[PATCH v2 6/6] d3dcompiler: Return an ID3D10Blob from parse_hlsl().

[Zebediah Figura]

On 6/11/20 3:04 PM, Matteo Bruni wrote:
> On Mon, Jun 8, 2020 at 11:50 PM Zebediah Figura <z.figura12 at gmail.com> wrote:
>>
>> Signed-off-by: Zebediah Figura <zfigura at codeweavers.com>
>> ---
>>  dlls/d3dcompiler_43/compiler.c            | 49 ++++++-----------------
>>  dlls/d3dcompiler_43/d3dcompiler_private.h |  4 +-
>>  dlls/d3dcompiler_43/hlsl.l                | 14 +++----
>>  dlls/d3dcompiler_43/hlsl.y                | 10 +++--
>>  dlls/d3dcompiler_43/tests/hlsl_d3d9.c     |  2 +-
>>  5 files changed, 29 insertions(+), 50 deletions(-)
>>
>> diff --git a/dlls/d3dcompiler_43/compiler.c b/dlls/d3dcompiler_43/compiler.c
>> index 9fc7c1d7734..78fb0c46cc0 100644
>> --- a/dlls/d3dcompiler_43/compiler.c
>> +++ b/dlls/d3dcompiler_43/compiler.c
>> @@ -752,12 +752,11 @@ static const struct target_info * get_target_info(const char *target)
>>  }
>>
>>  static HRESULT compile_shader(const char *preproc_shader, const char *target, const char *entrypoint,
>> -        ID3DBlob **shader_blob, ID3DBlob **error_messages)
>> +        ID3DBlob **shader, ID3DBlob **error_messages)
>>  {
>> -    struct bwriter_shader *shader;
>> +    DWORD size, major, minor;
>>      char *messages = NULL;
>>      HRESULT hr;
>> -    DWORD *res, size, major, minor;
>>      ID3DBlob *buffer;
>>      char *pos;
>>      enum shader_type shader_type;
>> @@ -787,7 +786,7 @@ static HRESULT compile_shader(const char *preproc_shader, const char *target, co
>>          }
>>      }
>>
>> -    shader = parse_hlsl_shader(preproc_shader, shader_type, major, minor, entrypoint, &messages);
>> +    hr = parse_hlsl_shader(preproc_shader, shader_type, major, minor, entrypoint, shader, &messages);
> 
> Nothing really new, but it feels a bit awkward that
> parse_hlsl_shader() is supposed to generate the complete shader. I
> think it's just a matter of naming, replacing parse_hlsl_shader() and
> parse_hlsl() with e.g. compile_hlsl_shader() / compile_hlsl() is
> probably all there is to it.
> 

Agreed; I'll queue a follow-up patch for that.