You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@subversion.apache.org by hw...@apache.org on 2011/05/23 15:11:10 UTC
svn commit: r1126481 - /subversion/trunk/subversion/libsvn_subr/svn_base64.c
Author: hwright
Date: Mon May 23 13:11:10 2011
New Revision: 1126481
URL: http://svn.apache.org/viewvc?rev=1126481&view=rev
Log:
Remove some integer size mismatch warnings by using size_t in places where
we calculate pointer differences.
* subversion/libsvn_subr/svn_base64.c
(encode_baton, encode_bytes, encode_partial_group, svn_base64_encode_string2,
base64_from_checksum): Use size_t in place of int.
Modified:
subversion/trunk/subversion/libsvn_subr/svn_base64.c
Modified: subversion/trunk/subversion/libsvn_subr/svn_base64.c
URL: http://svn.apache.org/viewvc/subversion/trunk/subversion/libsvn_subr/svn_base64.c?rev=1126481&r1=1126480&r2=1126481&view=diff
==============================================================================
--- subversion/trunk/subversion/libsvn_subr/svn_base64.c (original)
+++ subversion/trunk/subversion/libsvn_subr/svn_base64.c Mon May 23 13:11:10 2011
@@ -46,8 +46,8 @@ static const char base64tab[] = "ABCDEFG
struct encode_baton {
svn_stream_t *output;
unsigned char buf[3]; /* Bytes waiting to be encoded */
- int buflen; /* Number of bytes waiting */
- int linelen; /* Bytes output so far on this line */
+ size_t buflen; /* Number of bytes waiting */
+ size_t linelen; /* Bytes output so far on this line */
apr_pool_t *scratch_pool;
};
@@ -74,7 +74,7 @@ encode_group(const unsigned char *in, ch
STR. Include newlines every so often if BREAK_LINES is true. */
static void
encode_bytes(svn_stringbuf_t *str, const void *data, apr_size_t len,
- unsigned char *inbuf, int *inbuflen, int *linelen,
+ unsigned char *inbuf, size_t *inbuflen, size_t *linelen,
svn_boolean_t break_lines)
{
char group[4];
@@ -118,7 +118,7 @@ encode_bytes(svn_stringbuf_t *str, const
LEN must be in the range 0..2. */
static void
encode_partial_group(svn_stringbuf_t *str, const unsigned char *extra,
- int len, int linelen, svn_boolean_t break_lines)
+ size_t len, size_t linelen, svn_boolean_t break_lines)
{
unsigned char ingroup[3];
char outgroup[4];
@@ -204,7 +204,8 @@ svn_base64_encode_string2(const svn_stri
svn_stringbuf_t *encoded = svn_stringbuf_create("", pool);
svn_string_t *retval = apr_pcalloc(pool, sizeof(*retval));
unsigned char ingroup[3];
- int ingrouplen = 0, linelen = 0;
+ size_t ingrouplen = 0;
+ size_t linelen = 0;
encode_bytes(encoded, str->data, str->len, ingroup, &ingrouplen, &linelen,
break_lines);
@@ -396,7 +397,8 @@ base64_from_checksum(const svn_checksum_
{
svn_stringbuf_t *checksum_str;
unsigned char ingroup[3];
- int ingrouplen = 0, linelen = 0;
+ size_t ingrouplen = 0;
+ size_t linelen = 0;
checksum_str = svn_stringbuf_create("", pool);
encode_bytes(checksum_str, checksum->digest,
Re: svn commit: r1126481 - /subversion/trunk/subversion/libsvn_subr/svn_base64.c
Posted by Greg Stein <gs...@gmail.com>.
On Mon, May 23, 2011 at 09:29, Hyrum K Wright <hy...@hyrumwright.org> wrote:
> On Mon, May 23, 2011 at 6:16 AM, Greg Stein <gs...@gmail.com> wrote:
>>
>> On Mon, May 23, 2011 at 09:11, <hw...@apache.org> wrote:
>> > Author: hwright
>> > Date: Mon May 23 13:11:10 2011
>> > New Revision: 1126481
>> >
>> > URL: http://svn.apache.org/viewvc?rev=1126481&view=rev
>> > Log:
>> > Remove some integer size mismatch warnings by using size_t in places
>> > where
>> > we calculate pointer differences.
>>
>> Strictly speaking, it is best to use apr_size_t for those old
>> platforms where size_t is not available.
>>
>> That said, I doubt we're running on them nowadays...
>
> This came up in IRC just now, and we did the following on the 1.6.x branch:
> $ find . -name '*.c' | xargs grep -w 'size_t' | wc -l
> 219
> $
> It appears we've been releasing code with size_t for quite some time, and we
> haven't heard any complains, so I think we're safe. :)
"Strictly speaking" :-)
(and its also why pocore doesn't go down that nonsense road...)
Cheers,
-g
Re: svn commit: r1126481 - /subversion/trunk/subversion/libsvn_subr/svn_base64.c
Posted by Hyrum K Wright <hy...@hyrumwright.org>.
On Mon, May 23, 2011 at 6:16 AM, Greg Stein <gs...@gmail.com> wrote:
> On Mon, May 23, 2011 at 09:11, <hw...@apache.org> wrote:
> > Author: hwright
> > Date: Mon May 23 13:11:10 2011
> > New Revision: 1126481
> >
> > URL: http://svn.apache.org/viewvc?rev=1126481&view=rev
> > Log:
> > Remove some integer size mismatch warnings by using size_t in places
> where
> > we calculate pointer differences.
>
> Strictly speaking, it is best to use apr_size_t for those old
> platforms where size_t is not available.
>
> That said, I doubt we're running on them nowadays...
This came up in IRC just now, and we did the following on the 1.6.x branch:
$ find . -name '*.c' | xargs grep -w 'size_t' | wc -l
219
$
It appears we've been releasing code with size_t for quite some time, and we
haven't heard any complains, so I think we're safe. :)
-Hyrum
Re: svn commit: r1126481 - /subversion/trunk/subversion/libsvn_subr/svn_base64.c
Posted by Greg Stein <gs...@gmail.com>.
On Mon, May 23, 2011 at 09:11, <hw...@apache.org> wrote:
> Author: hwright
> Date: Mon May 23 13:11:10 2011
> New Revision: 1126481
>
> URL: http://svn.apache.org/viewvc?rev=1126481&view=rev
> Log:
> Remove some integer size mismatch warnings by using size_t in places where
> we calculate pointer differences.
Strictly speaking, it is best to use apr_size_t for those old
platforms where size_t is not available.
That said, I doubt we're running on them nowadays...
Cheers,
-g