Re: treelang patch parts n of 6

[Tim Josling <tej at melbpc dot org dot au>]

I was given approval to maintain treelang by the steering committee, but you
weren't to know that.

I should have updated MAINTAINERS to add myself (see extract below), but I
forgot to do so, mainly because this was such a protracted process (over 1
year, including ports to gcc3.1 and 3.2, two almost total rewrites and one
line by line reformat!).

I will fix MAINTAINERS shortly once I reverse engineer the format.

If I was ever going to be discouraged I would have been a long long time ago.
I made major changes to the code and doc based on suggestions which came out
of previously submitted versions.

I have committed to maintain this for at least one year, and I will address
any /specific/ comments. An example of a specific comment would be 'you are
not using the new garbage collection', or even 'please rewrite the manual
using X as a base'.

I am also fixing the copyrights as we speak (as far as I can tell these
relate to the test suite only). One file gcc/testsuite/lib/treelang.exp has a
long copyright provenance and hardly any content left any more. I will leave
that as it is. Probably all the old content has gone but I don't think it is
worth checking that it is so.

I have hoarded the other comments and will progressively address them. With
the manual I will add a disclaimer at the top that this is not a good example
as a manual and please seek guidance on this... this should limit the damage
until I fix it up properly.

If anyone feels the code itself is a bad example please provide me with
specifics (if you have not already done so) and I will work on that. The only
exception is the issue of the single union vs multiple structs for language
tree constructs, as previously discussed. I will accept a patch that changes
it to one union, but I don't have time to fix it myself.

Tim Josling

"David S. Miller" wrote:

> Actually, you aren't even listed in MAINTAINERS as "write after
> approval" nor anywhere else in that file, so you shouldn't be checking
> anything at all into the tree.

"Geoffrey Keating" wrote:

> This is why we require that patches are reviewed before they can be
> contributed and given positive approval.  It's quite possible that
> no-one ever looked at those parts of the patches you posted.

> --
> - Geoffrey Keating <geoffk@geoffk.org>

>From "Jerry" a while back:

> 1c. You update the MAINTAINERS of GCC accordingly, also testing your
 >    CVS write access...
> Cheers,
> Gerald
> --
> Gerald "Jerry" pfeifer@dbai.tuwien.ac.at
http://www.dbai.tuwien.ac.at/~pfeifer/