Now I'm even more confused.  In the past you had written this to a query of mine:
>On 01/12/2013 12:04 AM, Jonathan Wilkes wrote:
>>>>> In C would I just make a struct with fields of t_symbol,
>>>>>
>>>>> t_class, and a pointer to link to the next one?
>>>>
>>>>
>>>> Yeah, a linked list would work fine, probably not as efficient as >the c++ hash structure (but lots easier
>to maintain).  One nit-to-pick:  Use a t_class pointer, which is a >t_pd.
>>
>>
>> Hm... since the code to add new classes to the list will probably
>> end up looking exactly like the code to add symbols to the
>> symbol table, what if I just bloat the _symbol struct by adding
>> a t_class *s_class?  Would that affect performance?

>it would break binary compatibility.

But now you say the opposite in response to DesireData's _symbol struct which adds a refcount and a symbol size member "n".

How does the one break binary compatibility but the other does not?

-Jonathan


On Friday, September 26, 2014 1:48 AM, IOhannes m zmölnig <zmoelnig@iem.at> wrote:


On 09/26/2014 04:22 AM, Jonathan Wilkes via Pd-list wrote:
> On 09/25/2014 12:54 PM, Jonathan Wilkes via Pd-list wrote:
>> Um... have you actually read the source for DesireData?
>
> Just to clarify this-- from m_pd.h desiredata 2010.01.05:
>
> struct _symbol {
>    char *name;          /* the const string that represents this
> symbol */
>    t_pd *thing;          /* pointer to the target of a receive-symbol
> or to the bindlist of several targets */
>    struct _symbol *next; /* brochette of all symbols (only for
> permanent symbols) */
>    size_t refcount;      /* refcount<0 means that the symbol is
> permanent */
>    size_t n;            /* size of name (support for NUL characters) */
> #ifdef PD_PLUSPLUS_FACE
>    bool operator == (const char *s) const {return
> strcmp(this->name,s)==0;}
>    bool operator != (const char *s) const {return strcmp(this->name,s);}
> #endif
> };
>
> Desiredata's t_symbol has extra members that aren't in Pd Vanilla's
> t_symbol struct.  If there is any external out there that uses an array
> of symbols, then there will be problems due to this binary compatibility.
>

actually, i have yet to come across a *single* external that uses
(t_symbol) rather than (t_symbol*) - or, if you insist on arrays
(t_symbol[]) rather than (t_symbol*[]).

i don't see how this breaks binary compatibility - unless of course you
*use* these members¹...

fmgdsr
IOhannes


¹ that is, pass them around, in a "dosomething(s->foo)" sort of way (and
i don't know how to do this with an overloaded operator).
since the additional members are actually methods with an implementation
in the header file, i guess that any compiler would just inline them
when it comes to using them (in an "s->foo(z)" sort of way), rather than
forcing a resolving via dynamic lookup.



_______________________________________________
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list