Peter Rolf wrote:

Hi Hans,

can you implement *ordered* output with 'lpdf.dictionary'? There are two reasons why I ask.

- My md5 hashing mechanism depends on the order of the entries. I think its much easier to implement an ordered output than to create a hashing mechanism, that checks for equal content. I also doubt, that there is a faster alternative for this job than md5. - Also debugging is a pain with longer dictionaries (I have to use a check list).

as the dictionary is a hash, ordering would imply way more overhead than i'm willing to accept so why not use the serialized stream as hash, ie. just d() given that d is a dictionary; there is no gain in an md5 i might provide a 'shared object' mechanism in the lpdf code (i considered it for some mechanisms) Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------