More info-- when the "header" is set with an index of >= 16, the encoding
process appears to insert an extra byte in the encoded value, and a decode
attempt will fail, here is an example where an index of 15 can allow for an
encode and subsequent decode, but an index of 16 fails at the subsequent
decode:
using a header with an index = 15
obj to publish (typeof [object]): [TestResponse {
header: { timestamp: 'monkeybutt' },
mykey: 'hi',
myvalue: 'there'
}]
encodded as buff (typeof [object]): [<Buffer 12 02 68 69 1a 05 74 68 65 72
65 7a 0c 0a 0a 6d 6f 6e 6b 65 79 62 75 74 74>]
encoded as str (typeof [string]):
['\u0012\u0002hi\u001a\u0005therez\f\n\nmonkeybutt']
sanBinArray (typeof [object]): [[
18, 2, 104,
105, 26, 5,
116, 104, 101,
114, 101, 122,
12, 10, 10,
109, 111, 110,
107, 101, 121,
98, 117, 116,
116, [length]: 25
]]
Sanity check, decoding encoded msg: (typeof [object]): [TestResponse {
mykey: 'hi',
myvalue: 'there',
header: Monkey { timestamp: 'monkeybutt' }
}]
pinging
CONSUMER << RXD [25] elements (typeof [string]):
['\u0012\u0002hi\u001a\u0005therez\f\n\nmonkeybutt']
As binArray (typeof [object]): [[
18, 2, 104,
105, 26, 5,
116, 104, 101,
114, 101, 122,
12, 10, 10,
109, 111, 110,
107, 101, 121,
98, 117, 116,
116, [length]: 25
]]
(typeof [object]): [TestResponse {
mykey: 'hi',
myvalue: 'there',
header: Monkey { timestamp: 'monkeybutt' }
}]
using a header with index = 16
obj to publish (typeof [object]): [TestResponse {
header: { timestamp: 'monkeybutt' },
mykey: 'hi',
myvalue: 'there'
}]
encodded as buff (typeof [object]): [<Buffer 12 02 68 69 1a 05 74 68 65 72
65 82 01 0c 0a 0a 6d 6f 6e 6b 65 79 62 75 74 74>]
== == <-- two bytes
encoded as str (typeof [string]):
['\u0012\u0002hi\u001a\u0005there�\u0001\f\n\nmonkeybutt']
sanBinArray (typeof [object]): [[
18, 2, 104,
105, 26, 5,
116, 104, 101,
114, 101, 65533, <---- suspicious!
1, 12, 10,
10, 109, 111,
110, 107, 101,
121, 98, 117,
116, 116, [length]: 26
]]
Sanity check error: Could not decode msg {
decodeExc: Error: invalid wire type 7 at offset 18
at Reader.skipType (node_modules/protobufjs/src/reader.js:377:19)
at Type.TestResponse$decode [as decode] (eval at Codegen
(node_modules/@protobufjs/codegen/index.js:50:33), <anonymous>:20:5)
at buildEncodedStr (test/utils/amqprototest.js:94:34)
at Timeout._onTimeout (test/utils/amqprototest.js:57:17)
at listOnTimeout (internal/timers.js:531:17)
at processTimers (internal/timers.js:475:7)
}
On Friday, January 8, 2021 at 6:25:41 AM UTC-7 peo stri tise safe wrote:
> Gotcha, so the index value of 100 is not a non-standard value.
>
> So I'm seeing "invalid wire type" when using the decode method of
> protobuf.js of an AMQ body. This is in a Node using the protobuf.js,
> stomp-client.js and an AMQ broker.
>
> The GPB that I'm using has a basic format of this:
>
> syntax = "proto2";
> package testt;
> message Monkey
> {
> optional string timestamp = 1;
> }
> message TestResponse
> {
> optional Monkey header = 100;
> optional string mykey = 2;
> optional string myvalue = 3;
> }
>
> And the message is encoded and sent to the broker like this:
>
> // note TestRespGPBDef is created earlier
> let payload = {
> header: {
> timestamp: 'monkeybutt'
> },
> mykey: 'hi',
> myvalue: 'there'
> };
> let verifyErr = TestRespGPBDef.verify(payload);
> if( verifyErr ) {
> console.log(`GPB OBJECT/PAYLOAD ERROR`, {verifyErr});
> return;
> }
> let gmsg = TestRespGPBDef.create(payload);
> let buff = TestRespGPBDef.encode(gmsg).finish();
> let str = buff.toString();
> stompClient.publish('mytopic',str);
>
>
> And the message is rd'x and decoded by the consumer like this:
>
> // note TestRespGPBDef is created earlier
> client.subscribe('mytopic', function(body, headers) {
> utils.printObj(`CONSUMER << RXD AMQ header and body`);
> utils.printObj(`AMQ body\n [${body.length}] elements`, body);
> let binArray = utils.stringToBinArray(body);
> utils.printObj(`As binArray`,binArray);
> try {
> let gpbMsg = TestRespGPBDef.decode(binArray);
> utils.printObj('',gpbMsg);
> } catch (decodeExc) {
> console.log(`Could not decode msg`,{decodeExc});
> }
> });
>
>
> When the consumer receives the body from AMQ and tries to decode it,
> this is the exception:
>
> CONSUMER << RXD AMQ header and body
> AMQ body:
> [26] elements (typeof [string]):
> ['\u0012\u0002hi\u001a\u0005there�\u0006\f\n\nmonkeybutt']
> As binArray (typeof [object]): [[
> 18, 2, 104,
> 105, 26, 5,
> 116, 104, 101,
> 114, 101, 65533,
> 6, 12, 10,
> 10, 109, 111,
> 110, 107, 101,
> 121, 98, 117,
> 116, 116, [length]: 26
> ]]
> Could not decode msg {
> decodeExc: Error: invalid wire type 7 at offset 18
> at Reader.skipType
> (/mytest/node_modules/protobufjs/src/reader.js:377:19)
> at Type.TestResponse$decode [as decode] (eval at Codegen
> (/mytest/node_modules/@protobufjs/codegen/index.js:50:33), <anonymous>:20:5)
> at
> /Users/rcox/gitprojects/POSTAPOCALYPSE/speb-hmi/deviceModules/keplerAdapter/test/utils/amqprototest.js:37:39
> at
> /Users/rcox/gitprojects/POSTAPOCALYPSE/speb-hmi/deviceModules/keplerAdapter/node_modules/stomp-client/lib/client.js:206:9
> at Array.map (<anonymous>)
> at StompFrameEmitter.<anonymous>
> (/mytest/node_modules/stomp-client/lib/client.js:205:28)
> at StompFrameEmitter.emit (events.js:210:5)
> at StompFrameEmitter.parseBody
> (/mytest/node_modules/stomp-client/lib/parser.js:136:12)
> at StompFrameEmitter.handleData
> (/mytest/node_modules/stomp-client/lib/parser.js:42:12)
> at Socket.<anonymous>
> (/mytest/node_modules/stomp-client/lib/client.js:188:18)
> }
>
> When the "header" index is set to 1, and the experiment is repeated, the
> decoding does not throw the exception.
> On Friday, January 8, 2021 at 12:52:16 AM UTC-7 [email protected] wrote:
>
>> 100 isn't "non-standard" as such, and shouldn't cause anything to fail.
>> What exactly are you seeing?
>>
>> The valid range is 1-536870911, omitting 19000-19999 (and any reserved
>> areas in your specific messages) smaller numbers are cheaper (fewer bytes)
>> to encode, so are usually preferred - but: that's it.
>>
>> I'm guessing you're actually using some non-compliant code that is
>> assuming field-headers are single bytes? That is true for very low field
>> numbers (4 bytes, so: 1-15), but field 100 will take two bytes for the
>> header. Is that the problem here?
>>
>> On Thu, 7 Jan 2021, 22:00 peo stri tise safe, <[email protected]>
>> wrote:
>>
>>> What is the purpose of setting starting index values for a GPB message
>>> definition to something other than 1? I am seeing a value of 100 in a
>>> particular application and it is causing the encoding of the message to
>>> fail. Just wondered why the use of values other than 1 for the first item
>>> in the message.
>>>
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "Protocol Buffers" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to [email protected].
>>> To view this discussion on the web visit
>>> https://groups.google.com/d/msgid/protobuf/957aae3b-32f8-4d9d-8f5f-732e26e08cedn%40googlegroups.com
>>>
>>> <https://groups.google.com/d/msgid/protobuf/957aae3b-32f8-4d9d-8f5f-732e26e08cedn%40googlegroups.com?utm_medium=email&utm_source=footer>
>>> .
>>>
>>
--
You received this message because you are subscribed to the Google Groups
"Protocol Buffers" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/protobuf/2a726cea-c829-4527-9878-232f746b32b8n%40googlegroups.com.
