Re: EOF at PS2

[Chet Ramey]

On 4/26/23 5:38 PM, Grisha Levit wrote:
> A few issues with EOF being received at PS2:

I finally had a chance to check out the ksh88 behavior on a Solaris 10 VM.
It's pretty bizarre, but it does point to some improvements when ignoreeof
is enabled. It seems like the POSIX committee wanted the ksh88 behavior
without the weird aspects. I didn't check ksh93.

I think we can have bash honor ignoreeof and not exit after delimiting the
token or command while not acting quite as bizarrely as ksh88.

> The setting of ignoreeof is ignored at PS2:
>
> bash --norc -o ignoreeof
> $ uname \
> > ^D
> Linux
> (bash exits)

ksh88 lets ^D delimit the token, executes the command, then doesn't prompt
and reads input characters until a newline. If line editing is enabled, it
does not echo them. Then it either exits (no ignoreeof) or prints "Use
`exit' to terminate..." (ignoreeof) and executes the input it previoulsy
read while waiting for the newline.

> If the previous line didn't terminate the current token, ^D causes the
> token (rather than the command) to be terminated and a new PS2 to be
> printed. The history code treats them as one word.

ksh88 behaves similarly to the above. The ^D delimits the token and the
command, which executes and echoes `A'. Then it reads input as described
above until a newline and either exits or prints the ignoreeof message and
executes the input it read (except it drops the first character).

> And perhaps most surprising, if PS0 contains a command substitution,
> the wrong prompts are shown:

This has to do with saved parser state, not hard to fix after making the
changes above.

Chet

--
``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/