Extended ASCII characters

Extended ASCII characters are handled several different ways in the full-text filter. Some are special characters while others are treated as spaces. Some are treated as letters that depend on the spaces around the characters.

Special characters or punctuation are not indexed and cannot be searched for in filters. The following table describes what behavior applies to these characters in the filter.

Table 1. Extended ASCII Character syntax
Extended ASCII Character Syntax Description
Special characters Some special characters generate syntax errors because they cannot be processed. Either remove the character from the filter or surround the filter with single quotation marks. For example, a filter for hello! without quotations generates an error. However, if the filter is 'hello!' (with single quotation marks), the query can be processed.

In general, special characters are treated as a space in a filter and they are not indexed. For example, Hello! is indexed as Hello. Searching for 'Hello!' returns zero object. To search for 'hello', data object that has the word hello! or hello is returned. However, there are some exceptions, and they are discussed later in this table

Most special characters can also be searched for with quotation marks. For example, "12/24/2014" finds 12/24/2014. The difference between quotation marks and single quotation marks is that quotation marks also have another function: to find words that are next to each other. Because of this, special characters that are used elsewhere in the syntax, such as * and ?, are treated as the full-text operators for wildcards instead of as the characters * and ?.

These special characters generate errors unless surrounded by quotation marks: @ / & : ! ( ) { } [ ] ^ ~ , # =
Period (.) A period with no space on either side is treated like a letter. For example, a filter for 172.16.0.46 finds the whole number of 172.16.0.46, not partial numbers of 172, 16, 0, 46.

A period with a space on either side is treated as a space and ignored. For example, a filter for "1234 Mahana St." ignores the period and just finds 1234 Mahana St only without a period.
At (@) An @ symbol is indexed like a letter when there are no spaces on either side. For example, a filter for jdoe@yourcompany.com finds the entire email address. An @ symbol with a space on either side or one side is treated as a space and ignored. For example, @home finds home in the data object.
Hyphen (-) A hyphen is indexed like a letter when there are no spaces on either side. For example, a filter for 012-44-5668 finds the entire number, not 012, 44, 5668 as separate numbers. A hyphen with a space on either side is treated as a space and ignored.
Slash (/) A slash is indexed like a letter when there are no spaces on either side. For example, a filter for 10/30/1940 finds the entire expression 10/30/1940 (and not 10 30 1940) as separate numbers.
Ampersand (&) An ampersand is indexed like a letter when there are no spaces on either side. For example, a filter for S&P finds S&P in the data objects. However, a filter for Proctor & Gamble without quotation marks finds either Proctor or Gamble. A filter with quotation marks "Proctor & Gamble" finds the phrase Proctor Gamble only.
Colon (:) A colon is considered a special character. It is treated as a space in filter and is not indexed. For example, 21:35:15 is indexed as three consecutive words, that is, 21, 35, and 15. Searching for '21:35:15' returns zero object. To search for '21 35 15', data object that has the word 21:35:15 or that has three consecutive words as 21, 35, 15 is returned. Specifying terms with colons without the single quotation marks results in an invalid syntax error.
Apostrophe (') The apostrophe in the term is indexed as a whole word. For example, a filter for patient's returns results for patient's but not for patient. A filter for the name O’Hara returns results for O’Hara. An apostrophe is treated as a single quotation mark, so it must always be quoted with double quotation marks.
Comma (,) Commas are left in when surrounded by numerals, and those numerals make a legitimate number. For example, "1,000" is indexed as 1,000. To search this term, specify 1,000 in single quotation marks or quotation marks. Otherwise, it results in an invalid syntax error. For the term "10,00", it is indexed as 10 and 00. To search for this term, specify "10 00".
Number sign (#), dollar sign ($), percent (%), equal sign (=), underscore (_) These characters are treated as white space in the filter. For example, a search for $100 returns results for the number 100 whether 100 refers to dollars, a percentage, or a whole number. The term 'JAVA_HOME' is indexed as 'JAVA HOME'. To search for this term, specify JAVA HOME in quotation marks.