[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

The remarkable similarities between XSLT and Flex/Lex

• From: Roger L Costello <costello@mitre.org>
• To: "xml-dev@lists.xml.org" <xml-dev@lists.xml.org>
• Date: Fri, 24 Jun 2022 10:46:03 +0000

Hi Folks,

Flex and Lex are tools to auto-generate lexical analyzers. Flex is a successor to Lex. Lex was created 47 years ago. Flex was created about the same time as XSLT.

There is a remarkable similarity between Flex/Lex and XSLT. One wonders if some members of the XSLT 1.0 Working Group were Lex users and were influenced by its concepts?

Here are some of the similarities between Flex/Lex and XSLT:

- Both are pattern-matching languages, i.e., 

	pattern	  action
	pattern   action
	pattern   action

- Both allow you to create subsets:

	- The XSLT "mode" allows the program to effectively be broken into a set of mini XSLT programs
	- The Flex/Lex "start condition" allows the lexer to effectively be broken into a set of mini lexers

- Both have a default rule that is executed when no other rule matches

There are some differences, of course:

- XSLT is primarily for processing XML but it can process plain text, whereas Flex/Lex is primarily for processing plain text (esp. source code) but it can process XML

- XSLT uses XPath as its pattern-matching language, whereas Flex/Lex uses regular expressions as its pattern-matching language

Pretty interesting, I think!

/Roger

• Follow-Ups:
  • RE: The remarkable similarities between XSLT and Flex/Lex
    • From: Roger L Costello <costello@mitre.org>
  • Re: The remarkable similarities between XSLT and Flex/Lex
    • From: Michael Kay <mike@saxonica.com>