If by rebuilds you mean the forced reinstallation of a package version
that's already installed, I don't see what the problem is. The procedure I
outlined will do nothing in that case. Nothing will be pulled, so no new
heads will be created. `hg update -c` may fail but I don't see how that's
in
On Thursday, August 25, 2011 8:28:09 PM UTC-7, leif wrote:
>
> On 26 Aug., 04:35, John H Palmieri wrote:
> > On Thursday, August 25, 2011 6:08:54 PM UTC-7, leif wrote:
> > > John, is this still current, i.e., do we still need this (especially
> > > aborting in case there are uncommitted changes
On 26 Aug., 05:44, Keshav Kini wrote:
> Here's how I think upgrading should work:
>
> Don't check for cleanness in repos.
> Pull from the new SPKGs' repos into the current installation repos.
> Do *not* merge.
> `hg update -c tip` in each of the current installation repos; save resulting
> message
Here's how I think upgrading should work:
Don't check for cleanness in repos.
Pull from the new SPKGs' repos into the current installation repos.
Do *not* merge.
`hg update -c tip` in each of the current installation repos; save resulting
messages (in a file, say) to display at the end of the upg
On 26 Aug., 05:14, Kevin Horton wrote:
> It would certainly be useful if sage would tell the user exactly what the
> "uncommitted changes" were, and if there was a way to force sage to upgrade
> anyway, if the user was convinced that these changes were not relevant. As
> it is now, it just abo
On 26 Aug., 04:35, John H Palmieri wrote:
> On Thursday, August 25, 2011 6:08:54 PM UTC-7, leif wrote:
> > John, is this still current, i.e., do we still need this (especially
> > aborting in case there are uncommitted changes)?
>
> I don't know about "need", but for example uncommitted changes in
On 2011-08-25, at 21:35 , John H Palmieri wrote:
>
>
> On Thursday, August 25, 2011 6:08:54 PM UTC-7, leif wrote:
>
> Found the "culprit" (for the message; this is from 4.7.rc2's sage-
> update / #9433):
>
> # check sage root repo. if present, check status; if there are
> # any un
On Thursday, August 25, 2011 6:08:54 PM UTC-7, leif wrote:
>
>
> Found the "culprit" (for the message; this is from 4.7.rc2's sage-
> update / #9433):
>
> # check sage root repo. if present, check status; if there are
> # any unchecked in changes, abort.
> os.chdir(SAGE_ROOT)
>
> > [Or didn't get the tarball fully extracted, such that some spkgs are
> > missing, perhaps due to insufficient disk space left?]
>
> Hmm, THAT is a possibility, I suppose... although those spkgs were
> still there, since newest_version appeared to work...
No, there are at least 65 GB free on th
On Thursday, August 25, 2011 6:42:30 PM UTC-7, Kevin Horton wrote:
> I suggest that the check for uncommitted changes should be revised to allow
> .DS_Store files that are not known to Mercurial on OS X. Perhaps it should
> allow any unknown files on any platform.
>
Actually, we should ignor
On 2011-08-25, at 20:08 , leif wrote:
> On 25 Aug., 23:04, John H Palmieri wrote:
>> On Thursday, August 25, 2011 9:12:01 AM UTC-7, leif wrote:
>>
>>> On 25 Aug., 16:51, Jeroen Demeyer wrote:
On 2011-08-25 14:40, leif wrote:
>>
> What I can definitely say is that Sage 4.7.1.rc2's hg c
On 25 Aug., 23:04, John H Palmieri wrote:
> On Thursday, August 25, 2011 9:12:01 AM UTC-7, leif wrote:
>
> > On 25 Aug., 16:51, Jeroen Demeyer wrote:
> > > On 2011-08-25 14:40, leif wrote:
>
> > > > What I can definitely say is that Sage 4.7.1.rc2's hg cannot handle
> > > > Sage 4.7.2.alpha2's ro
On Thursday, August 25, 2011 9:12:01 AM UTC-7, leif wrote:
>
> On 25 Aug., 16:51, Jeroen Demeyer wrote:
> > On 2011-08-25 14:40, leif wrote:
> >
> > > What I can definitely say is that Sage 4.7.1.rc2's hg cannot handle
> > > Sage 4.7.2.alpha2's root repo, which *will* break upgrading...
> >
>
> > But newest_version seems to be working fine.
>
> Apparently not for all packages / package base names. Which fail?
Sorry, I think I didn't make it clear that I tested newest_version on
some of the spkgs that it had failed for earlier (observed by looking
at install.log and where the env var
On 25 Aug., 18:13, kcrisman wrote:
> On Aug 25, 11:43 am, kcrisman wrote:
> > > > Something must have gotten corrupted. Is there some file I can delete
> > > > to start over? There isn't even an spkg/installed directory yet...
>
> > > Hmmm, take a look at (or re-extract) spkg/install (the main
> > > > Something must have gotten corrupted. Is there some file I can delete
> > > > to start over? There isn't even an spkg/installed directory yet...
>
> > > Hmmm, take a look at (or re-extract) spkg/install (the main build
> > > script), which sets up all package-related variables to be used
On Aug 25, 11:43 am, kcrisman wrote:
> > > Something must have gotten corrupted. Is there some file I can delete
> > > to start over? There isn't even an spkg/installed directory yet...
>
> > Hmmm, take a look at (or re-extract) spkg/install (the main build
> > script), which sets up all packa
On 25 Aug., 16:51, Jeroen Demeyer wrote:
> On 2011-08-25 14:40, leif wrote:
>
> > What I can definitely say is that Sage 4.7.1.rc2's hg cannot handle
> > Sage 4.7.2.alpha2's root repo, which *will* break upgrading...
>
> If this is true, then #10594 needs work!
According to deps, it /should/^TM w
> > Something must have gotten corrupted. Is there some file I can delete
> > to start over? There isn't even an spkg/installed directory yet...
>
> Hmmm, take a look at (or re-extract) spkg/install (the main build
> script), which sets up all package-related variables to be used in the
> "Makefi
On 2011-08-25 14:40, leif wrote:
> What I can definitely say is that Sage 4.7.1.rc2's hg cannot handle
> Sage 4.7.2.alpha2's root repo, which *will* break upgrading...
If this is true, then #10594 needs work!
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe f
> and a huge amount of ready recipes and scripts on the
Internet for copy-pasting into their own programs.
This clearly exists, but then you don't have to search for sage, but search
for python.
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this gr
On 25 Aug., 15:28, kcrisman wrote:
> > > > 'make' is irrelevant here. Apparently your 'sed' -- used by the (poor)
> > > > spkg/standard/newest_version script -- is broken, as the first error
> > > > message suggests.
>
> > > > (The variables in 'deps' are then messed up with error output instead
>
On 8/25/11 4:42 AM, v...@ukr.net wrote:
Another example - one of them asked me: "Where can I find a complete
list of functions available in Sage?" I had to explain him that Sage
includes a number of mathematical packages each one having its own
documentation, reference manuals, examples and so
> > > 'make' is irrelevant here. Apparently your 'sed' -- used by the (poor)
> > > spkg/standard/newest_version script -- is broken, as the first error
> > > message suggests.
>
> > > (The variables in 'deps' are then messed up with error output instead
> > > of package names.)
>
> > Right.
>
> >
On 22 Aug., 14:07, Jeroen Demeyer wrote:
> On 2011-08-22 00:38, MartinX wrote:
>
> > Graphviz would not install as pkg-config was not reading the
> > pangocairo.pc in the grphviz configure file in the sage environment so
> > it was not setting the includes correctly. Cause is a duplication of
> >
On 25 Aug., 14:19, kcrisman wrote:
> > > Did something change in prereq in the last couple alphas? Through
> > > 4.7.1 I really had no trouble on Cygwin other than known problems.
> > > But now I get weird things like what is below
> > > (seehttp://sage.math.washington.edu/home/kcrisman/cygwin
On 25 Aug., 14:28, leif wrote:
> On 25 Aug., 13:20, Keshav Kini wrote:
>
> > Mercurial versions in Sage haven't changed between 4.7 and 4.7.1. In any
> > case, incompatible Mercurial repository formats won't cause the error
> > messages we're seeing here, afaik.
> > > "There are uncommitted chang
On 25 Aug., 13:20, Keshav Kini wrote:
> Mercurial versions in Sage haven't changed between 4.7 and 4.7.1. In any
> case, incompatible Mercurial repository formats won't cause the error
> messages we're seeing here, afaik.
> > "There are uncommitted changes in the Sage root repository. Aborting."
> > Did something change in prereq in the last couple alphas? Through
> > 4.7.1 I really had no trouble on Cygwin other than known problems.
> > But now I get weird things like what is below
> > (seehttp://sage.math.washington.edu/home/kcrisman/cygwin-1.7.9-winxp-sage...
> > for all details).
>
Mercurial versions in Sage haven't changed between 4.7 and 4.7.1. In any
case, incompatible Mercurial repository formats won't cause the error
messages we're seeing here, afaik. If I recall correctly, `sage -upgrade`
does seem to complain about "uncommitted changes" even if there are files
whic
On 25 Aug., 10:13, Pedro Cruz wrote:
> Dear Maarten Derickx (and Sage developers)
>
> thank you for reply.
>
> Using:
> sage -t /home/jpedro/sage/local/lib/python2.6/site-packages/
> meg/paramparse.py
> causes "ImportError" as before but if I use
> sage -python -m doctest paramparse.py
> using
On 24 Aug., 21:25, Kevin Horton wrote:
> On 2011-08-24, at 13:41 , Maarten Derickx wrote:
> I have a hard time believing it could be repository corruption, as I have the
> same symptoms on two sage installations - one on OS X and one on Linux. I
> have been using sage -upgrade for quite some ti
Hi list,
Is there any problem now with the patchbot?
I see that many tickets fail at the doctest on 4.7.1, however, if you
check the logs in most of them you will not find any failure in the
log. In some cases some tests have been killed, kind of timeout? or
there are genuine doctest failures but
On 25 Aug., 00:54, William Stein wrote:
> Hi,
>
> I reconfigured apache2 so now (way, way) more processes are allocated
> to serving trac. This means that instead of trac being temporarily
> "down" due to two users doing a search at the same time, many users
> can use trac at once.
>
> Let me kno
Hello!
Just in case somebody is still interested in opinions on the initial
topic.
Since I'm not familiar with Matlab myself, I asked some people who is
familiar "Why do they choose Matlab and not any other system, for
example, Sage".
- Most of them didn't know about Sage at all till that m
On 25 Aug., 11:20, leif wrote:
> On 25 Aug., 04:17, kcrisman wrote:
> > Did something change in prereq in the last couple alphas? Through
> > 4.7.1 I really had no trouble on Cygwin other than known problems.
> > But now I get weird things like what is below
> > (seehttp://sage.math.washington.
On 25 Aug., 04:17, kcrisman wrote:
> Did something change in prereq in the last couple alphas? Through
> 4.7.1 I really had no trouble on Cygwin other than known problems.
> But now I get weird things like what is below
> (seehttp://sage.math.washington.edu/home/kcrisman/cygwin-1.7.9-winxp-sage.
Dear Maarten Derickx (and Sage developers)
thank you for reply.
Using:
sage -t /home/jpedro/sage/local/lib/python2.6/site-packages/
meg/paramparse.py
causes "ImportError" as before but if I use
sage -python -m doctest paramparse.py
using ">>>" instead of "sage:" everything works on each modul
38 matches
Mail list logo
