Back to the roots of date

by jinroq

Embed

Start on current slide

Slide 1

Slide 1 text

Back to the roots of date ODA Hirohito 2026-04-22

Slide 2

Slide 2 text

Slide 3

Slide 3 text

Slide 4

Slide 4 text

Slide 5

Slide 5 text

Slide 6

Slide 6 text

Copyright © BizteX Inc All Rights Reserved. 6 h tt ps://bugs.ruby-lang.org/issues/21264 • There have been no primary maintainers for the date library and the datetime library for over 10 years. • Maintaining the date library and the datetime library are becoming a burden for Ruby commi tt ers.

Slide 7

Slide 7 text

Copyright © BizteX Inc All Rights Reserved. 7 Future proposals for the date library and the datetime library • The Date class will be deprecated. • Instead, a simpli fi ed The Date class will be created. • Move the Date class into a core class. • Keep the current situation.

Slide 8

Slide 8 text

Copyright © BizteX Inc All Rights Reserved. 8 Future proposals for the date library and the datetime library • The Date class will be deprecated. • Instead, a simpli fi ed The Date class will be created. • Move the Date class into a core class. • Currently, there are no major problems, so Ruby commi tt ers will continue with maintenance that only involves version updates. How did I approach this challenge?

Slide 9

Slide 9 text

Slide 10

Slide 10 text

Slide 11

Slide 11 text

Slide 12

Slide 12 text

Slide 13

Slide 13 text

Slide 14

Slide 14 text

Slide 15

Slide 15 text

Slide 16

Slide 16 text

Copyright © BizteX Inc All Rights Reserved. 16 Ghost said… • Moving away from date library: • Decoupling DateTime from Date. • Moving `Time.parse` to the time library to decouple it from date library. • strptime issue.

Slide 17

Slide 17 text

Copyright © BizteX Inc All Rights Reserved. 17 Ghost said… • Moving away from date library: • Decoupling DateTime from Date. -> DateTime is already deprecated, so no major action is required. • Moving `Time.parse` to the time library to decouple it from date library. -> Since `Time.parse` internally calls `Date._parse`, we should port the equivalent of `Date._parse` into the time library. • strptime issue. -> The current implementation is slightly biased toward date, so it should be made more neutral and compact, assuming Time.

Slide 18

Slide 18 text

Slide 19

Slide 19 text

Slide 20

Slide 20 text

Copyright © BizteX Inc All Rights Reserved. 20 h tt ps://bugs.ruby-lang.org/issues/21616 • date library will be deprecated. • Porting `Date._parse` to time library. • Porting `Date.strptime` to time library.

Slide 21

Slide 21 text

Slide 22

Slide 22 text

Slide 23

Slide 23 text

Slide 24

Slide 24 text

Slide 25

Slide 25 text

Slide 26

Slide 26 text

Slide 27

Slide 27 text

Copyright © BizteX Inc All Rights Reserved. 27 Reconsider about the date library • Should we implement a part of Date's functionality (or a similar independent function) in Time? • If Time can handle dates, how does it di ff er from the already deprecated DateTime?

Slide 28

Slide 28 text

Copyright © BizteX Inc All Rights Reserved. 28 Reconsider about the date library • Should we implement a part of Date's functionality (or a similar independent function) in Time? • If Time can handle dates, how does it di ff er from the already deprecated DateTime? My thoughts are wandering in a labyrinth.

Slide 29

Slide 29 text

Slide 30

Slide 30 text

Slide 31

Slide 31 text

Slide 32

Slide 32 text

Slide 33

Slide 33 text

Slide 34

Slide 34 text

Slide 35

Slide 35 text

Slide 36

Slide 36 text

Slide 37

Slide 37 text

Slide 38

Slide 38 text

Slide 39

Slide 39 text

Slide 40

Slide 40 text

Slide 41

Slide 41 text

Slide 42

Slide 42 text

Slide 43

Slide 43 text

Slide 44

Slide 44 text

Slide 45

Slide 45 text

Copyright © BizteX Inc All Rights Reserved. 45 The day before the closing of the RubyKaigi 2026 CFP • Development of RubyDate is complete. • Until the end of March 2026 • Date rewriting complete. • Until the end of March 2027

Slide 46

Slide 46 text

Copyright © BizteX Inc All Rights Reserved. 46 RubyDate development plan • Used an AI agent • Started by porting ruby/date/ext/date/date_core.c. • Implemented it by referring to the behavior described in the usage comments. • Ported some of the tests from ruby/date/test/date/test_date.rb, and then had an AI agent write the rest of the tests. • Development environment • 4.1-dev • Ubuntu 24.04 LTS

Slide 47

Slide 47 text

Copyright © BizteX Inc All Rights Reserved. 47 First commit • Class Methods • jd • gregorian_leap? • julian_leap? • ordinal • Instance Methods • initialize • year • month/mon • day/mday • jd • start • <=> • == • eql? • hash? • +(n) • in fi nite?

Slide 48

Slide 48 text

Slide 49

Slide 49 text

Slide 50

Slide 50 text

Slide 51

Slide 51 text

Slide 52

Slide 52 text

Slide 53

Slide 53 text

Copyright © BizteX Inc All Rights Reserved. 53 Investigation of causes • Tests that had been partially ported from test_date.rb were somehow deleted by the AI agent. • The AI agent was only adding tests that passed the logic it had implemented.

Slide 54

Slide 54 text

Copyright © BizteX Inc All Rights Reserved. 54 Investigation of causes • Tests that had been partially ported from test_date.rb were somehow deleted by the AI agent. • The AI agent was only adding tests that passed the logic it had implemented. • All test code under ruby/date/test/date has been moved to ruby_date. • Only the tests in test_date.rb have been fi xed to pass.

Slide 55

Slide 55 text

Slide 56

Slide 56 text

Slide 57

Slide 57 text

Copyright © BizteX Inc All Rights Reserved. 57 The ruby_date gem was complete! • Development of RubyDate is complete. • Until the end of March 2026 • Date rewriting complete. • Until the end of March 2027

Slide 58

Slide 58 text

Slide 59

Slide 59 text

Slide 60

Slide 60 text

Copyright © BizteX Inc All Rights Reserved. 60 RubyDate vs Date • Supported Ruby version • RubyDate: 3.0 or later • Date: 2.6 or later • Presence or absence of DateTime • RubyDate: absence • Date: presence

Slide 61

Slide 61 text

Slide 62

Slide 62 text

Slide 63

Slide 63 text

Slide 64

Slide 64 text

Slide 65

Slide 65 text

Slide 66

Slide 66 text

Slide 67

Slide 67 text

Slide 68

Slide 68 text

Slide 69

Slide 69 text

Slide 70

Slide 70 text

Slide 71

Slide 71 text

Slide 72

Slide 72 text

Slide 73

Slide 73 text

Slide 74

Slide 74 text

Slide 75

Slide 75 text

Slide 76

Slide 76 text

Slide 77

Slide 77 text

Slide 78

Slide 78 text

Slide 79

Slide 79 text

Slide 80

Slide 80 text

Slide 81

Slide 81 text

Slide 82

Slide 82 text

Slide 83

Slide 83 text

Slide 84

Slide 84 text

Slide 85

Slide 85 text

Slide 86

Slide 86 text

Slide 87

Slide 87 text

Copyright © BizteX Inc All Rights Reserved. 87 Pure Ruby Date Benchmark .FUIPE $FYU JT CFGPSF3VCZ JT BGUFS3VCZ JT BGUFS3VCZ$ %BUFUPEBZ , , %BUF@QBSTF JTP , , %BUFQBSTF JTP , , %BUFTUSQUJNF . , %BUFTUSGUJNF . . %BUFDJWJM . . %BUFKE . , . %BUFPSEJOBM . , , %BUFDPNNFSDJBM . , ,

Slide 88

Slide 88 text

Slide 89

Slide 89 text

Slide 90

Slide 90 text

Slide 91

Slide 91 text

Slide 92

Slide 92 text

Copyright © BizteX Inc All Rights Reserved. 92 The Trigger for a Clean Slate • AI agents boosted development speed more than I ever imagined. • Working with AI agents gave me a much higher resolution of how date works.

Slide 93

Slide 93 text

Copyright © BizteX Inc All Rights Reserved. 93 Shift in Implementation Approach • The Rules for AI agent: • Zero changes to tests. (Maintain the ground truth) • Target: 85%+ performance of C extension. • The Process: • Identify bo tt lenecks method by method. • Apply micro-optimizations.

Slide 94

Slide 94 text

Copyright © BizteX Inc All Rights Reserved. 94 Shift in Implementation Approach • The Rules for AI agent: • Zero changes to tests. (Maintain the ground truth) • Target: 85%+ performance of C extension. • The Process: • Identify bo tt lenecks method by method. • Apply micro-optimizations. These optimizations didn't move the needle as much as I expected.

Slide 95

Slide 95 text

Copyright © BizteX Inc All Rights Reserved. 95 Shift in Implementation Approach • The Rules for AI agent: • Zero changes to tests. (Maintain the ground truth) • Target: 85%+ performance of C extension. • The Process: • Identify bo tt lenecks method by method. • Apply micro-optimizations. And then, I arrived at a single truth.

Slide 96

Slide 96 text

Slide 97

Slide 97 text

Copyright © BizteX Inc All Rights Reserved. 97 Pure Ruby(+YJIT) Date Benchmark .FUIPE $FYU JT 3VCZ JT 3VCZ:+*5 JT 3VCZ:+*5$ %BUFUPEBZ , , , %BUF@QBSTF JTP , , . %BUFQBSTF JTP , , . %BUFTUSQUJNF . , . %BUFTUSGUJNF . . . %BUFDJWJM . . . %BUFKE . . . %BUFPSEJOBM . , . %BUFDPNNFSDJBM . , .

Slide 98

Slide 98 text

Copyright © BizteX Inc All Rights Reserved. 98 Pure Ruby(+YJIT) Date Benchmark .FUIPE $FYU JT 3VCZ JT 3VCZ:+*5 JT 3VCZ:+*5$ %BUFUPEBZ , , , %BUF@QBSTF JTP , , . %BUFQBSTF JTP , , . %BUFTUSQUJNF . , . %BUFTUSGUJNF . . . %BUFDJWJM . . . %BUFKE . . . %BUFPSEJOBM . , . %BUFDPNNFSDJBM . , . 🙌

Slide 99

Slide 99 text

Slide 100

Slide 100 text

Slide 101

Slide 101 text

Slide 102

Slide 102 text

Slide 103

Slide 103 text

Copyright © BizteX Inc All Rights Reserved. 103 Restraining optimization • Improve readability and maintainability • Eliminated `getbyte` byte-level manipulation. • Switched to StringScanner and regex for clearer logic.

Slide 104

Slide 104 text

Slide 105

Slide 105 text

Slide 106

Slide 106 text

Slide 107

Slide 107 text

Slide 108

Slide 108 text

Slide 109

Slide 109 text

Slide 110

Slide 110 text

Slide 111

Slide 111 text

Slide 112

Slide 112 text

Slide 113

Slide 113 text

Copyright © BizteX Inc All Rights Reserved. 113 MatchData#integer_at VS StringScanner#integer_at .FUIPE 3VCZ:+*5 JT .BUDI%BUBJOUFHFS@BU 4USJOH4DBOOFSJOUFHF S@BU %BUF@TUSQUJNF %BUFTUSQUJNF %BUFTUSQUJNF DPNQMFY

Slide 114

Slide 114 text

Slide 115

Slide 115 text

Slide 116

Slide 116 text

Slide 117

Slide 117 text

Copyright © BizteX Inc All Rights Reserved. 117 Conclusion • Jinroq would like to avoid deprecating the date library. • However, it's easier said than done. • AI agents enable us to build products with incredible speed. • Therefore, there is no compelling reason to wait. • AI agents are powerful partners, but NEVER our masters. • We MUST remain accountable for what we build.