On 08/30/2011 03:26 PM, C. Michael Pilato wrote:
> On 08/30/2011 10:11 AM, Daniel Shahaf wrote:
>> C. Michael Pilato wrote on Tue, Aug 30, 2011 at 10:01:26 -0400:
>>> Would it make more sense to just replace the warning at the end with a
>>> single instance of each of the various mid-stream warnings? Or are all such
>>> warnings not quite as redundant as this one?
>>
>> But then people have to finish the dump (successfully! ie, without
>> hitting any corruption or other error partway through) before they see
>> that "FYI, the data you've just dumped isn't self-contained" warning.
>>
>> Reporting error conditions as early as possible seems preferable.
>
> Yep, makes perfect sense. +1 to move forward per your plan.
Actually, I'll make one more suggestion: change the singularly printed
warning message to include a phrase such as "(Future notifications of this
same warning will be suppressed.)" Just so seasoned Subversionites (used to
seeing tons of these messages) don't make false assumptions about how
prevalent the problem is/isn't now.
--
C. Michael Pilato <cmpilato_at_collab.net>
CollabNet <> www.collab.net <> Distributed Development On Demand
Received on 2011-08-30 21:30:40 CEST