This repository has been archived by the owner on Apr 11, 2022. It is now read-only.
Full title match is better than split match #739
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
aslagle
reviewed
Nov 30, 2017
external_search.py
Outdated
| @@ -453,6 +453,13 @@ def make_target_age_query(target_age): | |||
| match_phrase = make_phrase_query(query_string, ['title.minimal', 'author', 'series.minimal']) | |||
| must_match_options.append(match_phrase) | |||
|
|
|||
| # An exact title or author match outweighs a match that is split | |||
| # across fields. | |||
| match_title = make_phrase_query(query_string, ['title.minimal'], 150) | |||
There was a problem hiding this comment.
What happened to title.standard?
…bject is created.
…as to having a single setting for the prefix. Index and alias are always calculated from the prefix.
…s the other queries get the same boost it works out.
|
I think you need a migration for this - for the configuration setting change, if nothing else. |
|
I forgot to add the migration script. All it does is create a brand new index using the default prefix. I don't think there are any existing instances where this won't work, and if there are any, we don't know enough about their setup to automatically migrate them. |
|
I could also add a migration script to remove the old configuration settings, just to avoid any confusion when looking at the database in the future. |
aslagle
reviewed
Dec 1, 2017
tests/test_external_search.py
Outdated
| ) | ||
|
|
||
| # TODO: Uncomment these lines and the 'modern romance' | ||
| # test fails for some reason. |
leonardr
commented
Dec 1, 2017
| "standard": { | ||
| "type": "string", | ||
| "analyzer": "standard" | ||
| } |
There was a problem hiding this comment.
This is the part that's different.
aslagle
approved these changes
Dec 1, 2017
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This branch adds two new ways a book might show up in search results: its title or one of its authors might be a near-exact match against the search term.
This should resolve the 'modern romance' and 'law of the mountain man' problems described in #482, where a book whose title includes a genre loses out to books that are in that genre. It may also resolve some of the 'game of thrones' problems described in that issue.
This branch changes the Elasticsearch configuration used by core during Travis runs so that everything runs in a single shard; this eliminates the possibility that search result ordering will be unpredictable because of how documents are assigned to shards.
This branch also changes the way Elasticsearch is configured: instead of separate configuration settings for the alias and the index, a single configuration setting controls the prefix used for both the alias and the index. This means that changing the alias to point to some other index is currently an unsupported operation -- it's supported by the code (transfer_current_alias) but nobody ever calls that method.
This is a good time to make that configuration change because this branch introduces a v3 of the search document, which requires that the entire index be rebuilt anyway. (The migration script should take care of this.)