You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@streampipes.apache.org by "obermeier (via GitHub)" <gi...@apache.org> on 2023/01/26 14:26:56 UTC
[GitHub] [streampipes] obermeier opened a new pull request, #1165: fix #1164
obermeier opened a new pull request, #1165:
URL: https://github.com/apache/streampipes/pull/1165
fix #1164
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: dev-unsubscribe@streampipes.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] More specific types for large numbers. (streampipes)
Posted by "obermeier (via GitHub)" <gi...@apache.org>.
obermeier closed pull request #1165: More specific types for large numbers.
URL: https://github.com/apache/streampipes/pull/1165
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: dev-unsubscribe@streampipes.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] More specific types for large numbers. (streampipes)
Posted by "obermeier (via GitHub)" <gi...@apache.org>.
obermeier commented on PR #1165:
URL: https://github.com/apache/streampipes/pull/1165#issuecomment-1407778085
> thanks a lot @obermeier 🙏🏼 Your changes look good to me Can you please try to address the code style issues: https://github.com/apache/streampipes/actions/runs/4016747501/jobs/6900267599#step:4:70961 You can read on how to apply our checkstyle config here: https://cwiki.apache.org/confluence/display/STREAMPIPES/Code+Style+-+Java
Thank you for this hint and the nice description!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: dev-unsubscribe@streampipes.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] More specific types for large numbers. (streampipes)
Posted by "obermeier (via GitHub)" <gi...@apache.org>.
obermeier commented on PR #1165:
URL: https://github.com/apache/streampipes/pull/1165#issuecomment-1407781682
> Hello @obermeier, thanks for opening the issue and providing a PR.
>
> I recently encountered the same issue and changed it in [PR-1099](https://github.com/apache/streampipes/pull/1099/files). However, I suspect that the changes only affect the CSV format, as I tried your example and got the same behavior.
>
> The default should be that we always assume a floating point number. If the number is too large for a float (e.g. a Unix timestamp in ms) it is assumed to be a double (see class 'DatatypeUtils'). We decided to do this because if the first value of the CSV file is represented as an integer 1 and not as 1.0, when another value is then a float (e.g. 2.4) we will get runtime exceptions when storing the data in the time series storage.
>
> Your implementation now implements the same behavior, correct? Do your changes also affect the values at runtime or only the description of the event?
> Hello @obermeier, thanks for opening the issue and providing a PR.
>
> I recently encountered the same issue and changed it in [PR-1099](https://github.com/apache/streampipes/pull/1099/files). However, I suspect that the changes only affect the CSV format, as I tried your example and got the same behavior.
>
> The default should be that we always assume a floating point number. If the number is too large for a float (e.g. a Unix timestamp in ms) it is assumed to be a double (see class 'DatatypeUtils'). We decided to do this because if the first value of the CSV file is represented as an integer 1 and not as 1.0, when another value is then a float (e.g. 2.4) we will get runtime exceptions when storing the data in the time series storage.
>
> Your implementation now implements the same behavior, correct? Do your changes also affect the values at runtime or only the description of the event?
Hi @tenthe, I like the Idear to use just Floats and Doubles. A timestamp in ms shoud fit to this model.
I ralized that I could not reproduce my "bug".
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: dev-unsubscribe@streampipes.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] More specific types for large numbers. (streampipes)
Posted by "tenthe (via GitHub)" <gi...@apache.org>.
tenthe commented on PR #1165:
URL: https://github.com/apache/streampipes/pull/1165#issuecomment-1406133679
Hello @obermeier,
thanks for opening the issue and providing a PR.
I recently encountered the same issue and changed it in [PR-1099](https://github.com/apache/streampipes/pull/1099/files). However, I suspect that the changes only affect the CSV format, as I tried your example and got the same behavior.
The default should be that we always assume a floating point number. If the number is too large for a float (e.g. a Unix timestamp in ms) it is assumed to be a double (see class 'DatatypeUtils'). We decided to do this because if the first value of the CSV file is represented as an integer 1 and not as 1.0, when another value is then a float (e.g. 2.4) we will get runtime exceptions when storing the data in the time series storage.
Your implementation now implements the same behavior, correct?
Do your changes also affect the values at runtime or only the description of the event?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: dev-unsubscribe@streampipes.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org