Hi Georg,
On Mon, May 10, 2010 at 4:11 PM, Georg S. Weber
wrote:
> sage -t -long "devel/sage/sage/matrix/matrix1.pyx"
> **
> File "/Users/Shared/sage/test/sage-4.4.2.alpha0/devel/sage/sage/matrix/
> matrix1.pyx", line 460:
>s
On Mon, May 10, 2010 at 4:47 PM, Georg S. Weber
wrote:
> Exactly, the Sage documentation is no longer built by "make"!
That's because I was being careless and accidentally replaced the
following line in makefile
all: build doc
with
all: build
It should never have been replaced in the first pl
On Mon, May 10, 2010 at 4:17 PM, Georg S. Weber
wrote:
> The second problem occurs in a "long" only doctest, but probably
> occurs now on any machine. From what I guess skipping through the
> patches newly applied to sage-4.4.2.alpha0, I'd suspect " #8479:
> Wilfried Huss: numpy support for more b
On 64-bit ubuntu, built from scratch, I had these (with make ptestlong):
The following tests failed:
sage -t -long devel/sage/sage/misc/sagedoc.py # 3 doctests failed
sage -t -long devel/sage/sage/libs/mwrank/all.py # 0 doctests failed
sage -t -long devel/sage/sage/cal
The third issue I se with sage-4.4.2-alpha0 is the following. I did
the usual " export SAGE_BINARY_BUILD=yes " (the PIL spkg does not
build for me without this additional environment variable, something
that is "accepted") and " export MAKE='make -j2' " (my machine has a
Core2Duo CPU, so there), an
The second problem occurs in a "long" only doctest, but probably
occurs now on any machine. From what I guess skipping through the
patches newly applied to sage-4.4.2.alpha0, I'd suspect " #8479:
Wilfried Huss: numpy support for more basic functions [Reviewed
by Burcin Erocal] ", but this might be
On MacIntel OS X 10.4, building sage-4.4.2.alpha0 from scratch, the
build finishes, but I see three new issues. I'll describe them in
three different posts. (Note that with sage-4.4.1, on the same
machine, I had done a full "make testlong", which succeeded
flawlessly). The first on seems to be a tr
Minh Nguyen wrote:
Hi folks,
This release of Sage 4.4.2.alpha0 is earlier than I scheduled [1]. But
I have found some doctest failures that could take a while to resolve.
So getting this release out early should help with resolving them.
On Fedora 12, 64 bit all troubles seem far away!
-