Hi Vladimir,
Thanks for your comments.
I'll try to answer few of your questions inline below, where you've
referred me within your mail within this mail thread.
On Thu, Oct 19, 2023 at 12:35 PM Vladimir Sitnikov
<sitnikov.vladi...@gmail.com> wrote:
> I would suggest deleting xslt3 branch from xalan-java repository and using a
> single long-lived PR instead.
> It would allow us to monitor progress, discuss the changes, and unsubscribe
> from all xslt3 notifications at once.
The XalanJ dev repos xalan-j_xslt3.0 branch (from XalanJ dev repos
'master' branch) was created many months ago (I guess during april,
may 2023), after few XalanJ committers and pmc members agreed on that.
I remember Gary Gregory (XalanJ's pmc chair) and Joseph Kesselman (at
least Joseph did not object to this. Joseph is currently XalanJ's pmc
member) supporting for that.
At this moment, XalanJ dev repos xalan-j_xslt3.0 branch has lot of new
codebase available within this branch, related to XSLT 3.0 and XPath
3.1 implementations, and we pass large number of XSLT 3.0 and and
XPath 3.1 language test cases. It seems to me that, its probably good
to keep making further progress on XalanJ's XSLT 3.0 and XPath 3.1
implementations, as much as we can.
I'm personally a member of XalanJ mailing lists since more than a
decade I guess, and I've seen various people on these lists (including
me), desiring an XSLT 3.0 implementation from XalanJ.
IHMO I think, the current codebase development work on XalanJ dev
repos xalan-j_xslt3.0 branch, is a result of these wishes and
requirements of XalanJ users.
The current XalanJ's XSLT 3.0 and XPath 3.1 implementations on dev
repos xalan-j_xslt3.0 branch, essentially builds completely over
XalanJ's current XSLT 1.0 processor design and codebase.
I've read earlier on this list (and not long back), Joseph Kesselman
suggesting ideas like he could help introduce within XalanJ concepts
like XCI apis and (XML Schema) type aware XalanJ DTM. If Joseph is
able to help, get these technical concepts implemented within XalanJ
(I believe, the XSLT 3.0 implementation) that'll be great.
> mukulga opens a PR and merges it within a couple of seconds.
> If a review is expected from someone, then the PR should be open at least for
> several days.
Personally, I currently don't seek reviews of my XalanJ XSLT 3.0 and
XPath 3.1 codebase contributions (on XalanJ dev repos branch
xalan-j_xslt3.0), whenever I make git PRs and myself within few mins
merge my PRs to the XalanJ dev repos branch xalan-j_xslt3.0. Since
IMHO, I'm currently a XalanJ committer, and XalanJ's XSLT 3.0 dev work
on branch xalan-j_xslt3.0 has an approval from current XalanJ PMC, I
feel myself 'empowered' to do these XalanJ implementation development
activities.
But when I've committed any of my codebase changes to XalanJ dev repos
xalan-j_xslt3.0, I desire constructive comments from members of this
list to solve any defects or to improve anything within the code that
I've committed.
> If a review is not expected, then, well, just commit to the branch and be
> done with it.
I think, this is a nice idea. I'll think about this, and shall act
accordingly (until I've made up my mind on this, I'll continue working
as before). I also seek Gary's and Joseph's feedback on this point
that you've made.
Having written, all of the above mentioned points about XalanJ's XSLT
3.0 and XPath 3.1 current dev implementations, I wish to say that, I
like very much (and consider useful as well) XSLT 1.0 and XalanJ's
implementation of that.
--
Regards,
Mukul Gandhi
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@xalan.apache.org
For additional commands, e-mail: dev-h...@xalan.apache.org
