2bf81922f7
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
{ ... }
62 lines
1.4 KiB
Plaintext
62 lines
1.4 KiB
Plaintext
SOURCE_FILE@0..60
|
|
FN@0..34
|
|
ATTR@0..12
|
|
POUND@0..1 "#"
|
|
L_BRACK@1..2 "["
|
|
META@2..11
|
|
PATH@2..5
|
|
PATH_SEGMENT@2..5
|
|
NAME_REF@2..5
|
|
IDENT@2..5 "cfg"
|
|
TOKEN_TREE@5..11
|
|
L_PAREN@5..6 "("
|
|
IDENT@6..10 "test"
|
|
R_PAREN@10..11 ")"
|
|
R_BRACK@11..12 "]"
|
|
WHITESPACE@12..13 "\n"
|
|
ATTR@13..22
|
|
POUND@13..14 "#"
|
|
L_BRACK@14..15 "["
|
|
META@15..21
|
|
PATH@15..21
|
|
PATH_SEGMENT@15..21
|
|
NAME_REF@15..21
|
|
IDENT@15..21 "Ignore"
|
|
R_BRACK@21..22 "]"
|
|
WHITESPACE@22..23 "\n"
|
|
FN_KW@23..25 "fn"
|
|
WHITESPACE@25..26 " "
|
|
NAME@26..29
|
|
IDENT@26..29 "foo"
|
|
PARAM_LIST@29..31
|
|
L_PAREN@29..30 "("
|
|
R_PAREN@30..31 ")"
|
|
WHITESPACE@31..32 " "
|
|
BLOCK_EXPR@32..34
|
|
STMT_LIST@32..34
|
|
L_CURLY@32..33 "{"
|
|
R_CURLY@33..34 "}"
|
|
WHITESPACE@34..36 "\n\n"
|
|
MODULE@36..59
|
|
ATTR@36..52
|
|
POUND@36..37 "#"
|
|
L_BRACK@37..38 "["
|
|
META@38..51
|
|
PATH@38..42
|
|
PATH_SEGMENT@38..42
|
|
NAME_REF@38..42
|
|
IDENT@38..42 "path"
|
|
WHITESPACE@42..43 " "
|
|
EQ@43..44 "="
|
|
WHITESPACE@44..45 " "
|
|
LITERAL@45..51
|
|
STRING@45..51 "\"a.rs\""
|
|
R_BRACK@51..52 "]"
|
|
WHITESPACE@52..53 "\n"
|
|
MOD_KW@53..56 "mod"
|
|
WHITESPACE@56..57 " "
|
|
NAME@57..58
|
|
IDENT@57..58 "b"
|
|
SEMICOLON@58..59 ";"
|
|
WHITESPACE@59..60 "\n"
|