From: Phillip Wood <phillip.wood123@gmail.com>
Date: Wed, 30 Nov 2022 10:28:17 GMT
Subject: Re: [PATCH v4 1/5] run-command: add duplicate_output_fn to run_processes_parallel_opts
Message-ID: <5f6f0181-c251-0a98-a39f-a910b93f8c1a@dunelm.org.uk>
In-Reply-To: <221130.86lensiwy0.gmgdl@evledraar.gmail.com>

On 30/11/2022 09:53, Ævar Arnfjörð Bjarmason wrote:
> 
> On Tue, Nov 29 2022, Glen Choo wrote:
> 
>> Calvin Wan <calvinwan@google.com> writes:
>> So this might be more idiomatically written as:
>>
>>          int n = strbuf_read_once(&pp->children[i].err,
>>          			 pp->children[i].process.err, 0);
>>   +      if (opts->duplicate_output) {
>>   +          struct strbuf buf = STRBUF_INIT;
>>   +          strbuf_addbuf(&buf, &pp->children[i].err);
>>   +        	opts->duplicate_output(&buf, &pp->children[i].err,
>>   +        		  opts->data,
>>   +        		  pp->children[i].data);
>>   +          strbuf_release(&buf);
>>   +      }
>>
>> which also has the nice benefit of not touching the strbuf_read_once()
>> line.
> 
> We should also use "size_t n" there, not "int n", which is what it
> returns.

It returns an ssize_t not size_t, lower down we test `n < 0` so we 
certainly should not be using an unsigned type.

Best Wishes

Phillip