Re: large integer truncation in regex module

2012-05-28 Thread Paul Eggert
On 05/28/2012 09:26 PM, Steven M. Schweda wrote: > any environment-specific code really needs to be tested in > the specific environment Sure, that's important, but it's also important to minimize the number of problems that can slip through and be caught only by testing. Our tests are not exhaus

Re: large integer truncation in regex module

2012-05-28 Thread Steven M. Schweda
From: Paul Eggert > [...] > It's maintainability. For example, if a typo had > been inadvertently put into the old 32-bit code > (a misspelled identifier, say, due to someone typing > a character by mistake), it would not have been caught > by someone compiling on the 64-bit platform. But with >

Re: test-system-quote-child.c build error on CentOS 5

2012-05-28 Thread Paul Eggert
On 05/28/2012 08:25 AM, Tom G. Christensen wrote: > As a workaround I've added $(LDADD) to test_system_quote_child_LDADD in > modules/system-quote-tests. That doesn't sound right, given the comments there. I don't understand the situation fully and do not have the patience to run that test, but d

Re: large integer truncation in regex module

2012-05-28 Thread Paul Eggert
On 05/28/2012 08:38 AM, Steven M. Schweda wrote: > I'm left wondering whether, say, "var = con1 << 31 << 1 | con0" offers > so great an improvement in clarity It's not simply clarity (which is a matter of style). It's maintainability. For example, if a typo had been inadvertently put into the old

Re: [PATCH] manywarnings: remove duplicate -Wmultichar entry

2012-05-28 Thread Jim Meyering
Paul Eggert wrote: > I noticed that Emacs's configure script was checking for > -Wmultichar twice and tracked it down to a gnulib inefficiency, > fixed as follows. ... > diff --git a/m4/manywarnings.m4 b/m4/manywarnings.m4 ... > @@ -143,7 +143,6 @@ AC_DEFUN([gl_MANYWARN_ALL_GCC], >for gl_manywa

Re: large integer truncation in regex module

2012-05-28 Thread Steven M. Schweda
From: Paul Eggert > Any decent compiler will generate exactly the same > code for both. No shifts or OR will be done at > runtime -- they'll all be done at compile time. > For example: > [...] Yes, while I tend to underestimate them, I do understand compile-time expression evaluation and dead

Re: test-system-quote-child.c build error on CentOS 5

2012-05-28 Thread Tom G. Christensen
On 05/22/2012 05:42 PM, Tom G. Christensen wrote: Hello, Since May 9 my gnulib daily build running on CentOS 5 has failed to complete like this: gcc -std=gnu99 -g -O2 -Wall -o test-system-quote-child test-system-quote-child.o -lm -lm -lm -lm -lm -lm -lm -lm -lm -lm -lm test-system-quote-child.o