Matching Strategies

Every rule has a strategy that determines how FunkArr extracts season/episode information from a Mediathek entry. Choosing the right strategy depends on how the show's titles are structured.

Choosing a Strategy

Strategy	Use when...	Example shows
seasonAndEpisodeNumber	Titles contain season and episode numbers, even in non-standard formats	Shows with "Staffel 3 Folge 12" or "S03E12" in the title
byAbsoluteEpisodeNumber	Episodes use a single running number with no seasons	Schloss Einstein (1000+ episodes), daily soaps
itemTitleExact	Episode titles are unique and must match exactly	Anthology series with distinct titles per episode
itemTitleIncludes	Episode titles are embedded in a longer string	Tatort ("Tatort: Borowski und die Kinder")
itemTitleEqualsAirdate	Episodes are identified by their broadcast date	heute-show, Die Sendung mit der Maus, talk shows

seasonAndEpisodeNumber

Extracts a season and episode number from the Mediathek title using regex patterns.

When to use: The title contains season and episode information in any format — German ("Staffel 3 Folge 12"), abbreviated ("S03/E12"), or embedded in parentheses.

Required fields:

• seasonRegex — pattern to extract the season number
• episodeRegex — pattern to extract the episode number
• captureGroup (optional) — which capture group to use (default: 1)

Example:

For a show where titles look like "Feuer und Flamme (S03/E05)":

{
  "id": "regex-se",
  "priority": 0,
  "strategy": "seasonAndEpisodeNumber",
  "seasonRegex": "S(\\d{2})",
  "episodeRegex": "E(\\d{2})",
  "filters": {
    "all": [
      { "field": "duration", "op": "greaterThan", "value": "20" }
    ]
  }
}

The seasonRegex captures 03 from "S03", and the episodeRegex captures 05 from "E05".

TIP

The regex must contain a capture group (). FunkArr extracts the content of the first capture group by default. Use captureGroup to select a different group.

byAbsoluteEpisodeNumber

Extracts a single absolute episode number with no season. Sonarr maps it to the correct season internally using its episode guide.

When to use: Long-running shows that use a single running episode counter. Common for German children's shows and daily series.

Required fields:

• episodeRegex — pattern to extract the episode number
• captureGroup (optional)

Example: Schloss Einstein

Mediathek titles: "Schloss Einstein (1042)", "Schloss Einstein (Folge 1043)"

{
  "id": "absolute-episode",
  "priority": 0,
  "filters": {
    "all": [
      { "field": "duration", "op": "greaterThan", "value": "15" }
    ]
  },
  "strategy": "byAbsoluteEpisodeNumber",
  "episodeRegex": "\\((\\d{3,5})\\)"
}

The regex \((\d{3,5})\) matches a 3-to-5-digit number inside parentheses and extracts it. For "Schloss Einstein (1042)", the extracted episode number is 1042.

A second rule with lower priority handles the "Folge" variant:

{
  "id": "absolute-episode-p1",
  "priority": 1,
  "strategy": "byAbsoluteEpisodeNumber",
  "episodeRegex": "\\(Folge (\\d{3,5})\\)"
}

itemTitleExact

Constructs a title string from parts and matches it exactly against the Sonarr/Radarr episode title. The constructed title must be an exact match.

When to use: Episode titles are unique and predictable. The Mediathek title contains the episode title verbatim or can be extracted cleanly.

Required fields:

• titleRules — array of title parts (static text and/or regex extractions)

Example:

For a show where the Mediathek title is exactly the episode title:

{
  "id": "title-exact",
  "priority": 0,
  "strategy": "itemTitleExact",
  "titleRules": [
    { "type": "regex", "field": "title", "pattern": "^(.+)$" }
  ]
}

itemTitleIncludes

Constructs a title string from parts and checks if the Sonarr/Radarr episode title contains it as a substring. More forgiving than exact matching.

When to use: Episode titles are embedded in the Mediathek title with extra formatting, or when exact matching is too brittle.

Required fields:

• titleRules — array of title parts

Example: Tatort

Mediathek titles: "Tatort: Borowski und die Kinder", "Tatort: Der letzte Schrei"

{
  "id": "title-includes",
  "priority": 0,
  "filters": {
    "all": [
      { "field": "duration", "op": "greaterThan", "value": "35" }
    ]
  },
  "strategy": "itemTitleIncludes",
  "titleRules": [
    { "type": "static", "value": " - " },
    {
      "type": "regex",
      "field": "title",
      "pattern": "(?<=Tatort:\\s*)\\S.*"
    }
  ]
}

How it works:

1. The static part appends " - " to the constructed title
2. The regex part extracts everything after "Tatort: " from the Mediathek title
3. Result: " - Borowski und die Kinder"
4. Sonarr searches for an episode title that contains this substring

The Tatort ruleset includes a second rule at priority 1 with a more lenient regex for edge cases where the title format varies.

Example: Movie matching (Apocalypse Now)

{
  "id": "title-match",
  "priority": 0,
  "filters": {
    "all": [
      { "field": "topic", "op": "contains", "value": "Spielfilm" }
    ]
  },
  "strategy": "itemTitleIncludes",
  "titleRules": [
    { "type": "static", "value": "Apocalypse Now" }
  ]
}

This matches any entry in the "Spielfilm" topic that contains "Apocalypse Now" in its title.

itemTitleEqualsAirdate

Extracts a broadcast date from the title. Used for shows identified by their air date rather than season/episode numbers.

When to use: Talk shows, news programs, weekly magazines, and daily shows where episodes are identified by broadcast date.

Required fields:

• titleRules — regex to extract the date from the Mediathek title

Example: heute-show

Mediathek titles: "heute-show vom 13. September 2026"

{
  "id": "airdate",
  "priority": 0,
  "filters": {
    "all": [
      { "field": "duration", "op": "greaterThan", "value": "25" }
    ]
  },
  "strategy": "itemTitleEqualsAirdate",
  "titleRules": [
    {
      "type": "regex",
      "field": "title",
      "pattern": "^heute-show vom (\\d{1,2}\\. \\w+ \\d{4})"
    }
  ]
}

The regex extracts "13. September 2026" from the title. FunkArr parses this into a date and matches it to the episode with that air date in Sonarr.

Example: Die Sendung mit der Maus

Mediathek titles use a numeric date format: "Die Sendung mit der Maus 15.09.2026"

{
  "id": "airdate",
  "priority": 0,
  "filters": {
    "all": [
      { "field": "duration", "op": "greaterThan", "value": "21" }
    ]
  },
  "strategy": "itemTitleEqualsAirdate",
  "titleRules": [
    {
      "type": "regex",
      "field": "title",
      "pattern": "\\b\\d{2}\\.\\d{2}\\.\\d{4}\\b"
    }
  ]
}

TIP

For airdate-matched shows, set the Series Type to Daily in Sonarr. Without this, Sonarr won't look up episodes by date.

Title Construction Deep-Dive

Title rules (titleRules) build a string by appending parts left to right. Each part is either:

• Static — appends literal text
• Regex — runs a regex against a Mediathek field and appends the match

Evaluation order

Part 1        Part 2        Part 3
[static]  →  [regex]   →  [static]  →  final title
" - "        "Borowski"    ""            " - Borowski"

Regex parts

A regex part has three components:

1. field — which Mediathek field to search (usually title)
2. pattern — the regex to run
3. captureGroup — which group to extract (default: 0 = full match)

If the regex doesn't match, the entire rule fails identification.

Multiple rules for robustness

Real-world Mediathek titles vary. A common pattern is to define multiple rules at increasing priority levels — the first handles the standard case, and fallback rules handle edge cases:

"rules": [
  {
    "id": "main",
    "priority": 0,
    "strategy": "itemTitleIncludes",
    "titleRules": [{ "type": "regex", "field": "title", "pattern": "strict-pattern" }]
  },
  {
    "id": "fallback",
    "priority": 1,
    "strategy": "itemTitleIncludes",
    "titleRules": [{ "type": "regex", "field": "title", "pattern": "lenient-pattern" }]
  }
]

Priority 0 is tried first. If it fails (regex doesn't match or filter rejects the candidate), priority 1 gets a chance.