Volker Braun]

How much RAM do you have? Do you set SAGE_NUM_THREADS? Whats the value of SAGE_NUM_THREADS in a Sage shell?

[Nathann Cohen]

Hellooooooo !!

> How much RAM do you have?

4GB

> Do you set SAGE_NUM_THREADS? Whats the value of SAGE_NUM_THREADS in a Sage shell?

It is set to 4.

Nathann

[Volker Braun]

On Monday, March 24, 2014 5:46:16 PM UTC, Nathann Cohen wrote:

> How much RAM do you have?
4GB

That should be enough if your swap is set up correctly. How much swap space do you have? (e.g. output of "top")

[Nathann Cohen]

> That should be enough if your swap is set up correctly. How much swap space do you have? (e.g. output of "top")

Ahem.

0GB.

I don't believe in SWAP.

Okay, so I should change the number of threads to 3 or 2 ? :-P

Nathann

[Volker Braun]

Unix needs a swap file, even if no swapping is being done (as a guarantee for anonymous mmaps). If you don't provide one then you force certain "virtual" allocations into real ram.

On Monday, March 24, 2014 6:02:58 PM UTC, Nathann Cohen wrote:

Okay, so I should change the number of threads to 3 or 2 ? :-P

You can try out for yourself how many processes fit into your ram ;-)

[kcrisman]

Just wanted to say that my issue on OS X 10.4 hasn't been resolved with this flint upgrade, though I didn't expect it to (at least, in building flint right now, it's giving the same messages about includes). 

[Bill Hart]

The only way we can fix the -include problem is to either 

* upgrade your version of make 

* remove dependency tracking altogether from flint (talk to Mike Hansen about that)

* or to put dummy.c files in all relevant directories (I obviously don't like to work around bugs with ugly hacks, so am not in favour of this)

* somehow have Make check for the existence of the directory first (again, talk to Mike Hansen about that)

But I agree with the opinion that this is not actually what is causing the actual problem, but it is instead some Mac related issue, as we do not have this issue anywhere else, including some very old, dusty museum pieces.

Bill.

--

[Bill Hart]

Actually, option 1 is not the solution. I use Make 3.81 as well, and the problem does not occur for me. So that leaves three possible solutions.

Bill.

[Ivan Andrus]

I haven’t really followed the discussion, but if the problem is because of lines in the Makefile like

-include

then you should be able to just add a dummy file on that line so that it’s 

-include dummy.mk …

where the … is of course the expression that will sometimes be empty.  But it’s also not hard to check for the existence of a directory.

-Ivan

[kcrisman]

On Thursday, March 27, 2014 1:56:21 AM UTC-4, Ivan Andrus wrote:

I haven’t really followed the discussion, but if the problem is because of lines in the Makefile like

-include

then you should be able to just add a dummy file on that line so that it’s 

-include dummy.mk …

where the … is of course the expression that will sometimes be empty.  But it’s also not hard to check for the existence of a directory.

-Ivan

I think the problem is that we don't even know that this *is* the problem (see Leif's reply in the other thread on beta4).  Is there any way to easily check the differences between make 3.80 and 3.81 on this issue?  I assume that people building with 3.81 don't see these warning messages.  

I think *if* this is the problem, then Ivan's solution of having this include line only added if there is a profile/ directory makes sense, as the syntax probably is fairly standard.

Oh, and the libs.mwrank error does still occur on beta5 (took this long just to get there).

[kcrisman]

Neither of http://www.cmcrossroads.com/article/whats-new-gnu-make-381 and https://notendur.hi.is/jonasson/software/make-book/make-news.pdf seem to have something that immediately shows this problem, though.    The full list of bugs does have the thing where -include with an empty or false list continues (http://savannah.gnu.org/bugs/index.php?13216) is fixed in 3.81, would this be enough to cause the problems in question?  Unfortunately I don't see a link to the actual diff.  http://savannah.gnu.org/bugs/?29074 and http://savannah.gnu.org/bugs/?27396 seem relevant but were never actually dealt with, and http://savannah.gnu.org/bugs/?102 is a really old version of something related...

[François Bissey]

The NEWS file for make 3.81 doesn't have anything that I can really relate
to the problem. However it would be nice to check if upgrading make on your
machine solves this particular problem. Or did I miss that someone else had
troubles with 3.81?

Francois

[leif]

François Bissey wrote:
> The NEWS file for make 3.81 doesn't have anything that I can really relate
> to the problem. However it would be nice to check if upgrading make on your
> machine solves this particular problem. Or did I miss that someone else had
> troubles with 3.81?

Not in this context. At some point we had trouble with 3.81 and 'make
ptest[long]' (GAP and/or ECL IIRC), which did not happen with 3.82
(although again, GNU Make's changelog or NEWS didn't mention anything
obviously relevant).


I don't think the *warnings* Karl-Dieter gets with 3.80 cause any harm
(other than probably annoying the user). The changes in 3.81 suppress
them, but do not change the behaviour w.r.t. '[-]include' and *empty*
include file lists; they mainly deal with how 'make' behaves if the
files to include don't exist and/or cannot be remade (according to [1]
and [2] at least).


-leif


[1] http://savannah.gnu.org/bugs/?13216

[2] http://savannah.gnu.org/bugs/?29074

--
() The ASCII Ribbon Campaign
/\ Help Cure HTML E-Mail

[François Bissey]

On Thu, 27 Mar 2014 23:04:56 leif wrote:
> François Bissey wrote:
> > The NEWS file for make 3.81 doesn't have anything that I can really relate
> > to the problem. However it would be nice to check if upgrading make on
> > your
> > machine solves this particular problem. Or did I miss that someone else
> > had
> > troubles with 3.81?
>
> Not in this context. At some point we had trouble with 3.81 and 'make
> ptest[long]' (GAP and/or ECL IIRC), which did not happen with 3.82
> (although again, GNU Make's changelog or NEWS didn't mention anything
> obviously relevant).
>
>
> I don't think the *warnings* Karl-Dieter gets with 3.80 cause any harm
> (other than probably annoying the user). The changes in 3.81 suppress
> them, but do not change the behaviour w.r.t. '[-]include' and *empty*
> include file lists; they mainly deal with how 'make' behaves if the
> files to include don't exist and/or cannot be remade (according to [1]
> and [2] at least).
>

I hadn't seen the stuff in the other thread when I wrote that answer.
Stuff split between two threads can be like that.

Francois

[kcrisman]

Yeah, sorry, that's my fault.  I'll clarify the situation in the other thread.  ;-) 

[kcrisman]

Aannd  - yet another build error, beta5.  (I suspect that ever since the 6.0 transition, or before, probably Sage fails to build on this platform, but I'm sure all the tweaks needed are just that, at this point.)

Now, before you read this, yes freetype-2.5.2.p0 built fine.   But maybe I need some kind of --with-freetype option in the configure to make sure it finds it?  This was not a problem in the past.  Unfortunately, I just walked in the door and have a lot of other things to do, so I can't search for similar errors yet, but at least wanted to record this before I forget.  Also note they didn't find pkg-config info for libpng, which I'm not sure is good.

 - kcrisman
***


matplotlib-1.3.1
====================================================
Setting up build directory for matplotlib-1.3.1
Finished set up
****************************************************
Host system:
Darwin Dasher-03.local 8.11.0 Darwin Kernel Version 8.11.0: Wed Oct 10 18:26:00 PDT 2007; root:xnu-792.24.17~1/RELEASE_PPC Power Macintosh powerpc
****************************************************
C compiler: gcc
C compiler version:
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/Users/student/Desktop/sage-6.2.beta5/local/libexec/gcc/powerpc-apple-darwin8.11.0/4.7.3/lto-wrapper
Target: powerpc-apple-darwin8.11.0
Configured with: ../src/configure --prefix=/Users/student/Desktop/sage-6.2.beta5/local --with-local-prefix=/Users/student/Desktop/sage-6.2.beta5/local --with-gmp=/Users/student/Desktop/sage-6.2.beta5/local --with-mpfr=/Users/student/Desktop/sage-6.2.beta5/local --with-mpc=/Users/student/Desktop/sage-6.2.beta5/local --with-system-zlib --disable-multilib --disable-nls  
Thread model: posix
gcc version 4.7.3 (GCC)
****************************************************
NOTE: Set SAGE_MATPLOTLIB_GUI to anything but 'no' to try to build the Matplotlib GUI.
Not building any matplotlib graphical backends.
patching file lib/matplotlib/tri/_tri.cpp
patching file lib/matplotlib/tri/_tri.h
patching file setup.py
============================================================================
Edit setup.cfg to change the build options

BUILDING MATPLOTLIB
            matplotlib: yes [1.3.1]
                python: yes [2.7.5 (default, Mar 26 2014, 08:50:37)  [GCC
                        4.7.3]]
              platform: yes [darwin]

REQUIRED DEPENDENCIES AND EXTENSIONS
                 numpy: yes [version 1.7.0]
              dateutil: yes [using dateutil version 2.2]
               tornado: yes [tornado was not found. It is required for the
                        WebAgg backend. pip/easy_install may attempt to
                        install it after matplotlib.]
             pyparsing: yes [using pyparsing version 2.0.1]
                 pycxx: yes [Couldn't import.  Using local copy.]
                libagg: yes [pkg-config information for 'libagg' could not
                        be found. Using local copy.]
              freetype: no  [The C/C++ header for freetype2 (ft2build.h)
                        could not be found.  You may need to install the
                        development package.]
                   png: yes [pkg-config information for 'libpng' could not
                        be found. Using unknown version.]

OPTIONAL SUBPACKAGES
           sample_data: yes [installing]
              toolkits: yes [installing]

OPTIONAL BACKEND EXTENSIONS
                macosx: no  [skipping due to configuration]
                qt4agg: no  [PyQt4 not found]
               gtk3agg: no  [Requires pygobject to be installed.]
             gtk3cairo: no  [Requires cairo to be installed.]
                gtkagg: no  [skipping due to configuration]
                 tkagg: no  [skipping due to configuration]
                 wxagg: no  [skipping due to configuration]
                   gtk: no  [skipping due to configuration]
                   agg: yes [installing]
                 cairo: no  [cairo not found]
             windowing: no  [skipping due to configuration]

OPTIONAL LATEX DEPENDENCIES
                dvipng: no
           ghostscript: no
                 latex: no
               pdftops: no

============================================================================
                        * The following required packages can not be built:
                        * freetype
Error building matplotlib package.

real    0m10.504s
user    0m4.290s
sys     0m2.525s
************************************************************************
Error installing package matplotlib-1.3.1
************************************************************************

[François]

And for when you come back:

Do you have either "local/include/ft2build.h" or "local/include/freetype2/ft2build.h" present.

Francois

[Anne Schilling]

Hi!

It looks like that after installing sage-6.2.beta5 dot2tex does not work any longer:

sage: C = KirillovReshetikhinCrystal(['A',5,2],1,1)
sage: view(C)
dot2tex not available.  Install after running 'sage -sh'

Is this related to matplotlib-1.3.1 and possibly #14594?

Anne

[Ben Salisbury]

Hi all,

It seems that my dot2tex no longer functions after upgrading to 6.2.beta5.

sage: C = crystals.Tableaux("B2",shape=[2,1])

sage: view(C)
dot2tex not available. Install after running 'sage -sh'

I attempted to force a reinstallation of dot2tex via sage -i -f dot2tex, which succeeded without any errors. However, the same error appears in Sage when trying to compile something with dot2tex.

Best,
Ben

[leif]

Well, (re)try this with SAGE_CHECK=yes:

...
Successfully installed dot2tex-2.8.7.p2
Running the test suite for dot2tex-2.8.7.p2...
############################################################
spkg-check: Running test_buildexamples.py...
############################################################

Traceback (most recent call last):

File "/Users/leif/Sage/sage-6.2.beta5/local/bin/dot2tex", line 2, in
<module>
from dot2tex.dot2tex import main
File
"/Users/leif/Sage/sage-6.2.beta5/local/lib/python2.7/site-packages/dot2tex/__init__.py",
line 36, in <module>
import dot2tex as d2t
File
"/Users/leif/Sage/sage-6.2.beta5/local/lib/python2.7/site-packages/dot2tex/dot2tex.py",
line 47, in <module>
import dotparsing
File
"/Users/leif/Sage/sage-6.2.beta5/local/lib/python2.7/site-packages/dot2tex/dotparsing.py",
line 26, in <module>
import matplotlib.pyparsing as pyparsing
ImportError: No module named pyparsing
...


This is (once again ;-) ) due to #14993 [1] (matplotlib upgrade), merged
into beta5.

Matplotlib no longer ships pyparsing, so we presumably have to update
the dot2tex package accordingly. And it would probably make sense to
add some tiny test to its spkg-install to avoid such issues in the
future (and/or slightly change the way dot2tex gets run from within
Sage, as the error message is misleading, not to say plain wrong).


-leif

[1] http://trac.sagemath.org/ticket/14993

[Francois Bissey]

I looked for dot2tex in the git tree and it looks like it hasn’t been updated to git yet.
So it will be the perfect opportunity to review it.

Francois
This email may be confidential and subject to legal privilege, it may
not reflect the views of the University of Canterbury, and it is not
guaranteed to be virus free. If you are not an intended recipient,
please notify the sender immediately and erase all copies of the message
and any attachments.

Please refer to http://www.canterbury.ac.nz/emaildisclaimer for more
information.

[leif]

P.S.: This is meanwhile http://trac.sagemath.org/ticket/16026 ("Upgrade
dot2tex spkg and make it work with most recent matplotlib").


-leif

[kcrisman]

On Friday, March 28, 2014 2:49:01 PM UTC-4, François wrote:

And for when you come back:

Do you have either "local/include/ft2build.h" or "local/include/freetype2/ft2build.h" present.

Thanks, I'll look at this.  I don't know whether http://trac.sagemath.org/ticket/14993 is the problem since I hadn't finished the previous build of beta4, but I note http://trac.sagemath.org/ticket/14993#comment:29 and http://trac.sagemath.org/ticket/15742 as just conceivably relevant.  Now that there is no indication of *which* beta things are merged in on Trac, it's a little harder to know where to look... anyway, I'll look on Monday, thanks. 

[kcrisman]

On Saturday, March 29, 2014 10:32:30 PM UTC-4, kcrisman wrote:

On Friday, March 28, 2014 2:49:01 PM UTC-4, François wrote:

And for when you come back:

Do you have either "local/include/ft2build.h" or "local/include/freetype2/ft2build.h" present.

Indeed, the latter is present, but not the former. 

[kcrisman]

Not only did mpl not build, but even git wouldn't!

I haven't built Sage-git unsuccessfully before, so I'm not sure where to look for this guy.  Where would this common crypto thing live?

    CC http-backend.o
    LINK git-http-backend
/usr/bin/ld: warning multiple definitions of symbol _regcomp
libgit.a(regex.o) definition of _regcomp in section (__TEXT,__text)
/usr/lib/libSystem.dylib(regcomp.So) definition of _regcomp
/usr/bin/ld: warning multiple definitions of symbol _regexec
libgit.a(regex.o) definition of _regexec in section (__TEXT,__text)
/usr/lib/libSystem.dylib(regexec.So) definition of _regexec
/usr/bin/ld: warning multiple definitions of symbol _regfree
libgit.a(regex.o) definition of _regfree in section (__TEXT,__text)
/usr/lib/libSystem.dylib(regfree.So) definition of _regfree
    CC imap-send.o
imap-send.c:33:37: fatal error: CommonCrypto/CommonHMAC.h: No such file or directory
compilation terminated.
make[3]: *** [imap-send.o] Error 1
Error building git.

real    9m46.658s
user    5m48.673s
sys     1m46.609s
************************************************************************
Error installing package git-1.8.4.4
************************************************************************

[Volker Braun]

Fix here: http://trac.macports.org/ticket/39848

[kcrisman]

On Monday, March 31, 2014 5:10:47 PM UTC-4, Volker Braun wrote:

Fix here: http://trac.macports.org/ticket/39848

Oh, great.  I've opened http://trac.sagemath.org/ticket/16041 for this and posted the (correct) diff there.  I'll try it tomorrow. 

[kcrisman]

And for when you come back:

Do you have either "local/include/ft2build.h" or "local/include/freetype2/ft2build.h" present.

Indeed, the latter is present, but not the former. 

Any ideas? 

[François Bissey]

Not right now in that your setup looks correct as expected
since you are NOT upgrading. I have to think a little bit.

Francois

[François Bissey]

On Mon, 31 Mar 2014 17:26:06 kcrisman wrote:

Would you happen to have another version of freetype2 installed
on the system along with a non-empty PKG_CONFIG_PATH?

Francois

[kcrisman]

In sage -sh, I have

PKG_CONFIG_PATH=/Users/student/Desktop/sage-6.2.beta5/local/lib/pkgconfig

And the pkgconfig for freetype2 is

SAGE_ROOT=/Users/student/Desktop/sage-6.2.beta5
prefix="${SAGE_ROOT}/local"
exec_prefix="${SAGE_ROOT}/local"
libdir="${SAGE_ROOT}/local/lib"
includedir="${SAGE_ROOT}/local/include/freetype2"

Name: FreeType 2
Description: A free, high-quality, and portable font engine.
Version: 17.1.11
Requires:
Libs: -L${libdir} -lfreetype
Libs.private: -lz -lbz2 -lpng12
Cflags: -I${includedir}


There is no such path outside sage -sh, though.

There is

$ ls /usr/X11R6/lib/libfreetype.
libfreetype.6.3.dylib  libfreetype.6.dylib    libfreetype.a          libfreetype.dylib

but I don't know where else to look for freetype2 on such an old machine.

[kcrisman]

On Tuesday, April 1, 2014 3:14:38 PM UTC-4, kcrisman wrote:

On Monday, March 31, 2014 8:53:43 PM UTC-4, François wrote:

On Mon, 31 Mar 2014 17:26:06 kcrisman wrote:
> And for when you come back:
> > Do you have either "local/include/ft2build.h" or
> >
> >>> "local/include/freetype2/ft2build.h" present.
> >
> > Indeed, the latter is present, but not the former.
>
> Any ideas?
Would you happen to have another version of freetype2 installed
on the system along with a non-empty PKG_CONFIG_PATH?

$ ls /Developer/SDKs/MacOSX10.4u.sdk/usr/X11R6/include/freetype2/freetype/
cache         ftchapters.h  ftincrem.h    ftoutln.h     ftsynth.h     ftwinfnt.h    tttags.h
config        fterrdef.h    ftlist.h      ftpfr.h       ftsysio.h     ftxf86.h     
freetype.h    fterrors.h    ftmac.h       ftrender.h    ftsysmem.h    internal     
ftbbox.h      ftglyph.h     ftmm.h        ftsizes.h     ftsystem.h    t1tables.h   
ftbdf.h       ftgzip.h      ftmoderr.h    ftsnames.h    fttrigon.h    ttnameid.h   
ftcache.h     ftimage.h     ftmodule.h    ftstroker.h   fttypes.h     tttables.h   

[kcrisman]

In the meantime, yet another build error - this time rpy.  What the heck?  After all,

$ ls local/lib/R/lib/              
libR.dylib              libRblas.dylib          libRlapack.dylib

$ otool -L local/lib/R/lib/libRblas.dylib
local/lib/R/lib/libRblas.dylib:
        libRblas.dylib (compatibility version 0.0.0, current version 0.0.0)
        /Users/student/Desktop/sage-6.2.beta5/local/lib/libgfortran.3.dylib (compatibility version 4.0.0, current version 4.0.0)
        /Users/student/Desktop/sage-6.2.beta5/local/lib/libgomp.1.dylib (compatibility version 2.0.0, current version 2.0.0)
        /usr/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current version 1.0.0)
        /Users/student/Desktop/sage-6.2.beta5/local/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current version 1.0.0)
        /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 88.1.12)


building 'rpy2.rinterface._rinterface' extension
creating build/temp.macosx-10.4-ppc-2.7
creating build/temp.macosx-10.4-ppc-2.7/rpy
creating build/temp.macosx-10.4-ppc-2.7/rpy/rinterface
gcc -fno-strict-aliasing -g -O2 -DNDEBUG -g -fwrapv -O3 -Wall -DR_INTERFACE_PTRS=1 -DHAVE_POSIX_SIGJMP=1 -DRIF_HAS_RSIGHAND=1 -DCSTACK_DEFNS=1 -DHAS_READLINE=1 -DRPY_BIGENDIAN=1 -I./rpy/rinterface -I/Users/student/Desktop/sage-6.2.beta5/local/include/python2.7 -I/Users/student/Desktop/sage-6.2.beta5/local/lib/R//include -c ./rpy/rinterface/_rinterface.c -o build/temp.macosx-10.4-ppc-2.7/./rpy/rinterface/_rinterface.o
In file included from ./rpy/rinterface/_rinterface.c:125:0:
./rpy/rinterface/sexp.c: In function 'Sexp_rclass_set':
./rpy/rinterface/sexp.c:369:5: warning: return makes integer from pointer without a cast [enabled by default]
./rpy/rinterface/sexp.c: In function 'Sexp_init':
./rpy/rinterface/sexp.c:711:13: warning: unused variable 'copy' [-Wunused-variable]
In file included from ./rpy/rinterface/_rinterface.c:104:0:
./rpy/rinterface/_rinterface.c: At top level:
./rpy/rinterface/embeddedr.h:20:20: warning: '_Rpy_PreserveObject' declared 'static' but never defined [-Wunused-function]
In file included from ./rpy/rinterface/_rinterface.c:129:0:
./rpy/rinterface/sequence.c:2190:1: warning: 'ComplexVectorSexp_AsSexp' defined but not used [-Wunused-function]
./rpy/rinterface/_rinterface.c: In function 'EmbeddedR_ShowFiles':
./rpy/rinterface/_rinterface.c:893:3: warning: 'gstate' may be used uninitialized in this function [-Wmaybe-uninitialized]
gcc -bundle -undefined dynamic_lookup -L/Users/student/Desktop/sage-6.2.beta5/local/lib build/temp.macosx-10.4-ppc-2.7/./rpy/rinterface/_rinterface.o -L/Users/student/Desktop/sage-6.2.beta5/local/lib/R//lib -L/Users/student/Desktop/sage-6.2.beta5/local/lib/R/modules -lR -lR -o build/lib.macosx-10.4-ppc-2.7/rpy2/rinterface/_rinterface.so
/usr/bin/ld: warning can't open dynamic library: libRblas.dylib referenced from: /Users/student/Desktop/sage-6.2.beta5/local/lib/R//lib/libR.dylib (checking for undefined symbols may be affected) (No such file or directory, errno = 2)
/usr/bin/ld: Undefined symbols:
_dgemm_ referenced from libR expected to be defined in libRblas.dylib
_dsyrk_ referenced from libR expected to be defined in libRblas.dylib
_dtrsm_ referenced from libR expected to be defined in libRblas.dylib
_zgemm_ referenced from libR expected to be defined in libRblas.dylib
_daxpy_ referenced from libR expected to be defined in libRblas.dylib
_dswap_ referenced from libR expected to be defined in libRblas.dylib
_ddot_ referenced from libR expected to be defined in libRblas.dylib
_dasum_ referenced from libR expected to be defined in libRblas.dylib
_dscal_ referenced from libR expected to be defined in libRblas.dylib
_dnrm2_ referenced from libR expected to be defined in libRblas.dylib
_dcopy_ referenced from libR expected to be defined in libRblas.dylib
_drot_ referenced from libR expected to be defined in libRblas.dylib
_drotg_ referenced from libR expected to be defined in libRblas.dylib
collect2: error: ld returned 1 exit status
error: command 'gcc' failed with exit status 1

real    0m39.928s
user    0m18.498s
sys     0m4.154s
************************************************************************
Error installing package rpy2-2.3.8
************************************************************************

[leif]

kcrisman wrote:
> In the meantime, yet another build error - this time rpy. What the
> heck?

Same as with eclib and FLINT...

> After all,
>
> $ ls local/lib/R/lib/
> libR.dylib libRblas.dylib libRlapack.dylib
>
> $ otool -L local/lib/R/lib/libRblas.dylib
> local/lib/R/lib/libRblas.dylib:
> libRblas.dylib (compatibility version 0.0.0, current version 0.0.0)

^^^^^^^^^^^^^^

You have to change this to an absolute path ($SAGE_LOCAL/lib/R/lib/...),
with 'install_name_tool -id ...'.

Presumably the same for libRlapack.dylib, probably libR.dylib as well.


Don't know what changed (since Sage 6.2.beta0, inclusive, or probably
later) that makes this necessary now.


-leif

[François Bissey]

On Tue, 01 Apr 2014 22:50:43 leif wrote:
> kcrisman wrote:
> > In the meantime, yet another build error - this time rpy. What the
> > heck?
>
> Same as with eclib and FLINT...
>
> > After all,
> >
> > $ ls local/lib/R/lib/
> > libR.dylib libRblas.dylib libRlapack.dylib
> >
> > $ otool -L local/lib/R/lib/libRblas.dylib
> >
> > local/lib/R/lib/libRblas.dylib:
> > libRblas.dylib (compatibility version 0.0.0, current version
> > 0.0.0)
>
> ^^^^^^^^^^^^^^
>
> You have to change this to an absolute path ($SAGE_LOCAL/lib/R/lib/...),
> with 'install_name_tool -id ...'.
>

Indeed from the corresponding gentoo ebuild:
if use prefix; then
if [[ ${CHOST} == *-darwin* ]] ; then
sed -i \
-e 's:-install_name libR.dylib:-install_name
${libdir}/R/lib/libR.dylib:' \
-e 's:-install_name libRlapack.dylib:-install_name
${libdir}/R/lib/libRlapack.dylib:' \
-e 's:-install_name libRblas.dylib:-install_name
${libdir}/R/lib/libRblas.dylib:' \
-e "/SHLIB_EXT/s/\.so/.dylib/" \
configure.ac || die
# sort of "undo" 2.14.1-rmath-shared.patch
sed -i \
-e "s:-Wl,-soname=libRmath.so:-install_name
${EROOT%/}/usr/$(get_libdir)/libRmath.dylib:" \
src/nmath/standalone/Makefile.in || die
else
append-ldflags -Wl,-rpath="${EROOT%/}/usr/$(get_libdir)/R/lib"
fi
fi


Francois

[kcrisman]

> Same as with eclib and FLINT...
>

And in retrospect this makes sense.

 

> > After all,
> >
> > $ ls local/lib/R/lib/
> > libR.dylib              libRblas.dylib          libRlapack.dylib
> >
> > $ otool -L local/lib/R/lib/libRblas.dylib
> >
> > local/lib/R/lib/libRblas.dylib:
> >          libRblas.dylib (compatibility version 0.0.0, current version
> >          0.0.0)
>
>             ^^^^^^^^^^^^^^
>
> You have to change this to an absolute path ($SAGE_LOCAL/lib/R/lib/...),
> with 'install_name_tool -id ...'.
>
Indeed from the corresponding gentoo ebuild:
        if use prefix; then
                if [[ ${CHOST} == *-darwin* ]] ; then
                        sed -i \
                                -e 's:-install_name libR.dylib:-install_name
${libdir}/R/lib/libR.dylib:' \
                                -e 's:-install_name libRlapack.dylib:-install_name
${libdir}/R/lib/libRlapack.dylib:' \
                                -e 's:-install_name libRblas.dylib:-install_name
${libdir}/R/lib/libRblas.dylib:' \

Nice.  At least this time I knew to use otool to ask...

I have a feeling R is going to be less excited about this downstream request.  What say I just open http://trac.sagemath.org/ticket/16044 for any remaining such things and we call it a day?  Hopefully this build finally will finish - I'll update tomorrow.

[kcrisman]

And finally, a doc build error!  I already did make doc-clean and started over, fails in exactly the same place.  My apologies if this was already mentioned somewhere.

[plotting ] loading pickled environment... not yet created
[plotting ] building [inventory]: targets for 27 source files that are out of date
[plotting ] updating environment: 27 added, 0 changed, 0 removed
[plotting ] reading sources... [  3%] index
[plotting ] reading sources... [  7%] sage/graphs/graph_plot
[plotting ] reading sources... [ 11%] sage/plot/animate
[plotting ] reading sources... [ 14%] sage/plot/arc
[plotting ] reading sources... [ 18%] sage/plot/arrow
[plotting ] reading sources... [ 22%] sage/plot/bar_chart
[plotting ] reading sources... [ 25%] sage/plot/bezier_path
[plotting ] reading sources... [ 29%] sage/plot/circle
[plotting ] reading sources... [ 33%] sage/plot/colors
[plotting ] Exception occurred:

Traceback (most recent call last):

  File "/Users/student/Desktop/sage-6.2.beta5/src/doc/common/builder.py", line 83, in f
    execfile(sys.argv[0])
  File "/Users/student/Desktop/sage-6.2.beta5/src/doc/common/custom-sphinx-build.py", line 185, in <module>
    sphinx.cmdline.main(sys.argv)
  File "/Users/student/Desktop/sage-6.2.beta5/local/lib/python2.7/site-packages/Sphinx-1.1.2-py2.7.egg/sphinx/cmdline.py", line 214, in main
    print >>error, red('Exception occurred:')
  File "/Users/student/Desktop/sage-6.2.beta5/src/doc/common/custom-sphinx-build.py", line 165, in write
    self._write(str)
  File "/Users/student/Desktop/sage-6.2.beta5/src/doc/common/custom-sphinx-build.py", line 139, in _write
    self._log_line(line)
  File "/Users/student/Desktop/sage-6.2.beta5/src/doc/common/custom-sphinx-build.py", line 108, in _log_line
    raise OSError(line)
OSError: [plotting ] Exception occurred:


Error building the documentation.

Traceback (most recent call last):

  File "/Users/student/Desktop/sage-6.2.beta5/src/doc/common/builder.py", line 1474, in <module>
    getattr(get_builder(name), type)()
  File "/Users/student/Desktop/sage-6.2.beta5/src/doc/common/builder.py", line 269, in _wrapper
    getattr(get_builder(document), 'inventory')(*args, **kwds)
  File "/Users/student/Desktop/sage-6.2.beta5/src/doc/common/builder.py", line 485, in _wrapper
    x.get(99999)
  File "/Users/student/Desktop/sage-6.2.beta5/local/lib/python/multiprocessing/pool.py", line 554, in get
    raise self._value
OSError: [plotting ] Exception occurred:

make: *** [doc-html] Error 1

[kcrisman]

And for when you come back:

Do you have either "local/include/ft2build.h" or "local/include/freetype2/ft2build.h" present.

Indeed, the latter is present, but not the former. 

See http://trac.sagemath.org/ticket/16047 for the resolution - this actually broke pip install for mpl as well, apparently - see the discussion around https://github.com/efiring/matplotlib/commit/be37790f449641a6ead3af2d2334cc544789e869 and a big thanks to ppurka for realizing this was the issue.

- kcrisman