Character-based wildcard scheme

The character-based wildcard scheme allows you to select topics based on traditional character matching.

You can select all topics at multiple levels in a topic hierarchy using the string '*'. Using '*' in the character-based wildcard scheme is equivalent to using the topic-based wildcard string '#'

' x/*/y ' is equivalent to ' x/#/y ' in the topic-based scheme, and selects all topics in the topic hierarchy between levels ' x and y ', where ' x ' and ' y ' are topic names that are not in the set of levels returned by the wildcard.

'/+/' in the topic-based scheme has no exact equivalent in the character-based scheme. 'IBM/*/Results' would also select 'IBM/Patents/Software/Results'. Only if the set of topic names at each level of the hierarchy are unique, can you always construct queries with the two schemes that yield identical matches.

Used in a general way, '*' and '?' in the character-based scheme have no equivalents in the topic-based scheme. The topic-based scheme does not perform partial matching using wildcards. The character based wildcard subscription 'IBM/*ware/Results' has no topic-based equivalent.
Note: Matches using character wildcard subscriptions are slower than matches using topic-based subscriptions.

Character-based wildcard string

Read syntax diagramSkip visual syntax diagramV6 literal%*?%1*2?3
V6 literal
Read syntax diagramSkip visual syntax diagramAny Unicode character except *, ? and % 
Notes:
  • 1 Means escape the following character, so that it is treated as a literal. '%' must be followed by either '*', '?' or '%'. See Examples of topic strings.
  • 2 Means match zero or more characters in a subscription.
  • 3 Means match exactly one character in a subscription.

Examples

IBM/*/Results
IBM/*ware/Results