Hi guys,
I have a suspicion that I have fallen foul of #2 in Evan's mail here.
Having upgraded to R16 (thankfully in a duplicate machine!) my app fails to
compile with the following . . .
(ege@testhost)1> 05:10:06.972 [error] gen_server boss_web terminated with
reason: no match of right hand value
{error,[{"/home/graeme/projects/egolf/dev/tournaments/src/controller/egolf_comp_controller.erl",[{{265,25},erl_lint,{undefined_function,{is_integer,2}}}]}]}
in boss_load:load_all_modules/3 line 29
05:10:07.060 [error] CRASH REPORT Process boss_web with 0 neighbours exited
with reason: no match of right hand value
{error,[{"/home/graeme/projects/egolf/dev/tournaments/src/controller/egolf_comp_controller.erl",[{{265,25},erl_lint,{undefined_function,{is_integer,2}}}]}]}
in boss_load:load_all_modules/3 line 29 in gen_server:terminate/6 line 744
05:10:07.149 [error] Supervisor boss_sup had child boss_web_controller
started with boss_web_controller:start_link([{ip,"0.0.0.0"},{port,8001}])
at <0.67.0> exit with reason no match of right hand value
{error,[{"/home/graeme/projects/egolf/dev/tournaments/src/controller/egolf_comp_controller.erl",[{{265,25},erl_lint,{undefined_function,{is_integer,2}}}]}]}
in boss_load:load_all_modules/3 line 29 in context child_terminated
!
Etc
!
Now "is_integer(23)" works just fine from the shell, so I am suspecting the
failure to compile is related to the chance in the internal representation,
since I m passing "fun is_integer/1" to lists:filter.
Has there been some further discussion on this that I missed? (I am on CB
0.8.9)
Any help would be greatly appreciated.
Thanks
g
On 7 April 2013 21:45, Evan Miller <[email protected]> wrote:
> Hi yoast,
>
> I suspect one of the following is the culprit:
>
> 1. Erlang's syntax changed slightly with R16, disallowing unquoted
> atoms with dots in them (e.g. foo.bar.baz ). This was partly a result
> of removing packages from Erlang. It could be that the parse transform
> relied on this atom syntax.
>
> 2. The internal representation of Erlang AST also changed slightly,
> but I've been unable to find any documentation on it. The issue that
> has bitten me is that the atom 'func' was replaced by 'function' in a
> few places. This might cause the parse transform to fail -- but that
> assumes the Erlang code was correctly parsed (which if #1 holds, it
> was not).
>
> Parse transforms are indeed a dark corner of Erlang... probably more
> P-T code out there is broken than is functioning. But they're fun,
> aren't they?
>
> Evan
>
> On Sun, Apr 7, 2013 at 8:52 AM, yoast <[email protected]> wrote:
> > in this blog
> http://blogtrader.net/blog/recbird_an_erlang_dynamic_record#
> > ,
> > caoyuan gives a parse-transform to make a better erlang record syntax,
> > i.e., using R.a.b.c and R.a.b.c=Sth.
> >
> > it may compiles in the early versions of erlang, but failed in recent
> > versions.
> > AFAIK, now the input to a P-T should be well-formed erlang AST,
> obviously,
> > a.b.c cannot parse.
> >
> > But CB used a lot of compiler magic, may source files cannot parse
> without
> > P-T.
> > So i think there is some way to work around the recbird P-T so that it
> can
> > work
> > with new erlang version.
> >
> > Can anyone(Evan? :-) give me some help?
> > There seems to be no good tutorial on this "secret" topic, to show
> newbie a
> > step by step way to master this nuclear weapon.
> >
> > thanks in advance~
> >
> >
> > --
> > You received this message because you are subscribed to the Google Groups
> > "ChicagoBoss" group.
> > To unsubscribe from this group and stop receiving emails from it, send an
> > email to [email protected].
> > For more options, visit https://groups.google.com/groups/opt_out.
> >
> >
>
>
>
> --
> Evan Miller
> http://www.evanmiller.org/
>
> --
> You received this message because you are subscribed to the Google Groups
> "ChicagoBoss" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>
--
You received this message because you are subscribed to the Google Groups
"ChicagoBoss" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
Visit this group at http://groups.google.com/group/chicagoboss.
To view this discussion on the web visit
https://groups.google.com/d/msgid/chicagoboss/CAKF5fiDvRj2gVLp5NWwY%3DFojcn5MZ0BKXf2GqxK5o24X58pbTw%40mail.gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.
