defmodule Memorex.Scheduler.Config do
@moduledoc """
The configuration for the [Anki SM-2 algorithm](https://faqs.ankiweb.net/what-spaced-repetition-algorithm.html).
See a detailed list of the Anki settings in this [Anki forum](https://forums.ankiweb.net/t/deck-options-explained/213)
The [Anki manual](https://docs.ankiweb.net/deck-options.html) also contains a lot of information.
These values are modified by deck config `toml` files, and also by environment variables (see `config/runtime.exs`).
| Memorex Setting | Anki Setting | Anki Default |
|:--------------------------|:--------------------------------|----------------:|
| new_cards_per_day | new cards per day | 20 |
| max_reviews_per_day | maximum reviews per day | 200 |
| - | - | - |
| learn_ahead_time_interval | learn ahead time (in settings?) | 20 minutes |
| - | - | - |
| learn_steps | learning steps | [1 min, 10 min] |
| graduating_interval_good | graduating interval | 1 day |
| graduating_interval_easy | easy interval | 4 days |
| - | - | - |
| relearn_steps | relearning steps | [10 min] |
| - | - | - |
| initial_ease | starting ease | 2.5 |
| - | - | - |
| easy_multiplier | easy bonus | 1.3 |
| hard_multiplier | hard interval | 1.2 |
| lapse_multiplier | new interval (? I think so) | 0.0 |
| interval_multiplier | interval modifier | 1.0 |
| - | - | - |
| ease_again | <not in settings> | -0.2 |
| ease_hard | <not in settings> | -0.15 |
| ease_good | <not in settings> | 0.0 |
| ease_easy | <not in settings> | 0.15 |
| ease_minimum | <not in settings> | 1.3 |
| - | - | - |
| max_review_interval | maximum interval | 100 years |
| min_review_interval | minimum interval | 1 day |
| - | - | - |
| leech_threshold | leech threshhold | 8 lapses |
| - | - | - |
| min_time_to_answer | <not in settings> | 1 sec |
| max_time_to_answer | maximum answer seconds | 1 min |
| - | - | - |
| relearn_easy_adj | NOT SURE WHERE THIS IS FROM | 1 day |
| - | - | - |
| timezone | timezone | 1 day |
"""
alias Timex.Duration
@type t() :: %__MODULE__{
new_cards_per_day: nil | non_neg_integer(),
max_reviews_per_day: nil | non_neg_integer(),
#
learn_ahead_time_interval: nil | Duration.t(),
#
learn_steps: nil | [Duration.t()],
graduating_interval_good: nil | Duration.t(),
graduating_interval_easy: nil | Duration.t(),
#
relearn_steps: nil | [Duration.t()],
#
initial_ease: nil | float(),
#
easy_multiplier: nil | float(),
hard_multiplier: nil | float(),
lapse_multiplier: nil | float(),
interval_multiplier: nil | float(),
#
ease_again: nil | float(),
ease_hard: nil | float(),
ease_good: nil | float(),
ease_easy: nil | float(),
ease_minimum: nil | float(),
#
max_review_interval: nil | Duration.t(),
min_review_interval: nil | Duration.t(),
#
#
leech_threshold: nil | non_neg_integer(),
#
min_time_to_answer: nil | Duration.t(),
max_time_to_answer: nil | Duration.t(),
#
relearn_easy_adj: nil | Duration.t(),
#
timezone: nil | String.t()
}
defstruct [
:new_cards_per_day,
:max_reviews_per_day,
#
:learn_ahead_time_interval,
#
:learn_steps,
:graduating_interval_good,
:graduating_interval_easy,
#
:relearn_steps,
#
:initial_ease,
#
:easy_multiplier,
:hard_multiplier,
:lapse_multiplier,
:interval_multiplier,
#
:ease_again,
:ease_hard,
:ease_good,
:ease_easy,
:ease_minimum,
#
:max_review_interval,
:min_review_interval,
#
:leech_threshold,
#
:min_time_to_answer,
:max_time_to_answer,
#
:relearn_easy_adj,
#
:timezone
]
@duration_fields [
:learn_ahead_time_interval,
#
:graduating_interval_good,
:graduating_interval_easy,
#
:max_review_interval,
:min_review_interval,
#
:min_time_to_answer,
:max_time_to_answer,
#
:relearn_easy_adj
]
@duration_array_fields [:learn_steps, :relearn_steps]
@doc """
Returns the default `Memorex.Schedler.Config` as returned from the application environment (e.g., environment
variables. Converts from the application env map to an actual `Memorex.Scheduler.Config` struct.
"""
@spec default() :: map()
def default() do
Application.get_env(:memorex, __MODULE__)
|> Enum.reduce(%__MODULE__{}, fn {key, value}, acc ->
Map.put(acc, key, value)
end)
end
@doc """
Merges in a map into the `Memorex.Scheduler.Config`. This is used to merge in deck-specific configuration from
toml config files.
"""
@spec merge(map(), t()) :: t()
def merge(default_config, deck_config) do
config = Map.merge(default_config, atomize_keys(deck_config))
config =
@duration_fields
|> Enum.reduce(config, fn field_name, config ->
Map.put(config, field_name, convert_string_to_duration(Map.get(config, field_name)))
end)
@duration_array_fields
|> Enum.reduce(config, fn field_name, config ->
converted_array_values = Map.get(config, field_name) |> Enum.map(&convert_string_to_duration(&1))
Map.put(config, field_name, converted_array_values)
end)
end
@spec atomize_keys(map()) :: map()
defp atomize_keys(map), do: map |> Enum.into(%{}, fn {key, value} -> {String.to_atom(key), value} end)
@spec convert_string_to_duration(String.t() | Duration.t()) :: Duration.t()
defp convert_string_to_duration(duration) when is_binary(duration), do: Duration.parse!(duration)
defp convert_string_to_duration(duration), do: duration
end
