From: Tobias Burnus <burnus@net-b.de>
To: fortran@gcc.gnu.org
Subject: [Patch, Fortran] PR fortran/83522 – reject array-valued substrings
Date: Mon, 08 Oct 2018 19:45:00 -0000 [thread overview]
Message-ID: <c0ed507e-bf17-ea71-6715-906c10ee054d@net-b.de> (raw)
[-- Attachment #1: Type: text/plain, Size: 1499 bytes --]
Hi all,
as the PR (and the attached testcase) shows, gfortran doesn't handle
noncontiguous access with substrings. Namely, "str_array(:)(1:5)" is
mishandled (segfault while building the "tree" as expr == NULL).
As Jerry has dug up in the standard (F2008, R610 alias F2018, R908),
substrings are only permitted on scalars (or array elements). â
[gfortran has some partial support for noncontiguous memory as actual
argument the now rejected test cases show.]
In words of the standard (here: F2018):
9.4.1Â Substrings
A substring is a contiguous portion of a character string (7.4.4).
R908  substring  is  parent-string ( substring-range )
R909Â Â parent-string
         is  scalar-variable-name
         or array-element
         or coindexed-named-object
         or scalar-structure-component
         or scalar-constant
The patch adds a check to the variable resolving â it comes too late to
print a nice error message for "string(:)(:)"; that statement gets
matched unsuccessfully and at the end the buffered error message of
match_variable (unexpected character "(") gets printed. â One could
reject it earlier but the code doesn't seem to be that nice. (See PR for
a variant.)
Build and regtested on x86_64-linux.
OK for the trunk?
Cheers
Tobias
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: array_substring.diff --]
[-- Type: text/x-patch; name="array_substring.diff", Size: 4992 bytes --]
2018-10-08 Tobias Burnus <burnus@net-b.de>
PR fortran/83522
* resolve.c (resolve_ref): Reject nonscalar
substring references.
PR fortran/83522
* gfortran.dg/actual_array_substr_1.f90: Add dg-error,
change to dg-do compile.
* gfortran.dg/actual_array_substr_2.f90: Ditto.
* gfortran.dg/array_initializer_1.f90: Use array
element not size-one section.
* gfortran.dg/array_substring.f90: New.
diff --git a/gcc/fortran/resolve.c b/gcc/fortran/resolve.c
index 87e65df5f4e..4ec881e5d78 100644
--- a/gcc/fortran/resolve.c
+++ b/gcc/fortran/resolve.c
@@ -5129,6 +5129,13 @@ resolve_ref (gfc_expr *expr)
break;
case REF_SUBSTRING:
+ /* F2008, R610 alias F2018, R908. */
+ if (current_part_dimension || seen_part_dimension)
+ {
+ gfc_error ("Substring reference of nonscalar not permitted at %L",
+ &expr->where);
+ return false;
+ }
break;
}
diff --git a/gcc/testsuite/gfortran.dg/actual_array_substr_1.f90 b/gcc/testsuite/gfortran.dg/actual_array_substr_1.f90
index 7c10c7020eb..d992f45f465 100644
--- a/gcc/testsuite/gfortran.dg/actual_array_substr_1.f90
+++ b/gcc/testsuite/gfortran.dg/actual_array_substr_1.f90
@@ -1,15 +1,17 @@
-! { dg-do run }
+! { dg-do compile }
! Test fix of PR28118, in which a substring reference to an
! actual argument with an array reference would cause a segfault.
!
+! Revised for PR fortran/83522
+!
! Contributed by Paul Thomas <pault@gcc.gnu.org>
!
program gfcbug33
character(12) :: a(2)
a(1) = "abcdefghijkl"
a(2) = "mnopqrstuvwx"
- call foo ((a(2:1:-1)(6:)))
- call bar ((a(:)(7:11)))
+ call foo ((a(2:1:-1)(6:))) ! { dg-error "Substring reference of nonscalar not permitted" }
+ call bar ((a(:)(7:11))) ! { dg-error "Substring reference of nonscalar not permitted" }
contains
subroutine foo (chr)
character(7) :: chr(:)
diff --git a/gcc/testsuite/gfortran.dg/actual_array_substr_2.f90 b/gcc/testsuite/gfortran.dg/actual_array_substr_2.f90
index fed51342ee0..9bcd19c3fb0 100644
--- a/gcc/testsuite/gfortran.dg/actual_array_substr_2.f90
+++ b/gcc/testsuite/gfortran.dg/actual_array_substr_2.f90
@@ -1,10 +1,12 @@
-! { dg-do run }
+! { dg-do compile }
! Tests the fix for pr28174, in which the fix for pr28118 was
! corrupting the character lengths of arrays that shared a
! character length structure. In addition, in developing the
! fix, it was noted that intent(out/inout) arguments were not
! getting written back to the calling scope.
!
+! Revised for PR fortran/83522
+!
! Based on the testscase by Harald Anlauf <anlauf@gmx.de>
!
program pr28174
@@ -20,7 +22,7 @@ program pr28174
n = m - 4
! Make sure that variable substring references work.
- call foo (a(:)(m:m+5), c(:)(n:m+2), d(:)(5:9))
+ call foo (a(:)(m:m+5), c(:)(n:m+2), d(:)(5:9)) ! { dg-error "Substring reference of nonscalar not permitted" }
if (any (a .ne. teststring)) STOP 1
if (any (b .ne. teststring)) STOP 2
if (any (c .ne. (/"ab456789#hij", &
@@ -37,8 +39,7 @@ contains
! This next is not required by the standard but tests the
! functioning of the gfortran implementation.
! if (all (x(:)(3:7) .eq. y)) STOP 5
- x = foostring (:)(5 : 4 + len (x))
- y = foostring (:)(3 : 2 + len (y))
+ x = foostring (:)(5 : 4 + len (x)) ! { dg-error "Substring reference of nonscalar not permitted" }
+ y = foostring (:)(3 : 2 + len (y)) ! { dg-error "Substring reference of nonscalar not permitted" }
end subroutine foo
end program pr28174
-
diff --git a/gcc/testsuite/gfortran.dg/array_initializer_1.f90 b/gcc/testsuite/gfortran.dg/array_initializer_1.f90
index d4f5e940c10..28c9272f3ac 100644
--- a/gcc/testsuite/gfortran.dg/array_initializer_1.f90
+++ b/gcc/testsuite/gfortran.dg/array_initializer_1.f90
@@ -18,7 +18,7 @@
character(4), parameter :: chr(4) = (/"abcd", "efgh", "ijkl", "mnop"/)
character(4), parameter :: chrs = chr(ii)(2:3)//chr(2)(ii-3:ii-2)
- character(4), parameter :: chrt(2) = (/chr(2:2)(2:3), chr(ii-1)(3:ii)/)
+ character(4), parameter :: chrt(2) = (/chr(2)(2:3), chr(ii-1)(3:ii)/)
character(2), parameter :: chrx(2) = (/(chr(i)(i:i+1), i=2,3)/)
if (any (y .ne. (/5., 6., 15., 16./))) STOP 1
diff --git a/gcc/testsuite/gfortran.dg/array_substring.f90 b/gcc/testsuite/gfortran.dg/array_substring.f90
new file mode 100644
index 00000000000..ea80556cfc1
--- /dev/null
+++ b/gcc/testsuite/gfortran.dg/array_substring.f90
@@ -0,0 +1,17 @@
+! { dg-do compile }
+!
+! PR fortran/83522
+!
+! Contributed by urbanjost and Jerry DeLisle
+!
+program testit
+ character(len=:),allocatable :: strings(:)
+ integer :: i
+ strings=[character(len=2) :: 'AA','BB']
+ write(*,*)strings(:)(:) ! { dg-error "Substring reference of nonscalar not permitted" }
+ !strings(:)(:) ! Parse error: "Invalid character in name"
+ strings(:)(:) = 'x' ! { dg-error "Substring reference of nonscalar not permitted" }
+ do i=1, size(strings)
+ write(*,*)strings(i)(:) ! This is valid and works
+ end do
+end program testit
next reply other threads:[~2018-10-08 19:45 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-08 19:45 Tobias Burnus [this message]
2018-10-08 22:15 ` Thomas Koenig
2018-10-10 23:12 ` Paul Richard Thomas
2018-10-11 13:56 ` David Edelsohn
2018-10-11 14:20 ` David Edelsohn
2018-10-11 14:39 ` Paul Richard Thomas
2018-10-11 10:46 ` Jakub Jelinek
2018-10-11 19:37 ` Tobias Burnus
2018-10-11 19:40 ` Paul Richard Thomas
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=c0ed507e-bf17-ea71-6715-906c10ee054d@net-b.de \
--to=burnus@net-b.de \
--cc=fortran@gcc.gnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).