On Wed, Oct 28, 2009 at 2:06 PM, Waldek Hebisch wrote:
>
> AFAICS the patch below is exactly the patch that I commited in
> revision 709.
I am sorry. You are absolutely right. I was dealing with the wrong
base. :-( Mystery solved: user error.
>>
>> sumOrParen(x) ==
>> @@ -724,7 +723,8 @@
>> null argl => '".."
>> (null rest argl) or (null first rest argl) =>
>> concat(first argl, '"..")
>> - concat(first argl, concat('"..", first rest argl))
>> +-- concat(first argl, concat('"..", first rest argl))
>> + concat("_(",concat(concat(first argl, concat('"..", first rest
>> argl)),"_)"))
>> concat(app2StringWrap(formWrapId op, linkInfo) ,
>> concat("_(",concat(tuple2String argl,"_)")))
>>
>> --
>>
>
> There is no need to escape parenthesis: "(" will work too. I suspect
> that single argument case also needs parenthesis.
>
Ok, I just copied that coding from some other part of format.boot.
You are also right about the single argument case:
(42) -> ex := (1..)::INFORM
(42) (($elt (UniversalSegment (Integer)) SEGMENT) 1)
Type: InputForm
(43) -> unparse %
(43) "SEGMENT(1)$UniversalSegment(Integer)"
Type: String
>>
>> We have been discussing quite a few patches here lately and might be
>> in danger of loosing track of some of them. I wish we had a designated
>> "patch co-coordinator" ... :-) Maybe it would be a good idea if I at
>> least took the time to add them to the axiom-wiki issue tracker, eh?
>> Then we could get back to using that to co-ordinate the bug and patch
>> queues.
>>
>
> Well, first I prefer to have good patches committed quickly instead
> of sitting in the patch queue. Second, it would be good if person
> working on a patch explicitely said that he/she supends it. For
> example, you Bill had patch adding 'Comparable' to some domains.
> The patch looks good but I do not know if you hit some problems
> during testing and are busy solving them or you are occupied with
> something else...
>
Time is always the constraint - never enough of it. So I think we need
something to help keep track of things in spite of that.
About Comparable: Yes, it was important for some things I was working
on at the time and I did commit at least part of these changes. But
they were also became part of something more ambitious involving
changing Void to a domain in SetCategory which did not work out so
well and I have set aside for now. I guess I still do not really know
how to properly organize and discipline my "play" so that it results
in some things that can be easily salvaged. I am inclined to blame the
tools such as limitations of svn but really it just comes down again
to a lack of time and excess of ambition. BTW, maybe I will start
using darcs again locally as a way of more easily staging patches.
> Concerning issue tracker on Axiom Wiki: I afraid it become
> unmanageble mess. One thing is that 'fixed somewhere' effectively
> obscures information.
I have to agree completely about the status "fixed somewhere". I think
that was definitely a bad idea. On the other hand a wiki is supposed
to be an interactive tool managed by a community of users and we have
not really had much of that.
> Second, separate problems are frequently mixed in a single issue.
This is easily solved by editing.
> Third it seems that we have non-trivial amount of change which makes
> no sense (either author make a mistake or possible the change is due
> to vandals). And issue tracker have technical problems: in Lynx still
> display is broken (first code segment looks OK, but the next is
> scrambled), in Firefox cut and paste is problematic.
>
I cannot imagine trying to create web sites that are fully compatible
with a text-only browser like lynx any more but I suppose that with
some effort it might be possible. I have no idea how many people might
be trying to use lynx at the axiom-wiki site.
I have no problem at all with cut-and-paste to/from Firefox on either
Linux or Windows and I do use it quite extensively. What sort of
problems do you see?
> AFAICS SourceForge bugzilla solves most of problems that
> issue tracker have. So I would encurage use of SourceForge
> for this purpose. They also offer separate trackes, it make
> sense to use them for non-bug related issues (we would have
> to activate them first).
>
Is this the same as what OpenAxiom uses. Personally I find the email
from that system very confusing. And of course there is no way to
easily illustrate problems. But if you insist then I might be able to
bring myself to start to use it. :-(
> Of course we could try to fix technical problems with issue
> tracker or setup something like Trac. But it seems that
> currently we are busy enough working on FriCAS itself and
> it seems that simply we have no time to setup something
> better than SourceForge facilities.
I do still occasionally make changes in the axiom-wiki code and would
be more motivated to do so if I received comments about it. But you
are right about not having enough time to do everything we want to do.
Regards,
Bill Page.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To post to this group, send email to fricas-devel@googlegroups.com
To unsubscribe from this group, send email to
fricas-devel+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/fricas-devel?hl=en
-~----------~----~----~----~------~----~------~--~---
