Deprecate symbol suport for tools/configure_server_settings

[Fryguy]

In the move away from allowing symbols in the settings (#20908), we should drop the symbol support from this tool as well.

@jrafanie @kbrock Please review.

[jrafanie]

Is there a link to where this was removed? Was it deprecated first?

[Fryguy]

#20908 - Updated the OP

Also the more general ManageIQ/manageiq-api#979

I think there still needs to be a database migration for any existing symbols, which is on @kbrock 's radar, I believe.

[jrafanie]

So, I'm guessing it would be hard to deprecate it in settings, or would it? But if so, shouldn't we deprecate the client here before we just drop support for it?

[Fryguy]

What's to deprecate? We don't support symbols anymore and all symbols have become strings. Are you suggesting keep the symbol and spit out a deprecation, then just convert to .to_s anyway?

Since it's a tool, I guess I don't really understand the purpose of deprecation?

[jrafanie]

What's to deprecate? We don't support symbols anymore and all symbols have become strings. Are you suggesting keep the symbol and spit out a deprecation, then just convert to .to_s anyway?

Since it's a tool, I guess I don't really understand the purpose of deprecation?

Yeah, but it seems odd to just drop support for -t symbol and suddenly exit with an error to use one of string integer boolean float array. Maybe this is overkill: honor -t symbol but treat it like a string with a warning that symbol support is deprecated as it's now treated like a string anyway? I don't know if anyone ever populated symbol values with this tool. Poll escalate seems like such a unlikely value to be set.

[Fryguy]

Yeah...I have no problem deprecating, but the audience for it seems really small if at all.

[Fryguy]

@jrafanie Updated.

[miq-bot]

Checked commit Fryguy@dd253c9 with ruby 2.6.3, rubocop 0.82.0, haml-lint 0.35.0, and yamllint
1 file checked, 0 offenses detected
Everything looks fine. 👍

[jrafanie]

Tests are failing with "pg_restore exit code: 1 error was: pg_restore: [archiver] could not open input file "foo/bar/mkfifo": No such file or directory"... should be fixed by #21009... kicking.