Add a new scored value to this collection. The score provided must be less than or equal to every other score in this collection, otherwise this meth

Consume this scored value and map its value using the function given, returning a new scored value with the result of the map and an unchanged score.

Returns true if and only if no similarity function was selected.

Execute the given function over each ngram in the text provided using the given size configuration. We don't use normal Rust iterators here because a

Return a reference to the kind of this error.

Return a reference to the underlying value.

Associate the given identifier with the given integer. If the given key is not strictly lexicographically greater than the previous key, then an erro

Returns the number of results in this collection.

Set the name to query by. The name given here is normalized and broken down into components automatically to facilitate fuzzy searching. Note that i

Return a reference to the underlying `Title`.

Consume this scored value and return a new scored value that drops the existing score and replaces it with the given score. This panics if the given

Return the integer associated with the given ID, if it exists.

Returns the directory containing this index's files.

Returns true if and only if this collection is empty.

Return the score with the current document ID. If this iterator has been exhausted, then this returns `None`.

Trim this collection so that it contains at most the first `size` results.

Return a string representation of this title kind. This string representation is intended to be the same string representation used in the IMDb data

Consume this collection and return the underlying sorted sequence of scored values.

Return the current posting. If this iterator has been exhausted, then this returns `None`.

Collect the top K results from the given searcher using the given index reader. Return the results with normalized scores sorted in order of best-to-w

A function for creating a CSV reader builder that is pre-loaded with the correct settings for reading all IMDb CSV files.

Finish writing the index. This must be called, otherwise the index will likely be unreadable.

Return true if and only if this query is empty. Searching with an empty query always yields no results.

Returns the document frequency for the term corresponding to these postings.

Consume this scored value and map its score using the function given, return a new `Scored` with an unchanged value. This panics if score returned by

Return the current position of this writer.

Builds a CSV reader (using `csv_reader_builder`) that is backed by a file. While this read can be seeked, it will be less efficient than using a memor

Returns the data directory that this index returns results for.

Inserts the given name to this index, and associates it with the provided `NameID`. Multiple names may be associated with the same `NameID`.

Parse a sequence of values from clap.

Recomputes the scores in this collection using the given function. The results are then re-sorted according to the new scores.

Return the score for this item. In general, no restrictions are placed on the range of scores, however most search APIs that use it will return score

Creates a file and truncates it.

Creates a name index, where each name provided is assigned its own unique ID, starting at 0.

Return the current document ID. If this iterator has been exhausted, then this returns `None`.

Returns the `MediaEntity` for the given `Title`. This is like the `entity` method, except it takes a `Title` record as given.

Return a reference to the underlying `Episode`, if it exists.

Return a (possibly empty) iterator over all AKA records for the given IMDb ID.

Consume this scored value and return the underlying pair of score and `T`.

Builds a file-backed memory map.

Build a name query, and disable the dynamic stop word detection. It would be nice to test the stop word detection, but it makes writing unit tests ve

Read the CSV `Title` record beginning at the given file offset. If no such record exists, then this returns `None`. If there was a problem reading t

Execute a search.

Set the score, replacing the existing value with the given value. This panics if the given score is `NaN`.

Return a slice of search results in order.

Returns the `MediaEntity` for the given IMDb ID. An entity includes an [`Episode`](struct.Episode.html) and [`Rating`](struct.Rating.html) records if

Creates an FST set builder for the given file path.

Open an FST set file for the given file path as a memory map.

Return a mutable reference to the underlying index for this searcher.

Consume this scored value, drop the score and return the underlying `T`.

Add a title kind to filter by. Multiple title kinds can be added to query, and search results must match at least one of them. Note that it is not p

Returns true if and only if the given entity matches this query. Note that this only applies filters in this query. e.g., The name aspect of the quer

Opens a file for reading.

Set the maximum number of results to be returned by a search. Note that setting this number too high (e.g., `> 10,000`) can impact performance. This

Make a choice among the search results given. If there is no clear winner, then a prompt is shown to the end user, where they must make a selection.

Execute a search against the given searcher with the given query and choose a single result from the search. If no obvious single result stands out, t

Create a new index using default settings. Calling this method is expensive, and one should expect this to take dozens of seconds or more to complete

Builds a CSV reader (using `csv_reader_builder`) that is backed by a seekable memory map. We use memory maps for this even though we could use a norm

Build the path on disk for a dataset, given the directory and the dataset name.

Return the length, in terms, of the given document. This panics if the given document id does not correspond to an indexed document.

Downloads a single data set, decompresses it and writes it to the corresponding file path in the given directory.

Creates an FST map builder for the given file path.

Open an FST map file for the given file path as a memory map.

Either open or create an index suitable for this specification. If no index exists in the expected sub-directory of `eval_dir`, then a new index is c

Returns true if and only if the given title matches this query. This ignores non-title filters.

Build a query and seed it with the given name, after sanitizing the name.

Set the scorer to use for name searches. The name scorer is used to rank results from searching the IMDb name index. If no name query is given, then

Returns true if and only if this query has an episode filter.

Returns true if and only if this query has a rating filter.

Set the ngram size on this index. When creating an index, ngrams with this size will be used.

Set the type of ngram generation to use. The default type is `Window`.

Normalizes the scores in this collection such that all scores are in the range `[0, 1]` where the top result always has score `1.0`. This operation i

Return the total number of documents have been assigned doc ids.

Return the underlying path to this temporary directory.

Return a reference to the underlying `Rating`, if it exists.

Execute a search with the given `Query`. Generally, the results returned are ranked in relevance order, where each result has a score associated with

Execute a search against the given searcher with the given query. If this exact query has been previously executed by this renamer, then a cache of r

Set the lower inclusive bound on a title's season. This automatically limits all results to episodes.

Set the upper inclusive bound on a title's season. This automatically limits all results to episodes.

Build all evaluation specifications as indicated by command line options.