On 02/25/2011 11:00 AM, Lyre wrote:
于 2011年02月25日 17:08, Michal Novotny 写道:
> On 02/25/2011 07:23 AM, Lyre wrote:
>> * tools/generate-api-docs.c: add initialize for private function names
>> ---
>> tools/generate-api-docs.c | 13 +++++++++----
>> 1 files changed, 9 insertions(+), 4 deletions(-)
>>
>> diff --git a/tools/generate-api-docs.c b/tools/generate-api-docs.c
>> index 322f3ef..8a16e22 100644
>> --- a/tools/generate-api-docs.c
>> +++ b/tools/generate-api-docs.c
>> @@ -43,10 +43,15 @@ void parse_comment(char *line, int func_num, int
>> *arg_num)
>> if (strncmp(ltrimmed, "Function name:", 14) == 0) {
>> functions[func_num].name = strdup( ltrim_string( ltrimmed + 14) );
>> }
>> - else
>> - if (strncmp(ltrimmed, "Description:", 12) == 0) {
>> - functions[func_num].desc = strdup( ltrim_string( ltrimmed + 12) );
>> - }
>> + else
>> + if (strncmp(ltrimmed, "Private function name:", 22) == 0)
>> + {
>> + functions[func_num].name = NULL;
>> + }
>> + else
>> + if (strncmp(ltrimmed, "Description:", 12) == 0) {
>> + functions[func_num].desc = strdup( ltrim_string( ltrimmed + 12) );
>> + }
>> else
>> if (strncmp(ltrimmed, "Arguments:", 10) == 0) {
>> char *str = ltrim_string(ltrimmed + 11);
> Thanks for your patch Lyre but I've already chosen and implemented
> another way of doing this to set the private flag in the func_t
> structure. Please see commit 851039c8. My approach could be useful to
> generate some internal (private) API guide that could be used to
> generate a libvirt-php bindings developer guide page if we decide to
> have it one day.
>
> In fact what do you think about this? Is it good to have such a guide
> one day ? I think it could be however not for just 3 private
> functions. I guess there will be more private functions one day.
>
> Thanks,
> Michal
>
Got it. Before your patch, I encountered SIGSEGV on x86_64 platform, so
I fixed it. I'm not sure your design and just try to make it work.
I believe it is a good to mark up private or public APIs.
Also, did you notice another patch from me yesterday? It modified many
files, seems not a good practice. Next time, should I Cc the patch to
you as well?
Well, it's better to put me to Cc. Also, did you have a look whether the
patch is committed to the Git repo or not?
If not, could you please resend it privately just to my e-mail now?
Thanks,
Michal
--
Michal Novotny<minovotn(a)redhat.com>, RHCE
Virtualization Team (xen userspace), Red Hat