Re: pageinspect: add tuple_data_record()

Hi,

On 2018-10-16 21:39:02 -0400, James Coleman wrote:
> Background:
> In my usage of pageinspect for debugging or other purposes I often find it
> frustrating that I don't have a way to easily view a heap page's tuple data
> as actual records rather than binary data. After encountering this again
> last week while doing some data recovery after an accidental delete (and
> manually parsing a few int and timestamp fields with get_byte magic) I
> decided to write up a patch to add a function to handle this case.

> Summary:
> The new function tuple_data_record() parallels the existing
> tuple_data_split() function, but instead of returning a bytea array of raw
> attribute heap values, it returns a row type of the relation being examined.

I don't think it's entirely safe to do so for invisible rows.  The toast
references could be reused, constraints be violated, etc.  So while I
could have used this several times before, it's also very likely a good
way to cause trouble.  It'd probably be ok to just fetch the binary
value of the columns, but detoasting sure ain't ok.


> This is my first patch submission, so I encountered a few questions that
> the coding style docs didn't seem to address, like whether it's preferable
> to line up variable declarations by name.

There's ./src/tools/pgindent/pgindent (which needs an external dep), to
do that kind of thing...

Greetings,

Andres Freund