Generate RPG data structure from result metadata

[janfh]

Feature to generate RPG data structure from column metadata.

Added statement qualifier "rpg". E.g:

rpg: select cast('' as varchar(100)) as my_text, current_timestamp as my_timestamp, cast(1.0 as numeric(13, 2)) as my_val from sysibm.sysdummy1;

generates:

**free

// statement: select cast('' as varchar(100)) as my_text, current_timestamp as my_timestamp, cast(1.0 as numeric(13, 2)) as my_val from sysibm.sysdummy1

// Row data structure
dcl-ds row_t qualified template;
  my_text varchar(100);
  my_timestamp timestamp;
  my_val zoned(13 : 2);
end-ds;

[worksofliam]

I left a comment about moving logic to another function.

I don't want the logic to generate the content in this function really. We should move it elsewhere so we can write test cases for it.

Please make this change and then lmk when to look again.

[worksofliam]

Left another review.

[janfh]

@worksofliam : Where do you want the tests? A new testsuite? Please continue to comment if you want me to change something else :) I will continue working on this tomorrow

[worksofliam]

@janfh For my own sake of mind, I have had to move this to fix day 2 (Wednesday 23rd) and this will be the first PR I look at then. I may get to it over the weekend if I find a moment. Thanks for your work here!

Also, let's chat on chat.ibmioss.org sometime.

[worksofliam]

@janfh Is there any reason you aren't using this switch statement on line 409 to check the .qualifier instead of adding a new else if block?

switch (statementDetail.qualifier) {

[janfh]

@janfh Is there any reason you aren't using this switch statement on line 409 to check the .qualifier instead of adding a new else if block?

switch (statementDetail.qualifier) {

@worksofliam : I think it was to avoid changing that runSql on 405. But I can change this if you like. I guess there is no reason why we can't always use the runSqlVerbose?

[worksofliam]

@janfh i guess runSqlVerbose can always be used, but i see your point. Leave it as it is for now!

Going to do some final tests in the morning on this one and it should be good to go.

[worksofliam]

@janfh Last thing now is to fix the conflict and we should be good to go.

[janfh]

Done :)