Ground rules

This is a public repo; make sure that linked data sources are secure and that outputs are properly ignored.

Data linkage

You will need to have access to data files in a Dropbox. The default location of this Dropbox is ~/Dropbox/Rabies_TZ/. If your data is elsewhere (or if you're using a different Dropbox), you will need to either set up links, or make a local configuration file. If ls ~/Dropbox/Rabies_TZ/Tanzania_Animal*.csv shows you a bunch of WiseMonkey files, you're probably OK.

If not, but you already have a personal .local file, say make <yourname>.config after make Makefile (see below). NB: There is a file called katie.local

Note that the Dropbox is also currently used to share and collate output files. If you just link Animal_CT.csv to a WiseMonkey file, you will be able to do checks, and examine local copies of the outputs, but not move the files to the shared Dropbox or use the current version of the report.

Getting started

File setup

1. Clone this repo
2. make Makefile
3. Sort out your Data linkage issues (see above)
4. make dropsetup
5. ls -l *.csv
   • This should show two data files in the main directory, and where they point to in the Dropbox

R setup

You will need:

1. The program R
2. The tidyverse set of packages
   • `install.packages("tidyverse")
3. The shellpipes package

R script workflow

The script SD_dogs.allchecks.Rscript sources all of the files needed to make the report, and moves them to their final locations at the end (the system call).

The generic script report.rmd is set up for the hack-a-thon to focus on SD_dogs. Thus, you should be able edit it and knit it however you like as long as you leave the few magic lines at the beginning.

After the hack-a-thon, I'll build a version that should be easier to modify for other branches.

Make workflow

The current report structure is based on configuration scripts in a subdirectory called (branch)[branch/]; so far this only one: SD_dogs (here is the selection file).

make SD_dogs.report.html should just work to make the report, and related .csv files linked from the report.

make SD_dogs.report.html.go will often work to make the report and automatically open it on your screen.

To add a new report to the pipeline, it should be sufficient to make (and commit) a new config file branch/.R, and then follow steps above.

Working on the pipeline

You should be able to edit any of the R scripts listed in the script file and then run any part of the R pipeline by using the script file, or by using make and the target name of your choice, e.g., make SD_dogs.wildlifeCheck.Rout.csv.

You should also be able to run any R script directly from rstudio, modify it and run it again. The rpcall statement at the top tells it where to read and save things. To save changes, re-run the save-like commands at the bottom, and to incorporate upstream changes, re-run the load-like commands at the top (or just always run the script top-to-bottom).

.csv files are made first in the main directory. The report right now can only be made by make; this has a step to copy the .csv files to where the report wants them. csv files in the main directory can be deleted without harming the report, once it's made.

Editing the report

You should be able to edit report.rmd in a straightforward manner in rstudio or any text editor. For now, you should use make to produce the html file. I'm still working on that.

Updating WiseMonkey

To change the focal WiseMonkey file, you need to delete Animal_CT.csv in the main directory, and make sure that the file you want to use is the most recent file matching Animal.csv in the datadir (you can access via here or your Dropbox folder). If you download a new WiseMonkey file and delete Animal_CT.csv, make should just work.

To focus another WiseMonkey file, you can update its modification time (use touch from the command line, or do something Mac-ish if you know how). This should rarely be necessary, and remember to touch the latest file when you're done whatever test you are doing.