cat test.sql | vendor/bin/hoa compiler:pp sql.pp 0 -s --visitor dump

actual.txt

  #  namespace     token name            token value                     offset
--------------------------------------------------------------------------------
  0  default       T_SELECT              SELECT                               0
  1  default       T_IDENTIFIER          foo                                  7
  2  default       T_AS                  AS                                  11
  3  default       T_IDENTIFIER          bar                                 14
  4  default       T_FROM                from                                18
  5  default       T_OPAREN              (                                   23
  6  default       T_SELECT              select                              24
  7  default       T_NUMBER              1                                   31
  8  default       T_AS                  as                                  33
  9  default       T_IDENTIFIER          foo                                 36
 10  default       T_CPAREN              )                                   39
 11  default       T_SEMICOLON           ;                                   40
 12  default       T_SELECT              SELECT                              42
 13  default       T_SQUOTE              '                                   49
 14  string        T_STRING              hello                               50
 15  string        T_SQUOTE              '                                   55
 16  default       T_SEMICOLON           ;                                   56
 17  default       T_SELECT              SELECT                              58
 18  default       T_DQUOTE              "                                   65
 19  string        T_STRING              world                               66
 20  string        T_DQUOTE              "                                   71
 21  default       T_SEMICOLON           ;                                   72
 22  default       EOF                                                       75


>  #queries
>  >  #query
>  >  >  #select
>  >  >  >  #fields
>  >  >  >  >  #field
>  >  >  >  >  >  token(T_IDENTIFIER, foo)
>  >  >  >  >  >  #alias
>  >  >  >  >  >  >  token(T_IDENTIFIER, bar)
>  >  >  >  #from
>  >  >  >  >  #select
>  >  >  >  >  >  #fields
>  >  >  >  >  >  >  #field
>  >  >  >  >  >  >  >  #value
>  >  >  >  >  >  >  >  >  token(T_NUMBER, 1)
>  >  >  >  >  >  >  >  #alias
>  >  >  >  >  >  >  >  >  token(T_IDENTIFIER, foo)
>  >  #queries
>  >  >  #query
>  >  >  >  #select
>  >  >  >  >  #fields
>  >  >  >  >  >  #field
>  >  >  >  >  >  >  #value
>  >  >  >  >  >  >  >  #string
>  >  >  >  >  >  >  >  >  token(string:T_STRING, hello)
>  >  >  #queries
>  >  >  >  #query
>  >  >  >  >  #select
>  >  >  >  >  >  #fields
>  >  >  >  >  >  >  #field
>  >  >  >  >  >  >  >  #value
>  >  >  >  >  >  >  >  >  #string
>  >  >  >  >  >  >  >  >  >  token(string:T_STRING, world)

expected.txt

  #  namespace     token name            token value                     offset
--------------------------------------------------------------------------------
  0  default       T_SELECT              SELECT                               0
  1  default       T_IDENTIFIER          foo                                  7
  2  default       T_AS                  AS                                  11
  3  default       T_IDENTIFIER          bar                                 14
  4  default       T_FROM                from                                18
  5  default       T_OPAREN              (                                   23
  6  default       T_SELECT              select                              24
  7  default       T_NUMBER              1                                   31
  8  default       T_AS                  as                                  33
  9  default       T_IDENTIFIER          foo                                 36
 10  default       T_CPAREN              )                                   39
 11  default       T_SEMICOLON           ;                                   40
 12  default       T_SELECT              SELECT                              42
 13  default       T_SQUOTE              '                                   49
 14  string        T_STRING              hello                               50
 15  string        T_SQUOTE              '                                   55
 16  default       T_SEMICOLON           ;                                   56
 17  default       T_SELECT              SELECT                              58
 18  default       T_DQUOTE              "                                   65
 19  string        T_STRING              world                               66
 20  string        T_DQUOTE              "                                   71
 21  default       T_SEMICOLON           ;                                   72
 22  default       EOF                                                       75


>  #queries
>  >  #query
>  >  >  #select
>  >  >  >  #fields
>  >  >  >  >  #field
>  >  >  >  >  >  token(T_IDENTIFIER, foo)
>  >  >  >  >  >  #alias
>  >  >  >  >  >  >  token(T_IDENTIFIER, bar)
>  >  >  >  #from
>  >  >  >  >  #select
>  >  >  >  >  >  #fields
>  >  >  >  >  >  >  #field
>  >  >  >  >  >  >  >  #value
>  >  >  >  >  >  >  >  >  token(T_NUMBER, 1)
>  >  >  >  >  >  >  >  #alias
>  >  >  >  >  >  >  >  >  token(T_IDENTIFIER, foo)
>  >  #query
>  >  >  #select
>  >  >  >  #fields
>  >  >  >  >  #field
>  >  >  >  >  >  #value
>  >  >  >  >  >  >  #string
>  >  >  >  >  >  >  >  token(string:T_STRING, hello)
>  >  #query
>  >  >  #select
>  >  >  >  #fields
>  >  >  >  >  #field
>  >  >  >  >  >  #value
>  >  >  >  >  >  >  #string
>  >  >  >  >  >  >  >  token(string:T_STRING, world)

sql.pp

%skip   space           [\s\t\r\n]

%token T_SELECT         (?i)select
%token T_UPDATE         (?i)update
%token T_INSERT         (?i)insert
%token T_DELETE         (?i)delete
%token T_FROM           (?i)from
%token T_AS             (?i)as
%token T_INTO           (?i)into
%token T_VALUES         (?i)values

%token T_COMMA          ,
%token T_SEMICOLON      ;
%token T_SQUOTE         ' -> string
%token T_DQUOTE         " -> string
%token T_OPAREN         \(
%token T_CPAREN         \)

%token T_NUMBER         \-?(0|[1-9]\d*)(\.\d+)?([eE][\+\-]?\d+)?
%token T_IDENTIFIER     [a-zA-Z0-9]+

%token string:T_STRING  [^"']+
%token string:T_SQUOTE  ' -> default
%token string:T_DQUOTE  " -> default

#queries:
    query() (::T_SEMICOLON:: queries()?)?

#query:
    select() | insert()

#select:
    ::T_SELECT:: fields() from()?

#insert:
    ::T_INSERT:: ::T_INTO:: <T_IDENTIFIER> ::T_VALUES:: values()

#from:
    ::T_FROM:: (<T_IDENTIFIER> | ::T_OPAREN:: select() ::T_CPAREN::)

#fields:
    field() (::T_COMMA:: field())*

#field:
    (<T_IDENTIFIER> | value()) alias()?

#alias:
    ::T_AS:: <T_IDENTIFIER>

#values:
    ::T_OPAREN:: value() (::T_COMMA:: value())* ::T_CPAREN::

#value:
    string() | <T_NUMBER>

#string:
    sstring() | dstring()

sstring:
    ::T_SQUOTE:: <T_STRING> ::T_SQUOTE::

dstring:
    ::T_DQUOTE:: <T_STRING> ::T_DQUOTE::

test.sql

SELECT foo AS bar from (select 1 as foo);
SELECT 'hello';
SELECT "world";

Question hoa/compiler :

• j'ai ce fichier SQL : https://gist.github.com/jubianchi/4eed70c361cb161c3c1a#file-test-sql
• et cette grammaire : https://gist.github.com/jubianchi/4eed70c361cb161c3c1a#file-sql-pp
• quand je lance cat test.sql | vendor/bin/hoa compiler:pp sql.pp 0 -s --visitor dump
• j'obtiens https://gist.github.com/jubianchi/4eed70c361cb161c3c1a#file-actual-txt mais j'aimerais obtenir https://gist.github.com/jubianchi/4eed70c361cb161c3c1a#file-expected-txt

Une idée ?