When do we ERR= on (internal) READ statements ?

[Toon Moene]

[ God rest ye ] Merry Gentlemen,

Erik Kruyt provided me with an interesting case:

program read
character(50) :: buf='0.D99999'
double precision val
READ (UNIT=buf, FMT='(D60.0)', ERR=800) Val
stop 'ok'
800 print *, 'error'
End

Apparently, the internal read aborts due to the large exponent:

$ ./a.out
a.out: ../../../src/libgfortran/io/read.c:1131: read_f: Assertion 
`exponent < 10000' failed.

Program received signal SIGABRT: Process abort signal.

Backtrace for this error:
#0  0x7F9E4EC10607
#1  0x7F9E4EC10C0E
#2  0x7F9E4E13124F
#3  0x7F9E4E1311D5
#4  0x7F9E4E134387
#5  0x7F9E4E12A251
#6  0x7F9E4E12A301
#7  0x7F9E4ECD051D
#8  0x7F9E4ECD2C8C
#9  0x40097F in MAIN__ at kruyt.f90:?
Aborted

Is that the behavior we want, or should the ERR= label be honored ?

Kind regards,

-- 
Toon Moene - e-mail: toon@moene.org - phone: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
At home: http://moene.org/~toon/; weather: http://moene.org/~hirlam/
Progress of GNU Fortran: http://gcc.gnu.org/wiki/GFortran#news