Re: [PATCH v3 4/7] gdb/python: allow ref_ptr<T, Policy>::new_reference to accept subclasses of T

[Matthieu Longo <matthieu.longo@arm.com>]

On 09/03/2026 19:40, Tom Tromey wrote:
>>>>>> Matthieu Longo <matthieu.longo@arm.com> writes:
> 
>> When ref_ptr<T,Policy>::new_reference() is specialized for 'PyObject'
>> (i.e. gdbpy_ref<>), it currently requires the argument type to be exactly
>> 'PyObject *'. As a result, pointers to subclasses of 'PyObject' must be
>> explicitly cast before being passed, making call sites unnecessarily
>> verbose.
> 
>> This patch makes ref_ptr<T,Policy>::new_reference() a template method
>> that accepts both T and subclasses of T, performing the cast to 'T *'
>> internally when needed. This removes redundant casts at call sites
>> without changing behavior.
> 
> This seems fine but I noticed something weird.
> 
> Approved-By: Tom Tromey <tom@tromey.com>
> 
>>     block_object *result = (block_object *) htab_find_with_hash (table, block,
>>   							       hash);
>>     if (result != nullptr)
>> -    return gdbpy_ref<>::new_reference ((PyObject *) result);
>> +    return gdbpy_ref<>::new_reference (result);
>   
> ... this part makes sense to me.
> 
>>     for (const thread_map_t::value_type &entry : *inf_obj->threads)
>>       {
>> -      auto thr = gdbpy_ref<>::new_reference ((PyObject *) entry.second.get ());
>> +      auto thr = gdbpy_ref<>::new_reference (entry.second.get ());
> 
> I guess it isn't super important but new_reference isn't really needed
> for a gdbpy_ref<> since we already have a copy constructor.  So just
> 
>      gdbpy_ref<> thr = entry.second;
> 
> ... should do it?
> 

I hadn't paid attention that entry.second was a gdbpy_ref<>.
Changed to what you suggested.

Matthieu

> Perhaps an explicit new_reference is more clear.
> 
> Tom