Skip to content

pacf.mts

3 messages · Martin Maechler, Paul Gilbert

Original
Paul Gilbert
# 
I am getting the following:

*** Weave Errors ***
Error in driver$runcode(drobj, chunk, chunkopts) : 
        Error in eval(expr, envir, enclos) : couldn't find function "pacf.mts"
*** Source Errors ***
Error in eval(expr, envir, enclos) : couldn't find function "pacf.mts"
make[1]: *** [checkVignettes] Error 1


I don't really understand the new namespace mechanism, so I am not sure if this
is a problem with R-1.7.0beta, or because of a change I should be making in my
code. Does it make sense that the generic pacf is found, but not the method
pacf.mts? (I am calling pacf and not pacf.mts.)

Paul
Martin Maechler
# 
PaulG> I am getting the following:

    PaulG> *** Weave Errors ***
    PaulG> Error in driver$runcode(drobj, chunk, chunkopts) : 
    PaulG> Error in eval(expr, envir, enclos) : couldn't find
    PaulG> function "pacf.mts"

    PaulG> *** Source Errors ***
    PaulG> Error in eval(expr, envir, enclos) : couldn't find
    PaulG> function "pacf.mts"
    PaulG> make[1]: *** [checkVignettes] Error 1


    PaulG> I don't really understand the new namespace
    PaulG> mechanism, so I am not sure if this is a problem with
    PaulG> R-1.7.0beta, or because of a change I should be
    PaulG> making in my code. Does it make sense that the
    PaulG> generic pacf is found, but not the method pacf.mts?

yes, that was the idea of "hiding" methods in the namespace,
such that users cannot call the methods directly.

    PaulG> (I am calling pacf and not pacf.mts.)

hmm, so I don't see why the problem above happens.
Can you give a bit more details on how you triggered the errors
above ?

Thanks (to everyone else who does) for beta testing R-beta !
Martin
1 day later
Paul Gilbert
#
Martin Maechler wrote:
First I just wanted to confirm that this is not something related to
namespaces that I don't understand. From your comment above, I think my
understanding is roughly correct, so I will try to isolate the problem a
bit more carefully.

I am a bit suspicious that this is related to the other sweave/gc
problem I am having, because the error message is somewhat similar to
the message I get when I don't get a segfault. (These errors have always
become segfaults when I set gctorture.) If this is the problem then it
likely has nothing to do with namespaces or sweave, other than the fact
that something tickles a deeper problem in R.

Paul