Angus Robertson - Magenta Systems Ltd wrote:
> The error handling issue is not only FtpCli, but also HttpProt. I was
> having an SSL issue with URL checker, always got return code 0, which
> I eventually discovered was because the SSL DLLs were not in the
> path.
Even this does not guarantee th
> I would also like to remind you (or Arno or Francois) of a planned
> change in error handling in FTPCli which we discussed a month ago.
Not forgotten, just too busy on other projects to spend time on ICS,
except for fire fighting customer issues, in this case an FTP server that
returned very st
> Sorry Anton, you'll need to change your application to use one of the new
> commands, but the change broke too many things.
Angus, no problem: I use HighLevelAsync directly as it much more flexible and
powerful than predefined methods
HighLevelAsync(ftpConnectAsync, [ftpFctOpen, ftpFctAuth, f
> In ICS v7, FTP client performs FEAT command during connect. If
> server does not support FEAT command, connect fails with: 500 FEAT:
This change was made in V7 in January 2010, and has now been reversed so
the FTP client is fully backward compatible.
There are four new methods, ConnectFeat, C
> That apears to works, thanks.
>
> But it's a bit counter intuitive, because Connect returns false.
> Maybe changing Connect to a procedure would be better, because that
> would be a breaking change
I've just rebuilt my own FTP client application for the first time in a
year, and a customer fo
Angus wrote:
>Connect does ignore the failed FEAT response, since it's the last action
>performed after logging onto the server. But you'll need to check
>FtpState to ensure the server connected. Or just do the various methods
>separately. It's debatable whether a failed FEAT command should be
>i
> > If FEAT failed it is the same as if there were no features so IMO
> > it should not trigger an error.
>
> Of course only when called internally in one of the HighLevel
> methods.
That's the hard part, and where making supposedly simply changes to ICS
often becomes very complicated!
Angus
-
Arno Garrels wrote:
> Angus Robertson - Magenta Systems Ltd wrote:
>> It's debatable whether a failed FEAT
>> command should be ignored or reported, but I agree it's now backward
>> compatible with v6 or v7 prior to this year.
>
> If FEAT failed it is the same as if there were no features so IMO
>
Angus Robertson - Magenta Systems Ltd wrote:
> It's debatable whether a failed FEAT
> command should be ignored or reported, but I agree it's now backward
> compatible with v6 or v7 prior to this year.
If FEAT failed it is the same as if there were no features so IMO
it should not trigger an error
> In ICS v7, FTP client performs FEAT command during connect. If
> server does
> not support FEAT command, connect fails with: 500 FEAT: command not
> understood. ICS v7 should either not do FEAT during connect (like
> ICS v6) or ignore its response and proceed.
Connect does ignore the failed FE
10 matches
Mail list logo
