public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug fortran/57633] New: I/O: Problem with formatted read: reading an incomplete record under Windows
@ 2013-06-17  8:52 burnus at gcc dot gnu.org
  2013-06-17  8:55 ` [Bug fortran/57633] I/O: Problem with formatted read: reading CR-LF files (\r\n) burnus at gcc dot gnu.org
                   ` (6 more replies)
  0 siblings, 7 replies; 8+ messages in thread
From: burnus at gcc dot gnu.org @ 2013-06-17  8:52 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57633

            Bug ID: 57633
           Summary: I/O: Problem with formatted read: reading an
                    incomplete record under Windows
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Keywords: wrong-code
          Severity: normal
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: burnus at gcc dot gnu.org
                CC: jb at gcc dot gnu.org, jvdelisle at gcc dot gnu.org

The following seems to be a Windows only problem as I cannot reproduce it under
Linux. See
https://groups.google.com/forum/?fromgroups#!topic/comp.lang.fortran/mtERtiPKqu0


testcase.txt file:

line1,1,
line2


Produces the following output
  line1                1
                       0

Expected:
  line1                1
  line2                5

Comment by BAF:
------------------------------------------------------------
"I can confirm these result with the exact same program under mingw32
gfortran 4.8.0 20130302 (experimental) [trunk revision 196403].

To eliminate the concern about the end of line/record character for the
second line of the file, I added a third line, so that the data file is now

line1,1,
line2
line3

The output remains the same, i.e.

  line1                1
                       0

Very strange.  If I change the second read statement to be be

   read(11,*)s1

(i.e., change the formatted read to a list directed form), the output
changes to what would be expected in either case

  line1                1
  line2                5

Seems like the "bug" is related to the formatted read. 
------------------------------------------------------------


Test program:

program teststuff
  implicit none
  integer::a
  character(len=10)::s1
  open(11,file="testcase.txt")
  read(11,*)s1,a
  write(*,*)s1,a
  read(11,"(a)")s1
  write(*,*)s1, len_trim(s1)
end program teststuff


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug fortran/57633] I/O: Problem with formatted read: reading CR-LF files (\r\n)
  2013-06-17  8:52 [Bug fortran/57633] New: I/O: Problem with formatted read: reading an incomplete record under Windows burnus at gcc dot gnu.org
@ 2013-06-17  8:55 ` burnus at gcc dot gnu.org
  2013-06-17  9:20 ` dominiq at lps dot ens.fr
                   ` (5 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: burnus at gcc dot gnu.org @ 2013-06-17  8:55 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57633

Tobias Burnus <burnus at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|I/O: Problem with formatted |I/O: Problem with formatted
                   |read: reading an incomplete |read: reading CR-LF files
                   |record under Windows        |(\r\n)

--- Comment #1 from Tobias Burnus <burnus at gcc dot gnu.org> ---
I can reproduce the issue using the following test file:

printf "line1,1,\r\nline2\r\nline3" > testcase.txt

It works if one has \n instead of \r\n


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug fortran/57633] I/O: Problem with formatted read: reading CR-LF files (\r\n)
  2013-06-17  8:52 [Bug fortran/57633] New: I/O: Problem with formatted read: reading an incomplete record under Windows burnus at gcc dot gnu.org
  2013-06-17  8:55 ` [Bug fortran/57633] I/O: Problem with formatted read: reading CR-LF files (\r\n) burnus at gcc dot gnu.org
@ 2013-06-17  9:20 ` dominiq at lps dot ens.fr
  2013-06-18  1:38 ` jvdelisle at gcc dot gnu.org
                   ` (4 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: dominiq at lps dot ens.fr @ 2013-06-17  9:20 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57633

Dominique d'Humieres <dominiq at lps dot ens.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2013-06-17
     Ever confirmed|0                           |1

--- Comment #3 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
Confirmed.


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug fortran/57633] I/O: Problem with formatted read: reading CR-LF files (\r\n)
  2013-06-17  8:52 [Bug fortran/57633] New: I/O: Problem with formatted read: reading an incomplete record under Windows burnus at gcc dot gnu.org
  2013-06-17  8:55 ` [Bug fortran/57633] I/O: Problem with formatted read: reading CR-LF files (\r\n) burnus at gcc dot gnu.org
  2013-06-17  9:20 ` dominiq at lps dot ens.fr
@ 2013-06-18  1:38 ` jvdelisle at gcc dot gnu.org
  2013-06-20  8:44 ` burnus at gcc dot gnu.org
                   ` (3 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: jvdelisle at gcc dot gnu.org @ 2013-06-18  1:38 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57633

--- Comment #4 from Jerry DeLisle <jvdelisle at gcc dot gnu.org> ---
We need to audit all places where '\n' is used and make sure we are taking care
of the '\r' properly in read.c , etc.

Then review what were are doing when we get a premature end of record.

Maybe we are skipping a code path related to read buffers, like fbuf.


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug fortran/57633] I/O: Problem with formatted read: reading CR-LF files (\r\n)
  2013-06-17  8:52 [Bug fortran/57633] New: I/O: Problem with formatted read: reading an incomplete record under Windows burnus at gcc dot gnu.org
                   ` (2 preceding siblings ...)
  2013-06-18  1:38 ` jvdelisle at gcc dot gnu.org
@ 2013-06-20  8:44 ` burnus at gcc dot gnu.org
  2013-06-20  9:12 ` burnus at gcc dot gnu.org
                   ` (2 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: burnus at gcc dot gnu.org @ 2013-06-20  8:44 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57633

--- Comment #5 from Tobias Burnus <burnus at gcc dot gnu.org> ---
The problem seems to be

2012    finish_list_read (st_parameter_dt *dtp)
...
2019      if (dtp->u.p.at_eol)
2020        {
2021          dtp->u.p.at_eol = 0;
2022          return;
2023        }
2024
2025      err = eat_line (dtp);

as at_eol == 1.


Draft patch:

The first hunk solves the actual problem of this PR.

The second one handles the case "abc\rdef" where \r is not followed by \n and
does not start a new record. (Note: gfortran does not handle pre-MacOS X line
breaks of the form "\r", only Windows \r\n and Unix (Linux, MacOS X,...) "\n".)
- that's unrelated to this PR.

--- a/libgfortran/io/list_read.c
+++ b/libgfortran/io/list_read.c
@@ -244,3 +244,3 @@ next_char (st_parameter_dt *dtp)
 done:
-  dtp->u.p.at_eol = (c == '\n' || c == '\r' || c == EOF);
+  dtp->u.p.at_eol = (c == '\n' || c == EOF);
   return c;
@@ -336,3 +336,2 @@ eat_separator (st_parameter_dt *dtp)
     case '\r':
-      dtp->u.p.at_eol = 1;
       if ((n = next_char(dtp)) == EOF)


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug fortran/57633] I/O: Problem with formatted read: reading CR-LF files (\r\n)
  2013-06-17  8:52 [Bug fortran/57633] New: I/O: Problem with formatted read: reading an incomplete record under Windows burnus at gcc dot gnu.org
                   ` (3 preceding siblings ...)
  2013-06-20  8:44 ` burnus at gcc dot gnu.org
@ 2013-06-20  9:12 ` burnus at gcc dot gnu.org
  2013-06-20 10:55 ` burnus at gcc dot gnu.org
  2013-06-20 10:55 ` burnus at gcc dot gnu.org
  6 siblings, 0 replies; 8+ messages in thread
From: burnus at gcc dot gnu.org @ 2013-06-20  9:12 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57633

--- Comment #6 from Tobias Burnus <burnus at gcc dot gnu.org> ---
Patch: http://gcc.gnu.org/ml/fortran/2013-06/msg00109.html


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug fortran/57633] I/O: Problem with formatted read: reading CR-LF files (\r\n)
  2013-06-17  8:52 [Bug fortran/57633] New: I/O: Problem with formatted read: reading an incomplete record under Windows burnus at gcc dot gnu.org
                   ` (4 preceding siblings ...)
  2013-06-20  9:12 ` burnus at gcc dot gnu.org
@ 2013-06-20 10:55 ` burnus at gcc dot gnu.org
  2013-06-20 10:55 ` burnus at gcc dot gnu.org
  6 siblings, 0 replies; 8+ messages in thread
From: burnus at gcc dot gnu.org @ 2013-06-20 10:55 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57633

--- Comment #7 from Tobias Burnus <burnus at gcc dot gnu.org> ---
Author: burnus
Date: Thu Jun 20 10:55:03 2013
New Revision: 200238

URL: http://gcc.gnu.org/viewcvs?rev=200238&root=gcc&view=rev
Log:
2013-06-20  Tobias Burnus  <burnus@net-b.de>

        PR fortran/57633
        * io/list_read.c (next_char, eat_separator): Don't set EOL for \r.

2013-06-20  Tobias Burnus  <burnus@net-b.de>

        PR fortran/57633
        * gfortran.dg/list_read_11.f90: New.


Added:
    trunk/gcc/testsuite/gfortran.dg/list_read_11.f90
Modified:
    trunk/gcc/testsuite/ChangeLog
    trunk/libgfortran/ChangeLog
    trunk/libgfortran/io/list_read.c


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug fortran/57633] I/O: Problem with formatted read: reading CR-LF files (\r\n)
  2013-06-17  8:52 [Bug fortran/57633] New: I/O: Problem with formatted read: reading an incomplete record under Windows burnus at gcc dot gnu.org
                   ` (5 preceding siblings ...)
  2013-06-20 10:55 ` burnus at gcc dot gnu.org
@ 2013-06-20 10:55 ` burnus at gcc dot gnu.org
  6 siblings, 0 replies; 8+ messages in thread
From: burnus at gcc dot gnu.org @ 2013-06-20 10:55 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57633

Tobias Burnus <burnus at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #8 from Tobias Burnus <burnus at gcc dot gnu.org> ---
FIXED on the 4.9 trunk.


^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2013-06-20 10:55 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-06-17  8:52 [Bug fortran/57633] New: I/O: Problem with formatted read: reading an incomplete record under Windows burnus at gcc dot gnu.org
2013-06-17  8:55 ` [Bug fortran/57633] I/O: Problem with formatted read: reading CR-LF files (\r\n) burnus at gcc dot gnu.org
2013-06-17  9:20 ` dominiq at lps dot ens.fr
2013-06-18  1:38 ` jvdelisle at gcc dot gnu.org
2013-06-20  8:44 ` burnus at gcc dot gnu.org
2013-06-20  9:12 ` burnus at gcc dot gnu.org
2013-06-20 10:55 ` burnus at gcc dot gnu.org
2013-06-20 10:55 ` burnus at gcc dot gnu.org

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).