Support DROP PROCEDURE statement

[LorrensP-2158466]

Potentially closes #1282

This introduces a new statement: DropProcedure. It has the same fields as DropFunction because it also has the same syntax.

The other way i could have implemented this was using the Drop statement and introducing a new ObjectType called ObjectType::Procedure, but because DROP PROCEDURE ... is the same (just different word) as DROP FUNCTION..., I thought this was weird to do.

I also added a test case in the same way as DropFunction.

Let me know what you guys think of this!

[coveralls]

Pull Request Test Coverage Report for Build 9666189324

Details

• 83 of 96 (86.46%) changed or added relevant lines in 3 files are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage decreased (-0.01%) to 88.997%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/parser/mod.rs	13	15	86.67%
src/ast/mod.rs	8	13	61.54%
tests/sqlparser_postgres.rs	62	68	91.18%

Files with Coverage Reduction	New Missed Lines	%
src/ast/mod.rs	1	81.37%

Totals
Change from base Build 9633122504:	-0.01%
Covered Lines:	26353
Relevant Lines:	29611

---

💛 - Coveralls

[iffyio]

was it a specific reason to rename DropFunctionDesc to DropFuncDesc? if not we can probably undo the change given its a breaking change and the benefit is arguable, (and we have other similarly named objects like DropFunctionOption, parse_drop_function_desc)

[LorrensP-2158466]

Ah yeah you're right. I playing around with some new names since DropProcedure uses it. I'll change it back.

[alamb]

done in bb0dad8

[iffyio]

Suggested change

	_ => None,
	Some(invalid) => return parser_err!(...)
	None => None,

Should this rather return an error if the action is unexpected?
seems otherwise e.g. DROP PROCEDURE testproc SET NULL will be accepted by the parser but it will silently discard the SET NULL part (we can also include a test case demonstrating the behavior)

[alamb]

I think they way that parse_one_of_keywords works, it will only return CASCADE or RESTRICT here (never Some(..)).

I added a test in e782bf5 and changed the code to try and make this clearer

[coveralls]

Pull Request Test Coverage Report for Build 9839024282

Details

• 83 of 99 (83.84%) changed or added relevant lines in 3 files are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage decreased (-0.02%) to 89.155%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/parser/mod.rs	11	14	78.57%
src/ast/mod.rs	8	13	61.54%
tests/sqlparser_postgres.rs	64	72	88.89%

Files with Coverage Reduction	New Missed Lines	%
src/ast/mod.rs	1	81.39%

Totals
Change from base Build 9838850938:	-0.02%
Covered Lines:	26668
Relevant Lines:	29912

---

💛 - Coveralls

[alamb]

Thank you @LorrensP-2158466 and @iffyio

[LorrensP-2158466]

Thanks for doing those final changes! It's very busy at the moment, sorry.

[alamb]

Thanks for doing those final changes! It's very busy at the moment, sorry.

No problems! The best software in my opinion is a good team effort. 🚀