Regular expressions 101

Regular expressions are under-valued and most developers tend to only know the basics. Having a thorough understanding of how regular expressions work, will be incredibly helpful when you need to parse structured data.

This presentation will assume you already know what regular expressions are, but will sum up (with an example) some fancy things you probably didn’t know were possible with regular expressions.

If you're interested in a more detailed write-up, I suggest you check out http://www.mullie.eu/regular-expressions-basics/ & http://www.mullie.eu/regular-expressions-advanced/

This presentation is based on the PHP-implementation of PCRE, but nearly all programming languages support the same functionality, albeit sometimes with their own twists.

Transcript

1. The Swiss Army knife of string manipulation
   REGEX 101
   

   View Slide

2. @matthiasmullie
   Regular expressions 101
   

   View Slide

3. Regular expressions 101
   INTRODUCTION
   What are regular expressions?
   

   View Slide

4. Regular expressions 101 » Introduction
   Google
   Regular expressions are special
   characters that match or capture
   portions of a field, as well as the rules
   that govern all characters.
   

   View Slide

5. Regular expressions 101 » Introduction
   Wikipedia
   A regular expression provides a
   concise and flexible means for
   "matching" strings of text, such as
   particular characters, words, or
   patterns of characters.
   

   View Slide

6. Regular expressions 101 » Introduction
   /\{\$([a-z0-9_]*)((\.[a-z0-9_]*)*)

   (-\>[a-z0-9_]*((\.[a-z0-9_]*)*))?

   ((\|[a-z_][a-z0-9_]*(:.*?)*)*)\}/i
   

   View Slide

7. Regular expressions 101 » Introduction
   Me
   Regular expressions find patterns in
   strings.
   

   View Slide

8. Regular expressions 101 » Introduction
   Neque porro quisquam est qui
   dolorem ipsum quia dolor sit amet,
   consectetur, adipisci velit...
   !
   ‣ /[a-z]/i
   ‣ /[^\w]/i
   !
   ‣ /ipsum/
   ‣ /(est|qui)/
   

   View Slide

9. Regular expressions 101
   BASICS
   The syntax everyone should know already
   

   View Slide

10. /Delimiter/
    Regular expressions 101 » Delimiter
    ‣ Any [^a-zA-Z0-9\\\s]
    character
    ‣ Opening char == terminating char
    ‣ Except for [ ]
    , ( )
    , { }
    and < >
    

    View Slide

11. Regular expressions 101 » Delimiter
    Use /
    (uniformity, you know)
    

    View Slide

12. ‣ .
    ‣ [ ]
    ‣ ^ $
    ‣ |
    Meta characters
    Regular expressions 101 » Meta characters
    ‣ ( )
    ‣ \
    ‣ * ? +
    ‣ {n} {n,m}
    

    View Slide

13. ‣ i
    ‣ m
    ‣ s
    ‣ x
    ‣ e
    ‣ A
    ‣ D
    ‣ U
    ‣ J
    ‣ ...
    Pattern modifiers //x
    Regular expressions 101 » Pattern modifiers
    

    View Slide

14. Character classes [ ]
    Regular expressions 101 » Character classes
    Ranges Inverse ranges
    ‣ [0-9]
    ‣ [a-zA-Z]
    ‣ [A-F0-9]
    ‣ [^0-9]
    ‣ [^a-zA-Z]
    ‣ [^A-F0-9]
    

    View Slide

15. Character classes [ ]
    Regular expressions 101 » Character classes
    No sequence of characters!
    !
    [lorem]
    ‣ l, o, r, e or m
    ‣ lorem
    

    View Slide

16. Character classes [ ]
    ‣ [:alnum:]
    ‣ [:blank:]
    ‣ [:lower:]
    ‣ ...
    Regular expressions 101 » Character classes
    POSIX
    

    View Slide

17. Greediness: greedy
    Regular expressions 101 » Greediness
    list-item1list-item2
    !
    /.*<\/li>/
    ‣ list-item1list-item2
    

    View Slide

18. Greediness: lazy
    Regular expressions 101 » Greediness
    list-item1list-item2
    !
    /.*?<\/li>/ /.*<\/li>/U
    ‣ list-item1
    ‣ list-item2
    or
    

    View Slide

19. Subpatterns
    Regular expressions 101 » Subpatterns
    /([a-z0-9]*)@([a-z0-9\.]*\.[a-z0-9]{2,3})/i email
    !
    hostname
    user
    Note: this regex only barely satisfies my needs for this particular example; do not use this really find occurrences of email addresses, it does not fully satisfy RFC5321 & RFC5322
    

    View Slide

20. Questions?
    Regular expressions 101
    

    View Slide

21. Regular expressions 101
    ADVANCED
    The juicy stuff you never knew about, until now
    

    View Slide

22. Regular expressions 101 » Back references
    Problem: /href=['"](.*?)['"]/i
    Matches:
    ‣ href="xxx"
    ‣ href='xxx'
    !
    !
    ‣ href="xxx'
    ‣ href='xxx"
    Back references
    

    View Slide

23. Regular expressions 101 » Back references
    Back references
    Solution: /href=(['"])(.*?)\1/i
    \1 references first subpattern!
    !
    Don’t forget to also string-escape in PHP:
    preg_match('/href=([\'"])(.*?)\\1/i', ...);
    

    View Slide

24. Regular expressions 101 » Named subpatterns
    Named subpatterns
    Scenario: parsing large CSV
    1,a title,5.00,92,green
    2,another title,3.50,4,blue
    3,one more,33699.99,15,white
    ...
    

    View Slide

25. /([0-9]+),(.*?),([0-9]+\.[0-9]{2}),([0-9]+),([a-z]+)/i
    !
    !
    Result excerpt:
    Regular expressions 101 » Named subpatterns
    Named subpatterns
    [1] => string(1) "1"
    [2] => string(7) "a title"
    [3] => string(4) "5.00"
    [4] => string(2) "92"
    [5] => string(5) "green"
    !
    !
    !
    !
    

    View Slide

26. /(?P[0-9]+),(?P.*?),(?P[0-9]+\.[0-9]
    {2}),(?P[0-9]+),(?P[a-z]+)/i
    !
    Result excerpt:
    Regular expressions 101 » Named subpatterns
    ["id"] => string(1) "1"
    [1] => string(1) "1"
    ["title"] => string(7) "a title"
    [2] => string(7) "a title"
    ["price"] => string(4) "5.00"
    [3] => string(4) "5.00"
    ["stock"] => string(2) "92"
    [4] => string(2) "92"
    ["color"] => string(5) "green"
    [5] => string(5) "green"
    Named subpatterns
    

    View Slide

27. Regular expressions 101 » Named subpatterns
    Named subpatterns
    ‣ (?Ppattern)
    ‣ (?pattern) & (?'name'pattern)
    since PHP 5.2.2
    

    View Slide

28. Regular expressions 101 » Named subpatterns
    Named subpatterns +
    back references
    !
    /href=(?P['"])(?P.*?)(?P=quotes)/i
    

    View Slide

29. Regular expressions 101 » Assertions
    “Take a peek, don’t eat it”
    Lookahead/-behind assertions
    

    View Slide

30. Scenario: find all occurrences of “here”
    
    !
    “Where can I find here, not there?”
    Regular expressions 101 » Assertions
    Lookahead/-behind assertions
    

    View Slide

31. Regular expressions 101 » Assertions
    Lookahead/-behind assertions
    Deduction:
    Find all here’s, not preceded or followed by
    an alphabetic character.
    !
    Solution: /(?

    View Slide

32. Regular expressions 101 » Assertions
    Lookahead/-behind assertions
    ‣ Positive lookahead: (?=expression)
    ‣ Negative lookahead: (?!expression)
    ‣ Positive lookbehind: (?<=expression)
    ‣ Negative lookbehind: (?

    View Slide

33. “lookbehind assertion is not fixed length...”
    In PHP, lookbehind can not contain repetition,
    while lookahead can.
    ‣ (?=.*)
    ‣ (?=abc)
    Regular expressions 101 » Assertions
    ‣ (?<=.*)
    ‣ (?<=abc)
    Lookahead/-behind assertions
    

    View Slide

34. Regular expressions 101 » Conditional subpatterns
    Conditional subpatterns
    if-then(-else) in regular expressions
    !
    !
    YES RLY!
    

    View Slide

35. Regular expressions 101 » Conditional subpatterns
    Conditional subpatterns
    Scenario: match all (x|ht)ml tags
    !
    Caution!
    ‣
    ‣
    

    View Slide

36. Solution: if then else
    /<(?P[a-z]+).*?(?P\/)?>(?(self)|.*?<\/(?P=tag)>)/i
    Named patterns
    If self-closing, then do nothing,

    else, find matching end tag
    Regular expressions 101 » Conditional subpatterns
    Conditional subpatterns
    

    View Slide

37. Regular expressions 101 » Conditional subpatterns
    Conditional subpatterns
    ‣ With subpattern (named or by id):
    ‣ (?(pattern)then)
    ‣ (?(pattern)then|else)
    ‣ With lookahead/-behind:
    ‣ (?(?=assertion)then)
    ‣ (?(?=assertion)then|else)
    

    View Slide

38. Regular expressions 101 » Comments
    Comments
    /
    # match currency symbols for USD, EUR, GBP & YEN
    [$€£¥]
    # must be followed by a number to indicate a price
    (?=[0-9])
    # pattern modifiers:
    # u for UTF-8 interpretation (currency symbols),
    # x to ignore whitespace (for comments)
    /ux
    

    View Slide

39. Regular expressions 101 » Comments
    Comments
    ‣ # Perl-style
    ‣ /x modifier
    ‣ Ignores unescaped whitespace
    

    View Slide

40. Presentation title
    

    View Slide

41. Questions?
    Regular expressions 101
    

    View Slide

42. mullie.eu
    ‣ www.mullie.eu/regular-expressions-basics/
    ‣ www.mullie.eu/regular-expressions-advanced/
    Regular expressions 101
    Resources
    

    View Slide