GNU bug report logs -
#77301
Is this a bug?
Previous Next
To reply to this bug, email your comments to 77301 AT debbugs.gnu.org.
There is no need to reopen the bug first.
Toggle the display of automated, internal messages from the tracker.
Report forwarded
to
bug-grep <at> gnu.org:
bug#77301; Package
grep.
(Thu, 27 Mar 2025 06:33:06 GMT)
Full text and
rfc822 format available.
Acknowledgement sent
to
George R Goffe <grgoffe <at> yahoo.com>:
New bug report received and forwarded. Copy sent to
bug-grep <at> gnu.org.
(Thu, 27 Mar 2025 06:33:06 GMT)
Full text and
rfc822 format available.
Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):
Howdy,
Apparently the gcc folks have been busy, tightening up the "rules". I'm not sure this isn't one so I'm reporting it.
Best regards,
George...
This is output from gcc (GCC) 15.0.1 20250325 (experimental) but the non-experimental version shows the same output.
mv fgrep-t fgrep
pcresearch.c: In function 'Pcompile':
pcresearch.c:222:23: error: initializer-string for array of 'char' is too long [-Werror=unterminated-string-initialization]
222 | wprefix[10] = "(?<!\\w)(?:", wsuffix[7] = ")(?!\\w)";
| ^~~~~~~~~~~~~
pcresearch.c:222:51: error: initializer-string for array of 'char' is too long [-Werror=unterminated-string-initialization]
222 | wprefix[10] = "(?<!\\w)(?:", wsuffix[7] = ")(?!\\w)";
| ^~~~~~~~~~
cc1: all warnings being treated as errors
Information forwarded
to
bug-grep <at> gnu.org:
bug#77301; Package
grep.
(Thu, 27 Mar 2025 15:58:05 GMT)
Full text and
rfc822 format available.
Message #8 received at 77301 <at> debbugs.gnu.org (full text, mbox):
George R Goffe via Bug reports for GNU grep <bug-grep <at> gnu.org> writes:
> Apparently the gcc folks have been busy, tightening up the
> "rules". I'm not sure this isn't one so I'm reporting it.
> pcresearch.c:222:23: error: initializer-string for array of 'char' is too long [-Werror=unterminated-string-initialization]
> 222 | wprefix[10] = "(?<!\\w)(?:", wsuffix[7] = ")(?!\\w)";
> | ^~~~~~~~~~~~~
That is strange: I can't find any statement that recent C standards
have removed the special case "initialize a char array that doesn't have
room for the final NULL from a string literal".
Dale
Information forwarded
to
bug-grep <at> gnu.org:
bug#77301; Package
grep.
(Thu, 27 Mar 2025 16:18:02 GMT)
Full text and
rfc822 format available.
Message #11 received at 77301 <at> debbugs.gnu.org (full text, mbox):
"Dale R. Worley" via Bug reports for GNU grep <bug-grep <at> gnu.org> wrote:
> George R Goffe via Bug reports for GNU grep <bug-grep <at> gnu.org> writes:
> > Apparently the gcc folks have been busy, tightening up the
> > "rules". I'm not sure this isn't one so I'm reporting it.
>
> > pcresearch.c:222:23: error: initializer-string for array of 'char' is too long [-Werror=unterminated-string-initialization]
> > 222 | wprefix[10] = "(?<!\\w)(?:", wsuffix[7] = ")(?!\\w)";
> > | ^~~~~~~~~~~~~
>
> That is strange: I can't find any statement that recent C standards
> have removed the special case "initialize a char array that doesn't have
> room for the final NULL from a string literal".
>
> Dale
The bug is that the constants given for the sizes are each to small by one.
Count the characters in each string, add 1 for the final '\0', and
you'll see.
Arnold
Reply sent
to
Paul Eggert <eggert <at> cs.ucla.edu>:
You have taken responsibility.
(Thu, 27 Mar 2025 17:56:02 GMT)
Full text and
rfc822 format available.
Notification sent
to
George R Goffe <grgoffe <at> yahoo.com>:
bug acknowledged by developer.
(Thu, 27 Mar 2025 17:56:02 GMT)
Full text and
rfc822 format available.
Message #16 received at 77301-done <at> debbugs.gnu.org (full text, mbox):
It's not a bug in grep. We pacified GCC's false alarm in grep commit
08c4ce064b7f42acc023a9b28d1076a81feb6143.
Information forwarded
to
bug-grep <at> gnu.org:
bug#77301; Package
grep.
(Fri, 28 Mar 2025 16:25:01 GMT)
Full text and
rfc822 format available.
Message #19 received at 77301 <at> debbugs.gnu.org (full text, mbox):
arnold <at> skeeve.com writes:
> "Dale R. Worley" via Bug reports for GNU grep <bug-grep <at> gnu.org> wrote:
...
>> That is strange: I can't find any statement that recent C standards
>> have removed the special case "initialize a char array that doesn't have
>> room for the final NULL from a string literal".
>
> The bug is that the constants given for the sizes are each to small by one.
> Count the characters in each string, add 1 for the final '\0', and
> you'll see.
Yes, but ... Back when I was programming in C (and when I was involved
in writing a C compiler), there was a specific clause in the C standard
for that case, saying that it was OK, the characters went into the
char[] variable, and the final NUL was to be ignored. When I said
"special case" above, I meant it.
I expect I've thrown away the copy I have of the draft of the original
ANSI C, so I can't check that, and ANSI is uptight so there aren't
copies of the current standard on online. But trying to search for
statements about the *idea*, I couldn't find any online reference saying
that some Cnn standard had removed that special case.
Dale
Information forwarded
to
bug-grep <at> gnu.org:
bug#77301; Package
grep.
(Fri, 28 Mar 2025 17:15:01 GMT)
Full text and
rfc822 format available.
Message #22 received at 77301 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
You can always init the char one at a time. :D
On Friday, March 28, 2025 at 12:25:34 PM EDT, Dale R. Worley" via Bug reports for GNU grep <bug-grep <at> gnu.org> wrote:
arnold <at> skeeve.com writes:
> "Dale R. Worley" via Bug reports for GNU grep <bug-grep <at> gnu.org> wrote:
...
>> That is strange: I can't find any statement that recent C standards
>> have removed the special case "initialize a char array that doesn't have
>> room for the final NULL from a string literal".
>
> The bug is that the constants given for the sizes are each to small by one.
> Count the characters in each string, add 1 for the final '\0', and
> you'll see.
Yes, but ... Back when I was programming in C (and when I was involved
in writing a C compiler), there was a specific clause in the C standard
for that case, saying that it was OK, the characters went into the
char[] variable, and the final NUL was to be ignored. When I said
"special case" above, I meant it.
I expect I've thrown away the copy I have of the draft of the original
ANSI C, so I can't check that, and ANSI is uptight so there aren't
copies of the current standard on online. But trying to search for
statements about the *idea*, I couldn't find any online reference saying
that some Cnn standard had removed that special case.
Dale
[Message part 2 (text/html, inline)]
Information forwarded
to
bug-grep <at> gnu.org:
bug#77301; Package
grep.
(Fri, 28 Mar 2025 21:54:02 GMT)
Full text and
rfc822 format available.
Message #25 received at 77301 <at> debbugs.gnu.org (full text, mbox):
On 3/28/25 11:14, David G. Pickett via Bug reports for GNU grep wrote:
> there was a specific clause in the C standard
> for that case, saying that it was OK
That clause is still there, and so the code is valid C code. However,
GCC warns about it anyway. We've changed grep (in savannah master) so
that GCC no longer warns.
Information forwarded
to
bug-grep <at> gnu.org:
bug#77301; Package
grep.
(Sat, 29 Mar 2025 17:07:02 GMT)
Full text and
rfc822 format available.
Message #28 received at 77301 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
Maybe put the compiler errors and warnings in a log and use a script to grep out the not previously identified worthless ones? Some warnings can be stifled to make cleaner code!
On Friday, March 28, 2025 at 05:53:18 PM EDT, Paul Eggert <eggert <at> cs.ucla.edu> wrote:
On 3/28/25 11:14, David G. Pickett via Bug reports for GNU grep wrote:
> there was a specific clause in the C standard
> for that case, saying that it was OK
That clause is still there, and so the code is valid C code. However,
GCC warns about it anyway. We've changed grep (in savannah master) so
that GCC no longer warns.
[Message part 2 (text/html, inline)]
Information forwarded
to
bug-grep <at> gnu.org:
bug#77301; Package
grep.
(Sat, 29 Mar 2025 17:16:02 GMT)
Full text and
rfc822 format available.
Message #31 received at 77301 <at> debbugs.gnu.org (full text, mbox):
On 3/29/25 11:06, David G. Pickett wrote:
> Maybe put the compiler errors and warnings in a log and use a script to grep out the not previously identified worthless ones?
We typically do better than that, by telling the compiler to not
generate the false alarms. Anyway this is is moot for this particular
false alarm as we've redone the code.
This bug report was last modified 26 days ago.
Previous Next
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.