#11926: "make build" should run Sage once
------------------------+---------------------------------------------------
Reporter: jdemeyer | Owner: GeorgSWeber
Type: defect | Status: needs_review
Priority: major | Milestone: sage-4.7.3
Component: build | Keywords: Makefile build
Work_issues: | Upstream: N/A
Reviewer: | Author: Jeroen Demeyer
Merged: | Dependencies:
------------------------+---------------------------------------------------
Comment(by leif):
Replying to [comment:21 jhpalmieri]:
> Jeroen's suggestion here, on the other hand, is a quick solution for a
problem with Sage now.
I don't see any problem, except that the current behaviour of Sage
regarding on-the-fly creation or modification of files in the "static"
installation tree is suboptimal and -- IMHO worse -- AFAIK not documented,
at least not very well.
And the "problem" is all but new.
Sysadmins -- at least those that have installed Sage before (others will
quickly notice) -- are likely to be aware of that; most other users don't
care at all, so this is rather a non-issue.
Adding running `sage` at the end of every (perhaps partial re-) build is
not the right way to solve this; in contrast, it is likely to obscure
errors that don't show up during the build but when you try to start Sage,
in addition with the potential to make the build hang (without any
output).
So the minimal things to do are -- if you really want to run Sage
''there'':
* Print a message that we're attempting to run `sage`.
* Disable keyboard interrupts during important operations (`sage-
location`, maybe elsewhere, too), and issue a warning that this process
must not be interrupted, at least for a few minutes.
* Perhaps set a reasonably long timeout for running `sage`.
* Log the output instead of redirecting it to `/dev/null`, such that we
can print at least a reference to the error log in case of failures.
[[BR]]
> I suggest we merge this now (once the typos are fixed, along with any
other simple fixes), and consider a longer term solution separately.
There are various possible long-term solutions; the primary goal should be
to ''document'' potential issues, i.e., at least prominently state that
Sage should be run at least once at its final location for use in a multi-
user environment, or more precisely, for a system-wide installation.
[[BR]]
> I wouldn't object if this were implemented by adding a new target in the
Makefile, but I think it should be run by default to address the issues
Jeroen has raised here.
An `echo` statement at the end of the build, perhaps in the `Makefile`,
would IMHO fully suffice for the moment; no need to hurry or for "quick
[and dirty] fixes" (which btw. often cause more trouble than they solve).
Have there been any "error" reports regarding this? Changing the
permissions of ''all'' files in a couple of spkgs wasn't necessary either.
Note that "ordinary" users won't be able to install / upgrade spkgs
anyway, so the `pkg-config` and `.la` files for example aren't an issue in
a system-wide installation, nor will such users be able to move Sage,
unless they copy the whole installation (in which case they're likely to
gain ownership, or at least can change the owner, such that they'll have
write permission on the copied installation).
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/11926#comment:22>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
