การรวมการสอบถามด้วย Bool

Welcome to combining queries! Up until now, we've looked at single search conditions. But real-world searches often need multiple criteria.

Elasticsearch's bool query is your go-to for building complex search logic. It lets you combine several queries using AND, OR, and NOT conditions.

A bool query acts like a container for other queries. It has four main clauses:

• must: All queries inside must match (AND).
• should: At least one query inside should match (OR).
• must_not: Queries inside must NOT match (NOT).
• filter: Queries inside must match, but don't affect relevancy score.

We'll explore each one!

The must clause is like an "AND" operator. All queries listed within must must be true for a document to be considered a match.

Documents that satisfy all must clauses contribute positively to the relevancy score.

This query finds products that are both "laptop" AND "gaming".

The should clause is like an "OR" operator. If a document matches any of the queries listed in should, it's considered a potential match.

The more should queries a document matches, the higher its relevancy score usually is.

By default, if there are no must clauses, at least one should clause must match for a document to be returned.

This query finds products that are either "laptop" OR "tablet".

The must_not clause is like a "NOT" operator. Documents matching any query inside must_not will be excluded from the results.

Queries in must_not do not contribute to the relevancy score.

This query finds all products EXCEPT those in the "refurbished" category.

The filter clause is similar to must in that all queries within it must match. However, there's a key difference:

• No Scoring: Filter queries don't affect the relevancy score of documents.
• Caching: Filter queries are often cached, making them very fast for frequently used criteria.

Use filter for "yes/no" conditions where you just want to include or exclude documents without influencing their rank.

You can combine all these clauses within a single bool query to build powerful, precise searches. This example finds:

• Products with "laptop" in the name (must)
• THAT are either "gaming" OR "ultrabook" (should)
• BUT are NOT "refurbished" (must_not)
• AND have a status of "available" (filter)

Consider a bool query designed to find articles:

• About "Elasticsearch" (must)
• Published in "2023" OR "2024" (should)
• But NOT written by "John Doe" (must_not)

Which of the following statements is TRUE regarding this query's behavior?

Great job! You've learned how to combine simple queries into powerful, complex search logic using Elasticsearch's bool query.

• must: All conditions must be met (AND).
• should: At least one condition should be met (OR).
• must_not: Documents must NOT match these conditions (NOT).
• filter: Conditions must be met, but don't affect relevancy (fast & cached).

Understanding bool queries is crucial for building precise and effective search applications!