Sage 7.0 released

[Volker Braun]

The "master" git branch has been updated to Sage-7.0. As always, you can get the latest beta version from the "develop" git branch. Alternatively, the self-contained source tarball is at http://www.sagemath.org/download-latest.html

f76401f Updated SageMath version to 7.2

17c23eb Trac #20604: failing doctest with coxeter3

7e96c30 Updated SageMath version to 7.2.rc2

[Volker Braun]

Obviously its "Sage 7.2 released"... it was a long day ;-)

[Andrey Novoseltsev]

http://sagecell.sagemath.org is now running this version!

[Justin C. Walker]

On May 15, 2016, at 14:52 , Volker Braun wrote:

> Obviously its "Sage 7.2 released"... it was a long day ;-)
>
>
>
> On Sunday, May 15, 2016 at 11:28:15 PM UTC+2, Volker Braun wrote:
>>
>> The "master" git branch has been updated to Sage-7.0. As always, you can
>> get the latest beta version from the "develop" git branch. Alternatively,
>> the self-contained source tarball is at
>> http://www.sagemath.org/download-latest.html

I discovered that, on my 10.6.8 system, I left the "MacPorts" commands open for access (via path). Since I've had quickly-aborted builds of the last several releases, I thought to try again, after moving into a corner.

I was able to get the build of 7.2 farther than before.

This time most packages were built, but the build did blow up on the package "pathlib2-2.1.0". The log is attached.

Anyone got a clue to spare? I won't have time to dig in for a couple of days...

Thanks for any pointers/suggestions.

Justin

[Volker Braun]

I openend http://trac.sagemath.org/ticket/20677

[Francois Bissey]

Looks like “six” needs to be a build dependency of pathlib2.
Installed /Users/Sage/sage-7.2/local/lib/python2.7/site-packages/pathlib2-2.1.0-py2.7.egg
Processing dependencies for pathlib2==2.1.0
Searching for six
Reading https://pypi.python.org/simple/six/
Download error on https://pypi.python.org/simple/six/: timed out -- Some packages may not be found!
Couldn't find index page for 'six' (maybe misspelled?)
Scanning index of all packages (this may take a while)
Reading https://pypi.python.org/simple/
Download error on https://pypi.python.org/simple/: timed out -- Some packages may not be found!
No local packages or download links found for six
error: Could not find suitable distribution for Requirement.parse('six')


François

> --
> 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 https://groups.google.com/group/sage-release.
> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> 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
> --------
>
>
> --
> 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 https://groups.google.com/group/sage-release.
> For more options, visit https://groups.google.com/d/optout.
> <pathlib2-2.1.0.log.gz>

[Justin C. Walker]

> On May 25, 2016, at 00:43, Francois Bissey <francoi...@canterbury.ac.nz> wrote:
>
> Looks like “six” needs to be a build dependency of pathlib2.
> Installed /Users/Sage/sage-7.2/local/lib/python2.7/site-packages/pathlib2-2.1.0-py2.7.egg
> Processing dependencies for pathlib2==2.1.0
> Searching for six
> Reading https://pypi.python.org/simple/six/
> Download error on https://pypi.python.org/simple/six/: timed out -- Some packages may not be found!
> Couldn't find index page for 'six' (maybe misspelled?)
> Scanning index of all packages (this may take a while)
> Reading https://pypi.python.org/simple/
> Download error on https://pypi.python.org/simple/: timed out -- Some packages may not be found!
> No local packages or download links found for six
> error: Could not find suitable distribution for Requirement.parse('six’)

Mayhap’s this is clear to the Initiata, but how is it that this just showed up in my case, on only one of three systems.

Admittedly, the system in question is kind of long in the tooth, but even so…

Thanks for clarifying.

--
Justin C. Walker, Curmudgeon-at-Large
() The ASCII Ribbon Campaign
/\ Help Cure HTML Email

[François Bissey]

On 05/26/16 04:16, Justin C. Walker wrote:
>
>> On May 25, 2016, at 00:43, Francois Bissey <francoi...@canterbury.ac.nz> wrote:
>>
>> Looks like “six” needs to be a build dependency of pathlib2.
>> Installed /Users/Sage/sage-7.2/local/lib/python2.7/site-packages/pathlib2-2.1.0-py2.7.egg
>> Processing dependencies for pathlib2==2.1.0
>> Searching for six
>> Reading https://pypi.python.org/simple/six/
>> Download error on https://pypi.python.org/simple/six/: timed out -- Some packages may not be found!
>> Couldn't find index page for 'six' (maybe misspelled?)
>> Scanning index of all packages (this may take a while)
>> Reading https://pypi.python.org/simple/
>> Download error on https://pypi.python.org/simple/: timed out -- Some packages may not be found!
>> No local packages or download links found for six
>> error: Could not find suitable distribution for Requirement.parse('six’)
>
> Mayhap’s this is clear to the Initiata, but how is it that this just showed up in my case, on only one of three systems.
>
> Admittedly, the system in question is kind of long in the tooth, but even so…
>

It is technically a race condition. Those are fiddly to
predict.

Francois

[Justin C. Walker]

Yep.

Six is a requirement for several packages, which explains the issue (when coupled with the race issue).

My next questions are:
1. After reviewing the doc for packaging, it's not clear to me how to set
up the correct package structure from the "plain vanilla" upstream package.
Pointers?
2. Is this a better forum than trac for this discussion?

Thanks, as always.

Justin (new to the packaging world)

--
Justin C. Walker, Curmudgeon at Large
Institute for the Absorption of Federal Funds
-----------
While creating wives, God promised men
that good and obedient wives would be
found in all corners of the world.
Then He made the earth round.
--

[Justin C. Walker]

One more thing:

On May 25, 2016, at 14:39 , Justin C. Walker wrote:

>
> On May 25, 2016, at 13:58 , François Bissey wrote:

[snip]

>> It is technically a race condition. Those are fiddly to
>> predict.
>
> Yep.
>
> Six is a requirement for several packages, which explains the issue (when coupled with the race issue).
>
> My next questions are:
> 1. After reviewing the doc for packaging, it's not clear to me how to set
> up the correct package structure from the "plain vanilla" upstream package.
> Pointers?
> 2. Is this a better forum than trac for this discussion?

I forgot two possibly important questions:

3. I checked (I think) all 'dependencies' files in $SAGE_ROOT/build/pkgs
on a system with a good build of 7.2. The log file for each of these
packages claims that
"six 1.10.0 is already the active version in easy-install.pth"
I'm not sure what that means, but the logs don't seem to show any further
concern about finding/installing "six".

4. I see an "egg" directory for "six" in
$SAGE_ROOT/local/local/lib/python2.7/site-packages/
On my 10.6.8 system, some of the packages with dependence on
"six" have successfully built. The others did not build prior
to the build halting.
What's up with that?

I'm clearly missing something...

Thanks!

Justin

--
Justin C. Walker
Curmudgeon at Large

Director
Institute for the Enhancement of the Director's Income
--

Build a man a fire and he'll be warm
for a night.
Set a man on fire and he'll be warm
for the rest of his life.

[François Bissey]

If you were building in parallel (and I suspect that was the case) the
thread building pathlib2 could have failed before the one that lead to
the building of six. You should check the time stamps in logs/pkgs/.

Francois

[kcrisman]

While trying binary-pkg for Sage 7.2 on OS X 10.7 (due to be upgraded to new computer within a week, so wanted to provide last binaries from that), apparently the tar and non-app dmg were created, but the app failed with the following.  Any ideas?  (Later today or tomorrow I can provide these for upload to mirrors, but not if I don't have the app version.  I'll note I rarely had trouble creating bdist with the sage-bdist script...)

total size is 2610034433  speedup is 1.00
test -d sage-7.2-OSX_10.7.5_x86_64/SageMath-7.2.app/Contents/Resources/sage && touch sage-7.2-OSX_10.7.5_x86_64/SageMath-7.2.app/Contents/Resources/sage
rm -f sage-7.2-OSX_10.7.5_x86_64/SageMath-7.2.app/sage
ln -s Contents/Resources/sage/sage sage-7.2-OSX_10.7.5_x86_64/SageMath-7.2.app/sage
cp /Users/karl.crisman/Downloads/binary-pkg/staging/OSX_mac_app/SageMath/src/bin/sage-README-osx.txt sage-7.2-OSX_10.7.5_x86_64/README.txt
rm -f sage-7.2-OSX_10.7.5_x86_64/.DS_Store
tools/createDSStore/createDSStore.py sage-7.2-OSX_10.7.5_x86_64 sage-7.2-OSX_10.7.5_x86_64 SageMath-7.2.app
Traceback (most recent call last):
  File "tools/createDSStore/createDSStore.py", line 146, in <module>
    app_name = sys.argv[3])
  File "tools/createDSStore/createDSStore.py", line 123, in createDSStore
    with ds_store.DSStore.open(filePath, 'w+') as d:
  File "/Users/karl.crisman/Downloads/binary-pkg/staging/OSX_mac_app/SageMath/src/mac-app/tools/createDSStore/ds_store/store.py", line 269, in open
    store = buddy.Allocator.open (file_or_name, mode)
  File "/Users/karl.crisman/Downloads/binary-pkg/staging/OSX_mac_app/SageMath/src/mac-app/tools/createDSStore/ds_store/buddy.py", line 221, in open
    return Allocator (f)
  File "/Users/karl.crisman/Downloads/binary-pkg/staging/OSX_mac_app/SageMath/src/mac-app/tools/createDSStore/ds_store/buddy.py", line 136, in __init__
    count, self._unknown2 = self._root.read('>II')
  File "/Users/karl.crisman/Downloads/binary-pkg/staging/OSX_mac_app/SageMath/src/mac-app/tools/createDSStore/ds_store/buddy.py", line 76, in read
    return struct.unpack (size_or_format, data)
struct.error: unpack requires a string argument of length 8
make[2]: *** [sage-7.2-OSX_10.7.5_x86_64/.DS_Store] Error 1
cp: SageMath/src/mac-app/sage-7.2-*.app.dmg: No such file or directory
ERROR:root:Script failed:
ERROR:root:DMG=/Users/karl.crisman/Downloads/binary-pkg/dist/sage-7.2-OSX_10.7.5-x86_64.app.dmg
rm -f $DMG
pwd
(cd SageMath/src/mac-app && make)
cp SageMath/src/mac-app/sage-7.2-*.app.dmg $DMG

Traceback (most recent call last):
  File "/Users/karl.crisman/Downloads/binary-pkg/tools/binary-pkg/lib/python3.4/runpy.py", line 170, in _run_module_as_main
    "__main__", mod_spec)
  File "/Users/karl.crisman/Downloads/binary-pkg/tools/binary-pkg/lib/python3.4/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/Users/karl.crisman/Downloads/binary-pkg/binary_pkg/cmdline.py", line 78, in <module>
    launch()
  File "/Users/karl.crisman/Downloads/binary-pkg/binary_pkg/cmdline.py", line 74, in launch
    package.dist_script.run()
  File "/Users/karl.crisman/Downloads/binary-pkg/binary_pkg/bash_script.py", line 37, in run
    ['bash', self._filename], cwd=self._cwd, env=self.env())
  File "/Users/karl.crisman/Downloads/binary-pkg/tools/binary-pkg/lib/python3.4/subprocess.py", line 561, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['bash', '/Users/karl.crisman/Downloads/binary-pkg/tmp/SageMath/tmp976sv49f.sh']' returned non-zero exit status 1
make[1]: *** [dist-sage] Error 1
make: *** [bdist-sage-osx] Error 2

[Volker Braun]

We improved the mac app bundling script recently, seems like it is not compatible with OSX 10.7

[kcrisman]

We improved the mac app bundling script recently, seems like it is not compatible with OSX 10.7

I saw some posts about 10.8 not working but that was post 7.2, it was 7.3 betas, so I didn't think there would be a problem. 

Was this related to this?  https://github.com/sagemath/binary-pkg/commit/4d827f34da028563e69b066bda5ffba91c937eaa

No: https://github.com/sagemath/sage/commit/7b66c50be6c44f97d9779b40aaca15cd2ee035e4
and specifically this line failed
https://github.com/sagemath/sage/blob/master/src/mac-app/tools/createDSStore/ds_store/buddy.py#L76

So is there any alternative?  In the past one could "just make the bundle" and then drop the binary in.

[kcrisman]

 > We improved the mac app bundling script recently, seems like it is not compatible with OSX 10.7

Volker: mgoerner helped me some at #20119:

"I think you can just remove $(TARGET)/.DS_store from the non_app_files: target in src/mac-app/Makefile and then build it on 10.7 -without the nice arrangement of the icons and background image :(

I am not sure how you can tell binary_pkg to pick up that change. I would commit that change to a local repository and point binary_pkg's sage_yaml to that local repository."

I did all this, and even

name: SageMath

repository: /Users/karl.crisman/Downloads/sage
branch: osx-10-7

doesn't seem to work.  Does the file assume an online repository?  I don't know how to make it look right but certainly

$ ls ../sage/.git

has the usual stuff.  Any help is greatly appreciated!

[Volker Braun]

You should be able to use local repos but only using the normal git syntax, i.e.

repository: file:///Users/karl.crisman/Downloads/sage

branch: osx-10-7

[kcrisman]

repository: file:///Users/karl.crisman/Downloads/sage

branch: osx-10-7

Got it - I've never seen that syntax before since I've never had to clone something already local before :) thank you very much.  Though unfortunately it still gives the same error message :(

 
http://trac.sagemath.org/ticket/20789

I think what is happening is that

def git_clone(config):
    if os.path.exists(os.path.join(config.source_path, '.git')):
        subprocess.check_call([
            'git', 'checkout', config.branch
        ], cwd=config.source_path)
        subprocess.check_call([
            'git', 'pull', '--ff-only'
        ], cwd=config.source_path)
    else:
        subprocess.check_call([
            'git', 'clone', config.repository, '-b', config.branch, config.source_path
        ])

is noting that my local path to the git repo does exist, but then

$ git checkout osx-10-7
error: pathspec 'osx-10-7' did not match any file(s) known to git.

which makes sense because that branch is not in binary-pkg (I couldn't figure out exactly where source_path was set).

But I'd rather patch binary-pkg than try to do it otherwise at this point.  Almost there!

[kcrisman]

On Wednesday, June 8, 2016 at 9:25:22 PM UTC-4, kcrisman wrote:

repository: file:///Users/karl.crisman/Downloads/sage

branch: osx-10-7

Got it - I've never seen that syntax before since I've never had to clone something already local before :) thank you very much.  Though unfortunately it still gives the same error message :(

 
http://trac.sagemath.org/ticket/20789

I think what is happening is that

def git_clone(config):
    if os.path.exists(os.path.join(config.source_path, '.git')):
        subprocess.check_call([
            'git', 'checkout', config.branch
        ], cwd=config.source_path)
        subprocess.check_call([
            'git', 'pull', '--ff-only'
        ], cwd=config.source_path)
    else:
        subprocess.check_call([
            'git', 'clone', config.repository, '-b', config.branch, config.source_path
        ])

is noting that my local path to the git repo does exist,

Actually, I realized this morning that probably the issue is I already build Sage once but there was the error.  So it finds that and of course there is no branch there.  I guess I could have worked on that one but mgoerner's solution seems reasonable, if I had known that there was still a 'local' Sage repo inside binary-pkg.  So I'll try again with this first.

[kcrisman]

Actually, I realized this morning that probably the issue is I already build Sage once but there was the error.  So it finds that and of course there is no branch there.  I guess I could have worked on that one but mgoerner's solution seems reasonable, if I had known that there was still a 'local' Sage repo inside binary-pkg.  So I'll try again with this first.

Yes, apparently that fixed it.

 (Harald, an app dmg 7.2 ready for mac/intel at http://www.math-cs.gordon.edu/~kcrisman/sage-7.2-OSX_10.7.5-x86_64.app.dmg - thanks!)

[Harald Schilly]

On Fri, Jun 10, 2016 at 3:01 PM, kcrisman <kcri...@gmail.com> wrote:
> (Harald, an app dmg 7.2 ready for mac/intel at
> http://www.math-cs.gordon.edu/~kcrisman/sage-7.2-OSX_10.7.5-x86_64.app.dmg -
> thanks!)

ok, done

-- h