Text Contains Clauses (Spider)

[Table of Contents](https://github.com/dell-oss/Doradus/wiki/Spider Databases: Table-of-Contents) | Previous | Next
Doradus Query Language: Text Contains Clauses

---

Text fields can be queried for values that contain specific terms. In contrast, equality clauses (described later) compare the entire field value.

Term Clauses

A term clause searches a specific field for one or more terms. To designate it as a contains clause, the field name must be followed by a colon. Example:

Name:Smith

This clause searches perspective objects’ Name field for the term Smith (case-insensitive). To specify multiple terms, enclose them in parentheses. Example:

Name:(John Smith)

To match this clause, an object’s Name field must contain both John and Smith, but they can appear in any order and be separated by other terms.

Be sure to enclose multi-term clauses in parentheses! The following query looks like it searches for John and Smith in the Name field:

Name:John Smith // doesn’t do what you think!

But, this sequence is actually interpreted as two clauses that are AND-ed together:

Name:John AND *:Smith

Matching objects must have John in the Name field and Smith in any field.

Phrase Clauses

A phrase clause is a contains clause that searches for a field for a specific term sequence. Its terms are enclosed in single or double quotes. For example:

Name:"John Sm*th"

This phrase clause searches the Name field for the term John immediately followed by a term that matches the pattern Sm*th. The matching terms may be preceded or followed by other terms, but they must be in the specified order and with no intervening terms. As with term clauses, phrases clauses can use wildcards, and searches are performed without case sensitivity.

Any-field Clauses

Term and phrase clauses can search all fields of the perspective table by eliminating the field name qualifier. The simplest form of an any-field clause is a term clause:

John

This clause searches all scalar fields of the perspective table for a value contains the term John (case-insensitive). All indexed scalar fields are searched, which means those whose analyzer is other than NullAnalyzer. Fields indexed with the OpaqueTextAnalyzer will match only if the field’s value equals John. Fields that use other analyzers will match if they contain the term John. However, Boolean, timestamp, and numeric fields do not contain text terms, so they will never contain a text term such as John.

However, suppose we perform an any-field search for an integer value:

12226

If an object has a text field called MessageID with the value "Host1-12226-Alpha1", 12226 is one the terms generated by the field, hence the above clause would match the field. If another object has an integer field called Size whose value is 12226, it would also match since the IntegerAnalyzer generates a term equal to the field’s value. Hence, objects could match a term clause based on the type of terms generated by their scalar field analyzers.

Multiple terms can be provided by separating them by at least one space. For example:

John Smith

This multi-term clause searches for the terms John and Smith in any order and across any field. For example, an object will match if it has a FirstName field that contains John and a LastName field that contains Smith. If all terms are contained in the same field, they can appear in any order and be separated by other terms. For example, an object will match the example above if it has a Name field whose value is "John Smith", “John Q. Smith”, or “Smith, John”.

An any-field clause can be requested using the field qualifier syntax by using the field name "*". Example:

*:(John Smith)

This is the same as the unqualified term clause:

John Smith

Phrase clauses can also use the field qualifier syntax and request an any-field search:

*:"John Smith"

This query finds object where the terms John and Smith appear in succession in any field. The phrase can be preceded or followed by other terms.