Does anyone else wish for a language construct like "define struct" which allowed one to define something like a buffer, but which was not associated with a table or temp-table. One should be able to do buffer-copy to and from buffers for tables, but I am thinking that there are lots of times when one has some collection of variables which is record-like, but where one knows that one is only ever going to have one instance. So, one really doesn't want to define a temp-table, but defining N indepedent variables isn't the right thing either.
My inclination is to simply overload define buffer to provide this functionality so that one can use all of the buffer-copy and similar buffer oriented functions.
Yes, that could be useful, since either packaging grouped variables within a TEMP-TABLE/WORKFILE or creating a new class to hold them produces too much overhead in ABL currently.
Agreed.
Any news on the ERS, by the way?
Who me?
Anybody from PSC reading this?
Didn't they have some security issues or something?
A leading technology provider should be able to resolve that faster.
Besides, wasn't the security issue a problem with the single sign-on, i.e., the same system we use to sign on to PSDN ... that is, after all, each time I go to log in it tells me the authentication has failed before I even supply the user name and password. I don't believe there was a security issue specific to ERS ... but I suppose this should go on your ERS thread instead!
True... maybe a switch to a more agile Web development language would help them get the site coded faster?
Sorry Thomas, just trying to get every possible bit of attention.
Experience suggests that PSDN posts are not a particularly good way to do that. Works sometimes, e.g., when there is a specific developer who follows the forum and gets involved, but I wouldn't count on the PSDN forum messages to reach administors or supervisors.