| From: | Fabien COELHO <coelho(at)cri(dot)ensmp(dot)fr> | 
|---|---|
| To: | David Fetter <david(at)fetter(dot)org> | 
| Cc: | PostgreSQL Development <pgsql-hackers(at)postgresql(dot)org> | 
| Subject: | Re: Make foo=null a warning by default. | 
| Date: | 2018-07-18 12:25:46 | 
| Message-ID: | alpine.DEB.2.21.1807180747040.11604@lancre | 
| Views: | Whole Thread | Raw Message | Download mbox | Resend email | 
| Thread: | |
| Lists: | pgsql-hackers | 
Hello David,
> I assure you that you expression yourself in English a good deal
> better than I do in Portuguese.
Alas, despite a Portuguese "rabbit" name, I cannot speak the language 
which got lost between generations.
About this v3: Patch applies, compiles, "make check" ok.
A few minor comments:
Variable "need_transform_null_equals" may be better named 
"is_null_equals", because there is no "need" of the transformation as 
such, and the expression just checks for the pattern, really.
I'm fine with the off/warn/error/on order.
Doc could mention that the transformation allows compatibility with other 
products, without naming them? Or not.
In doc, the list of valid values on a long line, where the practice seems 
to wrap around after about 80 columns in the XML file.
I notice that the feature was not tested at all before this patch:-(
Maybe there could be one test which results to true, which is the whole 
point of the transformation? eg "SELECT NULL = NULL".
-- 
Fabien.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Robert Haas | 2018-07-18 12:53:42 | Re: Libpq support to connect to standby server as priority | 
| Previous Message | Heikki Linnakangas | 2018-07-18 12:02:05 | Re: GiST VACUUM |