Bruce Momjian <pgman@candle.pha.pa.us> writes:
> Tom Lane wrote:
>> Which we do not have, because pg_dump doesn't use CSV. I do not think
>> this is a must-fix, especially not if the proposed fix introduces
>> inconsistencies elsewhere.
> Sure, pg_dump doesn't use it but COPY should be able to load anything it
> output.
I'd buy into that proposition if CSV showed any evidence of being a
sanely defined format, but it shows every indication of being neither
well-defined, nor self-consistent, nor even particularly portable.
I suggest adjusting your expectations. All I expect from that code is
being able to load the majority of data from the more popular Microsloth
applications. Trying to achieve 100% consistency for corner cases is
just going to interfere with the real use-case for the feature, which is
coping with output from applications that aren't very consistent in the
first place.
regards, tom lane
