pinchflat/lib/pinchflat/media_source.ex
Kieran dabec62e99
Redo indexing mechanism (#16)
* Bumped up the line length because I fear no man

* Refactored indexing

Previously, indexing worked by collecting the video IDs of only videos
that matched indexing criteria. This new model instead stores ALL videos
for a given source, but will only _download_ videos that meet that criteria.
This lets us backfill without indexing, makes it easier to add in other
backends, lets us download one-off videos for a source that don't quite
meet criteria, you name it.

* Updated media finders to respect format filters; Added credo file
2024-02-09 13:23:37 -08:00

165 lines
4.9 KiB
Elixir

defmodule Pinchflat.MediaSource do
@moduledoc """
The MediaSource context.
"""
import Ecto.Query, warn: false
alias Pinchflat.Repo
alias Pinchflat.Tasks
alias Pinchflat.Tasks.SourceTasks
alias Pinchflat.MediaSource.Source
alias Pinchflat.MediaClient.SourceDetails
@doc """
Returns the list of sources. Returns [%Source{}, ...]
"""
def list_sources do
Repo.all(Source)
end
@doc """
Gets a single source.
Returns %Source{}. Raises `Ecto.NoResultsError` if the Source does not exist.
"""
def get_source!(id), do: Repo.get!(Source, id)
@doc """
Creates a source. May attempt to pull additional source details from the
original_url (if provided). Will attempt to start indexing the source's
media if successfully inserted.
Returns {:ok, %Source{}} | {:error, %Ecto.Changeset{}}
"""
def create_source(attrs) do
%Source{}
|> change_source_from_url(attrs)
|> commit_and_start_indexing()
end
@doc """
Updates a source. May attempt to pull additional source details from the
original_url (if changed). May attempt to start indexing the source's
media if the indexing frequency has been changed.
Existing indexing tasks will be cancelled if the indexing frequency has been
changed (logic in `SourceTasks.kickoff_indexing_task`)
Returns {:ok, %Source{}} | {:error, %Ecto.Changeset{}}
"""
def update_source(%Source{} = source, attrs) do
source
|> change_source_from_url(attrs)
|> commit_and_start_indexing()
end
@doc """
Deletes a source and it's associated tasks (of any state).
Returns {:ok, %Source{}} | {:error, %Ecto.Changeset{}}
"""
def delete_source(%Source{} = source) do
Tasks.delete_tasks_for(source)
Repo.delete(source)
end
@doc """
Returns an `%Ecto.Changeset{}` for tracking source changes.
"""
def change_source(%Source{} = source, attrs \\ %{}) do
Source.changeset(source, attrs)
end
@doc """
Returns an `%Ecto.Changeset{}` for tracking source changes and additionally
fetches source details from the original_url (if provided). If the source
details cannot be fetched, an error is added to the changeset.
Note that this fetches source details as long as the `original_url` is present.
This means that it'll go for it even if a changeset is otherwise invalid. This
is pretty easy to change, but for MVP I'm not concerned.
NOTE: When operating in the ideal path, this effectively adds an API call
to the source creation/update process. Should be used only when needed.
IDEA: Maybe I could discern `collection_type` based on the original URL?
It also seems like it's a channel when the returned yt-dlp channel_id is the
same as the playlist_id - maybe could use that?
"""
def change_source_from_url(%Source{} = source, attrs) do
case change_source(source, attrs) do
%Ecto.Changeset{changes: %{original_url: _}} = changeset ->
add_source_details_to_changeset(source, changeset)
changeset ->
changeset
end
end
defp add_source_details_to_changeset(source, changeset) do
%Ecto.Changeset{changes: changes} = changeset
case SourceDetails.get_source_details(changes.original_url) do
{:ok, source_details} ->
add_source_details_by_collection_type(source, changeset, source_details)
{:error, runner_error, _status_code} ->
Ecto.Changeset.add_error(
changeset,
:original_url,
"could not fetch source details from URL",
error: runner_error
)
end
end
defp add_source_details_by_collection_type(source, changeset, source_details) do
%Ecto.Changeset{changes: changes} = changeset
collection_type = source.collection_type || changes[:collection_type]
collection_changes =
case collection_type do
:channel ->
%{
collection_id: source_details.channel_id,
collection_name: source_details.channel_name
}
:playlist ->
%{
collection_id: source_details.playlist_id,
collection_name: source_details.playlist_name
}
_ ->
%{}
end
change_source(source, Map.merge(changes, collection_changes))
end
defp commit_and_start_indexing(changeset) do
case Repo.insert_or_update(changeset) do
{:ok, %Source{} = source} -> maybe_run_indexing_task(changeset, source)
err -> err
end
end
defp maybe_run_indexing_task(changeset, source) do
case changeset.data do
# If the changeset is new (not persisted), attempt indexing no matter what
%{__meta__: %{state: :built}} ->
SourceTasks.kickoff_indexing_task(source)
# If the record has been persisted, only attempt indexing if the
# indexing frequency has been changed
%{__meta__: %{state: :loaded}} ->
if Map.has_key?(changeset.changes, :index_frequency_minutes) do
SourceTasks.kickoff_indexing_task(source)
end
end
{:ok, source}
end
end