peggy使用import引入编译后的peggy文件时的报错 (Error when peggy uses import to introduce compiled peggy files)

[cy-biggray]

import { select,ident_part,from,__,_ } from '../../common/common.js';

// select.peggy:
selectStatement
= select _ colunm_clause:colunm_clause _ from _ table_name:table_name __ ';'{ return column_name=${colunm_clause}, table_name=${table_name}}

colunm_clause
= column_name
/ '*'

column_name
= name:ident_part* {return name.join("")}

这段代码，当引入的ident_part后有其他属性时就会报错。如果 name:ident_part {return name.join("")}就不会报错。
错误内容： Possible infinite loop when parsing (repetition used with an expression that may not consume any input)；

// common.peggy

ident_part
= [A-Za-z0-9]

select
= 'select'i

from
= 'from'i

__ = whitespace*
_ = whitespace+

whitespace
= [ \t\r\n];

[cy-biggray]

但是如果使用括号并添加任意内容后就不会报错了 name:(__ ident_part)* {return name.join("")}

[hildjj]

We'll need someone's help translating to English if we're going to be able to help you for real. In the meantime, you might try reading this:

https://en.wikipedia.org/wiki/Left_recursion#Removing_direct_left_recursion

[hildjj]

这段代码，当引入的ident_part后有其他属性时就会报错。如果 name:ident_part {return name.join("")}就不会报错。 错误内容

Google Translate says:

This code will report an error if there are other attributes after the imported ident_part. If name:ident_part {return name.join("")}, no error will be reported. Error content:

[hildjj]

import statements in your peggy file aren't working very well yet, they still need a lot of work. I would recommend doing this instead at the moment:

peggy select.peggy ../../common/common.js

which will combine the grammars together at generation time.

[cy-biggray]

peggy select.peggy .. /.. /common/common.js If there are repeated rule definitions, the tree shake operation is not performed. This makes the packaged file larger.

[cy-biggray]

I self-test, even if the import does not report errors, he generated content is about only a large peggy file size.
And it uses require to associate other peggy files, and there is auxiliary code.

Can you use parse.parse to generate grammer first? Then create a rollup-like configuration file. Generate CPU based on file import
The dependency tree. Then merge the final peggy files from the bottom up