Single stat complaining about non-numeric values #13824
Comments
|
I believe this is related to influxdata/flux#1252. When the latest value of a numeric column is empty, our single stat computation returns the empty value which leads to the "Could not display single stat because your values are non-numeric" error message. I am going to update our single stat computation to find the latest non-empty value. |
chnn
added a commit
that referenced
this issue
May 8, 2019
The method we used to compute a single stat / gauge value previously had a few issues: - If the latest value was part of a numeric column but was null/NaN/not defined, the single stat computation would fail and a user would see an error message "Could not display single stat because your values are non-numeric" - If there were multiple latest values, an arbitrary selection would be made This commit updates the single stat computation to find the latest *defined* numeric values. If multiple latest valid numeric values are found, we will either: - Display an error message if using the compuation within a single stat visualization - Display nothing if using the computation within a within a line + single stat visualization (i.e. display the line vis only) The behavior is similar if no latest numeric value is found. This commit also updates the single stat computation to use the @influxdata/vis `Table` format as an intermediate/parsed representation of a Flux CSV response. This unlocks the possibility for performance gains in our CSV parsing. See #13852. Closes #13824
chnn
added a commit
that referenced
this issue
May 8, 2019
The method we used to compute a single stat / gauge value previously had a few issues: - If the latest value was part of a numeric column but was null/NaN/not defined, the single stat computation would fail and a user would see an error message "Could not display single stat because your values are non-numeric" - If there were multiple latest values, an arbitrary selection would be made This commit updates the single stat computation to find the latest *defined* numeric values. If multiple latest valid numeric values are found, we will either: - Display an error message if using the compuation within a single stat visualization - Display nothing if using the computation within a within a line + single stat visualization (i.e. display the line vis only) The behavior is similar if no latest numeric value is found. This commit also updates the single stat computation to use the @influxdata/vis `Table` format as an intermediate/parsed representation of a Flux CSV response. This unlocks the possibility for performance gains in our CSV parsing. See #13852. Closes #13824
chnn
added a commit
that referenced
this issue
May 8, 2019
The method we used to compute a single stat / gauge value previously had a few issues: - If the latest value was part of a numeric column but was null/NaN/not defined, the single stat computation would fail and a user would see an error message "Could not display single stat because your values are non-numeric" - If there were multiple latest values, an arbitrary selection would be made This commit updates the single stat computation to find the latest *defined* numeric values. If multiple latest valid numeric values are found, we will either: - Display an error message if using the compuation within a single stat visualization - Display nothing if using the computation within a within a line + single stat visualization (i.e. display the line vis only) The behavior is similar if no latest numeric value is found. This commit also updates the single stat computation to use the @influxdata/vis `Table` format as an intermediate/parsed representation of a Flux CSV response. This unlocks the possibility for performance gains in our CSV parsing. See #13852. Closes #13824
chnn
added a commit
that referenced
this issue
May 9, 2019
The method we used to compute a single stat / gauge value previously did account for missing data. If the latest value in a response was part of a numeric column but was null/NaN/not defined, the single stat computation would fail and a user would see an error message "Could not display single stat because your values are non-numeric". This commit updates the single stat computation to find the latest *defined* numeric values. If no latest valid numeric values are found, we will either: - Display an error message if using the compuation within a single stat visualization - Display nothing if using the computation within a within a line + single stat visualization (i.e. display the line vis only) If multiple latest values are found, we make an arbitrary selection (same as previous behavior). The goal is to eventually expose UI elements to the user so they can make this selection themselves. This commit also updates the single stat computation to use the @influxdata/vis `Table` format as an intermediate/parsed representation of a Flux CSV response. This unlocks the possibility for performance gains in our CSV parsing. See #13852. Closes #13824
chnn
added a commit
that referenced
this issue
May 9, 2019
The method we used to compute a single stat / gauge value previously did account for missing data. If the latest value in a response was part of a numeric column but was null/NaN/not defined, the single stat computation would fail and a user would see an error message "Could not display single stat because your values are non-numeric". This commit updates the single stat computation to find the latest *defined* numeric values. If no latest valid numeric values are found, we will either: - Display an error message if using the compuation within a single stat visualization - Display nothing if using the computation within a within a line + single stat visualization (i.e. display the line vis only) If multiple latest values are found, we make an arbitrary selection (same as previous behavior). The goal is to eventually expose UI elements to the user so they can make this selection themselves. This commit also updates the single stat computation to use the @influxdata/vis `Table` format as an intermediate/parsed representation of a Flux CSV response. This unlocks the possibility for performance gains in our CSV parsing. See #13852. Closes #13824
chnn
added a commit
that referenced
this issue
May 10, 2019
The method we used to compute a single stat / gauge value previously did account for missing data. If the latest value in a response was part of a numeric column but was null/NaN/not defined, the single stat computation would fail and a user would see an error message "Could not display single stat because your values are non-numeric". This commit updates the single stat computation to find the latest *defined* numeric values. If no latest valid numeric values are found, we will either: - Display an error message if using the compuation within a single stat visualization - Display nothing if using the computation within a within a line + single stat visualization (i.e. display the line vis only) If multiple latest values are found, we make an arbitrary selection (same as previous behavior). The goal is to eventually expose UI elements to the user so they can make this selection themselves. This commit also updates the single stat computation to use the @influxdata/vis `Table` format as an intermediate/parsed representation of a Flux CSV response. This unlocks the possibility for performance gains in our CSV parsing. See #13852. Closes #13824
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Details to reproduce coming shortly...
The text was updated successfully, but these errors were encountered: