Possible Bug in BASH
I think I may have found a bug in BASH 3.2.17(1)-release on a mac.
Below is a simple demonstration of the unexpected behavior:
SHELL 1:
mattsh$ alias c=clear
mattsh$ c () { echo foo; }
mattsh$ clear
foo
SHELL 2:
mattsh$ alias c=clear
mattsh$ function c () { echo foo; }
mattsh$ clear
[blanks screen]
The documentation for 'function' states:
function: function NAME { COMMANDS ; } or NAME () { COMMANDS ; }
Create a simple command invoked by NAME which runs COMMANDS.
Arguments on the command line along with NAME are passed to the
function as $0 .. $n.
According to this, "function NAME { COMMANDS ; }" and "NAME ()
{ COMMANDS ; }" should be equivalent.
If a change is made to either the documentation or code, I would
appreciate a mention, although this is not required. Regardless,
please let me know either way.
--
~Matthew Strax-Haber
Co-op, NASA, Langley Research Center
Undergrad, Northeastern University
Re: Possible Bug in BASH
On Jun 20, 2009, at 3:47 PM, Bob Proulx wrote:
> Matthew Strax-Haber wrote:
>> I think I may have found a bug in BASH 3.2.17(1)-release on a mac.
>
> I am not sure it is a bug but it does seem ood.
Perhaps it is not a bug, but if it is not then the documentation is
inconsistent with the behavior.
>
>> Below is a simple demonstration of the unexpected behavior:
>>
>> SHELL 1:
>> mattsh$ alias c=clear
>> mattsh$ c () { echo foo; }
>> mattsh$ clear
>> foo
>
> Here c is the first word. So it is replaced by the alias. (I didn't
> know this behavior before. This is actually really surprising
> behavior to me since I didn't expect a function definition to be a
> simple command.)
>
>> SHELL 2:
>> mattsh$ alias c=clear
>> mattsh$ function c () { echo foo; }
>> mattsh$ clear
>> [blanks screen]
>
> Here function is the first word.
>
> Useful in this context is 'type -a c' and type -a clear' to view the
> list of possible definitions of c and clear.
>
>> The documentation for 'function' states:
>> function: function NAME { COMMANDS ; } or NAME () { COMMANDS ; }
>>Create a simple command invoked by NAME which runs COMMANDS.
>>Arguments on the command line along with NAME are passed to the
>>function as $0 .. $n.
>>
>> According to this, "function NAME { COMMANDS ; }" and "NAME () { COMMANDS
>> ; }" should be equivalent.
>
> Yes. But the documentation for alias says:
>
>Aliases allow a string to be substituted for a word when it is
>used as the first word of a simple command.
>
> And the definition of reserved words says:
>
>Reserved words are words that have a special meaning to the
>shell. The following words are recognized as reserved when
>unquoted and either the first word of a simple command (see
>SHELL GRAMMAR below) or the third word of a case or for
>command:
>
>! case do done elif else esac fi for function if in select then
>until while { } time [[ ]]
>
> Therefore function is a reserved word when it is used as the first
> word of a simple command. Meaning to me that the name c can't be the
> first word in that case because function is the first word. Meaning
> that c isn't replaced by the alias in that case. It is only replaced
> when it is the first word.
I understand what is happening and why, but from my understanding either the
implementation should be changed or the documentation made more precise.
> Using aliases in this way seems really warped to me. I would avoid it
> for the clarity of code. Aliases are really an old paradigm from csh
> days. Functions in more modern shells completely replace the need for
> aliases.
Agreed.
> Bob
--
~ Matthew Strax-Haber
Northeastern University
