directory-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefan Seelmann <>
Subject Re: [Studio] Referring to Snapshot version of the API/ApacheDS + Some other things
Date Mon, 13 Mar 2017 20:43:11 GMT
On 03/13/2017 10:53 AM, Emmanuel L├ęcharny wrote:
> Hi,
> Stefan, for the record, I changed the pom files to point to SNAPSHOT of
> the API and ApacheDS libraries. I don't know if it's a problem per se,
> but it allowed me to use some of the changes I have introduced this
> week-end in the API. If needed, I can start to cut a release of thos lib
> and move back to a released version soon.

No need, we can work with the snapshot versions for a while. Only thing
to consider is that Tycho caches the snapshot version in
~/.m2/repository/.p2, so if you do furhter changes in the API make sure
to clean that cache.

> Otherwise, I have applied the workaround in the entry editor : you can
> now edit a Boolean value and the editor does not freeze on mac. This is
> a bit of a hack, as I'm injecting a 'TRUE' value as a default value. It
> works, and teh value is not really applied in the entry, until you
> validate it (it appears in red in the editor).
> This workaround is not perfect, and I discovered another issue : you
> can't inject a null value in, say, a 'dc' or a 'email' - which allows
> such null values -. It won't be sent to the server. I don't see easy way
> to deal with that at the moment, without having to refactor extensively
> the editor. I was expecting to be able to use teh LdapSyntax to verify
> if a null value is acceptable, but we can't, as we have no guarantee to
> have access to this LdapSyntax for every LdapServer we are connected to.
> So bottom lie, teh only way is to assume we can have a null value,
> unless we *know* we can't (like, if the LdapSyntax is available). We
> could also define a default list of AT that accept null values...
> Anyway, this is not that frequent, and we can live with such a
> limitation, as soon as we know it exists...
> Just let me know if you have some toughts about this problem.

I don't have a better idea currently, so let's keep it. Thanks for
handling it.

Kind Regards,

View raw message