Hi Alan,
> -----Original Message-----
> From: Alan W. Irwin [mailto:ir...@beluga.phys.uvic.ca]
> Sent: Tuesday, January 27, 2015 2:17 AM
> To: Arjen Markus; Phil Rosenberg; Jim Dishaw; PLplot development list
> Subject: Release plans
>
> To Arjen, Phil, and Jim:
>
> I am addressing this post mostly to you guys because you are the only ones I
> am
> aware of that are still working on substantial (i.e., non
> bug-fix) changes for this release cycle.
>
> As release manager, I really appreciate how hard all of you are working on
> improving
> PLplot in the areas of the Fortran binding, wxwidgets device driver, and
> plmeta/plbuf.
> And hopefully most or all of your substantial changes will get into the
> forthcoming
> release (scheduled for February 28th).
>
...
> The criteria you should consider about substantial (i.e, non bug-fix) master
> branch
> changes at any time but especially this close to release are the following:
>
> 1. Is your change thoroughly tested on Windows and Unix (typically
> Linux) with no build issues and no segfaults at run time?
>
> I think it is fair to say from the topic-branch commits I have seen via git
> format-patch
> (or other means), that the current state of the fortran binding and wxwidgets
> rewrites
> qualify on the build question but are disqualified by the run-time part of
> this criterion,
> but hopefully that will change soon in both cases.
>
The status at the moment (on my development machine) is that there are one or
two routines not behaving properly and some ten routines that still need to be
revised.
When that is finished, I will have eliminated the complicated ballet performed
in the plstubs.h header file and moreover the rather awkward way we pass
strings at the moment between Fortran and C. I will not have eliminated all the
pointer arithmetic on the C side. That is better left for a further cleanup -
if we want that at all.
> 2. Is it better than what it is replacing? I think it is fair to say from
> the topic-branch
> commits I have seen the current Fortran binding and wxwidgets rewrites are
> partially
> disqualified by this criterion, but hopefully that will change soon in both
> cases.
>
> Users of the Fortran binding rewrite should have an easier time of it (less
> worries
> about typing of variables) then the current Fortran binding. However, as I
> have told
> Arjen elsewhere, one of the big motivations for this rewrite is to make the
> Fortran
> bindings implementation much simpler. That goal has not been realized yet,
> and I
> hope to see that change soon.
The bindings will allow the use of single and double precision data from within
the same library, which is indeed a big advantage for the user. The bindings
require more code and in that sense it is not an improvement. On the other hand
we will be using a standard way to deal with the interfacing and that is an
improvement over the current compiler-dependent way.
>
> My impression is Phil has already achieved one of his desired goals which is a
> greatly simplified and easier to maintain wxwidgets device driver. Of
> course, the
> current rendering issues are of concern, but from what Phil has said, he is
> actively
> working on those so the prospects of meeting this criterion soon look fairly
> good.
>
> I don't know the status of the plmeta/plbuf changes with regard to these two
> criteria,
> but I strongly encourage Jim to make his topic branch commits available to the
> plplot-devel list via git format-patch so they can be evaluated by these
> criteria by
> everybody.
>
> <aside> And similarly for Arjen (who has been communicating with me by tarball
> because he has been having difficulty using git format-patch on Windows). Can
> someone here give Arjen advice about a good command-line git client to use on
> Windows where the format-patch command "just works"? </aside>
>
I probably have found it in the form of a Cygwin package for dealing with git.
I have not tried it yet, but I have a feeling it is more to my liking than the
one that I had which used the PowerShell.
>
> If you cannot meet that deadline, then I suggest you give your highest
> priority to
> finishing the small changes (such as Jim's plmem.c
> changes) that can be committed to master before that deadline.
Given that the changes to the Fortran bindings are huge (though I have tested
them on two platforms already - Cygwin and bare Windows), but also the fact
that the new geographic mapping API is waiting, I think I will postpone work on
the Fortran bindings for the moment and continue with these much more limited
changes.
Regards,
Arjen
DISCLAIMER: This message is intended exclusively for the addressee(s) and may
contain confidential and privileged information. If you are not the intended
recipient please notify the sender immediately and destroy this message.
Unauthorized use, disclosure or copying of this message is strictly prohibited.
The foundation 'Stichting Deltares', which has its seat at Delft, The
Netherlands, Commercial Registration Number 41146461, is not liable in any way
whatsoever for consequences and/or damages resulting from the improper,
incomplete and untimely dispatch, receipt and/or content of this e-mail.
------------------------------------------------------------------------------
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Plplot-devel mailing list
Plplot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot-devel
