The microdata table shows aggregates instead of microdata (and misses the row-section dim value)
In prod
https://data-explorer.oecd.org/vis?lc=en&fs[0]=Topic%2C0%7CFinance%23FIN%23&pg=0&fc=Topic&snb=2&vw=tb&df[ds]=dsDisseminateFinalDMZ&df[id]=DSD_DEBT_TRANS_DDOWN%40DF_DDOWN&df[ag]=OECD.DAF&df[vs]=1.0&pd=2020%2C&dq=........_T.&ly[cl]=TIME_PERIOD&ly[rs]=REF_AREA&mdc[REF_AREA]=THA&mdc[MEASURE]=GDO&mdc[TIME_PERIOD]=2020,
when clicking on a link in one of the cells (e.g. "Gross debt outstanding (USD)" for "Thailand" for 2020), then the DE fires the following NSI query:
Note that dimension "DD_DIM" is defined as the "DRILLDOWN" dimension, and that this dimension is at the second last position in the key. It means that the microdata query still uses "DD_DIM"="_T", which is incorrect.
The doc says: "the microdata preview-table is constructed out of the data resulting from the SDMX query that corresponds to the data selection (including also all respective child dimension values) of the related aggregated data value, but by replacing the selection “_T” for the DRILLDOWN dimension by “DD”. It thus only displays microdata, and no aggregated data."
The same issue can be reproduced in QA: https://de-qa.siscc.org/vis?lc=en&df[ds]=qa%3Astable&df[id]=DSD_DEBT_TRANS_DDOWN%40DF_DDOWN&df[ag]=OECD.DAF&df[vs]=1.0&av=true&pd=2020%2C&dq=........_T.&ly[cl]=TIME_PERIOD&ly[rs]=REF_AREA&vw=tb&mdc[REF_AREA]=SEN&mdc[MEASURE]=GDO&mdc[TIME_PERIOD]=2020
Here, in addition to miss the change from "DD_DIM"="_T" to "DD_DIM"="DD", the "MEASURE" dimension (which is in the row sections) misses its value in the generated microdata query: https://nsi-qa-stable.siscc.org/rest/data/OECD.DAF,DSD_DEBT_TRANS_DDOWN@DF_DDOWN,1.0/.UZB......._T.?startPeriod=2020&endPeriod=2020&dimensionAtObservation=AllDimensions.
Therefore the microdata table shows unrelated microdata.