I should mention too: I plan to eventually add a generated function
"item__init(Item *)" to initialize a message. i plan to add this when
i fix default values for non-fundamental types (which are currently
ignored).
(i haven't had a huge amount of time to deal with these few niggling
issues, but i've now got the design in my head, and the coding should
be pretty easy)
- dave
On Nov 27, 7:42 pm, daveb <[EMAIL PROTECTED]> wrote:
> The reason why it is crashing is that all messages have a few builtin
> members that are usually initialized via ITEM__INIT. If you look at
> the generated Item it looks like:
> struct _Item
> {
> ProtobufCMessage base_message;
> uint32_t id;
> };
> The best way to construct a message is to use C's assignment
> operator. Rewriting your code strictly:
> LookupReply nlr = LOOKUP_REPLY__INIT;
> Item item = ITEM__INIT;
>
> (&nlr)->replica_list = (Item**)malloc(1*sizeof(Item*));
> (&nlr)->n_replica_list = 1;
>
> Item* item1 = (Item*)malloc(sizeof(Item));
> item.id = 1;
> *item1 = item;
> (&nlr)->replica_list[0] = item1;
>
> int len = lookup_reply__get_packed_size(nlr);
>
> But that's kind of overkill in a way, it's shorter just to write:
> LookupReply nlr = LOOKUP_REPLY__INIT;
> Item item1 = ITEM__INIT;
> Item *items[1] = { &item };
>
> nlr.n_replica_list = 1;
> nlr.replica_list = items;
> item1.id = 1;
> int len = lookup_reply__get_packed_size(&nlr);
> Furthermore, using the latter approach, you can avoid malloc(), and
> therefore (most of) the possibility of memory leaks.
>
> On Nov 25, 8:54 am, [EMAIL PROTECTED] wrote:
>
> > Hi,
>
> > I'm using the protocol buffers package for c. It's very useful and
> > handy. However I've encountered a problem when trying to use a message
> > with a nested repeated field of an other message type.
>
> > This is the simplified version -
> > message Item {
> > int32 id = 1;
>
> > }
>
> > message LookupReply {
> > repeated Item replica_list = 1;
>
> > }
>
> > Here is the code for constructing the message:
> > ...
> > LookupReply nlr = LOOKUP_REPLY__INIT;
>
> > (&nlr)->replica_list = (Item**)malloc(1*sizeof(Item*));
> > (&nlr)->n_replica_list = 1;
>
> > Item* item1 = (Item*)malloc(sizeof(Item));
> > item1->id = 1;
> > (&nlr)->replica_list[0] = item1;
>
> > int len = lookup_reply__get_packed_size(nlr);
> > ...
>
> > The call to lookup_reply__get_packed_size causes a segmentation
> > fault.
> > I first tried using a string id field in the Item message, and that
> > failed too.
>
> > However when I used a string repeated field (with a few obvious
> > semantic changes) everything worked fine.
>
> > Anyone else has maybe used repeated message field in the c package and
> > got it to work (or if not, can explain why it doesn't work)?
>
> > Thank you,
> > Aviad
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Protocol Buffers" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/protobuf?hl=en
-~----------~----~----~----~------~----~------~--~---
