Support SELECT ALL

[benesch]

No description provided.

[coveralls]

Pull Request Test Coverage Report for Build 200

• 11 of 12 (91.67%) changed or added relevant lines in 2 files are covered.
• 7 unchanged lines in 2 files lost coverage.
• Overall coverage increased (+0.4%) to 89.49%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
tests/sqlparser_common.rs	8	9	88.89%

Files with Coverage Reduction	New Missed Lines	%
src/test_utils.rs	2	89.06%
src/sqlparser.rs	5	88.72%

Totals
Change from base Build 183:	0.4%
Covered Lines:	2878
Relevant Lines:	3216

---

💛 - Coveralls

[nickolay]

This is about <set quantifier> ::= DISTINCT | ALL in <query specification> ::= SELECT [ <set quantifier> ] <select list> ..., where ALL is the same as lack of an explicit set quantifier.

Given that parsing text->AST currently loses information unimportant to the query semantics, perhaps we shouldn't add the all flag to the AST? Did you need it for some reason?

(edit- if we do remove it from the AST, the test could be simplified to a simple one_statement_parses_to() call)

[benesch]

Given that parsing text->AST currently loses information unimportant to the query semantics, perhaps we shouldn't add the all flag to the AST? Did you need it for some reason?

No particular reason! I think we just hadn't fully adopted the "it's ok if queries don't roundtrip exactly, as long as they roundtrip to something equivalent" rule. I've updated the PR to drop the explicit all flag and simplified the test accordingly.

Thanks for the review!