sage-4.6.1.rc0 released
Jeroen Demeyer
We're releasing Sage 4.6.1.rc0.
Source archive:
http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0.tar
Upgrade path:
http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0/
Please build, test, and report! We'd love to hear about your
experiences with this release.
== Notes ==
Ticket #10303 (clean up sage-check-64 and use of SAGE64) was unmerged
because of issues with sage -ba (#10436).
There are 5 tickets which need review before sage-4.6.1 can be released:
#9163: Doctest error in expect.py on Cygwin and OS X
#9523: Upgrade the Readline spkg to 6.1
#10339: Simplify spkg/pipestatus
#10491: Dump environment before building
#10494: Upgrading 4.6->4.6.1 does not upgrade sagenb
Merry christmas and happy holidays to all! With some luck, we can
celebrate the new year with sage 4.6.1.
== Known issues ==
* On 64-bit OS X, there are doctest failures for sage/misc/trace.py
and sage/tests/cmdline.py related to sage-check-64 (see #10303).
Fixing this has been postponed to sage-4.6.2.
== Tickets ==
* For a more detailed overview of all tickets and patches which are
merged in this version, see
http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/tickets.html
Closed tickets:
#8088: sage library 4.3.1 failing to build on Open Solaris x64 [Reviewed
by David Kirkby]
#10185: ECL in Sage will not build on Fedora 14, which will be released
on 2nd November 2010 [Reviewed by David Kirkby]
Merged in sage-4.6.1.rc0:
#9163: Jeroen Demeyer: Doctest error in expect.py on Cygwin and OS X
[needs review]
#9434: John Palmieri: Stop greping for a non-existent sage-banner
[Reviewed by Jeroen Demeyer, David Kirkby]
#9863: Maarten Derickx, Mitesh Patel: Error in sage/graphs/genus.pyx on
ia64-Linux-suse [Reviewed by Mitesh Patel, Jeroen Demeyer]
#10177: Maite Aranes: Gamma0 equivalence function for number field cusps
returns wrong transformation matrix [Reviewed by David Loeffler]
#10339: Jeroen Demeyer: Simplify spkg/pipestatus [needs review]
#10367: Benjamin Jones: plot3d transformation documentation says
'independent' when it should be 'dependent' [Reviewed by Jason Grout]
#10427: John Palmieri: cloning is broken on Solaris [Reviewed by Leif
Leonhardy]
#10434: Minh Van Nguyen: add doctests from #8582 and other integration
improvements from Maxima 5.22.1 [Reviewed by Karl-Dieter Crisman]
#10491: Jeroen Demeyer: Dump environment before building [needs review]
#10494: Leif Leonhardy: Upgrading 4.6->4.6.1 does not upgrade sagenb
[needs review]
Justin C. Walker
On Dec 24, 2010, at 07:55 , Jeroen Demeyer wrote:
> Dear Sage lovers,
>
> We're releasing Sage 4.6.1.rc0.
>
> Source archive:
>
> http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0.tar
>
> Upgrade path:
>
> http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0/
Upgrade from 4.6.1.a3 failed with this complaint:
Failed to download 'http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0//spkg/VERSION.txt'.
Abort.
I am now trying a from-scratch build.
Justin
--
Justin C. Walker, Curmudgeon at Large
Institute for the Absorption of Federal Funds
-----------
If it weren't for carbon-14, I wouldn't date at all.
-----------
John H Palmieri
> Dear Sage lovers,
>
> We're releasing Sage 4.6.1.rc0.
>
> Source archive:
>
>
> Upgrade path:
>
> http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6....
> Please build, test, and report! We'd love to hear about your
> experiences with this release.
Upgraded from 4.6.1.alpha0 and it seemed to go okay except for some
merging issues: the application FileMerge starts up during a upgrades
of a few Sage packages (I think scripts and the main Sage repo). Sage
fails to start up:
ImportError: cannot import name Hasse_diagram_from_incidences
Error importing ipy_profile_sage - perhaps you should run %upgrade?
WARNING: Loading of ipy_profile_sage failed.
I'm going to try upgrading from 4.6 to see how that goes.
--
John
Volker Braun
John H Palmieri
upgraded. It worked fine when I upgraded from version 4.6, but not
from 4.6.1.alpha0. I have no idea why.
--
John
Florent Hivert
> We're releasing Sage 4.6.1.rc0.
>
> Source archive:
>
> http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0.tar
make ptestlong
[...]
----------------------------------------------------------------------
All tests passed!
Total time for all tests: 1970.5 seconds
openSUSE 11.3 (x86_64)
fresh parallel build from sources.
Good work guys,
Florent
Justin C. Walker
On Dec 24, 2010, at 07:55 , Jeroen Demeyer wrote:
> Dear Sage lovers,
>
> We're releasing Sage 4.6.1.rc0.
>
> Source archive:
>
> http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0.tar
Full build succeeded on Mac OS X, 10.6.5, Dual Quad Xeon. Testing (ptestlong) showed three failures:
sage -t -long -force_lib devel/sagenb-main/sagenb/misc/misc.py # 0 doctests failed
sage -t -long -force_lib devel/sage/sage/misc/trace.py # 1 doctests failed
sage -t -long -force_lib devel/sage/sage/tests/cmdline.py # 3 doctests failed
Full log is at
sage.math.washington.edu:~justin/logs/4.6.1.r0-full.log
Justin
--
Justin C. Walker
Curmudgeon-at-large
Director
Institute for the Absorption of Federal Funds
----
186,000 Miles per Second
Not just a good idea:
it's the law!
----
Justin C. Walker
On Dec 24, 2010, at 07:55 , Jeroen Demeyer wrote:
> Dear Sage lovers,
>
> We're releasing Sage 4.6.1.rc0.
>
> Source archive:
>
> http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0.tar
>
> Upgrade path:
>
> http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0/
Tried to upgrade a second time, and got the same failure. The transcript is below, and the problem is that the VERSION.txt is being accessed in ".../spkg", but it's not there.
Thoughts?
Justin
================================================================
./sage -upgrade http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0/
Downloading packages from 'http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0//spkg'.
Reading package lists... Done!
The following packages will be upgraded:
examples-4.6.1.rc0 extcode-4.6.1.rc0 sage-4.6.1.rc0
sage_scripts-4.6.1.rc0 sagenb-0.8.10
** WARNING: This is a source-based upgrade, which could take hours,
** fail, and render your Sage install useless!!
Do you want to continue [y/N]? y
http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0//spkg/standard/examples-4.6.1.rc0.spkg --> examples-4.6.1.rc0.spkg [..................................................]
Deleting old spkg 'examples-4.6.1.alpha3.spkg'...
http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0//spkg/standard/extcode-4.6.1.rc0.spkg --> extcode-4.6.1.rc0.spkg [..................................................]
Deleting old spkg 'extcode-4.6.1.alpha3.spkg'...
http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0//spkg/standard/sage-4.6.1.rc0.spkg --> sage-4.6.1.rc0.spkg [..................................................]
Deleting old spkg 'sage-4.6.1.alpha3.spkg'...
http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0//spkg/standard/sage_scripts-4.6.1.rc0.spkg --> sage_scripts-4.6.1.rc0.spkg [..................................................]
Deleting old spkg 'sage_scripts-4.6.1.alpha3.spkg'...
http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0//spkg/standard/sagenb-0.8.10.spkg --> sagenb-0.8.10.spkg [..................................................]
Deleting old spkg 'sagenb-0.8.9.spkg'...
http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0//spkg/VERSION.txt --> VERSION.txt [.]
Failed to download 'http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0//spkg/VERSION.txt'.
Abort.
================================================================
--
Justin C. Walker, Curmudgeon-At-Large
Institute for the Enhancement of the Director's Income
--------
When LuteFisk is outlawed,
Only outlaws will have LuteFisk
--------
leif
No patch to the Sage library should require 'sage -ba'.
If 'sage -b' isn't enough, dependencies should be added to
'module_list.py', or maybe the patch should 'touch' other files as well.
But perhaps that was just another weird MacOS X (only) issue.
-Leif
Volker Braun
leif
> For the record, the "Hasse_diagram_from_incidences" function was moved
> between two Python modules. So no change to module_list.py was necessary
> (or even possible). But if I understand John correctly, the patch was
> just not applied.
In which case 'sage -ba' wouldn't have helped either... ;-)
> Could be a weird mercurial issue where some versions
> are better applying patches than others...
Hmmm, the Mercurial in Sage should have been upgraded as well, I don't
recall in which release the new spkg was included.
But creating patches with the version contained in the latest Sage
release is unfortunately apparently safer.
-Leif
John H Palmieri
> On Dec 24, 2010, at 07:55 , Jeroen Demeyer wrote:
>
> > Dear Sage lovers,
>
> > We're releasing Sage 4.6.1.rc0.
>
> > Source archive:
>
>
> > Upgrade path:
>
> >http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6....
>
>
> Abort.
This is because of a bug in 4.6.1.alpha3 (or maybe alpha2?): in that
alpha version, sage-update was changed so that it tries to download a
nonexistent file. Updating from any other version of Sage should work
fine, and updating from rc0 to a later version I think should work
too.
--
John
Justin C. Walker
That will teach me to pay attention. Well, maybe...
Thanks; I'll try an earlier version.
Justin
--
Justin C. Walker, Curmudgeon-At-Large
Institute for the Enhancement of the Director's Income
--------
Experience is what you get
when you don't get what you want.
--------
leif
>
> On Dec 26, 2010, at 21:01 , John H Palmieri wrote:
>
>> On Dec 24, 2:42 pm, "Justin C. Walker" <jus...@mac.com> wrote:
>>> On Dec 24, 2010, at 07:55 , Jeroen Demeyer wrote:
>>>> Upgrade path:
>>>
>>>> http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6....
>>>
>>> Upgrade from 4.6.1.a3 failed with this complaint:
>>>
>>> Failed to download 'http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.....
>>> Abort.
>>
>> This is because of a bug in 4.6.1.alpha3 (or maybe alpha2?): in that
>> alpha version, sage-update was changed so that it tries to download a
>> nonexistent file. Updating from any other version of Sage should work
>> fine, and updating from rc0 to a later version I think should work
>> too.
>
> That will teach me to pay attention. Well, maybe...
>
> Thanks; I'll try an earlier version.
Just hack your current local/bin/sage-update: ;-)
@@ -351,7 +358,7 @@
version_file.close()
else:
old_version = ""
- download_file("VERSION.txt")
+ download_file("standard/VERSION.txt")
version_file = open("VERSION.txt")
new_version = version_file.read()
new_version = new_version.strip() # remove trailing newline
@@ -364,11 +371,18 @@
version_file = open("VERSION.txt", 'w')
version_file.write("%s%s" % (new_version, old_version))
version_file.close()
+ # shutil.copy raises an error if the destination already exists,
+ # so remove SPKG_ROOT/standard/VERSION.txt.
+ try:
+ os.remove(os.path.join(SPKG_ROOT, "standard", "VERSION.txt"))
+ except OSError:
+ pass
+ shutil.copy("VERSION.txt", os.path.join(SPKG_ROOT, "standard"))
(The first hunk should suffice I think. Just an excerpt from the patches.)
Cheers,
-Leif
Dima Pasechnik
without a parameter nuked the version (4.6.1.aplha3) I had, as it
started updating 4.6 instead....
I suppose this is a bug...
Dima
Adam Webb
On Dec 24, 4:55 pm, Jeroen Demeyer <jdeme...@cage.ugent.be> wrote:
> Dear Sage lovers,
>
> We're releasing Sage 4.6.1.rc0.
>
> Source archive:
Ubuntu 10.10 (AMD64) --> all tests passed
Scientific Linux 5.1 (32 bit) --> two tests failed
sage -t -long -force_lib devel/sage/sage/plot/plot.py # Time out
sage -t -long -force_lib devel/sage/sage/gsl/
probability_distribution.pyx # Time out
These passed when done separately.
This might be known but I found two possible conflicts with my system
python.
1) If a .pydistutils.cfg is present in the home directory it will
change the location of the installed python files.
2) If there is a .local directory in the home directory with anything
installed using --user than it will be found and might interfere.
Cheers,
Adam
Dima Pasechnik
On Dec 26, 5:08 am, "Justin C. Walker" <jus...@mac.com> wrote:
> On Dec 24, 2010, at 07:55 , Jeroen Demeyer wrote:
>
> > Dear Sage lovers,
>
> > We're releasing Sage 4.6.1.rc0.
>
> > Source archive:
>
>
> > Upgrade path:
>
> >http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6....
>
>
> Thoughts?
>
> Justin
>
> ================================================================
> Downloading packages from 'http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.....
> The following packages will be upgraded:
>
> examples-4.6.1.rc0 extcode-4.6.1.rc0 sage-4.6.1.rc0
> sage_scripts-4.6.1.rc0 sagenb-0.8.10
>
> ** WARNING: This is a source-based upgrade, which could take hours,
> ** fail, and render your Sage install useless!!
>
> Deleting old spkg 'examples-4.6.1.alpha3.spkg'...http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6....--> extcode-4.6.1.rc0.spkg [..................................................]
> Deleting old spkg 'extcode-4.6.1.alpha3.spkg'...http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6....--> sage-4.6.1.rc0.spkg [..................................................]
> Deleting old spkg 'sage-4.6.1.alpha3.spkg'...http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6....--> sage_scripts-4.6.1.rc0.spkg [..................................................]
> Deleting old spkg 'sage_scripts-4.6.1.alpha3.spkg'...http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6....--> sagenb-0.8.10.spkg [..................................................]
> Deleting old spkg 'sagenb-0.8.9.spkg'...http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6....--> VERSION.txt [.]
> Failed to download 'http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.....
> Abort.
I have exactly the same problem.
Hey Jeroen, prettige kerstdagen, hoor :-)
kcrisman
On Dec 29, 4:49 am, Dima Pasechnik <dimp...@gmail.com> wrote:
> Running ./sage --upgrade
> without a parameter nuked the version (4.6.1.aplha3) I had, as it
> started updating 4.6 instead....
>
> I suppose this is a bug...
x'.y'.z'? At the same time, the idea IS that upgrade without a
parameter defaults to the most recent official release, so perhaps
this is a feature of sorts.
- kcrisman
Keshav Kini
caused some weird stuff to happen to the hg repositories on my
machine. After sifting through $(hg glog) a bit, I found that the
changesets I had originally had from 4.6.1.alpha3 didn't have any tags
for 4.6.1.alpha0, alpha1, or alpha2. It looks like the repository was
branched from where alpha0 should have been, as all the changeset IDs
are different after a certain point. From #sage-devel:
[2010-12-30 18:15:16] <kini> well, how it looks in my log is that the
library diverged after revision df2578b0f467 - one child,
9ccb2df44e94, comment "#10174: speed up relative norm in relative
number fields" by lftabera, leads to 4.6.1.alpha3, and the other
child, 782092503e90 with the same comment, also by lftabera, leads to
4.6.1.rc0
[2010-12-30 18:16:57] <kini> indeed the difference between the
4.6.1.alpha3 head and the 4.6.1.alpha3 revision which is somewhere
along the ancestry chain of 4.6.1.rc0 is just a few lines:
http://pastebin.com/cbqG5qz7
Any stories about this one? Was the repository just rebooted from 4.6
for some reason? I assume it was a mistake that the release of alpha3
didn't have tags for alpha0-alpha2, and this was to fix that? Anyway,
just thought I'd post this here in case anyone else was scratching
their heads.
-Keshav
leif
> Even running `./sage -upgrade http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0/`
> caused some weird stuff to happen to the hg repositories on my
> machine. After sifting through $(hg glog) a bit, I found that the
> changesets I had originally had from 4.6.1.alpha3 didn't have any tags
> for 4.6.1.alpha0, alpha1, or alpha2. It looks like the repository was
> branched from where alpha0 should have been, as all the changeset IDs
> are different after a certain point. From #sage-devel:
>
> [2010-12-30 18:15:16] <kini> well, how it looks in my log is that the
> library diverged after revision df2578b0f467 - one child,
> 9ccb2df44e94, comment "#10174: speed up relative norm in relative
> number fields" by lftabera, leads to 4.6.1.alpha3, and the other
> child, 782092503e90 with the same comment, also by lftabera, leads to
> 4.6.1.rc0
> [2010-12-30 18:16:57] <kini> indeed the difference between the
> 4.6.1.alpha3 head and the 4.6.1.alpha3 revision which is somewhere
> along the ancestry chain of 4.6.1.rc0 is just a few lines:
> http://pastebin.com/cbqG5qz7
>
> Any stories about this one? Was the repository just rebooted from 4.6
> for some reason? I assume it was a mistake that the release of alpha3
> didn't have tags for alpha0-alpha2, and this was to fix that? Anyway,
> just thought I'd post this here in case anyone else was scratching
> their heads.
No, that's an intentional change by Jeroen (the current release manager)
since / for 4.6.1 that I don't like very much btw.
Every new alpha/rc is now based on the previous *official* release
rather than the previous developer one.
This makes sense to some extent since we now also have patches/tickets
still needing review in devel releases (alphas) that might get removed
in subsequent ones. (We occasionally had reverted / unmerged tickets in
earlier alpha releases, too, though, but not really intended as "trial"
tickets. Merged tickets still needing review shouldn't be expected to
actually "need work" on the other hand; they're just meant to be tested
more thoroughly, on a wider range of systems, as I understand this.)
I also suggested to output a warning if one updates / reinstalls the
Sage library spkg when the current branch isn't sage-main, as its
'spkg-install' always *silently* switches to the main branch and updates
that. (This behavior isn't documented at all afaik.)
-Leif
Keshav Kini
suddenly popped up during my upgrade process - it was trying to merge
my alpha3 into its 4.6.1.rc0, both of which were based somewhere a bit
after 4.6 - but I only figured that out after it finished installing
and I took a look through $(hg glog). For those who haven't set their
Mercurial to use a graphical client for three-way-diffs would have
diff3 or something appear in their terminal and thus be unable to
easily see the installation messages immediately prior, as well. Maybe
instead of forcing a merge, the upgrade process should just leave the
two heads in each version-controlled directory and print a message at
the end of the upgrade telling users, "The repositories have all been
updated to [version y]. Please rebase changes made to your previous
development version, if any, and (optionally) strip the previous
version's branch." or something? And yes, I noticed that the branch
was switched to sage-main as well. I agree that it should be mentioned
somewhere.
-Keshav
leif
> Keshav Kini wrote:
>> Even running `./sage -upgrade
>> http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.1.rc0/`
>>
>> caused some weird stuff to happen to the hg repositories on my
>> machine. After sifting through $(hg glog) a bit, I found that the
>> changesets I had originally had from 4.6.1.alpha3 didn't have any tags
>> for 4.6.1.alpha0, alpha1, or alpha2. It looks like the repository was
>> branched from where alpha0 should have been, as all the changeset IDs
>> are different after a certain point.
>
> No, that's an intentional change by Jeroen (the current release manager)
> since / for 4.6.1 that I don't like very much btw.
This makes it harder to find out which tickets were merged when (into
which devel release); one has to search trac or several release notes
(which are not even contained in the source tarballs!) which is quite
tedious and inconvenient.
But apparently the (devel version) tags are back in 4.6.1.rc0, so that's
not an issue if they're kept in future versions as well.
-Leif
Justin C. Walker
On Dec 26, 2010, at 22:54 , leif wrote:
> Justin C. Walker wrote:
>>
>> On Dec 26, 2010, at 21:01 , John H Palmieri wrote:
>>
>>> On Dec 24, 2:42 pm, "Justin C. Walker" <jus...@mac.com> wrote:
>>>> On Dec 24, 2010, at 07:55 , Jeroen Demeyer wrote:
>>>>> Upgrade path:
>>>>
>>>>> http://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6....
[snip]
>> That will teach me to pay attention. Well, maybe...
>>
>> Thanks; I'll try an earlier version.
>
> Just hack your current local/bin/sage-update: ;-)
>
> @@ -351,7 +358,7 @@
> version_file.close()
> else:
> old_version = ""
> - download_file("VERSION.txt")
> + download_file("standard/VERSION.txt")
> version_file = open("VERSION.txt")
> new_version = version_file.read()
> new_version = new_version.strip() # remove trailing newline
> @@ -364,11 +371,18 @@
> version_file = open("VERSION.txt", 'w')
> version_file.write("%s%s" % (new_version, old_version))
> version_file.close()
> + # shutil.copy raises an error if the destination already exists,
> + # so remove SPKG_ROOT/standard/VERSION.txt.
> + try:
> + os.remove(os.path.join(SPKG_ROOT, "standard", "VERSION.txt"))
> + except OSError:
> + pass
> + shutil.copy("VERSION.txt", os.path.join(SPKG_ROOT, "standard"))
>
> (The first hunk should suffice I think. Just an excerpt from the patches.)
Thanks for the suggestion. I tried this (just the first change). It was a change in alpha3, not alpha2). Worked like a champ.
In this case, all tests passed! I have tested (ptestlong) rc0 as follows:
As an upgrade to 4.6.1-a3 (Mac OS X, 10.5.6, Dual Quad Xeon):
all tests passed
As a from-scratch build (Mac OS X, 10.5.6, Dual Quad Xeon):
three tests failed:
sage -t -long -force_lib devel/sagenb-main/sagenb/misc/misc.py # 0 doct
ests failed
sage -t -long -force_lib devel/sage/sage/misc/trace.py # 1 doctests fai
led
sage -t -long -force_lib devel/sage/sage/tests/cmdline.py # 3 doctests
failed
As an upgrade from 4.6.rc0 (Mac OS X, 10.5.6, Core i7):
two tests failed
sage -t -long -force_lib devel/sage/sage/misc/trace.py # 1 doctests failed
sage -t -long -force_lib devel/sage/sage/tests/cmdline.py # 3 doctests failed
I'm redoing the last two 'ptestlong' runs to see how random this stuff is.
Justin
--
Justin C. Walker, Curmudgeon-at-Large
() The ASCII Ribbon Campaign
/\ Help Cure HTML Email
leif
> Finally got back to this...
[snip]
> As an upgrade to 4.6.1-a3 (Mac OS X, 10.5.6, Dual Quad Xeon):
> all tests passed
>
> As a from-scratch build (Mac OS X, 10.5.6, Dual Quad Xeon):
> three tests failed:
> sage -t -long -force_lib devel/sagenb-main/sagenb/misc/misc.py # 0 doct
> ests failed
> sage -t -long -force_lib devel/sage/sage/misc/trace.py # 1 doctests fai
> led
> sage -t -long -force_lib devel/sage/sage/tests/cmdline.py # 3 doctests
> failed
>
> As an upgrade from 4.6.rc0 (Mac OS X, 10.5.6, Core i7):
> two tests failed
> sage -t -long -force_lib devel/sage/sage/misc/trace.py # 1 doctests failed
> sage -t -long -force_lib devel/sage/sage/tests/cmdline.py # 3 doctests failed
>
> I'm redoing the last two 'ptestlong' runs to see how random this stuff is.
I guess you built the first (upgraded) one without SAGE64=yes, and the
second and third *with* SAGE64=yes.
The 4 failing doctests (trace & cmdline) should be deterministic IIRC.
-Leif
> () The ASCII Ribbon Campaign
> /\ Help Cure HTML Email
I do ;-)
leif
P.S.: Incidentally (i.e., as a side effect), #9960 should also fix these
doctest errors, currently needing review... ;-)
(#10303 was intended to fix them, but got unmerged again.)
-Leif
2011 mod 210 = 11^2
Justin C. Walker
On Dec 31, 2010, at 12:19 , leif wrote:
> Justin C. Walker wrote:
>> Finally got back to this...
[snip]
> I guess you built the first (upgraded) one without SAGE64=yes, and the second and third *with* SAGE64=yes.
I don't think I did. I think the build defaults to SAGE64=yes, and has for a while. It's probably too late to be sure, but I do keep my logs around:
On the Dual Quad Xeon:
sage-4.6.1-a3: install.log shows SAGE64=yes
On the Core i7:
sage-4.6-rc0: install.log shows building in 64-bit mode
The from-scratch build on the Xeon system was also 64-bit.
The redo of 'ptestlong' shows this:
Xeon (from-scratch build of 4.6.1-rc0):
Three tests failed (algebras/all.py, trace, cmdline)
Previously, in place of algebras/all.py, sagenb/misc.py had failed.
Core i7 (upgrade from 4.6-rc0):
Same two tests failed (trace, cmdline).
So, some randomness prevails...
Justin
--
Justin C. Walker, Curmudgeon-At-Large
Institute for the Absorption of Federal Funds
--------
Men are from Earth.
Women are from Earth.
Deal with it.
--------
Justin C. Walker
On Dec 31, 2010, at 12:27 , leif wrote:
> leif wrote:
>> Justin C. Walker wrote:
>>> Finally got back to this...
[snip]
>>> I'm redoing the last two 'ptestlong' runs to see how random this stuff is.
>> I guess you built the first (upgraded) one without SAGE64=yes, and the second and third *with* SAGE64=yes.
>> The 4 failing doctests (trace & cmdline) should be deterministic IIRC.
[snip]
> P.S.: Incidentally (i.e., as a side effect), #9960 should also fix these doctest errors, currently needing review... ;-)
I just tried this on two installs (from the full tarball) of 4.6.1.rc0, with the same results each time:
$ ../../sage -hg import /SandBox/DownLoads/trac_9960-scripts-SAGE_CHECK_rebased_to_4.6.1.rc0.patch
Detected SAGE64 flag
Building Sage on OS X in 64-bit mode
applying /SandBox/DownLoads/trac_9960-scripts-SAGE_CHECK_rebased_to_4.6.1.rc0.patch
unable to find 'sage-env' for patching
5 out of 5 hunks FAILED -- saving rejects to file sage-env.rej
unable to find 'sage-sage' for patching
1 out of 1 hunks FAILED -- saving rejects to file sage-sage.rej
unable to find 'sage-spkg' for patching
1 out of 1 hunks FAILED -- saving rejects to file sage-spkg.rej
abort: patch failed to apply
What am I missing?
Justin
--
Justin C. Walker
Curmudgeon-at-large
--
Network, n., Difference between work
charged for and work done
leif
Did you apply the patch(es) to the scripts repo (in SAGE_ROOT/local/bin)?
-Leif
Justin C. Walker
Oops.
--
Justin C. Walker, Curmudgeon at Large
Institute for the Absorption of Federal Funds
--
Democracy is two wolves and a lamb
voting on what to have for lunch.
Liberty is a well-armed lamb contesting
the vote.
Justin C. Walker
On Jan 1, 2011, at 20:06 , leif wrote:
> Justin C. Walker wrote:
>> Back from dancing the night away...
>> On Dec 31, 2010, at 12:27 , leif wrote:
[snip]
>>> P.S.: Incidentally (i.e., as a side effect), #9960 should also fix these doctest errors, currently needing review... ;-)
>>
>> I just tried this on two installs (from the full tarball) of 4.6.1.rc0, with the same results each time:
> Did you apply the patch(es) to the scripts repo (in SAGE_ROOT/local/bin)?
Yeah :-}
With #9960 in place (full build of 4.6.1.rc0, Mac OS X, 10.6.5, Dual Quad Xeon), sage-4.6.1.rc0 passes all tests!
With #9960 in place, (upgrade from sage-4.6.rc0, Mac OS X, 10.6.5, Core i7), sage-4.6.1.rc0 passes all tests!
Thanks for you comments.