[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Axiom-developer] Re: Axiom: fixed and fixed proposed bit vectors

From: William Sit
Subject: Re: [Axiom-developer] Re: Axiom: fixed and fixed proposed bit vectors
Date: Sun, 27 Jan 2008 01:28:52 -0500

I consider "fixed xxx" and "fixed proposed" as two statuses, not 16 (but that is up to your interpretation what counts as a status). One reason for proposing this "cryptic" version is to enable automatic sorting. One can write a program to search "fixed" and "fixed proposed" and then parse xxx to filter out those that are fixed or not fixed, or proposed or not proposed for any of the flavors. This program will allow users to do searches and be presented only with a particular xxx view.

As long as the bit vector is documented, it is no more cryptic than other code. A bit vector is the most efficient of data representation. Since each axiom has its own bit, it is easy to even spot the status visually.

If you use comments after "fixed" and "fixed proposed", someone will have to read the comments manually and sort the bug reports manually. It would be better to know ahead of reading the comments what one is about to read.

In the best scenario, the code xxx would only go through 4 values (starting with 000 and increasing by changing a 0 to a 1, at most three times).


On Sat, 26 Jan 2008 08:54:55 -0500
 "Bill Page" <address@hidden> wrote:
I must say that I do not like the sound of "fixed somewhere" - it sounds too vague - but I understand the intent. Maybe it would be
better to write:

  "fixed as noted ..."

And then expect to find a comment saying in which versions/forks the fix occurs. This is almost like you are doing now except for the distinction between "as noted" and "proposed" (patch attached to the
bug report).

I don't think William intended 7 different statuses. The parameter can be included as a comment. But I do not see any point in being so

Bill Page.

On 26 Jan 2008 09:47:43 +0100, Martin Rubey wrote:
Martin Rubey <address@hidden> writes:

> Yes, I like this proposal equally well.  Comments?

Oops, sorry, I don't understand it: it would imply having 7 additional statuses too. I don't think it's possible to have
a "variable" status?

The point is being able to filter out all issues that are
'fixed somewhere.'

> "William Sit" <address@hidden> writes:
> > Dear Martin:
> >
> > May I suggest using a bit vector to notate the fixes? I propose > > that the bits of the bit vector be assigned according to chronological > > order of axiom branches, so that bit 0 is for Tim's Axiom, bit 1 for > > FriCAS and bit 2 for OpenAxiom, etc. So we need only two status: > > fixed xxx and fixed proposed xxx for now. The bit vector is extensible
> > if more flavors come along.
> >
> > Example: "fixed 101" means patches exist and implemented for > > OpenAxiom and for Tim's Axiom, each patch according to the
> > requirements of the flavor.
> >
> > "fixed proposed 010" means a patch exist (proposed) but not
> > implemented (for whatever reasons) on FriCAS.
> >
> > "fixed 000" means no patch implemented
> >
> > "fixed proposed 000" means no patch proposed (implies of
> > course "fixed 000").
> >
> > You may forward this to the groups or disregard as you
> > please.
> >
> > William
> >
> >
> > On 25 Jan 2008 22:11:55 +0100
> >   Martin Rubey <address@hidden> wrote:
> > >I have just added a new bug status "fixed somewhere".
> > >
> > > Please use "fix proposed" and "fixed somewhere" according
> > > to the following rules:
> > >
> > >> 1) fix proposed
> > >> would be used if a patch is attached, but it hasn't been
> > >> fixed in one of the systems
> > >> 2) fixed somewhere would be used if the issue is fixed > > >> somewhere, regardless whether a patch is included or
> > >> not.
> > >> ...

Axiom-developer mailing list

William Sit, Professor of Mathematics, City College of New York Office: R6/202C Tel: 212-650-5179, Fax: 212-862-0004
Home Page:

reply via email to

[Prev in Thread] Current Thread [Next in Thread]