Instead of wildcards, you can specify a Perl regular expression to expand a query with all matching words.
Note: Regular expressions are used to find single words. A single regular expression cannot be used to find a phrase.
Due to the way data is pre-processed and stored in the CORE system, regular expression search does not work for:
- numbers
- any terms containing numbers
- any of the configured delimiter characters
- any terms containing any of the configured delimiter characters
Each regular expression in a query must be preceded by regex=.
Each regular expression is an exact match search, regardless of the query's search mode.
|
Regular expression |
Finds |
|---|---|
| regex=agre+ment |
agreement and several possible misspellings This will find agrement and agreement and agreeement, as the plus the regular expression matches the preceding element one or more times. |
| regex=licen.e | licence and license The dot in the regular expression matches a single character. |
Regular Expressions in the Axcelerate 5 User Interface
If you use regular expressions in Axcelerate 5, there are some additional items you need to be aware of.
- In
Axcelerate 5,
the regex operator is case sensitive. It must be written all lower-cased. If
you enter other spellings like
RegexorREGEXit will result in a syntax error message. - The Axcelerate 5 user interface only accepts a single regular expression in one keyword search. A regular expression must not be combined with other search expressions in one keyword search, not even with a second regular expression.
- You can nevertheless use regular expressions in conjunction with other search expressions by adding multiple keyword searches to the Current Criteria panel.
- Similarly, in the Search Query Editor, put any regular expression in a row of its own.
Example
A keyword search
crawler regex=r[ua]n is not possible in Axcelerate 5.
However, you can run two separate keyword searches and add them to the Current Criteria:
- Run a keyword search for
crawler. - Run a second keyword search for
regex=r[ua]n.
https://en.wikipedia.org/wiki/Regular_expression
Useful link for testing: https://regexr.com/