search for: mapping_matrix_init

thanks for all your feedback. here's the revised patch: On Wed, Oct 11, 2017 at 2:20 PM Timothy B. Terriberry <tterribe at xiph.org> wrote: > Jean-Marc Valin wrote: > > I think you'll want something like: > > (opus_int16)((unsigned)demixing_matrix[2*i+1] << 8) > > (though you might want to check it too) > > FWIW, we use the construct > int s =

...ems even earlier in the code? If not, then maybe an > assertion would be better. Also, is the size of the matrix the only > thing you can validate? > > Removed it. You're right, it's functionally impossible for the user to create an invalid matrix with the current API. > 6) mapping_matrix_init() returns an error code, but the code that's > calling it never checks it. Considering that it's an internal call, I > would say it's probably better not to return anything, but instead to > assert in the function. > > Done. > 7) Same for mapping_matrix_multiply_shor...

...assertion would be better. Also, is the size of the matrix the only > > thing you can validate? > > > > Removed it. You're right, it's functionally impossible for the user to > > create an invalid matrix with the current API. > > > > > > 6) mapping_matrix_init() returns an error code, but the code that's > > calling it never checks it. Considering that it's an internal call, I > > would say it's probably better not to return anything, but instead to > > assert in the function. > > > > Done. > >...

...ure I understand the purpose of mapping_matrix_validate(). Can the user really cause the matrix not to be valid? If yes, then can't this cause problems even earlier in the code? If not, then maybe an assertion would be better. Also, is the size of the matrix the only thing you can validate? 6) mapping_matrix_init() returns an error code, but the code that's calling it never checks it. Considering that it's an internal call, I would say it's probably better not to return anything, but instead to assert in the function. 7) Same for mapping_matrix_multiply_short() and mapping_matrix_multiply_float...

...If not, then maybe an > assertion would be better. Also, is the size of the matrix the only > thing you can validate? > > Removed it. You're right, it's functionally impossible for the user to > create an invalid matrix with the current API. >   > > 6) mapping_matrix_init() returns an error code, but the code that's > calling it never checks it. Considering that it's an internal call, I > would say it's probably better not to return anything, but instead to > assert in the function. > > Done. >   > > 7) Same for...

...>     > > >     > Removed it. You're right, it's functionally impossible > for the user to > >     > create an invalid matrix with the current API. > >     >   > >     > > >     >     6) mapping_matrix_init() returns an error code, but > the code > >     that's > >     >     calling it never checks it. Considering that it's an > internal > >     call, I > >     >     would say it's probably better not to re...

...impossible > > > for the user to > > > > > create an invalid matrix with the current API. > > > > > > > > > > > > > > > 6) mapping_matrix_init() returns an error > > code, but > > > the code > > > > that's > > > > > calling it never checks it. Considering > > that it's an > > > intern...

...39;s functionally > impossible > >         for the user to > >         >     > create an invalid matrix with the current API. > >         >     >   > >         >     > > >         >     >     6) mapping_matrix_init() returns an error > code, but > >         the code > >         >     that's > >         >     >     calling it never checks it. Considering > that it's an > >         internal > >         &...

...gt;> > > for the user to >> > > > > create an invalid matrix with the current API. >> > > > > >> > > > > >> > > > > 6) mapping_matrix_init() returns an error >> > code, but >> > > the code >> > > > that's >> > > > > calling it never checks it. Considering >> > that it's an >> >...

...   for the user to > >         >         >     > create an invalid matrix with the > current API. > >         >         >     >   > >         >         >     > > >         >         >     >     6) mapping_matrix_init() returns > an error > >         code, but > >         >         the code > >         >         >     that's > >         >         >     >     calling it never checks it. > Considering > &...