python-backend.md
In greta: Simple and 'Scalable' Statistical Modelling in R

setters warn when RETICULATE_PYTHON overrides the stored preference

Code
  greta_set_python()
Message
  v Stored preference: the managed (uv) Python environment.
Condition
  Warning:
  ! `RETICULATE_PYTHON` is set to '/fake/python' and takes precedence over the stored preference.
  i greta resolves Python in this order:
    1. `RETICULATE_PYTHON` - usually set in '~/.Renviron' or your shell environment
    2. Stored preference - set with `greta_set_python()`
    3. Auto-detected "greta-env-tf2" conda environment - created by `install_greta_deps()`
    4. The managed (uv) environment - the default, no setup needed
  i To use your stored preference, remove `RETICULATE_PYTHON` from '~/.Renviron' (or wherever it is set), then restart R.
  i See the installation vignette: `vignette(greta::installation)`.
Message
  i After you restart R, greta will use:
  * backend: "user-specified Python"
  * python: '/fake/python'
  * selected via: RETICULATE_PYTHON environment variable

setters do not warn when RETICULATE_PYTHON is unset

Code
  greta_set_python()
Message
  v Stored preference: the managed (uv) Python environment.
  i After you restart R, greta will use:
  * backend: "managed (uv) environment"
  * selected via: greta preference (see ?greta_set_python)

greta_set_python() hints when a path is passed as backend

Code
  greta_set_python("/usr/bin/python")
Condition
  Error in `greta_set_python()`:
  ! `backend` must be one of "uv", "conda", or "path".
  i To use the Python at '/usr/bin/python', run `greta_set_python("path", path = "/usr/bin/python")`.

greta_set_deps() reports what it stored

Code
  greta_set_deps()
Message
  v Stored dependency versions: TensorFlow "2.15.1", TensorFlow Probability "0.23.0", Python "3.11".
  i Restart R for this to take effect.

greta_set_deps(NULL) errors and points at greta_remove('deps')

Code
  greta_set_deps(NULL)
Condition
  Error in `greta_set_deps()`:
  ! `deps` must be a `greta_deps_spec()` object.
  i To clear stored dependency versions, use `greta_remove("deps")`.

report_offline_readiness() reports non-managed backends as ready

Code
  report_offline_readiness(plan = plan)
Message
  v offline-ready: this environment is already on disk; greta never downloads
  into it

report_offline_readiness() warns when a non-managed python is gone

Code
  report_offline_readiness(plan = plan)
Message
  x the selected Python environment no longer exists on disk (was it removed?);
  restart R to re-resolve it
  i See the installation vignette: `vignette(greta::installation)`.

report_offline_readiness() reports managed backend states

Code
  report_offline_readiness(plan, empty, uv_offline = "")
Message
  x will need internet on next start: uv cache not yet populated
  i See the installation vignette: `vignette(greta::installation)`.

Code
  report_offline_readiness(plan, empty, uv_offline = "1")
Message
  x `UV_OFFLINE`=1 is set but the uv cache is not yet populated, so the next
  start may fail to resolve dependencies
  i See the installation vignette: `vignette(greta::installation)`.

Code
  report_offline_readiness(plan, full, uv_offline = "")
Message
  v offline-ready: uv cache present, offline mode will engage

Code
  report_offline_readiness(plan, full, uv_offline = "1")
Message
  v offline-ready: `UV_OFFLINE`=1 is set and the uv cache is populated

Code
  report_offline_readiness(plan, full, uv_offline = "0")
Message
  x will need internet on next start: `UV_OFFLINE`=0 forces online resolution
  i See the installation vignette: `vignette(greta::installation)`.