Re: [protobuf] How to use SetString reflection method

The setters need a (non-const) pointer to the message object. It appears fractal_handle is a reference since the getter, which takes a const reference to a message, compiles. On Fri, Aug 6, 2010 at 4:42 AM, mark.t.macdon...@googlemail.com < mark.t.macdon...@googlemail.com> wrote: > Can you please

Re: [protobuf] Re: Java implementation questions

Groups are primarily deprecated because it used to be the case that groups could not be reused in other types. Now groups can be reused, but other messages must use the length-delimited format, rather than the group format. In theory, the fact that groups do not require a length prefix makes them

Re: [protobuf] Please add j2me implementation in ThirdPartAddOns page

Added to the wiki. On Thu, Jul 29, 2010 at 7:45 PM, Igor Gatis wrote: > Here it is: > > http://code.google.com/p/protobuf-j2me/ > > -- > You received this message because you are subscribed to the Google Groups > "Protocol Buffers" group. > To post to this group, send email to proto...@googlegro

Re: [protobuf] ByteString example

What are you trying to do? Does the Javadoc help? http://code.google.com/apis/protocolbuffers/docs/reference/java/com/google/protobuf/ByteString.html On Fri, Jul 30, 2010 at 8:07 AM, David wrote: > Hi, > I am a bit ashamed to ask this, but does anyone know how to use > ByteString in protocol buf

Re: [protobuf] libcurl integrtion

Not at all familiar with libcurl, but from lookng at man pages... On Wed, Jul 28, 2010 at 6:56 PM, Birch wrote: > I am not well versed in c++, just a disclamer. I have muddled through > creating a small client and I want to post the binary output via > HTTP. Here is the source I have so far: >

Re: [protobuf] Gzip

an error telling > me zlib.h could not be found and I was also wondering if I can use > GzipOutputStream along with CodedOutputStream? > > El 29 de julio de 2010 14:59, Jason Hsueh escribió: > > What exactly do you want to know about them? There are various details >> about

Re: [protobuf] Gzip

What exactly do you want to know about them? There are various details about how they adapt zlib to the ZeroCopy*Stream interfaces, but as a client you shouldn't need to worry about that. You can use them just like any other ZeroCopyStream. On Thu, Jul 29, 2010 at 8:58 AM, Julian González wrote:

Re: [protobuf] why ParseFromIstream using std::cin always returns true using a pipe?

Messages without any required fields are allowed to have an empty serialization, so the library cannot assume that parsing empty input is an error. You just need to test cin.eof() separately. On Wed, Jul 28, 2010 at 7:18 AM, jetcube wrote: > I have a simple app that received data using the stdin

Re: [protobuf] Extensions in Java

The C++ implementation does have a singleton registry, which allows it to magically parse extensions that are linked into the binary. This behavior can be somewhat confusing though since it relies on what the linker decides to do with generated code. The Java implementation uses an object oriented

Re: [protobuf] Building FileDescriptorSet out of FileDescriptor

epth-first. On Wed, Jul 21, 2010 at 11:38 PM, Nader Salehi wrote: > If any dependency needs to be build beforehand, then how is > topological ordering is supported? > > > On 7/21/2010 17:39 Jason Hsueh writes: > > Ordering is important because any dependencies need to be built

Re: [protobuf] Building FileDescriptorSet out of FileDescriptor

Ordering is important because any dependencies need to be built before they are used - if you give an arbitrary order there may be errors complaining about undefined symbols or imports not being found. A depth first ordering will work, and that's exactly what protoc does when producing FileDescript

Re: [protobuf] message->GetDescriptor() vs message->descriptor();

descriptor() is a static method only available in generated classes. If you have a pointer to a Message, there is no accessible descriptor() method, so generally you should use GetDescriptor(). On Thu, Jul 15, 2010 at 1:22 PM, mark.t.macdon...@googlemail.com < mark.t.macdon...@googlemail.com> wrot

Re: [protobuf] Protobuf as generic marshalling layer

On Tue, Jul 13, 2010 at 1:38 AM, Galder Zamarreno wrote: > Hi, > > I want to use protobuf as a generic marshalling mechanism and I'm > having some issues with the unmarshalling part, because I need > knowledge of what needs to be unmarshalled, i.e. in Java, access to > static parseFrom method in t

Re: [protobuf] Why do field default values not get serialised?

On Wed, Jul 14, 2010 at 1:47 PM, mark.t.macdon...@googlemail.com < mark.t.macdon...@googlemail.com> wrote: > I am new and want to understand the purpose/usage of default values in > the .proto file. E.g: > > test > { > optional string a 1 [default ="test"]; > } > > When I create a message object

Re: [protobuf] hasXXX methods

See http://code.google.com/apis/protocolbuffers/docs/cpptutorial.html under the section "The Protocol Buffer API" On Fri, Jul 2, 2010 at 11:08 AM, jd wrote: > Where are the hasXXX methods that are available in C++? > > -- > You received this message because you are subscribed to the Google Group

Re: [protobuf] custom protocol

The library provides a standard text format (TextFormat classes in C++ and Java, text_format.py for python). If that doesn't work for you, you could write your own serializers around reflection, much like TextFormat. On Wed, Jun 30, 2010 at 7:05 AM, Vladimir Danishevsky wrote: > Can I change the

Re: [protobuf] Re: using the new plugin infrastructure

Yes, you can write the plug-in in the language of your choice. On Wed, Jun 30, 2010 at 7:08 AM, qrilka wrote: > Is it possible to use other languages? > E.g. python? > > Kind regards, > Kirill Zaborsky > > On 2 июн, 20:25, Jason Hsueh wrote: > > Does the test plugin

Re: [protobuf] Re: Java: Deleting from lists in protobufs

On Wed, Jun 23, 2010 at 3:01 PM, Dheeraj Pandey wrote: > Java doesn't have such a thing. Is this by design? If yes, would love to > understand the rationale of why > clearing-everything-and-reinserting-(N-1)-values is so much better than: > (a) providing an API to do so, and > There is actually a

Re: [protobuf] Way to figure out the fields that are modified

On Mon, Jun 21, 2010 at 8:23 AM, Ananth Arockia wrote: > Hello, > > I am exploring protobuf to use it in my project. > > I would like to have a message defined in a .proto file. All the > fields in this message would be "optional" due to the fact that the > message producer in my project context w

Re: [protobuf] Python protobuf speed and latest Protobuf version

The last release, 2.3.0, includes significant improvments to Python: http://protobuf.googlecode.com/svn/trunk/CHANGES.txt There have also been further improvements since then (the C extension-based implementation mentioned in http://groups.google.com/group/protobuf/browse_thread/thread/6b5d6373b239

Re: [protobuf] question about SerializeToString output for cpp

Yes, '\0' may appear in the binary format. output->size() will return the correct result: C++ string can store null characters without any issue. However, strlen(output->c_str()) and other calls that assume null-terminated C strings will not. On Thu, Jun 10, 2010 at 11:30 PM, mistlike wrote: > b

Re: [protobuf] Options extensions for dynamic messages?

ons are defined. Rather convoluted, > but we don't have a better way. > > On Mon, Jun 7, 2010 at 3:25 PM, Jason Hsueh wrote: > >> Hm. Can you send the code snippet you are using to try to access your >> custom option? You should just be accessing the options proto from &g

Re: [protobuf] Options extensions for dynamic messages?

Hm. Can you send the code snippet you are using to try to access your custom option? You should just be accessing the options proto from FieldDescriptor::options(), and calling GetExtension() on that. Also, UnknownFields doesn't have anything to do with options extensions. Did you mean Uninterpret

Re: [protobuf] What is the default value for a message used as a field type within another message?

No, protobuf accessors never return null. For message fields, you get the default instance of the message type. On Fri, Jun 4, 2010 at 9:32 AM, Jax wrote: > > This should be pretty straightforward, but I couldn't find it in any > of the documentation on default values. > > A slight change from s

Re: [protobuf] Re: Questions/Ideas about Protobuf

Functionality -> Protobuf[I > have to implement stream by myself, but each stream packet is based on > protobuf. On the top,my app faces a lot of small stream packets in > protobuf]" > > Linkedin:http://www.linkedin.com/in/dirlt > > > 于 2010/6/4 0:21, Jason Hsueh 写道:

Re: [protobuf] Re: Questions/Ideas about Protobuf

o protobuf tells server that "Hi server,if you just > have 100MB memory,I can only hold 10,000 pageids" > > 4.so the server app knows it,so app just serialize 10,000 pageids into > memory instead of 1,000,000 pageids. > > I hope I clarify it now.. If the protobuf doesn&#

Re: [protobuf] Questions/Ideas about Protobuf

On Tue, Jun 1, 2010 at 6:21 AM, bnh wrote: > I'm using a protobuf as the protocol for a distributed system.But now > I > have some questions about protobuf > > a.Whether protobuf provides the inteface for user-defined allocator > because sometimes I find 'malloc' cost too much? I've tried TCmallo

Re: [protobuf] using the new plugin infrastructure

Does the test plugin help? There is a test code generator that is used in the unittest for command_line_interface.cc: http://code.google.com/p/protobuf/source/browse/trunk/src/google/protobuf/compiler/mock_code_generator.cc On Tue, Jun 1, 2010 at 4:00 AM, marcmo wrote: > hello, > i'm trying to m

Re: [protobuf] Protoc error on Linux Mint (protoc-gen-ccp: program not found)

On Wed, Jun 2, 2010 at 5:31 AM, mark.t.macdon...@googlemail.com < mark.t.macdon...@googlemail.com> wrote: > Hi, > > I'm running Linux Mint, and trying to get started with protobuf-2.3.0. > > My install was as follows (no problems reported that I could see): > > ./configure --prefix=/usr > make cle

Re: [protobuf] Implementing protobuf in symbian

Can you provide more details? What commands are you running, and what are your errors? In your previous thread it sounded like you were using C++, but now you mention Java - which language are you trying to use? Also, are you attempting to use the lite library or the full library? It seems most use

Re: [protobuf] Protocol buffers and large data sets

There is a default byte size limit of 64MB when parsing protocol buffers - if a message is larger than that, it will fail to parse. This can be configured if you really need to parse larger messages, but it is generally not recommended. Additionally, ByteSize() returns a 32-bit integer, so there's

Re: [protobuf] systesting custom utf8 validation on remote c++ node using protocol buffers from python

If you compile with the macro GOOGLE_PROTOBUF_UTF8_VALIDATION_ENABLED defined, the C++ code will do UTF8 validation. However, it doesn't prevent the data from serializing or parsing, it will simply log an error message. How would you like it to fail? On Mon, May 17, 2010 at 3:15 PM, JT Olds wrote

Re: [protobuf] Deadlock problems with protobuf static initialization in Java

Err. Got my mailing lists mixed up. I think this is due to http://bugs.sun.com/view_bug.do?bug_id=6501158 The suggested workaround has been to force initialization of your classes in main, before threads are spawned. On Sun, May 16, 2010 at 1:18 AM, Jason Hsueh wrote: > I think this is http:

Re: [protobuf] Deadlock problems with protobuf static initialization in Java

I think this is http://b/issue?id=2267627 On Sat, May 15, 2010 at 10:02 PM, Igor Gatis wrote: > Agreed. So I guess I need to start looking for the deadlock in my code. Do > you know any tool to help me out on that? > > > On Sun, May 16, 2010 at 1:38 AM, Christopher Smith wrote: > >> Never seen i

Re: [protobuf] Re: Can I add a new item into an existing list of enumerations safely?

erhead to using extensions; I think extensions become smaller when you have 8 fields or so. On Fri, May 14, 2010 at 5:44 PM, Jason Hsueh wrote: > It doesn't matter that much. I only set UNKNOWN = 2 on the assumption that > pop and push were already assigned and you needed to preserve wire

Re: [protobuf] Re: Can I add a new item into an existing list of enumerations safely?

as items after PUSH (meaning with values 2, 3 ... ). > > So, later we have: enum Type { POP = 0, PUSH = 1, TOP = 2, TAIL = 3, > UNKNOWN = 100 } and so on ... > > Or it really does not matter? > > Is this the recommended approach for this pattern? Or should I be > using extends? > >

Re: [protobuf] How to get rid of metadata only extensions dependencies

No, there isn't any way to remove the dependency. The compiler doesn't try to detect why you defined an import. On Fri, May 14, 2010 at 2:38 PM, Igor Gatis wrote: > So, let's say I'm developing a plugin which, say, add or removes builder > pattern to java generator (this is actually a real use c

Re: [protobuf] Can I add a new item into an existing list of enumerations safely?

Clients with the old binary will not be able to read messages containing the new enum value. This would get treated as an unknown field, meaning that the type field would not be sent in an older client. Since your field is also required, it would fail to parse. So, with this example, you would need

Re: [protobuf] Nested messages accessor methods

Yes, you can. What is your target language? The accessor names are generated differently - in C++ you would get lower-cased names. Have you looked at the generated header? You may also want to see http://code.google.com/apis/protocolbuffers/docs/style.html On Fri, May 14, 2010 at 1:29 AM, Jacques

Re: [protobuf] Detect type of message

Check out the docs: http://code.google.com/apis/protocolbuffers/docs/techniques.html#union On Fri, May 14, 2010 at 12:45 PM, Vince wrote: > Hi, > > If I want to serialize many message in the same pipe, > do I need one "master" message that includes all the others in > optional > and have a kind

Re: [protobuf] protobuf CXF RPC impl

Added to the list, thanks! On Wed, May 5, 2010 at 7:46 AM, orbang wrote: > Hi all, > > I've created a pluggable protobuf RPC impl based on Apache CXF: > http://code.google.com/p/cxf-protobuf/ > > Could you please list it on the > http://code.google.com/p/protobuf/wiki/ThirdPartyAddOns > twiki? >

Re: [protobuf] Re: Seg Fault with MutableRepeatedMessage()

and set the > name and batting_avg of the Player within that entry. > > This is my first time using protocol buffers, and I feel like I must > be missing a fundamental step here =) > > Thanks for the help! > > -Ryan > > > On May 4, 12:23 pm, Jason Hsueh wrote: >

Re: [protobuf] Avoid attributes/ Cyclic support

On Mon, May 3, 2010 at 11:29 PM, Fabio Maulo wrote: > - Is there a way to avoid attributes ? > I mean something like an "internal" metadata where we can define the > identifier of each member. > I'm not sure what you mean by this. By attributes do you mean field options? Can you describe in more

Re: [protobuf] Seg Fault with MutableRepeatedMessage()

Your msg is completely empty. There aren't any entries in the repeated field - FieldSize() should be returning 0. Your loop needs to terminate when i < size, not i <= size - accessing the 1st element of an empty repeated field is going to return an invalid pointer. FYI bounds checking is done in d

Re: [protobuf] packed repeated fields / extensions

The packed option is specified in the FieldOptions proto: const FieldDescriptor* f = ...; if (f->options().packed()) { // field is packed } On Mon, Apr 26, 2010 at 2:29 PM, adamdms wrote: > how to check whether the field / expansion (FieldDescriptor *) is > repeated with "packed = true" optio

Re: [protobuf] parseFrom with ExtensionRegistryLite when upgrading from 2.0.3 to 2.3

ionRegistry. > > m = returnType.getMethod("parseFrom", Array.newInstance(Byte.TYPE, > 0).getClass(), ExtensionRegistry.class); > > Hence I was wondering why things changed. > > > On Wed, Apr 21, 2010 at 12:24 PM, Jason Hsueh wrote: > > ExtensionRegist

Re: [protobuf] protobuf <==> SQL

You could write a bridge between the database and protobufs using reflection. It is probably simplest if each field in the protobuf corresponds to a column in the database table. Then when processing the result of a SQL query, you can just iterate over the columns, find the corresponding field in t

Re: [protobuf] parseFrom with ExtensionRegistryLite when upgrading from 2.0.3 to 2.3

ExtensionRegistry extends ExtensionRegistryLite, and the generated parseFrom()/mergeFrom() routines don't need the descriptor-based features that are added by ExtensionRegistry. Is this causing problems for you? On Wed, Apr 21, 2010 at 9:35 AM, skay wrote: > I'm upgrading from 2.0.3 to 2.3, and

Re: [protobuf] How to handle options in compiler plugin?

Your options extensions can live in a .proto file in the location of your choosing. protoc doesn't need to know about your options: it will simply parse the options into the UninterpretedOptions message. Then in the .proto files where you want to use your options, you need to put the option names i

Re: [protobuf] Usage of Protocol Buffer with map reduce framework

As far as I recall, mapreduce just works with raw bytes. If you want to use protocol buffers within map reduce, you just need to use the serialization routines to convert from the raw bytes to your proto. On Mon, Apr 19, 2010 at 4:06 AM, stuti awasthi wrote: > Hi all, > > I just started looking

Re: [protobuf] Suggesting a new Wire Type which encodes length by nested field count rather than bytes

It's not clear to me what the number of fields buys you. How do you intend to serialize this large data if it doesn't fit into memory? The usual solution to this is to split the large message into smaller fragments - in your example, say, 10 messages each containing 100 nested messages. You can ser

Re: [protobuf] parseFrom suggestion

The builders already have a mergeFrom() method with the second signature: http://code.google.com/apis/protocolbuffers/docs/reference/java/com/google/protobuf/AbstractMessage.Builder.html#mergeFrom(byte[], int, int) On Fri, Apr 16, 2010 at 3:02 PM, Fishtank wrote: > Hello, > I have java byte[] (

Re: [protobuf] Protobuf 2.3 and compiler warnings

Yes, these are safe to ignore. This is also addressed in r302: http://code.google.com/p/protobuf/source/detail?r=302 On Thu, Apr 15, 2010 at 12:40 PM, Fishtank wrote: > Hello, > I ran protoc --cpp_out on a proto file using 2.3.0 compiler . I > replaced the my previously generated cpp and header

Re: [protobuf] RepeatedPtrField&::iterator ?

On Wed, Apr 14, 2010 at 1:15 PM, Andrew Kimpton wrote: > Do I have a typo (or some other misunderstanding) or is there no support > for iterators with RepeatedPtrFields ? Iterators are only available in > RepeatedFields ? > > google::protobuf::RepeatedPtrField&::iterator it = > message->locations

Re: [protobuf] Importing one proto file in another

When you invoke the proto compiler are you passing a value for -I (the import path) such that "abc/first.proto" can be found? It defaults to the current working directory. On Wed, Apr 14, 2010 at 9:20 AM, DK wrote: > Hi, > > I have a proto file (first.proto) in folder abc and package abc. I > n

Re: [protobuf] compile issue with import statements having path info

I think if you run everything from the root of your imports this all goes away. The protobuf_AddDesc_* functions are file-level methods used to initialize various bits of static data. The names for these methods are generated based on the pathname of the .proto. When b.proto imports a.proto, it nee

Re: [protobuf] i am not able to compile the files which includes import statements

Can you provide more information? What are the commands you ran? What are the errors? On Fri, Apr 9, 2010 at 4:09 AM, prashant waykar wrote: > Hello, > can someone please help me out as i am new to PB. > i am getting compile errors in the files which consists import > statements > Thanks in advan

Re: [protobuf] Minimum set of libraries/files to run Python and Google Protocol Buffer (CentOS)

It looks like you just copied over the source code. You need to install the modules in a place where python can find them: the python/setup.py script does this for you. On Wed, Apr 7, 2010 at 3:29 PM, Sherry Yang wrote: > Kenton, > Thanks for the followup. I actually copied over the complete i

Re: [protobuf] multi arch build of protobuf on macosx

Kenton, do you have any experience compiling protobufs on OSX? On Wed, Apr 7, 2010 at 9:34 AM, Romain Francois < romain.francois.r.enthusi...@gmail.com> wrote: > Hello, > > Please bear with me if this is obvious ... I'm new to OSX. > > Are there instructions somewhere to build protobuf for both i

Re: [protobuf] protobuf-2.3.0 : Problem generating python files with protoc

What error do you get when trying to use the generated class? With python 2.5.2 I was able to execute the sample add_person.py program. On Wed, Apr 7, 2010 at 11:27 AM, dear chap wrote: > I have this weird problem when generating python files from .proto > files using protoc. I am running protoc

Re: [protobuf] Protocol buffers and CentOS

ck (most recent call last): > File "", line 1, in > File "google/__init__.py", line 1, in > __import__('pkg_resources').declare_namespace(__name__) > ImportError: No module named pkg_resources > > > thanks again, > Eric > > > On Tue, A

Re: [protobuf] Re: Tutorial example make failure

The test you mentioned verifies that descriptor.pb.{h,cc} and plugin.pb.{h,cc} that's in the source matches what the compiler would generate. The test can fail if you've made changes affecting what the compiler outputs, or if you've made changes to those files directly. On Tue, Apr 6, 2010 at 6:20

Re: [protobuf] Protocol buffers and CentOS

How did you install the python protobuf library, and what was the output? On Tue, Apr 6, 2010 at 11:15 AM, Eric wrote: > Hello, > I have installed python (2.6.5) and GPB on CentOS (it's a VM not > sure if that's relevant) but am not able to run. In it's simplest > version I do: > python -c "im

Re: [protobuf] 65k string limit in MS Visual C++ 2005

This sounds like a C++ version of the Java issue fixed with r189: the encoded descriptor data in the generated file exceeded the Java string literal limit. This large descriptor was previously built in C++ but was removed from the test suite in

Re: [protobuf] Re: c++: Print contents of a serialized file using its descriptor file

< > std::endl; >return -1; >} > my_message->ParseFromIstream(&serialized_data_file); > >//Print out the message >my_message->PrintDebugString(); > >// Optional: Delete all global

Re: [protobuf] Re: c++: Print contents of a serialized file using its descriptor file

gt; I'm not sure what you mean by this. In the snippet above you're adding all the files in the FileDescriptorSet to the same instance of DescriptorPool. What would you like to do instead? > > Thanks a lot. Protobuf should add an example to their package. Would the use in dynami

Re: [protobuf] Re: c++: Print contents of a serialized file using its descriptor file

o do is given ANY serialized message in a file, > I want to load it and print its contents out. That all. > > Thanks again. > > > > > > > > On Apr 2, 1:42 pm, Jason Hsueh wrote: > > On Fri, Apr 2, 2010 at 9:13 AM, Navin wrote: > > > 1. For the person

Re: [protobuf] c++: Print contents of a serialized file using its descriptor file

On Fri, Apr 2, 2010 at 9:13 AM, Navin wrote: > 1. For the person example, I create a descriptor as follows: > "E:\Desktop\prototype_MAIN\protoc.exe" person.proto -- > descriptor_set_out=msg.protocomp > > 2. I create a Person object and serialized it to file. "person.pb" > > 3. I want to write a p

[protobuf] Re:

+protobuf mailing list This is a bit outside the scope of protocol buffers - the core library does not provide any such functionality. I am personally not that familiar with their usage, but I don't see any technical limitations to using protocol buffers with shared memory. It would be a bit trick

Re: [protobuf] Why getting errors when exanging large messages

The first message from message_lite.cc:106 is telling you that some required fields were missing. Have you verified that each of the fields listed was set? If you don't want required fields checking, you need to use one of the ParsePartial* to skip required field checking. The second message is fr

Re: [protobuf] File size of the serialized records

If you're measuring using sizeof(), you won't account for memory allocated by subobjects (strings and submessages are stored as pointers). You should use Message::SpaceUsed() instead. The inmemory vs serialized size is going to depend on your proto definition and how you use it. If you have a lot o

Re: [protobuf] modifying lists within a message

Messages are indeed immutable once built. getXXXList() for both the builder and the message should return unmodifiable lists: if you try to modify them you should get an exception. If you haven't already, take a look at the Java tutorial here: http://code.google.com/apis/protocolbuffers/docs/javatu

Re: [protobuf] How to get optional data?

The getters for embedded messages return a reference - see http://code.google.com/apis/protocolbuffers/docs/reference/cpp-generated.html On Tue, Mar 16, 2010 at 9:02 PM, jungle wrote: > here is the .proto > > message Bar{ >required int32 l = 1; > } > > message Foo{ >option Bar bar = 1; >

Re: [protobuf] How to get the number of the repeated data?

This is covered in the tutorial - select the one for your language from http://code.google.com/apis/protocolbuffers/docs/overview.html On Tue, Mar 16, 2010 at 8:57 PM, jungle wrote: > here is .proto > message foo{ > required xxx > repeated xxx > } > > how to figure out the exact number of th

Re: [protobuf] Disabling the 'gen_cpp' output of Services?

In your .proto file, just specify the file-level option cc_generic_services = false. Comments from descriptor.proto: // Should generic services be generated in each language? "Generic" services // are not specific to any particular RPC system. They are generated by the // main code genera

Re: [protobuf] Variant-length encoding : is it OK to be wasteful?

On Wed, Mar 10, 2010 at 7:48 AM, Marc Gravell wrote: > The variable-length encoding allows multiple representations of the > same value - for example, 1 could be written as: > > 0001 > > or it could be (I believe): > > 1000 00011000 > > Is there anything in the core (o

Re: [protobuf] Using std::string as a zero-copy buffer

n't > use -fno-strict-aliasing. > > On Thu, Mar 4, 2010 at 10:09 AM, Jason Hsueh wrote: > >> I don't know about any other implementations, but a word of caution: I >> think that this would be especially complicated/hacky for gcc 4.3, which as >> I recall swit

Re: [protobuf] Using std::string as a zero-copy buffer

(__len); >_M_rep()->_M_set_length_and_sharable(__len); > } > } > > See the code for basic_string::append, for example. Given that, it > shouldn't be too ugly to do it from outside, but these things are > always easy to mess up. > > Sean > > > > On T

Re: [protobuf] Using std::string as a zero-copy buffer

I don't know about any other implementations, but a word of caution: I think that this would be especially complicated/hacky for gcc 4.3, which as I recall switched to a copy-on-write implementation for std::string. You'd probably want to make sure that the performance difference is really worth it

Re: [protobuf] Invalid embedded descriptor && Dependencies passed to FileDescriptor.buildFrom() don't match

I believe you should be running protoc from the same directory, using the same --proto_path, for both commands. The imports should be relative to the common root directory. modify the import in project2\proto\TaskProtos.proto: import "project1/proto/ProtoException.proto" $ protoc --proto_path=. pr

Re: [protobuf] "Third-Party Add-ons" - Update Request - for Protocol Buffer Editor

Updated the wiki page. Thanks for writing the tool! On Wed, Feb 17, 2010 at 1:40 AM, Bruce wrote: > "Third-Party Add-ons" - Update Request - for Proto Buffer Editor > > Could you please add my project http://code.google.com/p/protobufeditor/ > to the > "Third-Party Add-ons" ("Other Utilities). >

Re: [protobuf] how to get RepeatedField object

> On the documentation it says : > > " > Most users will not ever use a RepeatedField directly; they will use the > get-by-index, set-by-index, and add accessors that are generated for all > repeated fields > " > > What if I do want to use RepeatedField ? > > Romain &g

Re: [protobuf] how to get RepeatedField object

No, there isn't a way to get the RepeatedField from the reflection interface. You can only do so via the generated interface. On Thu, Feb 11, 2010 at 5:57 AM, Romain Francois < romain.francois.r.enthusi...@gmail.com> wrote: > Hello, > > How can I get hold of a RepeatedField object to manage a rep

Re: [protobuf] how to define a one-way rpc message?

No, each method needs to specify a response type. The workaround you suggested is the way to go. However, you may want to define your own message that is empty (rather than a generic Void that might get reused by other services) so that you can extend your message in the future. On Wed, Feb 3, 201

Re: [protobuf] analysing field sizes

(Since you said getserializedsize I'm assuming Java) There isn't a precanned way to get a breakdown of the serialized sizes, but I don't think you need to reimplement the entire parser: if you take a serialized message, you could set up a CodedInputStream and use a fairly simple loop of readTag(),

Re: [protobuf] Any j2me implementations that support extensions?

It appears the ones listed at http://code.google.com/p/protobuf/wiki/ThirdPartyAddOns do not, though I've only taken a cursory glance at their pages. Anyone on the mailing list holding out on us? :-) On Wed, Jan 27, 2010 at 2:11 AM, Juan Amiguet wrote: > Hi All, > > Does any one know of or has a

Re: [protobuf] Message limit

The limit applies to the data source from which a message is parsed. So if you want to parse a serialization of Foo, it applies to Foo. But if you parse a bunch of Bar messages one by one, and add them individually to Bar, then the limit only applies to each individual Bar. You can change the limi

Re: [protobuf] Builders containing builders

You should be able to pass a builder to some method, and have the method modify some value of the builder just fine. Modifying a value in a sub-message is a bit more inconvenient, but still doable. If you have some type Bar in type Foo like: message Bar { optional int32 a = 1; optional int32 b

Re: [protobuf] Re: Build problem with nested enum

Thanks. Is there an ADD or DELETE macro that gets defined somewhere in your MSVC? It doesn't seem like the compiler has a problem with previous enum typedefs, so I'm guessing that there may be a name conflict with your enum names, and it's expanding to code which no longer compiles. Can you try mak

Re: [protobuf] Reading Multiple Protobufs from a File

Indeed, CodedOutputStream and CodedInputStream aren't presently available in Python. Want to add it? You should also be able to implement what you want without the help of these classes. http://code.google.com/apis/protocolbuffers/docs/techniques.html#streamingdescribes what to do - once you have

Re: [protobuf] does java api have something like DescriptorPool

No, there isn't an equivalent to the DescriptorPool in Java. If you know the types that you want you can build a mapping yourself. Or if you'd be interested in porting the C++ DescriptorPool to java that would be great! 2009/12/9 Romain François > Hello, > > Given a service/method full name, I'd

Re: [protobuf] Extensions not serializing properly?

In Java you need to provide an ExtensionRegistry to the parse method Something like: ExtensionRegistry registry = ExtensionRegistry.newInstance(); Test.registerAllExtensions(registry); Test.BsgRequest req = Test.BsgRequest.parseFrom(payload, registry); On Thu, Dec 3, 2009 at 8:01 AM, Michael La

Re: [protobuf] Re: Java -using GPB -which type of stream should I be using over sockets?

On Wed, Dec 2, 2009 at 4:15 PM, Nigel Pickard wrote: > When the Java test app works, I'd say the IO stream is transparent. I > use the same simple object instance each time (have a hard coded > method that simply returns the same object so it's identical), and I > simply call the toString of the

Re: [protobuf] Re: Java -using GPB -which type of stream should I be using over sockets?

On Wed, Dec 2, 2009 at 12:13 PM, Nigel Pickard wrote: > Thanks for the reply Adam -I did a simple sub into my code with your > suggestion, but it doesn't work. However, now it's made me think of > more questions. > If protocol buffers doesn't use object I/O streams, which streams does > it

Re: [protobuf] Java -using GPB -which type of stream should I be using over sockets?

Protocol buffers implement their own serialization routines: the writeTo(OutputStream output) method will write the binary format to the OutputStream. From a socket, just do: MyProtoBuf pb = new MyProtoBuf; ... pb.writeTo(socket.getOutputStream()); On Wed, Dec 2, 2009 at 7:47 AM, Nigel Pickard wr

Re: [protobuf] Memory leak (fragmentation?)

Are you sure the leak is due to persistCacheToFile()? It looks like your FileManager::loadCacheFromFile() method is leaking the protocol buffer when you are reading the cache back in. If that's not the case, can you send a small reproduction of the problem? On Tue, Nov 24, 2009 at 2:03 AM, Greg B

Re: [protobuf] Apache Module + PB

It looks like you're running gcc on a .c file, so it's trying to compile as C, not C++. You can either invoke g++ or name your file with one of the recognized extensions: http://gcc.gnu.org/onlinedocs/gcc/Invoking-G_002b_002b.html#Invoking-G_002b_002b On Wed, Nov 18, 2009 at 7:56 AM, Mauro wrote:

[protobuf] Re: Regd: Resolving Wire type ambiguities

You can decode the protocol buffer with just wire type + tag number, but you won't know the original types without a proto definition. Everything would be treated as an unknown field. You could access these by iterating through the UnknownFieldSet, but again, you can't recover the original types.

[protobuf] Re: Seek in a codedoutputstream

the size of data1(and it is not from > protobuf parse) > I'm using protobuf for the stream libraries. > I can either a) loop through the calculation to get the total data > size, then loop again writing size and data > or b) loop once, write, get size, seek and adjust size.

[protobuf] Re: Seek in a codedoutputstream

Why can't you write the actual size of data rather than a 0? Just use the ByteSize() method to find out how much data will be written for a protocol buffer. On Sat, Nov 14, 2009 at 11:39 AM, Fishtank wrote: > > Hello, > I'm using the following idiom > ZeroCopyOutputStream* raw_output = new Fil

<    1   2   3   4   >