Re: What does `echo xxx 1>&2xxx` do?

[Chet Ramey]

On 5/8/21 8:01 PM, Peng Yu wrote:


> > You can get a long way by simply reading the POSIX grammar and rules for
> > recognizing and classifying tokens. These are the two relevant rules from
>
> I don't see the definition of metacharacters in the POSIX document. So
> this is definition is from the nomenclature used in bash?

You might consider looking at the bash manual for the definition.


> So the manual defines how to determine what are words and operators.
> It says how to determine operators. But the definition seems to be
> recursive making it hard to directly tell what are operators.

The definition of `control operator' is right there in the manual. If you
want to follow POSIX, and there's no reason not to, an operator is a
control operator or a redirection operator. The redirection operators are
listed in the REDIRECTION section.

> Are there only a limited number of operators? If so, is there a way to
> print all the operators supported by bash?

No.

> Are `${`, `$(` and `$((` operators?

No.

> So this means that a character is always preferred to be associated
> with the following characters, rather than with the preceding
> characters, to form a token.

Obviously not. Consider a string of digits forming a number, for example,
delimited by a space. The last digit is associated with the previous
digits to form a token.

> Could you show some examples for this
> case?

What case?

> Does this mean that one will always look ahead to determine what is a
> token? 

Sometimes. It depends on the delmiter and what the grammar is expecting.

> What is the number of characters to look ahead?

It varies, but usually one. In some circumstances, two.

> Could you give examples for the NAME and ASSIGNMENT_WORD cases?

Come on. A NAME is as defined in the man page and POSIX. An ASSIGNMENT_WORD
is an assignment statement. It's in

https://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_10_02

> So because this context-dependency, it is not possible to determine
> the token just by lexical analysis? What specific syntactic
> information is fed back to the lexer in order to determine whether it
> is a WORD, a NAME, or an ASSIGNMENT_WORD?

Look at the link above. A token is a NAME or ASSIGNMENT_WORD in specific
circumstances, and a reserved word or WORD otherwise.

The lexical analyzer knows the parser state encoded in the parser_state
variable and the previous two tokens.

--
``The lyf so short, the craft so long to lerne.'' - Chaucer
                 ``Ars longa, vita brevis'' - Hippocrates
Chet Ramey, UTech, CWRU    chet@case.edu    http://tiswww.cwru.edu/~chet/