[prev in list] [next in list] [prev in thread] [next in thread] 

List:       apr-dev
Subject:    Re: svn commit: r1835979 - in /apr/apr/trunk: include/apr_json.h json/apr_json.c json/apr_json_decod
From:       Yann Ylavic <ylavic.dev () gmail ! com>
Date:       2018-07-16 9:47:21
Message-ID: CAKQ1sVM-+05y5DkhS7bH3_AMM87cCsYCEk4YFhWMEWuLbCbWOg () mail ! gmail ! com
[Download RAW message or body]

On Sun, Jul 15, 2018 at 4:00 PM,  <minfrin@apache.org> wrote:
>
> ==============================================================================
> --- apr/apr/trunk/include/apr_json.h (original)
> +++ apr/apr/trunk/include/apr_json.h Sun Jul 15 14:00:04 2018
> @@ -101,8 +101,8 @@ typedef enum apr_json_type_e {
>  typedef struct apr_json_string_t {
>      /** pointer to the string */
>      const char *p;
> -    /** string length */
> -    apr_size_t len;
> +    /** string length, or APR_JSON_VALUE_STRING to compute length automatically */
> +    apr_ssize_t len;
>  } apr_json_string_t;

Can an apr_json_string_t have APR_JSON_VALUE_STRING in its 'len' field?
I suppose apr_json_string_create() can take APR_JSON_VALUE_STRING to
compute the length by itself, but finally each constructed or parsed
apr_json_string_t should have the real 'len' right?

Regards,
Yann.
[prev in list] [next in list] [prev in thread] [next in thread]