Re: [PATCH] Improve Configuration Checking and Parsing

From: Tianyin Xu <tixu_at_cs.ucsd.edu>
Date: Tue, 15 Jan 2013 22:47:22 -0800

Thanks, Amos.

What do you mean "Just check that unrelated changes it finds caused by
other peoples bad patches are left out of your patch."?

What should I do at my side?

T

On Tue, Jan 15, 2013 at 7:37 PM, Amos Jeffries <squid3_at_treenet.co.nz> wrote:
> On 16/01/2013 2:01 p.m., Tianyin Xu wrote:
>>
>> Sounds great! Thanks a lot, Alex! I'll take care about the formatting
>> issues next time.
>
>
> Best way is to get the particular version of astyle which the squid-cache
> server uses and run scripts/source-maintenance.sh on your checkout prior to
> submitting. That will catch most of the style issues and generate any
> automated changes such as profiler and debugs info files which may be
> changed by your patch. Just check that unrelated changes it finds caused by
> other peoples bad patches are left out of your patch.
>
> Amos
>
>
>>
>> Best,
>> T
>>
>> On Mon, Jan 14, 2013 at 11:17 AM, Alex Rousskov
>> <rousskov_at_measurement-factory.com> wrote:
>>>
>>> On 01/14/2013 11:55 AM, Tianyin Xu wrote:
>>>
>>>> BTW, how to identify these TAB and WHITESPACE issues by search the
>>>> patch?
>>>
>>> There are many ways. I usually search for the tab character and trailing
>>> space (" $") when looking at the patch file with "less" before sending
>>> the patch file to the mailing list, but I suspect there are better ways
>>> to do this.
>>>
>>> Configuring your editor to insert spaces instead of a tab character may
>>> also be a good idea.
>>>
>>>
>>> Cheers,
>>>
>>> Alex.
>>>
>>>
>>>
>>>> On Fri, Jan 11, 2013 at 4:18 PM, Alex Rousskov
>>>> <rousskov_at_measurement-factory.com> wrote:
>>>>>
>>>>> On 01/11/2013 01:06 AM, Tianyin Xu wrote:
>>>>>>
>>>>>> Hi, Amos, Alex,
>>>>>>
>>>>>> Thanks a lot for your comments! They are really good advice.
>>>>>>
>>>>>> The updated patch is attached. It addresses all the issues you
>>>>>> mentioned except whether to deprecate the "enable"/"disable" options,
>>>>>> which is worth of discussion.
>>>>>>
>>>>>> Thanks,
>>>>>> Tianyin
>>>>>>
>>>>>>
>>>>>> On Thu, Jan 10, 2013 at 8:14 AM, Alex Rousskov
>>>>>> <rousskov_at_measurement-factory.com> wrote:
>>>>>>>
>>>>>>> On 01/05/2013 05:12 PM, Tianyin Xu wrote:
>>>>>>>
>>>>>>>> @@ -193,6 +290,8 @@
>>>>>>>> return false;
>>>>>>>> } else if ((port = strtol(token, &tmp, 10)), !*tmp) {
>>>>>>>> /* port */
>>>>>>>> + port = xatos(token);
>>>>>>>> +
>>>>>>>> } else {
>>>>>>>> host = token;
>>>>>>>> port = 0;
>>>>>>>
>>>>>>> Please do not set "port" twice because it is confusing and the comma
>>>>>>> operator in the expression only makes things worse. Do something like
>>>>>>> this instead:
>>>>>>>
>>>>>>> else if (strtol(token, &tmp, 10) && !*tmp) {
>>>>>>> port = xatos(token);
>>>>>>> }
>>>>>
>>>>>
>>>>>
>>>>>> + } else if (strtol(token, &tmp, 10), !*tmp) {
>>>>>> /* port */
>>>>>> + port = xatos(token);
>>>>>> +
>>>>>> } else {
>>>>>
>>>>> Please replace the comma operator with the "&&" operator.
>>>>>
>>>>> Please remove the empty line added in the patch.
>>>>>
>>>>> Please remove the /* port */ comment as no longer needed.
>>>>>
>>>>>
>>>>>> - } else if ((port = strtol(token, &junk, 10)), !*junk) {
>>>>>> + } else if (strtol(token, &junk, 10), !*junk) {
>>>>>> /* port */
>>>>>> + port = xatos(token);
>>>>>
>>>>> Same comments apply here, except there is no extra empty line to
>>>>> remove.
>>>>>
>>>>>
>>>>>> while (port && i < WCCP2_NUMPORTS) {
>>>>>> - p = strtol(port, &end, 0);
>>>>>> + p = xatoi(port);TABHERE
>>>>>>
>>>>>> +unsigned int
>>>>>> +xatoui(const char *token)
>>>>>> +{
>>>>>> + int64_t input = xatoll(token, 10);
>>>>>> + if (input < 0) {
>>>>>> +TABHEREdebugs(0, DBG_PARSE_NOTE(DBG_IMPORTANT), "ERROR: The input
>>>>>> value '" << token << "' cannot be less than 0.");
>>>>>> + self_destruct();
>>>>>> + }
>>>>>
>>>>> Please replace the leading tab with spaces and remove trailing
>>>>> whitespace. BTW, such things are easy to find by searching the patch.
>>>>>
>>>>> The above polishing touches can be done during commit IMO if you do not
>>>>> want to resubmit.
>>>>>
>>>>>
>>>>> Thanks a lot,
>>>>>
>>>>> Alex.
>>>>>
>>>>
>>>>
>>
>>
>

-- 
Tianyin XU,
http://cseweb.ucsd.edu/~tixu/
Received on Wed Jan 16 2013 - 06:47:30 MST

This archive was generated by hypermail 2.2.0 : Wed Jan 16 2013 - 12:00:06 MST