Doctrine Lexer Use Case

Transcript

1. 1.

   Doctrine Lexer use case

2. 2.

   None
3. 3.

   Olivier Dolbeau @odolbeau Web Architect About me

4. 4.

   RIDESHARING Cost of motoring 100€ 25€ 25€ 25€ 25€ @BlaBlaCar_FR

   ALONE
5. 5.

   A fast growing community

6. 6.

   A European phenomenon

7. 7.

   None
8. 8.

   I want to be able to search tweets using a

   human language.
9. 9.

   None
10. 10.

    None
11. 11.

    None
12. 12.

    None
13. 13.

    Where are my tweets ?

14. 14.

    None
15. 15.

    None
16. 16.

    None
17. 17.

    None
18. 18.

    None
19. 19.

    None
20. 20.

    Other solutions Logstash Home made external river

21. 21.

    distributed restful search and analytics elasticsearch

22. 22.

    None
23. 23.

    None
24. 24.

    None
25. 25.

    None
26. 26.

    None
27. 27.

    We know what we want. We know where our data

    is.
28. 28.

    Now let’s see HOW we are doing to retrieve it.

29. 29.

    How... … to translate a human language into a technical

    one.
30. 30.

    “ A retweet FROM @CameronNash23 ”

31. 31.

    None
32. 32.

    What’s a Lexer ? In computer science, lexical analysis is

    the process of converting a sequence of characters into a sequence of tokens, i. e. meaningful character strings.
33. 33.

    Tokens ! Now !

34. 34.

    None
35. 35.

    None
36. 36.

    None
37. 37.

    None
38. 38.

    “ A retweet FROM @CameronNash23 ” T_SELECTOR T_RETWEET T_FROM T_USERNAME

39. 39.

    How... … to use these tokens to generate an ElasticSearch

    query.
40. 40.

    None
41. 41.

    None
42. 42.

    None
43. 43.

    None
44. 44.

    It works!

45. 45.

    None
46. 46.

    None
47. 47.

    None
48. 48.

    And what about Doctrine ORM ? Let’s take a look

    at : Doctrine\ORM\Query\Lexer
49. 49.

    From DQL to SQL SELECT t0.id AS id1, t0.name AS

    name3, t0. username AS username4, t0.email AS email5 FROM mos_users t0 WHERE t0.username = “foobar” SELECT u FROM MyProject\Model\User u WHERE u.username = ‘foobar’
50. 50.

    77 tokens !!

51. 51.

    Some patterns...

52. 52.

    And a switch (true) !

53. 53.

    Forget the Parser !

54. 54.

    None
55. 55.

    None
56. 56.

    None
57. 57.

    Example A trip OR A search FROM "Paris, France" OR

    TO "Paris, France" BETWEEN 12/01/2013 AND 26/05/2014 A trip OR A search FROM "Paris, France" OR TO "Paris, France" BETWEEN 12/01/2013 AND 26/05/2014
58. 58.

    Lexer to GET tokens. Parser to USE tokens.

59. 59.

    @odolbeau On recrute ! :) MERCI !

60. 60.

    @odolbeau https://joind.in/ 11246 https://github.com/odolbeau/elasticsearch-sandbox https://speakerdeck.com/odolbeau/doctrine-lexer-use-case