* [Bug fortran/27138] gfortran: Fails to
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
@ 2006-04-12 20:45 ` tobias dot burnus at physik dot fu-berlin dot de
2006-04-12 21:05 ` [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error tobias dot burnus at physik dot fu-berlin dot de
` (10 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: tobias dot burnus at physik dot fu-berlin dot de @ 2006-04-12 20:45 UTC (permalink / raw)
To: gcc-bugs
------- Comment #1 from tobias dot burnus at physik dot fu-berlin dot de 2006-04-12 20:45 -------
Created an attachment (id=11252)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=11252&action=view)
test.tar.gz containing inp-test and test.F
This is SUSE's gcc-fortran-4.1.0-15, "Merge[d] up to rev 112748 for gfortran
changes on the branch."
The attached program "test.F" reads an inputfile (also attached), which
contains three lines:
1| igrd=1,lwb=F,ndvgrd=6,idsprs=0,chng=-0.100D-11
2|
3| 12
This file is read by three "READ (5,*,END=<n>,ERR=<n>) ntype" lines. With
Ifort, the first READ has an error and the second read gets the "12".
With gfortran, all three reads get an error.
Expected: Either the second or the third read picks up the 12.
The attached program prints with the Intel Fortran Compiler 9.0.033:
Second read: ntype = 12
whereas with gfortran it only shows:
Error glancing at inp-file
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread* [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
2006-04-12 20:45 ` [Bug fortran/27138] " tobias dot burnus at physik dot fu-berlin dot de
@ 2006-04-12 21:05 ` tobias dot burnus at physik dot fu-berlin dot de
2006-04-13 0:22 ` jvdelisle at gcc dot gnu dot org
` (9 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: tobias dot burnus at physik dot fu-berlin dot de @ 2006-04-12 21:05 UTC (permalink / raw)
To: gcc-bugs
------- Comment #2 from tobias dot burnus at physik dot fu-berlin dot de 2006-04-12 21:05 -------
If one replaces the third read by
character(len=300) :: str
READ (5,*,END=100,ERR=99) str
WRITE(*,*) trim(str)
one sees that the first two failing READs advance by one character each
('i','g').
The third (string) read then reads up to the comma ('td=1')
I would expect (as well as the program writer) that after the failed read one
is still in the next line (as ADVANCE=YES is implied).
Why the read(5,*) str only reads up to the comma and not the whole line, is
also an enigma for me.
G95 (GCC 4.0.3 (g95!) Apr 4 2006,
pgf95 6.0-5
PathScale EKOPath Version 2.1
(and as mentioned) Intel Fortran Compiler 9.0.033
all read the "12" with the second read.
--
tobias dot burnus at physik dot fu-berlin dot de changed:
What |Removed |Added
----------------------------------------------------------------------------
Summary|gfortran: Fails to |gfortran: read(*,*) myInt
| |advances only one character
| |on error
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread* [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
2006-04-12 20:45 ` [Bug fortran/27138] " tobias dot burnus at physik dot fu-berlin dot de
2006-04-12 21:05 ` [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error tobias dot burnus at physik dot fu-berlin dot de
@ 2006-04-13 0:22 ` jvdelisle at gcc dot gnu dot org
2006-04-13 4:26 ` jvdelisle at gcc dot gnu dot org
` (8 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: jvdelisle at gcc dot gnu dot org @ 2006-04-13 0:22 UTC (permalink / raw)
To: gcc-bugs
------- Comment #3 from jvdelisle at gcc dot gnu dot org 2006-04-13 00:22 -------
I will investigate.
--
jvdelisle at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
AssignedTo|unassigned at gcc dot gnu |jvdelisle at gcc dot gnu dot
|dot org |org
Status|UNCONFIRMED |ASSIGNED
Ever Confirmed|0 |1
Last reconfirmed|0000-00-00 00:00:00 |2006-04-13 00:22:23
date| |
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread* [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
` (2 preceding siblings ...)
2006-04-13 0:22 ` jvdelisle at gcc dot gnu dot org
@ 2006-04-13 4:26 ` jvdelisle at gcc dot gnu dot org
2006-04-13 7:20 ` jvdelisle at gcc dot gnu dot org
` (7 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: jvdelisle at gcc dot gnu dot org @ 2006-04-13 4:26 UTC (permalink / raw)
To: gcc-bugs
------- Comment #4 from jvdelisle at gcc dot gnu dot org 2006-04-13 04:26 -------
I have looked this over. Its a matter of what behavior is acceptable for each
type of error we get. My interpretation of your testcase is that we should do
a next_record when we detect the error before we do the generate_error if we
are in ADVANCE_YES. So if its an error due to an I/O problem such as hardware
failure we will just fail out like now. If its a "content of the file" error
we do a next_record before going into th error handling code. This will
presumably take us past the current line, ready for another read.
I just wonder if this is the universal behavior we should have. Regardless, I
will see if I can work up a patch and give it a spin.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread* [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
` (3 preceding siblings ...)
2006-04-13 4:26 ` jvdelisle at gcc dot gnu dot org
@ 2006-04-13 7:20 ` jvdelisle at gcc dot gnu dot org
2006-04-14 5:50 ` jvdelisle at gcc dot gnu dot org
` (6 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: jvdelisle at gcc dot gnu dot org @ 2006-04-13 7:20 UTC (permalink / raw)
To: gcc-bugs
------- Comment #5 from jvdelisle at gcc dot gnu dot org 2006-04-13 07:20 -------
The "engima" referred to in comment#2 is because it is a list directed read and
commas and spaces are deliminators. If you turn this into a read with a format
of '(a)' you get the whole string.
I am still working on the remaining dubious behavior. :)
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread* [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
` (4 preceding siblings ...)
2006-04-13 7:20 ` jvdelisle at gcc dot gnu dot org
@ 2006-04-14 5:50 ` jvdelisle at gcc dot gnu dot org
2006-04-15 3:00 ` patchapp at dberlin dot org
` (5 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: jvdelisle at gcc dot gnu dot org @ 2006-04-14 5:50 UTC (permalink / raw)
To: gcc-bugs
------- Comment #6 from jvdelisle at gcc dot gnu dot org 2006-04-14 05:50 -------
Created an attachment (id=11263)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=11263&action=view)
Proposed Patch
This is a proposed patch you can try out while I work up a more complete test
case. The example given only addresses an integer read. The same thing
happens for real, complex, or logical I think.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread* [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
` (5 preceding siblings ...)
2006-04-14 5:50 ` jvdelisle at gcc dot gnu dot org
@ 2006-04-15 3:00 ` patchapp at dberlin dot org
2006-04-17 2:34 ` jvdelisle at gcc dot gnu dot org
` (4 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: patchapp at dberlin dot org @ 2006-04-15 3:00 UTC (permalink / raw)
To: gcc-bugs
------- Comment #7 from patchapp at dberlin dot org 2006-04-15 03:00 -------
Subject: Bug number PR27138
A patch for this bug has been added to the patch tracker.
The mailing list url for the patch is
http://gcc.gnu.org/ml/gcc-patches/2006-04/msg00547.html
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread* [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
` (6 preceding siblings ...)
2006-04-15 3:00 ` patchapp at dberlin dot org
@ 2006-04-17 2:34 ` jvdelisle at gcc dot gnu dot org
2006-04-17 2:37 ` jvdelisle at gcc dot gnu dot org
` (3 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: jvdelisle at gcc dot gnu dot org @ 2006-04-17 2:34 UTC (permalink / raw)
To: gcc-bugs
------- Comment #8 from jvdelisle at gcc dot gnu dot org 2006-04-17 02:33 -------
Subject: Bug 27138
Author: jvdelisle
Date: Mon Apr 17 02:33:51 2006
New Revision: 112999
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=112999
Log:
2006-04-14 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR libgfortran/27138
* io/list_read.c (eat_line): New function.
(parse_repeat): Use new function and free_saved.
(read_logical): Same.
(read_integer): Use new function.
(parse_real): Use nml_bad_return and new function.
(read_complex): Use new function and free_saved.
(read_real): Same.
Modified:
trunk/libgfortran/ChangeLog
trunk/libgfortran/io/list_read.c
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread* [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
` (7 preceding siblings ...)
2006-04-17 2:34 ` jvdelisle at gcc dot gnu dot org
@ 2006-04-17 2:37 ` jvdelisle at gcc dot gnu dot org
2006-04-21 3:47 ` jvdelisle at gcc dot gnu dot org
` (2 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: jvdelisle at gcc dot gnu dot org @ 2006-04-17 2:37 UTC (permalink / raw)
To: gcc-bugs
------- Comment #9 from jvdelisle at gcc dot gnu dot org 2006-04-17 02:37 -------
Subject: Bug 27138
Author: jvdelisle
Date: Mon Apr 17 02:37:15 2006
New Revision: 113000
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=113000
Log:
2006-04-16 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR libgfortran/27138
* gfortran.dg/read_bad_advance.f90: New test.
Added:
trunk/gcc/testsuite/gfortran.dg/read_bad_advance.f90
Modified:
trunk/gcc/testsuite/ChangeLog
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread* [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
` (8 preceding siblings ...)
2006-04-17 2:37 ` jvdelisle at gcc dot gnu dot org
@ 2006-04-21 3:47 ` jvdelisle at gcc dot gnu dot org
2006-04-21 3:51 ` jvdelisle at gcc dot gnu dot org
2006-04-21 3:53 ` jvdelisle at gcc dot gnu dot org
11 siblings, 0 replies; 13+ messages in thread
From: jvdelisle at gcc dot gnu dot org @ 2006-04-21 3:47 UTC (permalink / raw)
To: gcc-bugs
------- Comment #10 from jvdelisle at gcc dot gnu dot org 2006-04-21 03:47 -------
Subject: Bug 27138
Author: jvdelisle
Date: Fri Apr 21 03:47:23 2006
New Revision: 113131
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=113131
Log:
2006-04-20 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR libgfortran/27138
* io/list_read.c (eat_line): New function.
(parse_repeat): Use new function and free_saved.
(read_logical): Same.
(read_integer): Use new function.
(parse_real): Use nml_bad_return and new function.
(read_complex): Use new function and free_saved.
(read_real): Same.
Modified:
branches/gcc-4_1-branch/libgfortran/ChangeLog
branches/gcc-4_1-branch/libgfortran/io/list_read.c
branches/gcc-4_1-branch/libgfortran/io/unit.c
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread* [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
` (9 preceding siblings ...)
2006-04-21 3:47 ` jvdelisle at gcc dot gnu dot org
@ 2006-04-21 3:51 ` jvdelisle at gcc dot gnu dot org
2006-04-21 3:53 ` jvdelisle at gcc dot gnu dot org
11 siblings, 0 replies; 13+ messages in thread
From: jvdelisle at gcc dot gnu dot org @ 2006-04-21 3:51 UTC (permalink / raw)
To: gcc-bugs
------- Comment #11 from jvdelisle at gcc dot gnu dot org 2006-04-21 03:51 -------
Subject: Bug 27138
Author: jvdelisle
Date: Fri Apr 21 03:51:44 2006
New Revision: 113132
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=113132
Log:
2006-04-20 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR libgfortran/27138
* gfortran.dg/read_bad_advance.f90: New test
Added:
branches/gcc-4_1-branch/gcc/testsuite/gfortran.dg/read_bad_advance.f90
Modified:
branches/gcc-4_1-branch/gcc/testsuite/ChangeLog
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread* [Bug fortran/27138] gfortran: read(*,*) myInt advances only one character on error
2006-04-12 20:24 [Bug fortran/27138] New: gfortran: Fails to tobias dot burnus at physik dot fu-berlin dot de
` (10 preceding siblings ...)
2006-04-21 3:51 ` jvdelisle at gcc dot gnu dot org
@ 2006-04-21 3:53 ` jvdelisle at gcc dot gnu dot org
11 siblings, 0 replies; 13+ messages in thread
From: jvdelisle at gcc dot gnu dot org @ 2006-04-21 3:53 UTC (permalink / raw)
To: gcc-bugs
------- Comment #12 from jvdelisle at gcc dot gnu dot org 2006-04-21 03:53 -------
Fixed on 4.1 and 4.2.
--
jvdelisle at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution| |FIXED
Target Milestone|--- |4.1.1
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27138
^ permalink raw reply [flat|nested] 13+ messages in thread