[ros-users] CSV with time from bag files

Cedric Pradalier cedric.pradalier at mavt.ethz.ch
Wed Jun 9 15:52:27 UTC 2010


On 06/09/2010 05:07 PM, cuenta altas wrote:
> Hi Cedric,
>
> I don't know if what you say applies here. If I understood you well 
> when you mention 'testing software', that would be achieved using 
> 'rosbag play'. If you are using 'rostopic echo' you'll just get a 
> bunch of data printed on screen (or to a text file).
>
> Miguel.

Yes, I think you're right. I had read the email a bit too quickly.
I actually did not realise you could use rostopic to directly output a bag.
Good to know...


>
> 2010/6/9 Cedric Pradalier <cedric.pradalier at mavt.ethz.ch 
> <mailto:cedric.pradalier at mavt.ethz.ch>>
>
>     On 06/09/2010 04:31 PM, Jeremy Leibs wrote:
>     > Agreed.  I was also surprised to find it didn't print bag time.  I
>     > created ticket: https://code.ros.org/trac/ros/ticket/2780 to track
>     > this issue.
>     >
>
>     I actually implemented a playback functionality for another middleware
>     some time ago, and I had put that as an option.
>     In some context, one would want to have the time at which the data was
>     recorded (for plotting for instance).
>     On the other hand, if you test some software, then it's better to put
>     the writing time, to have a feeling of real-time.
>
>     Then I also had an option for time scaling (fast-forward or
>     slow-motion). Then again, the choice of old time-stamp, write
>     time-stamp
>     becomes tricky, especially if you introduce a pause mechanism in the
>     playback.
>
>     Even without the special features above, I think both type of
>     timestamp
>     are sometimes useful, and only a command-line option can select
>     between
>     them.
>
>     HTH
>
>     > On Wed, Jun 9, 2010 at 2:14 AM, Miguel
>     Prada<miguel.register at gmail.com
>     <mailto:miguel.register at gmail.com>>  wrote:
>     >
>     >> Hi Jeremy,
>     >>
>     >> Thanks for the tip. That's exactly what I've been doing after
>     writing to the list and it works great.
>     >>
>     >> However, shouldn't rostopic output the bagfile's timestamps
>     when echoing from a bagfile? I find it much more reasonable than
>     having to programatically read the data using python to achieve
>     this result. Am I missing some important reason for rostopic echo
>     to behave this way?
>     >>
>     >> Miguel.
>     >>
>     >> El 09/06/2010, a las 10:33, Jeremy Leibs escribió:
>     >>
>     >>
>     >>> Assuming you are still using boxturtle, rosrecord has a
>     rosrecord.py
>     >>> programmatic API
>     >>>
>     >>> Take a look at:
>     >>>
>     >>> http://www.ros.org/wiki/rosrecord#rosrecord.py
>     >>>
>     >>> It should be fairly straight forward to write a small program
>     which
>     >>> opens the bag and prints exactly what you want.
>     >>>
>     >>> On Wed, Jun 9, 2010 at 1:29 AM, Miguel
>     Prada<miguel.register at gmail.com
>     <mailto:miguel.register at gmail.com>>  wrote:
>     >>>
>     >>>> Hi,
>     >>>>
>     >>>> I want to manipulate some data from a bag file with a
>     spreadsheet. I'm exporting from the bag to CSV using 'rostopic
>     echo -p -b BAGFILE TOPIC>>  file.csv' but I just realised that the
>     time field you get from doing that is the time corresponding to
>     when rostopic is run, not the time stored in the bag file. That
>     means that if I use this method to output messages in one topic
>     and then repeat for another topic (e.g. controller commands and
>     robot positions) the time fields on both CSV files are of no use
>     at all to merge the data in those two files.
>     >>>>
>     >>>> I tried using 'rostopic echo -p -b BAGFILE -a>>  file.csv' to
>     print all topics at once, but the output is quite messed up and
>     doesn't clearly state which data corresponds to which topic.
>     >>>>
>     >>>> Does anyone have any suggestion on how should I proceed with
>     this?
>     >>>>
>     >>>> Miguel.
>     >>>>
>     >>>> _______________________________________________
>     >>>> ros-users mailing list
>     >>>> ros-users at code.ros.org <mailto:ros-users at code.ros.org>
>     >>>> https://code.ros.org/mailman/listinfo/ros-users
>     >>>>
>     >>>>
>     >>> _______________________________________________
>     >>> ros-users mailing list
>     >>> ros-users at code.ros.org <mailto:ros-users at code.ros.org>
>     >>> https://code.ros.org/mailman/listinfo/ros-users
>     >>>
>     >> _______________________________________________
>     >> ros-users mailing list
>     >> ros-users at code.ros.org <mailto:ros-users at code.ros.org>
>     >> https://code.ros.org/mailman/listinfo/ros-users
>     >>
>     >>
>     > _______________________________________________
>     > ros-users mailing list
>     > ros-users at code.ros.org <mailto:ros-users at code.ros.org>
>     > https://code.ros.org/mailman/listinfo/ros-users
>     >
>
>
>     --
>     Dr. Cedric Pradalier
>     http://www.asl.ethz.ch/people/cedricp
>
>     _______________________________________________
>     ros-users mailing list
>     ros-users at code.ros.org <mailto:ros-users at code.ros.org>
>     https://code.ros.org/mailman/listinfo/ros-users
>
>
>
> _______________________________________________
> ros-users mailing list
> ros-users at code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
>    


-- 
Dr. Cedric Pradalier
http://www.asl.ethz.ch/people/cedricp

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ros.org/pipermail/ros-users/attachments/20100609/3f3d9c65/attachment-0003.html>


More information about the ros-users mailing list