Regex positive lookahead when lookahead data doesn’t exist (PCRE)

regex

I'm trying to match out a string which will have a format similar to the following:

username=joe password='this! is a password with "quotes" and spaces#913 custom1 customd afd weirdhuh? custom1=myvalue

To explain in more detail, this is a list of variables separated by equal signs. The valid variables that can be passed are: username, password, and customX (where X is any number of digit(s)).

I am specifically trying to match out the password field as the passed variable can technically have any number of quotes, spaces, etc that are valid characters. I have decided that the best "break point" to use to determine when the password string terminates is the existence of a "customX=" string which infers that the next variable is starting.

Therefore, in the above example the actual password would be:

'this! is a password with "quotes" and spaces#913 custom1 customd afd weird huh?

I have arrived at the following regex:

(?i)password(?-i)=.+?(?= (?i)custom(?-i)\d+=)

This appears to match the following:

password='this! is a password with "quotes" and spaces#913 custom1 customd afd weird huh?

This is in effect what I want (I can easily parse out the "password="), but the problem is this regex only seems to work if there is an existence of that final "custom1=myvalue" at the end of the string. If that is removed, then there is no match at all.

I need to be able to match the password string regardless if that final value is there or not.

Bonus points if you can strip out the "password=" to end up only with the actual password.

Best Answer

but the problem is this regex only seems to work if there is an existence of that final "custom1=myvalue"

You can add an optional end of string to match instead of custom1=myvalue:
(?i)password(?-i)=.+?((?= (?i)custom(?-i)\d+=)|$)

Bonus points if you can strip out the "password=" to end up only with the actual password.

Use lookbehind when you match password=:
(?<=(?i)password(?-i)=).+?((?= (?i)custom(?-i)\d+=)|$)

Related Solutions

Regex to parse URLs from text

If you don't want to include the trailing whitespace in a match, use a negative lookahead:

[a-z]+[:.].*?(?=\s)

In your example, this would match:

regexbuddy.com
www.regexbuddy.com
http://regexbuddy.com
http://www.regexbuddy.com
http://www.regexbuddy.com/
http://www.regexbuddy.com/index.html
http://www.regexbuddy.com/index.html?source=library
http://www.regexbuddy.com/download.html.
www.domain.com/quoted
http://10.2.2.1.2/ttxx/txt/gg

To further match only http or https, and optional www use something like:

(https?):\/\/(www\.)?[a-z0-9\.:].*?(?=\s)

Here's John Gruber's regex to check for what looks like an URL, which appears to work quite well in your case:

(?i)\b((?:[a-z][\w-]+:(?:\/{1,3}|[a-z0-9%])|www\d{0,3}[.]|[a-z0-9.\-]+[.][a-z]{2,4}\/)(?:[^\s()<>]+|\(([^\s()<>]+|(\([^\s()<>]+\)))*\))+(?:\(([^\s()<>]+|(\([^\s()<>]+\)))*\)|[^\s`!()\[\]{};:'".,<>?«»“”‘’]))

But honestly, all those approaches will only get you false matches sooner or later. If you need a regular expression to parse URLs, see this Stack Overflow question: What is the best regular expression to check if a string is a valid URL?

Word – Regular Expression to match zero or more spaces in Microsoft Word

The document you linked to shows that Microsoft's "regular expressions" aren't really regular expressions at all; they're a bizarre hybrid (bastard child, rather) of shell-style globbing (http://www.tldp.org/LDP/GNU-Linux-Tools-Summary/html/x11655.htm) and true regular expressions.

Since the glob syntax makes use of the * character as a synonym for the regex .*, and Microsoft decided (as mentioned in a comment) to make @ equivalent to the regex quantifier + instead of * (which is stupid since a+ is equivalent to aa* for any atom a, making + unnecessary), it looks like you're out of luck.

My personal opinion is that (1) this is stupid and (2) calling these patterns "regular expressions" is misleading at best, but unfortunately I don't see any way around this except for abandoning Word in favor of a tool that properly supports regex. (Though I suppose in theory you could try to parse the xml-ish format of the docx file itself, extract the text, and then apply your regex....)

Best Answer

Related Solutions

Regex to parse URLs from text

Word – Regular Expression to match zero or more spaces in Microsoft Word

Related Question