[LTP] [PATCH 2/3] lib: Add TST_ASSERT_FILE_INT and TST_ASSERT_FILE_STR

[Yang Xu]

Hi Cyril


> Hi!
>>   #endif /* TST_ASSERT_H__ */
>> diff --git a/lib/tst_assert.c b/lib/tst_assert.c
>> index 8ef3cd72e..65ee76473 100644
>> --- a/lib/tst_assert.c
>> +++ b/lib/tst_assert.c
>> @@ -4,6 +4,7 @@
>>    * Author: Yang Xu <xuyang2018.jy@cn.fujitsu.com>
>>    * Copyright (c) 2020 Cyril Hrubis <chrubis@suse.cz>
>>    */
>> +#include <stdio.h>
>>   #define TST_NO_DEFAULT_MAIN
>>   #include "tst_assert.h"
>>   #include "tst_test.h"
>> @@ -22,6 +23,22 @@ void tst_assert_int(const char *file, const int lineno, const char *path, int va
>>   	tst_res_(file, lineno, TFAIL, "%s != %d got %d", path, val, sys_val);
>>   }
>>   
>> +void tst_assert_file_int(const char *file, const int lineno, const char *path, const char *buf, int val)
>> +{
>> +	int sys_val;
>> +	char fmt[1024];
>> +
>> +	sprintf(fmt, "%s: %%d", buf);
> 
> If we want to keep the function as generic as possible we shouldn't add
> the colon and space after the %s here.
> 
> There is no standard on proc files, for instance it wouldn't be possible
> to parse /proc/vmstat if we hardcode the format string like that.
> 
> So I would just change this to "%s%%d" instead and pass "Foo: " instead
> just "Foo" in the testcases.
Yes, good suggestion!!
> 
> Also I guess that we should call it prefix rather than buf, but that's
> minor.
Agree. I will send a v2 patch.

Best Regards
Yang Xu
>