Dev hiv ped 24

[mwalte10]

24

Set up storage for stratified population of pediatric HIV (hivstrat_paeds). Have aging mapped out as well, but will wait to confirm that that is working as expected until natural deaths are working appropriately.

It looks like for nosocomial infections in spectrum all infections are put in the highest CD4 category at initial distribution, I've replicated that here.

[mwalte10]

Also, have commented out the transmission matches tests because of the connection with demographics not matching (bc aging / natural deaths not integrated)

[mwalte10]

@r-ash, I'm not sure why these checks are now failing. To me it looks like there are just warnings. Do you see anything I should change?

[r-ash]

@r-ash, I'm not sure why these checks are now failing. To me it looks like there are just warnings. Do you see anything I should change?

You're right they are just warnings, current settings are to fail on warning and it is worth fixing warnings. Best to fix them by qualifying calls to unzip as utils::unzip, read.csv as read.csv and utils::read.csv and setNames as stats::setNames to avoid any ambiguity about what package the functions are from.

[mwalte10]

Looks like it isn't liking the ::: import of an internal function, I found this but not quite sure how to implement https://github.com/drsimonj/twidlr/issues/16 @r-ash @jeffeaton

Also: https://stackoverflow.com/questions/2675502/what-does-the-symbol-mean-in-r

[r-ash]

Looks like it isn't liking the ::: import of an internal function, I found this but not quite sure how to implement drsimonj/twidlr#16 @r-ash @jeffeaton

Also: https://stackoverflow.com/questions/2675502/what-does-the-symbol-mean-in-r

As the packages it is complaining about are under our control we can choose to export those functions.

‘epp:::.parse_array’ - @jeffeaton can we export this?

eppasm:::create_beers - @jeffeaton can we export this? Or use an already exported function from https://github.com/mrc-ide/beers ?

If these aren't really things we don't want to export we could refactor some of this into a helper package which we then use here and elsewhere. That is a bit more faff than you probably want though so we can turn off fail on warning if this really is a pain.

[jeffeaton]

Thanks @r-ash, sorry had parallel discussions going on this.

The ::: issue is not new and itself only gives a note (but yes we should resolve them).

What has escalated it to a Warning is pulling in the read-spectrum-files.R script, which references a bunch of packages that are not declared in Imports:, e.g. abind here: https://github.com/mrc-ide/leapfrog/blob/dev_hiv_ped_24/R/read-spectrum-files.R#L1073

@mwalte10 -- let's chat about if we want to copy that full script in.