sage-devel
https://groups.google.com/d/forum/sage-devel
This email list is for discussion of Sage (<a href="http://www.sagemath.org">www.sagemath.org</a>) development issues. This list usually has heavy traffic.enRe: sage always executes the binary found in SAGE_ROOT
https://groups.google.com/d/msg/sage-devel/D6uFTYxONEw/a_ztYpGLAXIJ
That's exactly the purpose of a *user-set* SAGE_ROOT: To run a Sage installation *somewhere else* in the filesystem hierarchy. It's probably a bit surprising that that even works with (e.g.) './sage', but in principle intentional. -leif > On 19 April 2015 at 17:33, ggrafendorfer <http://georg.grhttps://groups.google.com/d/topic/sage-devel/D6uFTYxONEw
leifSun, 19 Apr 2015 18:28:49 UTCRe: [sage-devel] sage always executes the binary found in SAGE_ROOT
https://groups.google.com/d/msg/sage-devel/D6uFTYxONEw/FsEa8mm4XGwJ
Do you have a reason for having SAGE_ROOT set at all? I don't. John On 19 April 2015 at 17:33, ggrafendorfer <georg.gra...@gmail.com> wrote: > Hi > > When I build a new version of sage in a separate directory, and then, while > beeing in that directory, execute > > ./sage > > then thehttps://groups.google.com/d/topic/sage-devel/D6uFTYxONEw
John CremonaSun, 19 Apr 2015 16:43:27 UTCsage always executes the binary found in SAGE_ROOT
https://groups.google.com/d/msg/sage-devel/D6uFTYxONEw/yrizHnZ_taIJ
Hi When I build a new version of sage in a separate directory, and then, while beeing in that directory, execute ./sage then the current "old" version of sage, which can be found in SAGE_ROOT is executed. E.g., this looks as follows: .../data/sage-6.6$ pwd /mnt/data/sage-6.6 .../data/sage-6.6$https://groups.google.com/d/topic/sage-devel/D6uFTYxONEw
ggrafendorferSun, 19 Apr 2015 16:33:24 UTCRe: pushing to tickets after setting it to positive_review is incompatible with the current workflow
https://groups.google.com/d/msg/sage-devel/b8r5tAhaYO0/DeHkj5rgIy4J
agreed >> that the changes were good to be integrated. But it might still interest >> other to have a look (or even the reviewer might think of something in the >> morning after setting the day before 'positive review'). So I would prefer a >> stand by period between the 'positive review' andhttps://groups.google.com/d/topic/sage-devel/b8r5tAhaYO0
leifSun, 19 Apr 2015 16:32:20 UTCRe: [sage-devel] Re: pushing to tickets after setting it to positive_review is incompatible with the current workflow
https://groups.google.com/d/msg/sage-devel/b8r5tAhaYO0/Qu5OhAH8l5cJ
Hellooooooooooooooooooo, > I often have a look at positively reviewed tickets and sometimes ask > questions about the review. Positive review just mean that one person agreed > that the changes were good to be integrated. But it might still interest > other to have a look (or even the reviewerhttps://groups.google.com/d/topic/sage-devel/b8r5tAhaYO0
Nathann CohenSun, 19 Apr 2015 15:41:09 UTCRe: [sage-devel] Re: pushing to tickets after setting it to positive_review is incompatible with the current workflow
https://groups.google.com/d/msg/sage-devel/b8r5tAhaYO0/1KyhsFpAQbEJ
Hello, I agree with the fact that it would be better to move ticket to 'closed' as soon as the release manager is working on it. But I do not agree that this should be automatic after one second! I often have a look at positively reviewed tickets and sometimes ask questions about the review.https://groups.google.com/d/topic/sage-devel/b8r5tAhaYO0
vdelecroixSun, 19 Apr 2015 15:24:42 UTCRe: pushing to tickets after setting it to positive_review is incompatible with the current workflow
https://groups.google.com/d/msg/sage-devel/b8r5tAhaYO0/Rr9e-rcIH9cJ
ok. +1 > The downside is that you'll get more emails as I'll inevitably have to reopen > tickets, of course. > but merging/unmerging is scripted anyways. Eventually the merge will be > done by a script, so ideally tickets will only stay for seconds in the > positive_review state then. Eshttps://groups.google.com/d/topic/sage-devel/b8r5tAhaYO0
Clemens HeubergerSun, 19 Apr 2015 15:12:56 UTCRe: matrices: matrix space vs matrix group
https://groups.google.com/d/msg/sage-devel/UXzQb1v0PzI/i_QpfwxaQlsJ
I agree, coercion G -> M is probably the right thing to do here. On Sunday, April 19, 2015 at 7:47:11 AM UTC-4, vdelecroix wrote: > > Hello, > > I really do not like > > sage: M = MatrixSpace(QQ,3) > sage: G = SL(3, QQ) > sage: m1 = M(1) > sage: m2 = G(1) > sage: m1 > [1 0 0] > [0 1 0] >https://groups.google.com/d/topic/sage-devel/UXzQb1v0PzI
Volker BraunSun, 19 Apr 2015 14:18:12 UTCRe: [sage-devel] matrices: matrix space vs matrix group
https://groups.google.com/d/msg/sage-devel/UXzQb1v0PzI/DN1t3dtbMY0J
to M > which looks natural to me. Is it? > I agree that there should be coercion from a group of units (or a subgroup of that, as here) into a ring. But automatic? You can do A(G(1)) and G(A(1)) here, so manual coercion is possible. John > Best, > Vincent > > -- > You received this mhttps://groups.google.com/d/topic/sage-devel/UXzQb1v0PzI
John CremonaSun, 19 Apr 2015 13:59:28 UTCRe: [sage-devel] Re: help for patchbot beginners
https://groups.google.com/d/msg/sage-devel/Ircha-HJ3IQ/gcRE2OV88bwJ
On 19 April 2015 at 10:16, 'Martin R' via sage-devel <sage-...@googlegroups.com> wrote: > I made distclean and started from scratch, now tests pass. Apparently, > something went wrong with the copying. I'll report as soon as the tests are > finished and patchbot is running... It is hard tohttps://groups.google.com/d/topic/sage-devel/Ircha-HJ3IQ
John CremonaSun, 19 Apr 2015 13:54:11 UTCShowing poset, which way up?
https://groups.google.com/d/msg/sage-devel/6xNK1Du-rOg/wc4Qd3t32C4J
"Can I just say I'm getting tired of people changing things in visual representation (e.g. show)" Somewhat related: how should Sage plot Poset([[1,2],[[1,2]]]).show() ? I made http://trac.sagemath.org/ticket/16865 from this 8 months ago, but nobody has said which should be defined correct wahttps://groups.google.com/d/topic/sage-devel/6xNK1Du-rOg
Jori MantysaloSun, 19 Apr 2015 13:22:41 UTCmatrices: matrix space vs matrix group
https://groups.google.com/d/msg/sage-devel/UXzQb1v0PzI/OymQXCKTUnQJ
Hello, I really do not like sage: M = MatrixSpace(QQ,3) sage: G = SL(3, QQ) sage: m1 = M(1) sage: m2 = G(1) sage: m1 [1 0 0] [0 1 0] [0 0 1] sage: m2 [1 0 0] [0 1 0] [0 0 1] sage: m1 == m2 False Shouldn't it be True? One way would be to have a coerce embeddig from G to M which lohttps://groups.google.com/d/topic/sage-devel/UXzQb1v0PzI
vdelecroixSun, 19 Apr 2015 11:47:11 UTCupdate patchbot Gentoo Base System/2.2/x86_64/3.2.1-gentoo-r2/sage4
https://groups.google.com/d/msg/sage-devel/-JA4heo_xBU/-dRJb5Eu0aoJ
Hello, It seems that the patchbot Gentoo Base System/2.2/x86_64/3.2.1-gentoo-r2/sage4 is somewhat broken! (it fails at a very early step while other build bot did well) Please update it or shut it down! Vincenthttps://groups.google.com/d/topic/sage-devel/-JA4heo_xBU
vdelecroixSun, 19 Apr 2015 11:21:47 UTCRe: help for patchbot beginners
https://groups.google.com/d/msg/sage-devel/Ircha-HJ3IQ/xkk5bHvjNmMJ
I made distclean and started from scratch, now tests pass. Apparently, something went wrong with the copying. I'll report as soon as the tests are finished and patchbot is running... Thanks, Martinhttps://groups.google.com/d/topic/sage-devel/Ircha-HJ3IQ
Martin RSun, 19 Apr 2015 09:16:04 UTCRe: [sage-devel] Re: [ARM] sage 6.6 -- bdist?
https://groups.google.com/d/msg/sage-devel/eFA5B8weoMA/N-WzRJWfV-UJ
note that the buildbot builds Atlas and gcc, AFAIK.https://groups.google.com/d/topic/sage-devel/eFA5B8weoMA
Dima PasechnikSun, 19 Apr 2015 08:28:36 UTCRe: [sage-devel] Re: [ARM] sage 6.6 -- bdist?
https://groups.google.com/d/msg/sage-devel/eFA5B8weoMA/2vl2iAEhxOkJ
Le Sat, 18 Apr 2015 16:17:53 -0700 (PDT), Volker Braun <vbrau...@gmail.com> a écrit : > Its a quad-core ARMv7 Cortex-A9 with 4gb ram iirc. (Calxeda) That's much more powerful than what I have here, and it shouldn't take that long to build sage 6.6 : the chromebook I (try to) use has an exynohttps://groups.google.com/d/topic/sage-devel/eFA5B8weoMA
SnarkSun, 19 Apr 2015 08:05:18 UTCRe: [sage-devel] Re: Deprecate or just remove
https://groups.google.com/d/msg/sage-devel/SeNoMnGSLE0/7GwsdW8VsPIJ
the somewhat arbitrary 1 year. Although I would agree a year is somewhat arbitrary, a couple of things are worth bearing in mind about the year. 1) If a lecturer uses Sage in a course, having a year of stability would be good, rather than functionality being broken half way through a course.https://groups.google.com/d/topic/sage-devel/SeNoMnGSLE0
Dr. David Kirkby (Kirkby Microwave Ltd)Sun, 19 Apr 2015 08:01:39 UTCRe: [sage-devel] Re: pushing to tickets after setting it to positive_review is incompatible with the current workflow
https://groups.google.com/d/msg/sage-devel/b8r5tAhaYO0/W_Kw7M3Ks3IJ
Hello Volker, > I'm happy to switch to closed when I merge it, and not only when it tests > ok. Works for me. This way we will know the last commit that you consider in your script, and we run no danger of adding a new one after that. > Eventually the > merge will be done by a script, so idhttps://groups.google.com/d/topic/sage-devel/b8r5tAhaYO0
Nathann CohenSun, 19 Apr 2015 06:52:49 UTCRe: [sage-devel] installation of optional spkg's on binary Sage releases
https://groups.google.com/d/msg/sage-devel/fEoDWZdkdUo/iD7DUACJ_VQJ
BTW, a bit OT but any enhancements there that could be reported downstream to "Sage proper"? (I mean without someone doing a lot of digging for them.) I assume that some like three.js stuff would take some work, but maybe there is some other stuff.https://groups.google.com/d/topic/sage-devel/fEoDWZdkdUo
kcrismanSun, 19 Apr 2015 00:35:20 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/8-zZnBH5oWAJ
On Sat, Apr 18, 2015 at 7:44 PM, Volker Braun <vbrau...@gmail.com> wrote: > x.cardinality().factor() should always work (unless it is inifinite). > Or zero. :) If we mention one pitfall, let's not forget about the other. def add_two_numbers(x,y): > return x+y # no need to convert explicitlyhttps://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
Darij GrinbergSat, 18 Apr 2015 23:47:09 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/xlTjByH28H0J
Yes, imho those are all bugs. The only place where Sage should return int is in Python magic methods, e.g. __len__ > I fear this kind of stuff exists throughout sage-combinat... and I have > never encountered it causing problems. > x.cardinality().factor() should always work (unless it is ihttps://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
Volker BraunSat, 18 Apr 2015 23:44:53 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/3mKlt_y5QDMJ
On Sat, Apr 18, 2015 at 7:15 PM, Volker Braun <vbrau...@gmail.com> wrote: > On Friday, April 17, 2015 at 11:54:29 PM UTC-4, Darij Grinberg wrote: >> >> Can anyone explain in a noob-friendly way how an implementer should >> decide between returning ints and Integers >> > > Sage functions should alwhttps://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
Darij GrinbergSat, 18 Apr 2015 23:36:36 UTCRe: [ARM] sage 6.6 -- bdist?
https://groups.google.com/d/msg/sage-devel/eFA5B8weoMA/Zbpa18987A4J
Its a quad-core ARMv7 Cortex-A9 with 4gb ram iirc. (Calxeda) On Saturday, April 18, 2015 at 4:46:45 AM UTC-4, mmarco wrote: > > Just out of curiosity: what is the setup of the ARM buildbot? > > I recently setup a qemu virtual machine (over my x86 computer, so no > virtualization, just cpu emulatihttps://groups.google.com/d/topic/sage-devel/eFA5B8weoMA
Volker BraunSat, 18 Apr 2015 23:17:53 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/_wSCd7R2dlcJ
Sage functions should always return Sage integers. A kind of corner case is if your code just works by duck-typing without ever needing a literal number. Then you should probably rely on coercion to do the right thing, so the output type matches the input type. when it is OK to treat them as ehttps://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
Volker BraunSat, 18 Apr 2015 23:15:10 UTCRe: pushing to tickets after setting it to positive_review is incompatible with the current workflow
https://groups.google.com/d/msg/sage-devel/b8r5tAhaYO0/Ru8sxM8YI1UJ
I'm happy to switch to closed when I merge it, and not only when it tests ok. The downside is that you'll get more emails as I'll inevitably have to reopen tickets, but merging/unmerging is scripted anyways. Eventually the merge will be done by a script, so ideally tickets will only stay for shttps://groups.google.com/d/topic/sage-devel/b8r5tAhaYO0
Volker BraunSat, 18 Apr 2015 23:10:47 UTCRe: Documentation of Python builtin functions broken in Sage 6.6
https://groups.google.com/d/msg/sage-devel/35Hzjk6K9rc/fJsW9mYnr3sJ
It is #18249 and needs review! Best regards, Simonhttps://groups.google.com/d/topic/sage-devel/35Hzjk6K9rc
Simon KingSat, 18 Apr 2015 22:58:10 UTCRe: help for patchbot beginners
https://groups.google.com/d/msg/sage-devel/Ircha-HJ3IQ/QouCsCubetAJ
Maybe a good solution is to have another *user* for the patchbot. so: 1) create a new user on your computer 2) copy your sage to this user home 3) run ./sage in there, it should (?) take care of the relocation once again, no garantee.. Frederic Le samedi 18 avril 2015 20:59:03 UTC+2, Frédéric Chhttps://groups.google.com/d/topic/sage-devel/Ircha-HJ3IQ
Frédéric ChapotonSat, 18 Apr 2015 20:23:11 UTCRe: Documentation of Python builtin functions broken in Sage 6.6
https://groups.google.com/d/msg/sage-devel/35Hzjk6K9rc/2PXFORdF50cJ
Hi Eric, On 2015-04-18, Eric Gourgoulhon <egourg...@gmail.com> wrote: > It seems that the inline documentation of Python builtin functions is > broken in Sage 6.6: Thank you for the report! I just finished #17814 (another introspection ticket), and based on it I will soon open a ticket for thhttps://groups.google.com/d/topic/sage-devel/35Hzjk6K9rc
Simon KingSat, 18 Apr 2015 20:13:11 UTCAvoid assignment to numeric literals after preparse
https://groups.google.com/d/msg/sage-devel/prm3Jv4Vnsk/xGM_zer_XsQJ
Hi! http://trac.sagemath.org/ticket/11542 deals with the fact that in preparsed code, assigments to numeric literals will not cause syntax errors since the literals get replaced by variables which are assumed to be constants but which are not constant in this scenario. I've got a branch attachedhttps://groups.google.com/d/topic/sage-devel/prm3Jv4Vnsk
martin....@gmx.netSat, 18 Apr 2015 20:12:04 UTCRe: help for patchbot beginners
https://groups.google.com/d/msg/sage-devel/Ircha-HJ3IQ/xtyta3gJXa8J
Hello, ok, I guess this is the source of the problem. Could you 1) copy again your original sage dir into a new dir 2) type ./sage inside this new dir maybe it will take care of the change of dir ? I am not sure.. Could some readers of the devel-group help us here ? How to safely duplicate a shttps://groups.google.com/d/topic/sage-devel/Ircha-HJ3IQ
Frédéric ChapotonSat, 18 Apr 2015 18:59:03 UTCRe: help for patchbot beginners
https://groups.google.com/d/msg/sage-devel/Ircha-HJ3IQ/BkU2lkNnfpwJ
> Do you confirm that the branch patchbot/ticket_merged > is the same as 6.7.beta1 ? > you can check that using > "git log " > the top line should have the tag 6.7.beta1. > If so the failure has probably nothing to do with the patchbot, but means that your sage is somewhat broken. It would probahttps://groups.google.com/d/topic/sage-devel/Ircha-HJ3IQ
Martin RSat, 18 Apr 2015 18:31:16 UTCRe: help for patchbot beginners
https://groups.google.com/d/msg/sage-devel/Ircha-HJ3IQ/edXcgu3A-gYJ
Hello, Do you confirm that the branch patchbot/ticket_merged is the same as 6.7.beta1 ? you can check that using "git log " the top line should have the tag 6.7.beta1. If so the failure has probably nothing to do with the patchbot, but means that your sage is somewhat broken. It would probably thttps://groups.google.com/d/topic/sage-devel/Ircha-HJ3IQ
Frédéric ChapotonSat, 18 Apr 2015 18:11:12 UTCRe: [sage-devel] Re: pushing to tickets after setting it to positive_review is incompatible with the current workflow
https://groups.google.com/d/msg/sage-devel/b8r5tAhaYO0/tEpIZG60tjAJ
Le Sat, 18 Apr 2015 11:20:59 +0200, Nathann Cohen <nathan...@gmail.com> a écrit : > Hello, > > > We somehow have to come to a conclusion. > +1 > > > Never change positive_review_tickets (and adapt developer guide, > > see #18228) > -1 > > > Cool-off period > +0.5 > > Freezing tickets ohttps://groups.google.com/d/topic/sage-devel/b8r5tAhaYO0
SnarkSat, 18 Apr 2015 17:48:24 UTCDocumentation of Python builtin functions broken in Sage 6.6
https://groups.google.com/d/msg/sage-devel/35Hzjk6K9rc/41nAxFEcAyMJ
Hi, It seems that the inline documentation of Python builtin functions is broken in Sage 6.6: sage: range? TypeError Traceback (most recent call last) ... TypeError: <built-in function range> is not a module, class, method, function, traceback, frame, or code objehttps://groups.google.com/d/topic/sage-devel/35Hzjk6K9rc
Eric GourgoulhonSat, 18 Apr 2015 17:25:11 UTCRe: help for patchbot beginners
https://groups.google.com/d/msg/sage-devel/Ircha-HJ3IQ/IP-eATUNaPAJ
It begins with the below. In case it helps, the banner prints as martin@Martin-Laptop:~/sage-develop$ ./sage ┌────────────────────────────────────────────────────────────────────┐ │ SageMath Version 6.7.beta1, Release Date: 2015-04-15 │ │ Type "notebook()" for the browser-based notehttps://groups.google.com/d/topic/sage-devel/Ircha-HJ3IQ
Martin RSat, 18 Apr 2015 17:11:58 UTCRe: [sage-devel] Re: help for patchbot beginners
https://groups.google.com/d/msg/sage-devel/Ircha-HJ3IQ/4VEVqtOUresJ
I don't know about the patchbot but the version of libec.so which is built by Sage-6.6 should be libec.so.1. John On 18 April 2015 at 16:31, Frédéric Chapoton <fchap...@gmail.com> wrote: > Hello Martin, > > ok, thanks for trying the bot, and let me try to investigate. > > Could you run onehttps://groups.google.com/d/topic/sage-devel/Ircha-HJ3IQ
John CremonaSat, 18 Apr 2015 15:57:18 UTCRe: help for patchbot beginners
https://groups.google.com/d/msg/sage-devel/Ircha-HJ3IQ/xOsqbXWpmkAJ
Hello Martin, ok, thanks for trying the bot, and let me try to investigate. Could you run one or several of the failing doctests by yourself (on the develop branch), to see if your sage install is sane ? for example sage -t --long src/sage/misc/cython.py Frederic Le samedi 18 avril 2015 15:07:https://groups.google.com/d/topic/sage-devel/Ircha-HJ3IQ
Frédéric ChapotonSat, 18 Apr 2015 15:31:55 UTChelp for patchbot beginners
https://groups.google.com/d/msg/sage-devel/Ircha-HJ3IQ/ecGPyP0ceFsJ
Hi there, motivated by Frederic's recent post, I installed the patchbot 2.3.3 on my laptop and ran it (selfishly, only on my favourite ticket for now). However, I'm absolutely clueless what its output really means. More precisely: what should I do? There are several complaints about Impohttps://groups.google.com/d/topic/sage-devel/Ircha-HJ3IQ
Martin RSat, 18 Apr 2015 13:07:17 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/RkfIt7Fs5pMJ
I had a discussion regarding the use of python integers some time ago with Daniel Krenn. Basically, range and xrange produce Python integers, whereas srange and sxrange (from sage.misc.misc) produce Integers. Also, when dealing with lists, len usually produces Python int's. (Maybe Daniel has sohttps://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
Benjamin HacklSat, 18 Apr 2015 09:40:26 UTCRe: [sage-devel] Re: pushing to tickets after setting it to positive_review is incompatible with the current workflow
https://groups.google.com/d/msg/sage-devel/b8r5tAhaYO0/Br2KdZPy47YJ
Hello, > We somehow have to come to a conclusion. +1 > Never change positive_review_tickets (and adapt developer guide, see #18228) -1 > Cool-off period +0.5 > Freezing tickets once release manager starts merging them > (various variants proposed: > - post a comment +1https://groups.google.com/d/topic/sage-devel/b8r5tAhaYO0
Nathann CohenSat, 18 Apr 2015 09:21:03 UTCRe: pushing to tickets after setting it to positive_review is incompatible with the current workflow
https://groups.google.com/d/msg/sage-devel/b8r5tAhaYO0/N8nQe06ik_cJ
What about blockers? Would they also have to wait two weeks? Even after .rc3 is out with one trivial bug? > Ideally, > continuous integration would start merging your ticket the second you set it to > positive review. We are not there yet, but we'll do it eventually. even when we are in rehttps://groups.google.com/d/topic/sage-devel/b8r5tAhaYO0
Clemens HeubergerSat, 18 Apr 2015 08:47:05 UTCRe: [ARM] sage 6.6 -- bdist?
https://groups.google.com/d/msg/sage-devel/eFA5B8weoMA/bCOVTk4L7iwJ
Just out of curiosity: what is the setup of the ARM buildbot? I recently setup a qemu virtual machine (over my x86 computer, so no virtualization, just cpu emulation) emulating a raspberry pi. Compiling Sage 6.5 in it took almost a week. I don't know if using the real hardware would be better orhttps://groups.google.com/d/topic/sage-devel/eFA5B8weoMA
mmarcoSat, 18 Apr 2015 08:46:45 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/vihZXfSSd0MJ
This stuff is spooky... I've always been treating int and Integer as interchangeable when coding for Sage. Maybe I've introduced several such bugs. Can anyone explain in a noob-friendly way how an implementer should decide between returning ints and Integers, when it is OK to treat them as equihttps://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
Darij GrinbergSat, 18 Apr 2015 03:54:29 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/VPwEiJopJk4J
And note that sage: DyckWords(4r, 2r).cardinality() 9 sage: type(_) <type 'int'> Which might explain part of the behavior. Vincent PS: the change certainly comes from #17852 which makes now binomial(3,2) return a Python int if the first argument is a Python int. On 18/04/15 01:05, Benjhttps://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
vdelecroixFri, 17 Apr 2015 23:15:18 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/XLTr-UXaUt0J
Well done! And with the verbose output from #18244 I got AssertionError: expected a Sage Integer and got 9 of type <type 'int'> which does not help a lot, but still it is some indication Vincent On 18/04/15 01:05, Benjamin Hackl wrote: > I take everything back; I have no idea how I passed "https://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
vdelecroixFri, 17 Apr 2015 23:08:05 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/mJOXbsAW7KsJ
I take everything back; I have no idea how I passed "make ptestlong" earlier -- but the failure is still very much here. I think it can be triggered somewhat reliably with ./sage -bt src/sage/combinat/dyck_word.py while at the same time ./sage -t src/sage/combinat/dyck_word.py passes withouthttps://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
Benjamin HacklFri, 17 Apr 2015 23:05:39 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/VCWWmbmUK9UJ
At least I opened #18244... On 18/04/15 00:02, Benjamin Hackl wrote: > Vincent, did you also upgrade to 6.7.beta0 or 6.7.beta1 from a previous > version? > > I just finished building 6.7.beta1 after "make distclean", and this time I > can't reproduce the error. > Unfortunately, this doesn'thttps://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
vdelecroixFri, 17 Apr 2015 22:13:59 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/fZYNfPQ00zoJ
I did upgrade. > I just finished building 6.7.beta1 after "make distclean", and this time I > can't reproduce the error. > Unfortunately, this doesn't really explain what happened earlier, and there > is still the possibilty that > the error is random. > > I will continue to investigate thttps://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
vdelecroixFri, 17 Apr 2015 22:09:14 UTCRe: [sage-devel] message to patchbot breeders
https://groups.google.com/d/msg/sage-devel/sD5ov3IzL_I/2DS1MUtWKiEJ
It's a mistake on my part, sorry! Martin Am Freitag, 17. April 2015 23:55:04 UTC+2 schrieb Martin R: > > Hi Frederic! > > It seems to me that the patchbot 2.3.3 reports itself as patchbot 2.2, is > this possible or a mistake on my part? > > Martin > > Am Mittwoch, 15. April 2015 22:55:34 UTC+2 schttps://groups.google.com/d/topic/sage-devel/sD5ov3IzL_I
Martin RFri, 17 Apr 2015 22:02:29 UTCRe: [sage-devel] Re: doctest failure in dyck_word.py
https://groups.google.com/d/msg/sage-devel/7BQPmEgfA8M/37ckqXSq7YQJ
Vincent, did you also upgrade to 6.7.beta0 or 6.7.beta1 from a previous version? I just finished building 6.7.beta1 after "make distclean", and this time I can't reproduce the error. Unfortunately, this doesn't really explain what happened earlier, and there is still the possibilty that the errhttps://groups.google.com/d/topic/sage-devel/7BQPmEgfA8M
Benjamin HacklFri, 17 Apr 2015 22:02:27 UTC