rust/crates/syntax/test_data/parser/err/0029_field_completion.rast

SOURCE_FILE@0..24
  FN@0..23
    FN_KW@0..2 "fn"
    WHITESPACE@2..3 " "
    NAME@3..6
      IDENT@3..6 "foo"
    PARAM_LIST@6..12
      L_PAREN@6..7 "("
      PARAM@7..11
        IDENT_PAT@7..8
          NAME@7..8
            IDENT@7..8 "a"
        COLON@8..9 ":"
        WHITESPACE@9..10 " "
        PATH_TYPE@10..11
          PATH@10..11
            PATH_SEGMENT@10..11
              NAME_REF@10..11
                IDENT@10..11 "A"
      R_PAREN@11..12 ")"
    WHITESPACE@12..13 " "
    BLOCK_EXPR@13..23
      STMT_LIST@13..23
        L_CURLY@13..14 "{"
        WHITESPACE@14..19 "\n    "
        FIELD_EXPR@19..21
          PATH_EXPR@19..20
            PATH@19..20
              PATH_SEGMENT@19..20
                NAME_REF@19..20
                  IDENT@19..20 "a"
          DOT@20..21 "."
        WHITESPACE@21..22 "\n"
        R_CURLY@22..23 "}"
  WHITESPACE@23..24 "\n"
error 21..21: expected field name or number
Convert tests to text-size 2020-04-24 16:51:02 -05:00			`SOURCE_FILE@0..24`
Rename FnDef -> Fn 2020-07-30 07:51:08 -05:00			`FN@0..23`
Convert tests to text-size 2020-04-24 16:51:02 -05:00			`FN_KW@0..2 "fn"`
			`WHITESPACE@2..3 " "`
			`NAME@3..6`
			`IDENT@3..6 "foo"`
			`PARAM_LIST@6..12`
			`L_PAREN@6..7 "("`
			`PARAM@7..11`
Rename BindPat -> IdentPat 2020-07-31 13:09:09 -05:00			`IDENT_PAT@7..8`
Convert tests to text-size 2020-04-24 16:51:02 -05:00			`NAME@7..8`
			`IDENT@7..8 "a"`
			`COLON@8..9 ":"`
			`WHITESPACE@9..10 " "`
			`PATH_TYPE@10..11`
			`PATH@10..11`
			`PATH_SEGMENT@10..11`
			`NAME_REF@10..11`
			`IDENT@10..11 "A"`
			`R_PAREN@11..12 ")"`
			`WHITESPACE@12..13 " "`
			`BLOCK_EXPR@13..23`
internal: more reasonable grammar for blocks Consider these expples { 92 } async { 92 } 'a: { 92 } #[a] { 92 } Previously the tree for them were BLOCK_EXPR { ... } EFFECT_EXPR async BLOCK_EXPR { ... } EFFECT_EXPR 'a: BLOCK_EXPR { ... } BLOCK_EXPR #[a] { ... } As you see, it gets progressively worse :) The last two items are especially odd. The last one even violates the balanced curleys invariant we have (#10357) The new approach is to say that the stuff in `{}` is stmt_list, and the block is stmt_list + optional modifiers BLOCK_EXPR STMT_LIST { ... } BLOCK_EXPR async STMT_LIST { ... } BLOCK_EXPR 'a: STMT_LIST { ... } BLOCK_EXPR #[a] STMT_LIST { ... } 2021-09-26 04:12:57 -05:00			`STMT_LIST@13..23`
			`L_CURLY@13..14 "{"`
			`WHITESPACE@14..19 "\n "`
			`FIELD_EXPR@19..21`
			`PATH_EXPR@19..20`
			`PATH@19..20`
			`PATH_SEGMENT@19..20`
			`NAME_REF@19..20`
			`IDENT@19..20 "a"`
			`DOT@20..21 "."`
			`WHITESPACE@21..22 "\n"`
			`R_CURLY@22..23 "}"`
Convert tests to text-size 2020-04-24 16:51:02 -05:00			`WHITESPACE@23..24 "\n"`
			`error 21..21: expected field name or number`