Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
LALR parser generatorの作り方
Search
yui-knk
January 20, 2024
Programming
6
7k
LALR parser generatorの作り方
BuriKaigi 2024
https://burikaigi.dev/
yui-knk
January 20, 2024
Tweet
Share
More Decks by yui-knk
See All by yui-knk
Ruby Parser progress report 2025
yui_knk
1
860
Understanding Ruby Grammar Through Conflicts
yui_knk
1
790
Ruby's Line Breaks
yui_knk
4
5.6k
What is Parser
yui_knk
11
5.5k
Ruby Parser progress report 2024
yui_knk
2
490
最高の構文木の設計 2024年版
yui_knk
9
6.6k
Converting AST
yui_knk
4
410
My favorite script, "dsl.rb"
yui_knk
2
1.4k
Rearchitect Ripper
yui_knk
2
1.4k
Other Decks in Programming
See All in Programming
Fundamentals of Software Engineering In the Age of AI
therealdanvega
2
260
エンジニアの「手元の自動化」を加速するn8n 2026.02.27
symy2co
0
160
The free-lunch guide to idea circularity
hollycummins
0
270
Claude Code の Skill で複雑な既存仕様をすっきり整理しよう
yuichirokato
1
410
ベクトル検索のフィルタを用いた機械学習モデルとの統合 / python-meetup-fukuoka-06-vector-attr
monochromegane
2
480
「接続」—パフォーマンスチューニングの最後の一手 〜点と点を結ぶ、その一瞬のために〜
kentaroutakeda
3
970
How to stabilize UI tests using XCTest
akkeylab
0
130
コーディングルールの鮮度を保ちたい / keep-fresh-go-internal-conventions
handlename
0
220
CS教育のDX AIによる育成の効率化
niftycorp
PRO
0
140
モックわからないマン卒業記 ~振る舞いを起点に見直した、フロントエンドテストにおけるモックの使いどころ~
tasukuwatanabe
3
400
米国のサイバーセキュリティタイムラインと見る Goの暗号パッケージの進化
tomtwinkle
2
610
LangChain4jとは一味違うLangChain4j-CDI
kazumura
1
200
Featured
See All Featured
Why Our Code Smells
bkeepers
PRO
340
58k
SEO in 2025: How to Prepare for the Future of Search
ipullrank
3
3.4k
How Fast Is Fast Enough? [PerfNow 2025]
tammyeverts
3
490
Testing 201, or: Great Expectations
jmmastey
46
8.1k
How to make the Groovebox
asonas
2
2k
Music & Morning Musume
bryan
47
7.1k
The Psychology of Web Performance [Beyond Tellerrand 2023]
tammyeverts
49
3.3k
Producing Creativity
orderedlist
PRO
348
40k
Utilizing Notion as your number one productivity tool
mfonobong
4
260
Agile Leadership in an Agile Organization
kimpetersen
PRO
0
110
Technical Leadership for Architectural Decision Making
baasie
3
300
Collaborative Software Design: How to facilitate domain modelling decisions
baasie
0
160
Transcript
LALR parser generatorͷ࡞Γํ January 20, 2024 BuriKaigi 2024 @yui-knk Yuichiro
Kaneko
About me • Yuichiro Kaneko • yui-knk (GitHub) / spikeolaf
(Twitter) • Treasure Data • Engineering Manager of Applications Backend • The author of ruby/lrama LALR parser generator • CRuby committer, mainly develop parser related features
PR: We are hiring!! • https://www.treasuredata.com/company/jobs/
Parserͱ ͍ͧ͘ɺ͍ͧ͘ࢁ
Parserͷׂ • ೖྗ͞ΕͨจࣈྻʹߏΛ༩͑Δ Class Method Method Assignment @name Call Name
capitalize
Lexerͷׂ • ॲཧܥ(ruby)͔ΒΈΔͱͨͩͷόΠτྻ • ·ͣจࣈྻΛదͳ୯ҐͰ۠Δඞཁ͕͋Δ • Ruby͞·͟·ͳEncodingΛαϙʔτ͍ͯ͠Δ 636c61737320477265657465720a2020646566 20696e697469616c697a65286e616d65290a20 202020406e616d65203d206e616d652e636170
6974616c697a650a2020656e640a0a2020646 5662073616c7574650a2020202070757473202 248656c6c6f20237b406e616d657d21220a202 0656e640a656e640a class Greeter def
ParserͱLexer • Lexer͕tokenΛΓग़͠ɺParser͕ߏԽ͢Δ Class Method Method Assignm @name Call Name
capitaliz class Greeter def Lexer Parser
͍·RubyͰParser͕͍ https://twitter.com/OssVision/status/1735433960191299602
͍·RubyͰParser͕͍ • Bison͕ਏ͍ • BisonͷversionڥʹΑͬͯҟͳΔ • ෳͷBisonͷversionΛαϙʔτ͢ΔͨΊʹɺ৽͍͠ػೳ͕͑ͳ͍ • RubyͷparserෳࡶͰɺϝϯςφϯεੑ͕Β͘ࢹ͞Ε͖ͯͨ •
Language Server Protocolͷ಄ • RubyͰRBS(ܕ)RubyCop(੩తղੳث)ͱ͍ͬͨπʔϧ͕׆ൃʹ։ൃ ͞Ε͍ͯΔ • ͦͷ݁ՌɺϢʔβʔͷೖྗதͷϓϩάϥϜͱ͍͏ෆશͳೖྗΛύʔε͢ Δඞཁ͕Ͱ͖ͯͨ (ΤϥʔτϨϥϯτͳύʔαʔ) • ͜ΕΒΛղܾ͢ΔͨΊʹParserʹେنͳվળΛ͍Ε͍ͯΔ
Parserͷ࡞Γํ • खॻ͖parser • Parser GeneratorΛར༻ͯ͠ੜ͢Δ • Ruby Ͱͪ͜Βͷํ๏Λ࠾༻͍ͯ͠Δ •
Yacc, Bison, ANTLR ͳͲ • Lramaparser generator
Parser Generator • ઃఆϑΝΠϧΛͱʹparserΛੜ͢Δπʔϧ • RubyͰGNU BisonΛ͍··Ͱ͖ͬͯͨ • Ruby 3.3ͰBisonΛLramaʹஔ͖͑ͨ
• https://github.com/ruby/lrama ઃఆϑΝΠϧ parse.c Bison Lrama
ઃఆϑΝΠϧͷྫ • BNFͰจ๏Λهड़͢Δ • ͱͯΘ͔Γ͍͢
Parser Generatorͷར • จ๏͕ཧղ͍͢͠ • จ๏ఆٛͱύʔαʔͷ࣮ʹဃ͕ͳ͍ • จ๏ͷมߋʹରͯ͠ϑΟʔυόοΫΛಘΔ͜ͱ͕Ͱ͖Δ • ίϯϐϡʔλαΠΤϯεͷཧʹج͍͍ͮͯΔ
• ΤϥʔτϨϥϯτͳύʔαʔΛจ๏ఆ͔ٛΒࣗಈੜͰ͖Δ
LR parser͍͍ͧ • Bisonʹ͍Ζ͍Ζͱػೳ͕ෆ͍ͯ͠ΔͷͰ • RubyͰLrama LR parser generatorΛ࣮ͯ͠ •
RubyͰLramaΛ͏Α͏ʹ͢Δ͘Β͍ʹਪ͍ͯ͠Δ
Parser Generatorͷ ࡞Γํ ࢁʹண͍ͨʂ
ΞʔΩςΫνϟ • Frontend, Backend, Code Generator͔ΒͳΔ ઃఆϑΝΠϧ Parser Frontend Backend
Code Generator Parser Generator
Frontend • LexerͱParserΛ༻͍ͯઃఆϑΝΠϧΛ෦తͳσʔλߏʹม͢Δ • σʔλߏͷओRule ઃఆϑΝΠϧ ෦දݱ
Action • Actionͷ෦ղੳ͢Δඞཁ͕͋Δ • $$ͳͲͷಛघͳมͰtokenͷҐஔใʹΞΫηεͰ͖Δ • จ๏ϑΝΠϧͱผͷLexerΛ༻ҙ͢Δͷ͕Α͍ • มCodeੜ࣌ʹparser internalͳมʹஔ͖͑Δ
Backend • Rule͔ΒState MachineΛੜ͢Δ • ߏจղੳදͱ͍͏ͷཁ͢ΔʹΦʔτϚτϯ ߏจղੳද
• ֤RuleΛΦʔτϚτϯʹม͢Δ • શͯͷΦʔτϚτϯΛ߹ͨ͠ͷ͕ߏจղੳද LR parserstackΛͬͨDAF class A body end
def m1 body end class B body end
LALRҎ֎ͷΞϧΰϦζϜ • ߏจղੳͷͨΊͷΦʔτϚτϯͷ࡞Γํ͍Ζ͍Ζ͋Δ • LR(0), SLR(1), LALR(1), LR(1), IELR(1) ͳͲͷΞϧΰϦζϜ͕͋Δ
• ղੳՄೳͳݴޠඞཁͳϝϞϦ͕ͦΕͧΕҟͳΔ • Rule͔ΒΦʔτϚτϯΛ࡞ΔͷͰɺΞϧΰϦζϜͷબ͕จ๏ϑΝΠϧͷ γϯλοΫεͱಠཱ͍ͯ͠Δ
ߴͳLook-Aheadू߹ͷܭࢉ • LALR(1)Λ࣮͢Δͱ͖ʹʹͳΔͷ͕ɺޮతʹLook-Aheadू߹Λ ܭࢉ͢Δ͜ͱ • “Ef fi cient Computation of
LALR(1) Look-Ahead Sets” ͱ͍͏จͷΞ ϧΰϦζϜΛ༻͢ΔͱΑ͍ • https://dl.acm.org/doi/pdf/10.1145/69622.357187
Code Generator • State MachineΛλʔήοτͷݴޠʹ߹Θͤ ࣮ͯ͢Δ • tableΛࠓͷstateͱtokenͰݕࡧͯ࣍͠ʹΔ ͖͜ͱΛܾΊΔ •
shift, reduce, accept, error
TemplateʹΛຒΊࠐΉ • ࣮ࡍʹΔ͜ͱඞཁͳมΛཧͯ͠templateʹຒΊࠐΉ࡞ۀ • LramaͩͱERB, Bisonͩͱm4 • ERBҒେ
εύʔε(ૄ)ͳߏจղੳද • ॎ͕ঢ়ଶɺԣ͕τʔΫϯͷछྨͱ͍͏େ͖ͳςʔϒϧ • ઌ΄Ͳͷྫͩͱ70/238Ϛε͔͍ͬͯ͠ͳ͍ (29%͘Β͍)
εύʔε(ૄ)ͳߏจղੳද • ޓ͍ҧ͍ʹͯ͠1ͭͷྻʹ·ͱΊΔ • ίϯύΫτσʔλߏͰΑ͘ͳΒͳ͍ͩΖ͏͔?
ΞʔΩςΫνϟʔ • RuleͱState Machine͕ͦΕͧΕͷίϯϙʔωϯτؒͷΠϯλʔϑΣΠε ઃఆϑΝΠϧ ύʔαʔ Frontend Backend Code
Generator Rule State Machine
Parser GeneratorΛ ֦ு͢Δ ͔ʹඒຯͦ͠͏…
Named References • TokenͷͳͲʹΞΫηε͢Δͱ͖ʹ$1, $2Ͱͳ͘ɺ$cpath, $bodyͱ ໊લͰΞΫηεͰ͖Δ • Lexerͷ࣮͚ͩͰ࣮ݱͰ͖ΔͷͰFrontend͚ͩͷมߋͰ࣮ݱͰ͖ͨ
Parameterizing Rules • ෳճͷ܁Γฦ͠ͱ͍͏ͷจ๏ఆ্ٛΑ͘ग़ ͯ͘Δ • ॻ͖ํͷύλʔϯ͕ܾ·͍ͬͯΔͷͰ͋Εɺͦ ΕΛநԽͯ͠ॻ͖͍ͨ • LramaͰ࣮ͣΈ
Parameterizing Rules • ઃఆϑΝΠϧ͔ΒRuleͷσʔλߏΛͭ͘Δͱ͖ʹల։͢Δ͚ͩͳͷͰɺ Frontend͚ͩͷมߋͰ࣮ݱͰ͖Δ
%after-shift • RipperͷΑ͏ͳ໘ന͍ػೳΛ࣮͠Α͏ͱ͢ΔͱShift͢Δॠؒ Reduce͢ΔॠؒʹcallbackΛ͜͞͠Έͨ͘ͳΔ • Frontendͷparser/lexerͱCode GeneratorͷtemplateΛ͍͡ΕͰ͖ Δ
·ͱΊ ͦΖͦΖᲳ͕৯͍ͨ…
·ͱΊ • LR parser͍͍ͧ • LR parser generator3ͭͷίϯϙʔωϯτ͔ΒͳΓɺݴޠॲཧܥʹߏ͕ࣅ ͍ͯΔ •
ׂ͞Ε͍ͯΔͨΊػೳՃͷࡍʹඞཁͳίϯϙʔωϯτ͚ͩΛมߋ͢Ε͍͍ • Lrama parser generatorΨϯΨϯ։ൃத • ͔Ͷ͜ʹ͖ͬͰRubyͷparserͷ։ൃঢ়گΛ·ͱΊ͍ͯΔ • https://yui-knk.hatenablog.com/ • ruby-jpͱ͍͏slackͷ #lr-parser νϟωϧʹීஈ͍Δ
RubyKaigi 2024
Thank you!!