On 10/10/10 11:27 PM, Mitesh Patel wrote:
On 10/10/2010 11:54 AM, Jeroen Demeyer wrote:
As part of the "rewrite interrupt handling" ticket #9678, I decided that
_sig_on should be an *inline function* instead of a macro (one very good
reason for this would be that we can declare it cdef inline int sig_on()
except -1 in order to use Cython's exception propagation mechanism).
For this to be possible, the syntax of _sig_on must be changed by adding
function-call parentheses. I also suggest to drop the leading
underscore and make it "sig_on()" instead of "_sig_on".
Right now, I would like to simply change "_sig_on" to "sig_on()" without
changing the meaning, only the syntax. At ticket #10115 I did just
that. After macro expansion, the generated code with #10115 should be
exactly the same as before.
I hope this patch #10115 can be in sage-4.6 because it would make
development of #9678 a lot easier. Since it doesn't actually change the
code (after macro expansion), I think it is also safe to include it at
this point in the sage-4.6 development process. I hope the release
manager agrees with me.
Are there any objections to making Jeroen's proposed change?
The sage repository patch at #10115 touches 106 files. We could merge
it late in 4.6 or very early in 4.6.1. The latter seems safer, but I
*might* be willing to consider the former, if #10115 has a positive
review soon and indeed does not cause any problems.
http://trac.sagemath.org/sage_trac/ticket/10115
http://trac.sagemath.org/sage_trac/ticket/9678
IMHO, something that touches so many files, that's not a bug fix but just
renaming a command, has a fairly large risk/benefit factor.
So I intend to agree with you - leave it to 4.6.1.alpha0
dave
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
