Howard Chu wrote:
ITS#4467 seems to have gone nowhere. Whether or not any patches have been committed for it isn't obvious because it hasn't been updated since it was submitted. There's no evidence that this ITS has any impact.
I'm (very slowly) working at it. I also believe it has no impact right now, it's mostly a matter of good programming. I'll commit what I've done so far (mainly, a lutil_snprintf() that should be used consistently throughout the code instead of snprintf(), and sprintf()!). But right now it's not used in many places, yet. That's why I didn't commit anything yet.
Either it's been fixed, or it's not a problem.
#4591 seems to have aged a bit too.
I haven't addressed this yet mainly because I've no clue about how to proceed. Since back-meta potentially collects several search responses, and only in some cases aborts when one is received, it might need to merge several response controls, and this could be tricky because they might be incompatible or so. I'd let the ITS there to note this design issue in back-meta. If distributed operations were in place, probably there would be a means to indicate that multiple sets of controls were responded, and propagate this information to distributed operation-aware clients, but the issue of returning something useful without assuming too much on the capabilities of clients would remain.
p.
Ing. Pierangelo Masarati OpenLDAP Core Team
SysNet s.r.l. via Dossi, 8 - 27100 Pavia - ITALIA http://www.sys-net.it --------------------------------------- Office: +39 02 23998309 Mobile: +39 333 4963172 Email: pierangelo.masarati@sys-net.it ---------------------------------------