2018-06-12 Meeting notes

• George Swinnerton has done the NHSExtract for this month
• Nurture Dashboard updates are configured using cron jobs but have not run correctly. George Swinnerton will sort this. Currently no log file is created and this will be added to ensure we have documented log of data flow. The documentation will also need updating to document the process.
• George Swinnerton has freed up some space on HS. There are still issues that patients don't seem to have had their current result despite new files being uploaded since it had issues. The exact cause of this is unclear. George Swinnerton will run a script to reload the latest file uploaded to PV again for each patient to ensure they should have their latest results.
• The demo server has wordpress on just needs configuring for Rhodri to use.
• Formstorm continues to be the main delay in the survey scanning. Currently it is not correctly creating the csv/database entries from every 6 pages resulting in some partial/part surveys. Awaiting fixes from Richard of the config files to sort this. Once this is sorted data should be available for Rapolas to run his report generator against. He has found the PAM api can support only 50 char ids so plans to use a combination of hashing and uuid to create the unique identifier required from the NHS number. This code will be put in a module so that the survey import code can make use of the procedures for calling the api.@Rapolas will finish off his report and look at populating the PAM Level and PAM score fields in the surveys table so the database holds a complete reference set of data.
• foghorn_leghorn (Unlicensed) is working on the validation counts and has a standalone script which he is now integrating in to the main codebase. George Swinnerton raised issue that the counts stored in the database should only reflect the counts when a file is uploaded to the database and not be updated every time validation is done. Currently the counts are being corrupted when a validation is re-run after an upload has occurred. A test version should be ready for the next validation meeting.
• George Swinnerton has reviewed the UKT Data ticket and it now complete

foghorn_leghorn (Unlicensed)

Roger has arranged a meeting with NHSBT on 21 Jun 2018 at 10:00am. This is to discuss the NHSBT matching process and fields that are required etc. with an aim to making the process smoother.

 Tim Whitlock (Deactivated)

We continue to have issues with working code being changed because it has not been understood properly by the individual doing the changes. This has led to time being wasted and broken systems since dependencies could be broken. We need to realise we are all part of a team with different knowledge and communicate directly to ensure we understand what is happening. It is important to ensure you understand why something has been done before deciding to change it especially if it is not your code. This means talking to other team members and maybe delaying some work until this can be done.

It was suggested that more use of pulls and branching would help clarify that these changes are being made. It was laso noted the repository for the UKRDC is gaining a lot of comments that say fixing test or flake8 but give no useful help. In general these sorts of issues shoul dbe fixed in local testing before uploading to the repository.