Fix Version/s: None
Sprint:SUIT Sprint 2018-07
Team:Science User Interface
In the PDAC instance of Firefly I have now a few times run into a problem with downloading tables that contain very wide values in exponential notation. Sometimes, not always, the last digit of the exponent is truncated.
For instance, I have seen "2.606998244411E-04" get truncated to "2.606998244411E-0".
I think this is reproducible right now if you search for
in the NEOWISE-R Year 1 Single Exposure (L1b) Source Table (neowiser_yr1_00.neowiser_yr1_p1bs_psd).
The value for the attribute "y" for the selected row comes back from dbserv as "0.0002606998244411" and is then written out in exponential notation to the IPAC Table download, sometimes one way, sometimes the other.
It seems to depend on whether other rows are also being written, so it may have something to do with the calculation of the required field width based on the observed values.
- relates to
DM-15089 Display format for numerical values for LSST data
In the current dev version, the number is not truncated any more.
In IRSA meta data, there is information 'dbtype', which give out the display format for numbers, e.g. 'number(9,7)'. Firefly will display 6 digits after decimal point if there is no specific format information. LSST needs to think about the display precision of different numbers.
The bug reported here will be fixed in next update of SUIT portal. A new ticket will be created for display information gathering.
please see my comment and let me know if it is OK to close this ticket.
I reproduced the reported bug in LSST PDAC. The attached two files are the saved search result, one from LSST with the bug, one from OPS IRSA viewer without bug.