Re: function defined (or not), more worries on version 10->14 upgrade

On 4/15/22 14:22, Adrian Klaver wrote:
> On 4/15/22 10:14, Tom Lane wrote:
>> Rob Sargent <robjsargent(at)gmail(dot)com> writes:
>>> The function in question is working but I'm having trouble getting at
>>> it's current definition and the possibility that there are more
>>> v10->v14
>>> issues in this database.  The last ERROR: below seems to me "a system
>>> thing"(tm)
>>
>> Hmm, it looks like \df does not cope well with spaces in the argument
>> list:
>>
>> regression=# \df genome_threshold_mono(text,text,double)
>>                         List of functions
>>   Schema | Name | Result data type | Argument data types | Type
>> --------+------+------------------+---------------------+------
>> (0 rows)
>>
>> regression=# \df genome_threshold_mono(text,text, double)
>> ERROR:  invalid regular expression: parentheses () not balanced
>>
>> It's sending the server a bogus pattern in the second case.
>> I've not looked at the code yet, but this does seem like a
>> psql (not server) bug.
>>
>
> Seems it also has another white space problem:
>
> \df upc_check_digit()
>                               List of functions
>  Schema |      Name       | Result data type  |  Argument data types 
> | Type
> --------+-----------------+-------------------+-----------------------+------
>
>  public | upc_check_digit | character varying | upc character varying
> | func
>
> \df upc_check_digit ()
>                        List of functions
>  Schema | Name | Result data type | Argument data types | Type
> --------+------+------------------+---------------------+------
> (0 rows)
>
> \df upc_check_digit(varchar)
>                        List of functions
>  Schema | Name | Result data type | Argument data types | Type
> --------+------+------------------+---------------------+------
> (0 rows)
>
> \df upc_check_digit (varchar)
>                               List of functions
>  Schema |      Name       | Result data type  |  Argument data types 
> | Type
> --------+-----------------+-------------------+-----------------------+------
>
>  public | upc_check_digit | character varying | upc character varying
> | func
> (1 row)
>
>

Perhaps beating a dead horse, but

barnard=# \df genome_threshold_mono(text,text,double precision);
ERROR:  invalid regular expression: parentheses () not balanced
barnard=# \df genome_threshold_mono(text,text,double precision);
ERROR:  invalid regular expression: parentheses () not balanced
barnard=# \df genome_threshold_mono(text,text,float);
                       List of functions
 Schema | Name | Result data type | Argument data types | Type
--------+------+------------------+---------------------+------
(0 rows)

barnard=# \df genome_threshold_mono(text,text,'double precision');
                       List of functions
 Schema | Name | Result data type | Argument data types | Type
--------+------+------------------+---------------------+------
(0 rows)

barnard=# \df genome_threshold_mon*
                                                 List of functions
 Schema |         Name          | Result data type
|                  Argument data types                   | Type
--------+-----------------------+------------------+--------------------------------------------------------+------
 public | genome_threshold_mono | uuid             | pbs_name text,
genome_name text                        | func
 public | genome_threshold_mono | uuid             | pbs_name text,
genome_name text, conf double precision | func
(2 rows)

space after function name

barnard=# \df genome_threshold_mono (text,text,double precision);
                                                 List of functions
 Schema |         Name          | Result data type
|                  Argument data types                   | Type
--------+-----------------------+------------------+--------------------------------------------------------+------
 public | genome_threshold_mono | uuid             | pbs_name text,
genome_name text                        | func
 public | genome_threshold_mono | uuid             | pbs_name text,
genome_name text, conf double precision | func
(2 rows)

\df: extra argument "(text,text,double" ignored
\df: extra argument "precision);" ignored
barnard=# \df genome_threshold_mono (text,text,float);
                                                 List of functions
 Schema |         Name          | Result data type
|                  Argument data types                   | Type
--------+-----------------------+------------------+--------------------------------------------------------+------
 public | genome_threshold_mono | uuid             | pbs_name text,
genome_name text                        | func
 public | genome_threshold_mono | uuid             | pbs_name text,
genome_name text, conf double precision | func
(2 rows)

\df: extra argument "(text,text,float);" ignored

I can get some definitions, eg

barnard=# \sf shoc.pvr(shoc.segment, float)
CREATE OR REPLACE FUNCTION shoc.pvr(seg shoc.segment, plus double
precision DEFAULT 0.0)
 RETURNS double precision
 LANGUAGE sql
AS $function$
  select
((1.0*seg.events_equal)+seg.events_greater+plus)/((1.0*seg.events_less)+seg.events_equal+seg.events_greater+plus)::float
as result;
$function$

but have not found a combination of name/args for
genome_threshold_mono(text,text,double precision)

I'm about to unleash new versions of the above and related functions
(args will change), so a drops are imminent. Any reason to hold off on that?

>>
>>             regards, tom lane
>>
>>
>
>