On Sun, Nov 7, 2010 at 5:18 PM, Jeff Trawick <[email protected]> wrote:
> On Sun, Nov 7, 2010 at 5:07 PM, Sander Temme <[email protected]> wrote:
>>
>> On Nov 4, 2010, at 12:15 PM, William A. Rowe Jr. wrote:
>>
>>> Looks good here.
>>
>> If folks find this unproblematic, can someone please commit it? I don't
>> have karma here.
>>
>> Thanks,
>
> looks fine to me; starting to try it out now
The patch is apparently not needed to make %lld work in 32-bit mode on
Leopard since APR_INT64_T_FMT is %lld.
It didn't make %lld work in 64-bit mode.
What are the cases where you see this help out?
I guess you're on Snow Leopard?
Here's what I tried in 64-bit mode:
Index: strings/apr_snprintf.c
===================================================================
--- strings/apr_snprintf.c (revision 1032408)
+++ strings/apr_snprintf.c (working copy)
@@ -832,6 +832,11 @@
else if (*fmt == 'l') {
var_type = IS_LONG;
fmt++;
+ /* Catch the %lld type modifier for long long and its ilk */
+ if (*fmt == 'l') {
+ var_type = IS_QUAD;
+ fmt++;
+ }
}
else if (*fmt == 'h') {
var_type = IS_SHORT;
ndex: testfmt.c
===================================================================
--- testfmt.c (revision 1032408)
+++ testfmt.c (working copy)
@@ -126,6 +126,9 @@
apr_snprintf(buf, sizeof buf, "%" APR_INT64_T_FMT, ibig);
ABTS_STR_EQUAL(tc, "-314159265358979323", buf);
+
+ apr_snprintf(buf, sizeof buf, "%lld", ibig);
+ ABTS_STR_EQUAL(tc, "-314159265358979323", buf);
}
static void error_fmt(abts_case *tc, void *data)
./testall -v testfmt
testfmt : /Line 131: expected <-314159265358979323>, but saw <%ld>
FAILED 1 of 10
Failed Tests Total Fail Failed %
===================================================
testfmt 10 1 10.00%
