u first define func
then u use aliases
On Wed, Jun 18, 2025, 10:05 PM Alberto Millán <[email protected]> wrote:
> First of all, I'd like to thank you for your responses.
>
> Okay, now I understand a little better what's going on with recursion.
>
> In most cases, I can now see what's wrong, but there's one specific case
> where I can't quite figure out what's wrong.
>
> bash -ic $'alias x=:\nx(){ id;};x' # This case crash.
> bash -ic $'alias x=:;x(){ id;};x' # This case do not crash.
>
> I understand that the crash is caused by recursion, but I don't quite
> understand why the expansion of '\n' works as expected (crashes), but it
> doesn't with the semicolon.
>
> Please, can you explain me that?
>
> El mié, 18 jun 2025 a las 20:30, Chet Ramey (<[email protected]>)
> escribió:
>
> > On 6/18/25 11:38 AM, Alberto Millán wrote:
> >
> > > Bash Version: 5.2
> > > Patch Level: 37
> > > Release Status: release
> > >
> > > Description:
> > > When I define an alias and a function with the same name, I
> > experience
> > > abnormal behavior.
> > > It seems to be especially dangerous when using a colon.
> >
> > I think there is a misunderstanding about exactly when alias expansion
> > happens in the parsing process.
> >
> > The current version of the man page says:
> >
> > "Aliases allow a string to be substituted for a word that is in a posi-
> > tion in the input where it can be the first word of a simple command.
> > Aliases have names and corresponding values that are set and unset us-
> > ing the alias and unalias builtin commands (see SHELL BUILTIN COMMANDS
> > below).
> >
> > If the shell reads an unquoted word in the right position, it checks
> > the word to see if it matches an alias name. If it matches, the shell
> > replaces the word with the alias value, and reads that value as if it
> > had been read instead of the word. The shell doesn't look at any char-
> > acters following the word before attempting alias substitution."
> >
> > So as part of the tokenizing process, the shell reads words (and
> operators,
> > etc.), one at a time. If it reads a word, and it's in the position where
> > it could be the first word in a simple command, it tries alias expansion.
> >
> > Given that, we can see what happens here.
> >
> > > alias x='echo foo'
> > > x(){ :;}
> > >
> > > # It return:
> > > # -bash: syntax error near unexpected token `('
> >
> > Because what you get in the input after alias expansion is basically
> >
> > echo foo(){ :;}
> >
> > which is a syntax error.
> >
> >
> > > alias x='echo foo;'
> > > x(){ :;}
> > >
> > > # It return similar but diferent error:
> > > # -bash: syntax error near unexpected token `)'
> >
> > This time what you get is
> >
> > echo foo;(){ :;}
> >
> > Which is a valid simple command, followed by a left paren, which can
> > introduce a subshell or arithmetic command, followed by a right paren,
> > which is a syntax error.
> >
> >
> > > alias x=:
> > > x(){ :;} # Not error yet but...
> > >
> > > x
> > > # It return:
> > > # Segmentation fault (core dumped)
> >
> > You have managed to define and execute an infinitely recursive function:
> >
> > : () { :; }
> > :
> >
> >
> > >
> > > And when you put a colon at the end of the alias, the function
> > > definition executes the alias.
> > > ```bash
> > > alias x='id;:'
> > > x(){ :;}
> > >
> > > # It return the execution of the alias:
> >
> > You have again managed to define a recursive function; this time, you
> > just executed a simple command before you did so.
> >
> >
> > > bash -i -c $'alias x=:\n x(){ :;};x' # This crash.
> > > bash -c $'alias x=:\n x(){ :;};x' # This do not crash.
> > > bash -i -c $'alias x=:\nfunction x(){ :;};x' # This do not crash.
> > > bash -i -c 'alias x=:; function x(){ :;};x' # This do not crash.
> > > bash -i -c 'alias x=:; x(){ :;};x' # This do not crash.
> >
> > Given the above, you should be able to figure out what these are doing.
> > Keep in mind that bash doesn't expand aliases by default in non-
> > interactive shells.
> >
> > --
> > ``The lyf so short, the craft so long to lerne.'' - Chaucer
> > ``Ars longa, vita brevis'' - Hippocrates
> > Chet Ramey, UTech, CWRU [email protected]
> http://tiswww.cwru.edu/~chet/
> >
>
