Re: Installation of regress.so? - Mailing list pgsql-hackers

Hi,

On 2021-05-20 09:16:50 -0400, Andrew Dunstan wrote:
> We certainly shouldn't want that.  But we do need it for the target
> unless we wipe out everything in the source that refers to it.

Is there a reason not to go for the wipe? I don't think the type of
functions we have in regress.so are necessarily ones we'd even expect to
work in the next version?

Here's references to explicit files I see after an installcheck:

SELECT oid::regproc, prosrc, probin FROM pg_proc WHERE probin IS NOT NULL AND probin NOT LIKE '$libdir%';

┌───────────────────────────┬───────────────────────────┬────────────────────────────────────────────────────────────────────────────┐
│            oid            │          prosrc           │                                   probin
            │
 

├───────────────────────────┼───────────────────────────┼────────────────────────────────────────────────────────────────────────────┤
│ check_primary_key         │ check_primary_key         │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/refint.so │
 
│ check_foreign_key         │ check_foreign_key         │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/refint.so │
 
│ autoinc                   │ autoinc                   │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/autoinc.so│
 
│ trigger_return_old        │ trigger_return_old        │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ ttdummy                   │ ttdummy                   │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ set_ttdummy               │ set_ttdummy               │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ make_tuple_indirect       │ make_tuple_indirect       │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ test_atomic_ops           │ test_atomic_ops           │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ test_fdw_handler          │ test_fdw_handler          │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ test_support_func         │ test_support_func         │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ test_opclass_options_func │ test_opclass_options_func │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ test_enc_conversion       │ test_enc_conversion       │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ binary_coercible          │ binary_coercible          │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ widget_in                 │ widget_in                 │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ widget_out                │ widget_out                │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ int44in                   │ int44in                   │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ int44out                  │ int44out                  │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ pt_in_widget              │ pt_in_widget              │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ overpaid                  │ overpaid                  │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ interpt_pp                │ interpt_pp                │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 
│ reverse_name              │ reverse_name              │
/home/andres/build/postgres/dev-optimize/vpath/src/test/regress/regress.so│
 

└───────────────────────────┴───────────────────────────┴────────────────────────────────────────────────────────────────────────────┘
(21 rows)

Testing the pg_upgrade path for these doesn't seem to add meaningful
coverage, and several seem likely to cause problems across versions?

Greetings,

Andres Freund