[Monetdb-developers] [Monetdb-sql-checkins] sql/NT configure.py, 1.51, 1.52 rules.msc, 1.37, 1.38
Stefan.Manegold at cwi.nl
Sun Dec 31 16:41:13 CET 2006
On Sun, Dec 31, 2006 at 03:56:04PM +0100, Martin Kersten wrote:
> Stefan Manegold wrote:
> > Update of /cvsroot/monetdb/sql/NT
> > In directory sc8-pr-cvs7.sourceforge.net:/tmp/cvs-serv13165/NT
> > Modified Files:
> > configure.py rules.msc
> > Log Message:
> > cleaning-up & preparing the "monet5" to "MonetDB5" move:
> > - renamed all remaining MONET5_* (enironnment-) variables to MONETDB5_*
> > - renamed all remaining *_monet5 aliases to *_MonetDB5
> > In case you use any MONET5_* (enironnment-) variables or
> > *_monet5 aliases in your own scripts,
> > you need to rename them to MONETDB5_* respectively *_MonetDB5,
> > too.
> Oops, it seems you have been out of the loop or overwhelmed with
> emails regarding the BIG BANG cleanup.
either or both, apparently ...
> I already mentioned to sjoerd/niels/fabian/... that the naming
> of packages should be given some thought and discussed.
> - move the package names to lower-case instead of a mixture of upper/lower
> - suggested packages names (if feasible on SF), buildtools,gdkcommon, clients,
> monet4, monet5, sql, and pathfinder.
> This propably won't affect your Environment variables.
It will and should:
For simplicity of use, the variable names are (or at least should be)
(easily) derived from (and hence related to) the package names.
(and btw, the environment variable are not "mine", they are inherent and
important in the code ;-))
With ever more packages to handle, they should be aligned as much as
possible, and similar things should be kept similar across packages, e.g.,
variable names and configure options should follow a (simple!) canonical
schema, so that they can (1) easily be guessed, and (2) maintained (grep,
find & replace, etc.)
However, my changes (IMHO) don't cause any conflict wrt. changes of package
names -- they rather help to make such changes simpler in the future:
all I did was to clean-up a mixture of MONET5_* & MONETDB5_* (and *_monet5 &
*_MonetDB5) that (at least for me) did not follow any recognizable rules any
more --- my changes simply unify all MONET5_* & MONETDB5_* into MONETDB5_*
(and all *_monet5 & *_MonetDB5 into *_MonetDB5).
Thus, a potential later renaming of MONETDB5/MonetDB5 to MONET5/monet5
(and likewise for 4) should now/then be possilbe with a simple find/replace,
not requiring another week (of holidays) to be spent on this.
To avoid further confusion and or extra work, I strongly recommend to plan a
new renameing carefully, test it properly before checkin and check it in
atomically in one go (--- possibly not in a rush just before Xmas ;-)).
Moreover, I *personally* prefer MonetDB4/5 (possible in lower case if
"necessary") over monet4/5 --- but that's just one vote out of many ...
ps: I guess, I'm done and off for this year. --- Guten Rutsch!
| Dr. Stefan Manegold | mailto:Stefan.Manegold at cwi.nl |
| CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ |
| 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 |
| The Netherlands | Fax : +31 (20) 592-4312 |
More information about the developers-list