Hi. * hans <hans@beehive.de> [020314 10:40]:
Christian Theune wrote:
Just for the record (can't repeat it too often), used the excellent http://www.zope.org/Members/spinwing/ZSQL_Results i have read also the source files (Shared.DC.ZRDB.Results and related), but it turns out to be obfuscated python (a *real* programmer can program assembler in any language) like a class r() w empty body in the middle of statements and one-lettered identifiers. Whats the use of open source if its hard to understand?
Actually this is a part of zope i really would like to see beeing cleaned up, as i fixed already 3 issues on sqltest, 1 outstanding and most time i don't see a real "strategy" someone had writing this code. And almost all of the sql code looks like this (hundreds of single letter identifiers, about 0 doc strings per class) and making patches over patches on them doesn't make them easier to read, and also not more stable. Hope this will get fixed at least in zope 3 ... :( COMMENT FOR ZC PPL: sqltest.py is f*$@d up very ugly. collector #6 is one proof of that!!!
FACT IS: result objects can't be pickled which implies can't be made an attribute of a persistent object and live through several transactions. The dictionaries() result however can (and thats all i needed).
cool idea ... let me think how one could abuse that ... :)
(hey, what is a zbrain and what is it good for?) dunno, i never cared for that and it simply doesn't disturb you. :)
Well, part of UNIX understanding is reading M.J.Bach's The Design of the UNIX Operating System, and i'd really would appreciate something similar for Zope.
cheers hans
Greetings Christian -- Christian Theune - ct@gocept.com gocept gmbh & co.kg - schalaunische strasse 6 - 06366 koethen/anhalt tel.+49 3496 3099112 - fax.+49 3496 3099118 mob. - 0178 48 33 981 reduce(lambda x,y:x+y,[chr(ord(x)^42) for x in 'zS^BED\nX_FOY\x0b'])