Nikhil Benesch
187376e657
Support DROP [TABLE|VIEW]
...
Co-authored-by: Jamie Brandon <jamie@scattered-thoughts.net>
2019-05-26 19:57:33 -04:00
Jamie Brandon
143846d333
Don't panic on weird infix garbage
...
Co-authored-by: Nikhil Benesch <nikhil.benesch@gmail.com>
2019-05-21 11:59:45 -04:00
Nickolay Ponomarev
5f3150e39a
Add IsOptional enum to use instead of optional: bool
...
The meaning of `self.parse_parenthesized_column_list(false)?` was not
very obvious.
2019-05-08 03:23:44 +03:00
Nickolay Ponomarev
f93e69d1d4
Add parse_parenthesized_column_list to reduce code duplication
2019-05-06 22:20:29 +03:00
Nickolay Ponomarev
cccf7f0d8e
Parse an optional column list after a CTE
2019-05-06 22:20:29 +03:00
Nickolay Ponomarev
f859c9b80e
Support COLLATE in expressions
...
Roughly the <character factor> production - https://jakewheat.github.io/sql-overview/sql-2011-foundation-grammar.html#character-factor
If an expression is followed by the keyword `COLLATE`, it must be
followed by the collation name, which is optionally schema-qualified
identifier.
The `COLLATE` keyword is not a regular binary operator in that it can't
be "nested": `foo COLLATE bar COLLATE baz` is not valid. If you prefer
to think of it as an operator, you might say it has the highest
precedence (judging from the spec), i.e. it binds to the smallest valid
expression to the left of it (so in `foo < bar COLLATE c`, the COLLATE
is applied first).
2019-05-06 22:20:29 +03:00
Nickolay Ponomarev
e7949d493c
Reduce indentation in parse_prefix()
2019-05-06 22:20:29 +03:00
Nickolay Ponomarev
1347ca0825
Move the rest of tests not specific to PG from the sqlparser_postgres.rs
2019-05-04 01:00:13 +03:00
Nickolay Ponomarev
de177f107c
Remove dead datetime-related code
...
1) Removed unused date/time parsing methods from `Parser`
I don't see how the token-based parsing code would ever be used: the
date/time literals are usually quoted, like `DATE 'yyyy-mm-dd'` or
simply `'YYYYMMDD'`, so the date will be a single token.
2) Removed unused date/time related variants from `Value` and the
dependency on `chrono`.
We don't support parsing date/time literals at the moment and when we
do I think we should store the exact String to let the consumer parse
it as they see fit.
3) Removed `parse_timestamps_example` and
`parse_timestamps_with_millis_example` tests. They parsed as
`number(2016) minus number(02) minus number(15) <END OF EXPRESSION>`
(leaving the time part unparsed) as it makes no sense to try parsing
a yyyy-mm-dd value as an SQL expression.
2019-05-04 01:00:13 +03:00
Nickolay Ponomarev
364f62f333
Parse table-valued functions and MSSQL-specific WITH hints
...
1) Table-valued functions (`FROM possibly_qualified.fn(arg1, ...)`) is
not part of ANSI SQL, but is supported in Postgres and MSSQL at least:
- "38.5.7. SQL Functions as Table Sources" <https://www.postgresql.org/docs/current/xfunc-sql.html#XFUNC-SQL-TABLE-FUNCTIONS >
- `user_defined_function` in "FROM (Transact-SQL)" <https://docs.microsoft.com/en-us/sql/t-sql/queries/from-transact-sql?view=sql-server-2017 >
I've considered renaming TableFactor::Table to something else (Object?),
now that it can be a TVF, but couldn't come up with a satisfactory name.
2) "WITH hints" is MSSQL-specific syntax
<https://docs.microsoft.com/en-us/sql/t-sql/queries/hints-transact-sql-table?view=sql-server-2017 >
Note that MSSQL supports the following ways of specifying hints, which
are parsed with varying degrees of accuracy:
- `FROM tab (NOLOCK)` -- deprecated syntax, parsed as a function with a `NOLOCK` argument
- `FROM tab C (NOLOCK)` -- deprecated syntax, rejected ATM
- `FROM TAB C WITH (NOLOCK)` -- OK
2019-04-27 21:14:18 +03:00
Nickolay Ponomarev
e5e3d71354
Support CASE operand WHEN expected_value THEN ..
...
Another part of #15
2019-04-27 21:14:18 +03:00
Nickolay Ponomarev
2aa4c267e7
Simplify CASE parsing
2019-04-27 21:14:18 +03:00
Nickolay Ponomarev
e5c0c426b7
Simplify parse_infix()
...
- ASTNode::SQLBinaryExpr is now constructed in one location
- There's no duplicate calls to parse_in/parse_between
- The list of tokens that correspond to operators is not duplicated
in parse_infix / to_sql_operator
- ...which made it easy to distinguish between errors that could be
caused by invalid input and internal errors. I switched the latter
to `panic!()`
2019-04-27 21:14:18 +03:00
Nickolay Ponomarev
d65274bc31
Improve ALTER error messages for unsupported syntax
2019-04-27 21:14:18 +03:00
Nickolay Ponomarev
787cd9a717
Add Parser::expected()
for more succinct error handling
2019-04-27 21:14:18 +03:00
Nickolay Ponomarev
6bb2acc9f8
Merge pull request #50 from nickolay/window-functions
...
Support OVER clause for window/analytic functions, add support for qualified function names
2019-04-27 21:12:08 +03:00
Nickolay Ponomarev
50a2310173
Rename SQLStatement::SQLSelect to SQLQuery
...
The name was confusing:
SQLStatement::SQLSelect(
SQLQuery {
body: SQLSetExpr::Select(SQLSelect)
}
)
Fix the `large_enum_variant` clippy lint for `SQLStatement::SQLQuery`
`SQLStatement::SQLCreateView`, and `SQLSetExpr::Select`, while we're
changing the AST anyway
https://rust-lang.github.io/rust-clippy/master/index.html#large_enum_variant
2019-04-21 04:46:19 +03:00
Nickolay Ponomarev
c223eaf0aa
Fix a bunch of trivial clippy lints
2019-04-21 04:46:19 +03:00
Nickolay Ponomarev
3df2223d95
Fix clippy if_same_then_else lint
...
https://rust-lang.github.io/rust-clippy/master/#if_same_then_else
There was a bug in parsing `TIME` (not followed by {WITH | WITHOUT}
TIME ZONE) -- that parsed as SQLType::Timestamp instead of SQLType::Time
2019-04-21 04:41:11 +03:00
Nickolay Ponomarev
0634ec4a83
Apply suggestions from cargo fix --edition-idioms
2019-04-21 04:41:11 +03:00
Nickolay Ponomarev
b12a19e197
Switch to the Rust 2018 edition
...
This requires Rust 1.31 (from last year) to build, but is otherwise
compatible with the 2015-edition code.
2019-04-21 04:41:11 +03:00
Nickolay Ponomarev
9a244e059e
Fix clippy lints in "Support qualified function names"
2019-04-21 02:05:36 +03:00
Nickolay Ponomarev
085e8a6b04
Use named fields in ExpressionWithAlias instead of a tuple
...
(This produces more natural JSON representation when serializing AST
with serde.)
2019-04-20 22:51:29 +03:00
Nickolay Ponomarev
4a5dc8dd4b
Support qualified function names
...
...e.g. `db.schema.func()`
2019-04-20 22:51:29 +03:00
Nickolay Ponomarev
d4de248c73
Support OVER clause for window/analytic functions
...
Since this changes SQLFunction anyway, changed its `id` field to `name`,
as we don't seem to use "id" to mean "name" anywhere else.
2019-04-20 22:51:29 +03:00
Nickolay Ponomarev
bbf1805729
Support SELECT DISTINCT
2019-04-20 14:21:26 +03:00
Nickolay Ponomarev
e113dee7e8
Fix error text for unexpected token after NOT
2019-04-20 14:21:26 +03:00
Nickolay Ponomarev
451513db5b
Simplify parse_function()
2019-04-20 14:21:26 +03:00
Zhiyuan Zheng
d8f824c400
merge CreateExternalTable & CreateTable.
2019-04-14 01:05:26 +08:00
Zhiyuan Zheng
26940920ac
Add unit tests.
2019-04-09 13:28:01 +08:00
Zhiyuan Zheng
f0f6082eff
support create external table
2019-04-08 21:36:02 +08:00
Nickolay Ponomarev
f30ab89ad2
Re-run cargo fmt
2019-03-08 15:46:40 +03:00
Nikhil Benesch
23a0fc79f5
Support CREATE MATERIALIZED VIEW
2019-03-07 13:14:33 -05:00
Nickolay Ponomarev
52e0f55b6f
Support UNION/EXCEPT/INTERSECT
2019-02-11 05:14:36 +03:00
Nickolay Ponomarev
533775c0da
Support CHAR synonym for CHARACTER
...
https://jakewheat.github.io/sql-overview/sql-2011-foundation-grammar.html#character-string-type
2019-02-11 05:14:34 +03:00
Nickolay Ponomarev
23a0d032bd
Support NUMERIC without precision or scale
...
https://jakewheat.github.io/sql-overview/sql-2011-foundation-grammar.html#exact-numeric-type
2019-02-11 05:13:48 +03:00
Nickolay Ponomarev
54c9ca8619
Support unary + / -
2019-02-11 05:13:48 +03:00
Nickolay Ponomarev
786b1cf18a
Support BETWEEN
2019-02-11 05:13:48 +03:00
Nickolay Ponomarev
264319347d
Support IN
2019-02-11 05:13:48 +03:00
Nickolay Ponomarev
bed03abe44
Support AS
and qualified wildcards in SELECT
2019-02-11 05:13:48 +03:00
Nickolay Ponomarev
bf0c07bb1b
Support basic CTEs (WITH
)
...
Some unsupported features are noted as TODOs.
2019-02-11 05:13:48 +03:00
Nickolay Ponomarev
35dd9342e2
Support national string literals (N'...')
...
Widely used in MS SQL and specified in ANSI.
2019-02-07 05:34:33 +03:00
Nickolay Ponomarev
0c0cbcaff4
Support basic CREATE VIEW
2019-02-07 05:34:12 +03:00
Nickolay Ponomarev
346d1ff2e4
Improve error messages in parse_create()
...
By not swallowing the Err from parse_data_type().
Also switch to `match` to enable parsing table-level constraints in this
loop later.
2019-02-07 05:34:09 +03:00
Nickolay Ponomarev
a0f625b949
Simplify parse_create() a little
...
Don't need the duplicate `columns.push()` + we advance the tokenizer,
so no need to peek first.
2019-02-07 05:34:06 +03:00
Nickolay Ponomarev
89602dc044
Fix a typo in parse_value error message
2019-02-07 05:34:03 +03:00
Nickolay Ponomarev
b716ffb937
Simplify JOIN USING (columns)
...
...by reusing `parse_column_names` instead of extracting identifiers
out of the `parse_expr_list`s result.
2019-02-07 05:34:00 +03:00
Nickolay Ponomarev
05a70a358a
Assert when an unknown keyword was passed to parse_keyword()
...
This happens all the time when I forget to check that the keyword I wanted
to use is actually listed in keywords.rs, this should help with debugging.
2019-02-07 05:33:57 +03:00
Nickolay Ponomarev
e3b981a0e2
Don't Box<ASTNode> in SQLSelect
...
Instead change ASTNode::SQLSubquery to be Box<SQLSelect>
2019-02-07 05:33:51 +03:00
Nickolay Ponomarev
c5bbfc33fd
Don't Box<ASTNode> in SQLStatement
...
This used to be needed when it was a variant in the ASTNode enum itself.
2019-02-07 05:33:46 +03:00