Re: PSQL suggested enhancement

On 2005-10-19 23:52, Michael Glaesemann wrote:
>
> On Oct 20, 2005, at 15:45 , Roger Hand wrote:
>
>>
>>> On Oct 20, 2005, at 14:50 , Dean Gibson (DB Administrator) wrote:
>>>
>>>> PSQL has the option to output the result of queries in several
>>>> different formats, including HTML.  Suggestion:  have an option to
>>>> output query results in XML format.
>>>>
>>>
>>> My personal opinion on this is that there are a lot of different
>>> ideas about how the XML should/could be written, and a the current
>>> output can be piped to a script in <insert favorite scripting
>>> language here> to format to match anyone's wish. Rather than have
>>> psql decree the one and true format or include so many as to make
>>> maintainability burdensome, further formatting is best left to the
>>> end user.
>>>
>>
>> By that reasoning there shouldn't be html format either.
>
> Agreed :) I'd submit a patch to remove it if I thought it'd be
> accepted--and more importantly, if I thought it were worth the time
> it'd take me (I'm still learning C, and the little bit of PostgreSQL
> code I'm familiar with is not the psql client) (excuses excuses :).
> Though justification for removing a feature is different from that of
> adding or extending one: removing one breaks backwards compatibility.
>
> Michael Glaesemann
> grzm myrealbox com

So, which other features THAT YOU DON"T PERSONALLY WANT OR SEE THE NEED
FOR, would you remove from PostgreSQL ???  Do you personally not use the
expanded, or unaligned outputs ???  By all means rip them out !!!
(sarcasm off)

While not every suggested feature needs to be in software, the idea that
you'd remove a useful feature that someone else found valuable enough to
spend the time coding, testing, etc, is anathema to me.  Remind me to
NEVER let you touch the open-source projects I control.  I'm looking for
contributors, not hackers or saboteurs.  Not to mention people that are
open to new ideas.

Second, surrounding field values with XML tags having the name of the
field, would be BY FAR the most natural way of representing the data.
The only "variable" in what I suggested was the name of the row tag,
which could either be a formatting parameter, or (my preference) easily
changed by the user in the script which added the surrounding XML (which
I left off because THAT PORTION is highly variable).

I just find it surprising that XML is not one of the formats provided,
considering that XML is considered a data interchange format (much more
than HTML, which is a representation format).

-- Dean