First commit

readme.md

@@ -0,0 +1,63 @@
+Training Database
+=================
+
+Toolset for registrering weight training with a focus on ease of registrering from
+a command line.
+
+To install, start by creating the database:
+
+```
+create database training;
+create user 'jakob';
+```
+
+Edit `pg_hba.conf` to have 'jakob' access the training database directly:
+
+```
+# TYPE  DATABASE        USER            ADDRESS                 METHOD
+local   training        jakob                                   peer
+```
+
+As postgres user, do a config reload:
+
+```
+select pg_reload_conf();
+```
+
+Then install rust toolchain by rustup.rs -- make and install the binary (I am assuming you have
+a `bin` directly in your homedir, that is in your path):
+
+```
+cd train-cli
+cargo build --release
+cp target/release/train-cli $HOME/bin/train
+```
+
+Right now the executable is hardcoded with database name `training` and the use
+of unix domain socket in `/var/run/postgresql`.
+
+Now training can be invoked with:
+
+```
+train squat 3 10 60
+```
+
+Which would register squauts, 3 runs of 10 reps of 60kg -- at local time and date. Optionally a
+time, date time or rfc3339 timestamp can be specified:
+
+```
+train squat 3 10 60 "12:05:00"
+train squat 3 10 60 "2025-12-24 18:00:00"
+train squat 3 10 60 "2025-12-24T18:00:00+0200"
+```
+
+The two former will source missing date and timezone information from the user session, i.e.
+type `date` on your commandline to see what you have.
+
+When travelling you might opt for specifying a specific location on the command line, like:
+
+```
+TZ=Australia/Sydney train squat 3 10 60
+```
+
+Data in the database can be visualized with, for example, Grafana, more info to follow.