[Rock-dev] Problem with mixed sizes of typekit types
Thomas Roehr
thomas.roehr at dfki.de
Fri Jul 21 12:35:10 CEST 2017
Hey,
then, how about following Sylvain's suggestion for solving this?
Best
Thomas
On 21.07.2017 12:22, Dennis Hemker wrote:
> Hey Thomas,
>
> yes, it's related. Most of the "corrupted" logs are the ones with
> different types of string offsets. As Janosch mentioned, the isSame
> method in C++ fails, but that was a thing we observed while we were
> struggling with the non-loadable streams in logfiles.
>
> Best,
>
> Dennis
>
>
> Am 20.07.2017 um 19:21 schrieb Thomas Roehr:
>> Hey Dennis,
>>
>> can you check if that is related:
>>
>> https://github.com/orocos-toolchain/typelib/issues/95
>>
>> Best
>> Thomas
>>
>> On 09.06.2017 14:06, Alexander Fabisch wrote:
>>> Hi Dennis,
>>>
>>> I think I have a similar problem: the problem seems to be the offset of
>>> strings, right? I assume that if you subtract the offset of sourceFrame
>>> from the offset of targetFrame you SHOULD get something like the size of
>>> the the type that has been used to encode the length of the string (8
>>> Byte vs. 32 Byte)? I have a logfile that looks like type a and a logfile
>>> that looks like type b but the actual length is encoded in the first 8
>>> Byte of the string in both cases. You can check that if you look
>>> directly at the binary data in the logfile with "hexdump -C <file>":
>>>
>>> > 00000820 ... 0c 00 00 00 00 00 00 00 |9...............| <- 0c 00
>>> 00 00 00 00 00 00 == 12
>>> > 00000830 68 65 6c 6c 6f 77 20 77 6f 72 6c 64 02 ff 01 00 |hellow
>>> world....|
>>>
>>> Best regards,
>>>
>>> Alexander
>>>
>>>
>>> Am 09.06.2017 um 12:35 schrieb Dennis Hemker:
>>>> Hey Guys,
>>>>
>>>> I encountered a weird problem during the use of the C++-based
>>>> rock-replay2, which uses the local typekit definitions to load
>>>> marshalled types from the log files. In some logfiles, the local and
>>>> the
>>>> marshalled typekit type mismatch in their sizes, for instance the
>>>> RigidBodyState_m type. The log was recorded with the Sherpa-Bot in
>>>> Utah:
>>>>
>>>> type a:
>>>> compound /base/samples/RigidBodyState_m [416] {
>>>> (+0) compound /base/Time [8] {
>>>> (+0) sint(8) (/int64_t) microseconds
>>>> }; time
>>>> (+8) container /std/string</int8_t> sourceFrame
>>>> (+16) container /std/string</int8_t> targetFrame
>>>> (+24) compound /wrappers/Matrix</double,3,1> [24] {
>>>> (+0) array[3] of
>>>> float(8) (/double) data
>>>> }; position
>>>> (+48) compound /wrappers/Matrix</double,3,3> [72] {
>>>> (+0) array[9] of
>>>> float(8) (/double) data
>>>> }; cov_position
>>>> (+120) compound /wrappers/Quaternion</double> [32] {
>>>> (+0) array[3] of
>>>> float(8) (/double) im
>>>> (+24) float(8) (/double) re
>>>> }; orientation
>>>> (+152) compound /wrappers/Matrix</double,3,3> [72] {
>>>> (+0) array[9] of
>>>> float(8) (/double) data
>>>> }; cov_orientation
>>>> (+224) compound /wrappers/Matrix</double,3,1> [24] {
>>>> (+0) array[3] of
>>>> float(8) (/double) data
>>>> }; velocity
>>>> (+248) compound /wrappers/Matrix</double,3,3> [72] {
>>>> (+0) array[9] of
>>>> float(8) (/double) data
>>>> }; cov_velocity
>>>> (+320) compound /wrappers/Matrix</double,3,1> [24] {
>>>> (+0) array[3] of
>>>> float(8) (/double) data
>>>> }; angular_velocity
>>>> (+344) compound /wrappers/Matrix</double,3,3> [72] {
>>>> (+0) array[9] of
>>>> float(8) (/double) data
>>>> }; cov_angular_velocity
>>>> };
>>>> type b:
>>>> compound /base/samples/RigidBodyState_m [464] {
>>>> (+0) compound /base/Time [8] {
>>>> (+0) sint(8) (/int64_t) microseconds
>>>> }; time
>>>> (+8) container /std/string</int8_t> sourceFrame
>>>> (+40) container /std/string</int8_t> targetFrame
>>>> (+72) compound /wrappers/Matrix</double,3,1> [24] {
>>>> (+0) array[3] of
>>>> float(8) (/double) data
>>>> }; position
>>>> (+96) compound /wrappers/Matrix</double,3,3> [72] {
>>>> (+0) array[9] of
>>>> float(8) (/double) data
>>>> }; cov_position
>>>> (+168) compound /wrappers/Quaternion</double> [32] {
>>>> (+0) array[3] of
>>>> float(8) (/double) im
>>>> (+24) float(8) (/double) re
>>>> }; orientation
>>>> (+200) compound /wrappers/Matrix</double,3,3> [72] {
>>>> (+0) array[9] of
>>>> float(8) (/double) data
>>>> }; cov_orientation
>>>> (+272) compound /wrappers/Matrix</double,3,1> [24] {
>>>> (+0) array[3] of
>>>> float(8) (/double) data
>>>> }; velocity
>>>> (+296) compound /wrappers/Matrix</double,3,3> [72] {
>>>> (+0) array[9] of
>>>> float(8) (/double) data
>>>> }; cov_velocity
>>>> (+368) compound /wrappers/Matrix</double,3,1> [24] {
>>>> (+0) array[3] of
>>>> float(8) (/double) data
>>>> }; angular_velocity
>>>> (+392) compound /wrappers/Matrix</double,3,3> [72] {
>>>> (+0) array[9] of
>>>> float(8) (/double) data
>>>> }; cov_angular_velocity
>>>> };Checking /body_joint.state
>>>>
>>>> Differences can be found in the source- and targetFrame definitions.
>>>>
>>>> Thank you and best regards,
>>>>
>>>> Dennis
>>>>
>>>> _______________________________________________
>>>> Rock-dev mailing list
>>>> Rock-dev at dfki.de
>>>> http://www.dfki.de/mailman/cgi-bin/listinfo/rock-dev
>>
>
--
Thomas Röhr (M.Sc.)
Space Robotics
Besuchsadresse der Nebengeschäftstelle:
DFKI GmbH
Robotics Innovation Center
Robert-Hooke-Straße 5
28359 Bremen, Germany
Postadresse der Hauptgeschäftsstelle Standort Bremen:
DFKI GmbH
Robotics Innovation Center
Robert-Hooke-Straße 1
28359 Bremen, Germany
Tel.: +49 421 178 45-4151
Zentrale: +49 421 178 45-0
Fax: +49 421 178 45-4150 (Faxe bitte namentlich kennzeichnen)
E-Mail: thomas.roehr at dfki.de
Weitere Informationen: http://www.dfki.de/robotik
-----------------------------------------------------------------------
Deutsches Forschungszentrum fuer Kuenstliche Intelligenz GmbH
Firmensitz: Trippstadter Straße 122, D-67663 Kaiserslautern
Geschaeftsfuehrung: Prof. Dr. Dr. h.c. mult. Wolfgang Wahlster
(Vorsitzender) Dr. Walter Olthoff
Vorsitzender des Aufsichtsrats: Prof. Dr. h.c. Hans A. Aukes
Amtsgericht Kaiserslautern, HRB 2313
Sitz der Gesellschaft: Kaiserslautern (HRB 2313)
USt-Id.Nr.: DE 148646973
Steuernummer: 19/673/0060/3
-----------------------------------------------------------------------
More information about the Rock-dev
mailing list