В Fri, 1 Mar 2024 10:54:08 -0500
Benjamin Tyner <bty...@gmail.com> пишет:
> > dput(letters, file = stderr())
> c("a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l",
> "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y",
> "z")
> Warning message:
> In dput(letters, file = stderr()) : wrote too few characters
Sounds like a bug.
This is where the warning is produced:
int res = Rconn_printf(con, "%s\n", CHAR(STRING_ELT(tval, i)));
if(!havewarned &&
res < strlen(CHAR(STRING_ELT(tval, i))) + 1) {
warning(_("wrote too few characters"));
havewarned = TRUE;
}
At this point, `res` is 0, and CHAR(STRING_ELT(tval, i)) is what you'd
expect:
(gdb) x /s (void*)(STRING_ELT(tval, i))+48
0x5555561bf0a8: "c(\"a\", \"b\", \"c\", \"d\", \"e\", \"f\", \"g\",
\"h\", \"i\", \"j\", \"k\", \"l\", "
`res` is 0 because that's what (con->vfprintf)(...) returns:
2813 static int stderr_vfprintf(Rconnection con, const char *format, va_list
ap)
2814 {
2815 REvprintf(format, ap);
2816 return 0;
2817 }
Why doesn't it warn for stdout()? Doesn't stdout_vfprintf(...) behave
the same? That's because stdout() (connection #1) is special for dput(),
and instead of going through the connection system, it goes straight for
Rprintf():
426 else { // ifile == 1 : "Stdout"
427 for (int i = 0; i < LENGTH(tval); i++)
428 Rprintf("%s\n", CHAR(STRING_ELT(tval, i)));
429 }
Not sure what the right fix is. There may be code depending on
(con->vfprintf)(...) returning 0.
--
Best regards,
Ivan
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
