Duplicate recording detection and episode identity

Duplicate recording detection compares intended content identities and existing capture outcomes so repeat airings can be skipped without confusing different episodes or incomplete recordings.

A repeat airing can be a second opportunity to capture the same episode. Duplicate detection decides whether that opportunity adds something wanted or merely creates another copy.

The decision depends on what “same” means and on whether the earlier recording is adequate.

Equal titles and equal content are different tests

Consider four invented airings:

Airing Display title Content key
A100 Harbour Stories P/episode/E1
A240 Harbour Stories P/episode/E1
A250 Harbour Stories P/episode/E2
A300 Coast Weather Unknown

A100 and A240 describe the same intended episode. A250 has the same series title but another episode. A300 cannot be given a stable episode identity from its title alone.

A title-only rule merges E1 and E2 incorrectly. An airing-ID-only rule fails to recognize that A100 and A240 repeat E1.

A useful content key retains its provider/type scope. Programme identifiers explains that distinction.

Existing does not necessarily mean sufficient

Use these original prior-capture records:

[
  {"content": "P/episode/E1", "outcome": "complete"},
  {"content": "P/episode/E2", "outcome": "partial"},
  {"content": null, "title": "Coast Weather", "outcome": "complete"}
]

Under a declared teaching policy, a complete matching content key suppresses a repeat, a partial match remains eligible for replacement and an unknown key requires another decision.

Candidate Result under that policy
Repeat of E1 Skip: adequate matching capture
Repeat of E2 Keep eligible: prior capture partial
Another Coast Weather Identity uncertain

The table illustrates policy rather than claiming Emby uses these exact outcome strings or completeness tests.

An application can instead prioritize avoiding duplicates, filling gaps, obtaining a better-quality edition or retaining every airing. Those choices have different expected results.

Numbering and edition affect identity

Season/episode coordinates need the correct series and episode order. A missing episode number is unknown rather than episode one.

A two-part airing, an edited repeat and a combined multi-episode programme add further distinctions. They can belong to the same story without being interchangeable captured artifacts.

Likewise, identical bytes provide strong evidence of equal artifacts, but do not identify another encoding of the same episode. Content metadata and byte equality answer different questions.

The library check is one input to a series rule

Emby’s series-rule contract includes skipping episodes in the library. The usefulness of that option depends on the identity information available for the airing and existing item.

If the guide supplies only a series title, confidently distinguishing individual daily episodes becomes harder. If an existing item was misidentified, a correct future airing can be skipped for the wrong reason.

See Also

References