Only the documents that are indexable can be searched. For example, pictures or binary content cannot be searched because they are not indexable.

Moreover, not all characters are searchable. Searchable characters are alphabetic, numeric, extender, and custom characters. Custom characters enclose Chinese, Japanese, Korean letters, and months.

Other characters, including punctuation, accent, and diacritical marks, and characters such as | and #, are not indexed or searched. Such nonsearchable characters are removed from the indexed text, and treated as if they are blank spaces. The index server treats these characters as white space: !@#$%^_,.&;:()+=<

When these characters appear in indexable content, they are replaced by white space. For example, when the email address [email protected] is indexed, it appears as “MyName company com” in the index. The text is treated as three words. Documents returned by a search for [email protected] are treated as if they contain the words “MyName company com”.

If a special character is included in a query, it is removed. For example, querying on Richard+Dodd would return a document containing the text Richard=Dodd because the + and = signs are both replaced by a blank space. If a search term includes an accent or diacritical mark, the search returns all matching words with or without the accent or diacritical mark.