[RPL/2] Autoconf error in finding libm

BERTRAND Joël joel.bertrand at systella.fr
Wed Jan 16 18:08:19 CET 2019


Jonathan Busby a écrit :
>     > "
>     > +++Error : Undefined implicit name [92178]
>     > "
> 
>             arith.rpl has to abort with this error. If you want to run
>     arith.rpl,
>     you have to add "all" implicit after first '<<'. But there is a bug,
>     arith.rpl aborts with "access violation" error. I'll fix this bug as
>     soon as possible.
> 
> 
> I tried placing an "all" as you described ( "ARITH all << ..." ) but I
> now just receive a "+++Fatal : Invalid definition name [20637]" error. I
> don't know if I'm doing something stupid or not.

ARITH
<<
    "all" implicit

    1 1 + disp
    1 1.1 + disp
    1.1 1 + disp
...

	But this test aborts with "access violation error". I don't have time
to check before next week.

> 
>     > Also, when I run "rpl --help" I get an error after the license
>     text that
>     > reads :
>     >
>     > "
>     > +++Error : Any executable definition
>     > "
> 
>             What do you expect ? If you want to obtain a prompt, you
>     have to start
>     RPL/2 with rpl -i. 
> 
> 
> Well, RPL/2's command line behavior needs to be modified to correspond
> with common Unix command line conventions. That is, if I type "cp
> --version", I don't receive an error. Running "rpl  -v" should behave
> the same way. In addition, I had no idea what the error message was
> indicating -- it should be changed to something such as "Error : No
> RPL/2 program found" or something like that. If one just runs "rpl" then
> the command line syntax is indicated by "rpl [-options] [program]",
> which, in Unix based systems, means that both "options" and "program"
> are optional. I think it should be changed to something like
> "
> Usage : rpl [OPTIONS] FILE
> With no FILE, or when FILE is -, read standard input
> "

	You can try to run "man rpl" ;-)

> But, all the above is just my opinion, and you'd probably have to do
> some recoding and it could break compatibility... I just think that the
> error messages are not very understandable...

	RPL/2 acts as regular RPL. You define some "definitions" in one or
several files (#include can be used in RPL/2) :

DEF1
<<
	...
>>

DEF2
<<
	...
>>

DEF3
<<
	...
>>

Program entry point is the first definition in file _after_
preprocessing stage.

	Regards,

	JKB


More information about the RPL2 mailing list