Appendix F
URL Splitting Algorithm Pseudo-Code
Below is the pseudo-code defining the algorithm used to split URLs within links into a domain, path and filename. It re-builds relative URLs in sub-pages, and re-combines the individual parts of the URL into a complete URL for download.
Function is passed: url, is_subpage Global variables used: stored_domain, stored_path Function returns: new_url domain := NULL path := NULL filename := NULL IF url STARTS WITH "http://" THEN url := RIGHT OF url AFTER AND NOT INCLUDING "http://" IF "/" NOT IN url THEN domain := url path := "/" filename := NULL ELSE domain := LEFT OF url UP TO AND NOT INCLUDING FIRST "/" url := RIGHT OF url AFTER AND INCLUDING FIRST "/" filename := RIGHT OF url AFTER AND NOT INCLUDING LAST "/" url := LEFT OF url UP TO AND INCLUDING LAST "/" path := url END IF ELSE IF url STARTS WITH "mailto:" THEN ERROR "This is a link to an email address" EXIT ELSE IF ":" IN url THEN ERROR "Can only download pages using HTTP" EXIT END IF IF "/" NOT IN url THEN path := "/" filename := url ELSE filename := RIGHT OF url AFTER AND NOT INCLUDING LAST "/" url := LEFT OF url UP TO AND INCLUDING LAST "/" path := url END IF END IF IF is_subpage = TRUE AND domain = NULL THEN domain := stored_domain path := stored_path + path ELSE IF is_subpage = TRUE AND NOT domain = stored_domain THEN ERROR "Sub page was not within domain of the main page" ELSE IF is_subpage = FALSE AND domain = NULL THEN ERROR "The URL specified is invalid" ELSE stored_domain := domain stored_path := path END IF new_url := "http://" + domain + path + filename