
ãSieve-Scopeãæ±ºå®è«çã¹ã³ãŒã管çãšã³ã³ããã¹ãå¢çã®å®çŸ© â ASTã»èå¥åã»æ§é æçŽã«ããã³ãŒãé¡äŒŒæ§ã®å€è»žèŠ³æž¬
ãã³ãŒãã®ãã©ãŒããããééããã¢ã«ãŽãªãºã ã®çŽç²ãªããžãã¯ã ããèŠã蟌ãã
å®å
šããŒã«ã«å®è¡ã»å€éšäŸåãŒãã»æ±ºå®è«çïŒ100%åçŸå¯èœïŒãªã³ãŒãççšã»æ§é é¡äŒŒæ€èšŒãšã³ãžã³ã§ãã
Sieve-Core ã®æ±ºå®è«çé¡äŒŒåºŠãšã³ãžã³ãšãSieve-Referee ã®å€è»žä»®èª¬ã»LUTå€å®ææ³ãããœãŒã¹ã³ãŒããã¡ã€ã³ãžæ¡åŒµãããããžã§ã¯ãã§ãã
ã³ãŒãã®ãèŠãç®ããšãæ§é ããåããŠèŠ³æž¬ãã
ããã°ã©ã ã®é¡äŒŒæ§ãè©äŸ¡ãããšããäœããã£ãŠã䌌ãŠããããšå€æããã°ããã®ã§ããããã
倿°åãåãã§ããã°ãåãã¢ã«ãŽãªãºã ãªã®ã§ãããããå察ã«ã倿°åãã³ã¡ã³ãã空çœããã¹ãŠå€æŽããã°ãç°ãªãããã°ã©ã ãšå€æãããã¹ããªã®ã§ããããã
åçŽãªæå忝èŒã§ã¯ããã®åé¡ã«ååã«å¯Ÿå¿ã§ããŸããã
ããšãã°ã次ã®2ã€ã®ã³ãŒããèããŠã¿ãŸãã
def calculate_total(values):
result = 0
for value in values:
result += value
return resultdef compute(items):
answer = 0
for item in items:
answer += item
return answer颿°åã倿°åãåŒæ°åã¯ç°ãªã£ãŠããŸããããããã©ã¡ããæ°å€ãé çªã«å ç®ããŠåèšãè¿ãæ§é ãæã£ãŠããŸãã
æååãšããŠæ¯èŒããã°å·®åãå€ãååšããŸãããASTïŒæœè±¡æ§ææšïŒãå©çšããŠæ§é ã芳枬ããã°ãå ±éããç¹åŸŽãæœåºã§ããŸãã
äžæ¹ãASTã®æ§é ã䌌ãŠãããããšãã£ãŠãããã°ã©ã ã®æå³ãŸã§åäžã§ãããšã¯éããŸãããçã颿°ã§ã¯ãç°ãªãåŠçã§ãäŒŒãæ§ææ§é ã«ãªãå ŽåããããŸãã
ãã®åé¡ã«å¯ŸããŠéçºããŠããã®ããSieve-Scope ã§ãã
Sieve-Scopeã¯ãPythonã³ãŒãã察象ã«ãæ§ææ§é ãèå¥åãäœçœ®çãªASTãã£ãŒã«ãæ§é ã宿°ãèå¥åã®é£ç¶æ§ãšããè€æ°ã®ç¹åŸŽã芳枬ããå®çŸ©æžã¿ã®ã«ãŒã«ã«ãã£ãŠé¡äŒŒæ§ãåé¡ããæ±ºå®è«çãªã³ãŒãæ¯èŒãšã³ãžã³ã§ãã
åäžã®é¡äŒŒåºŠã¹ã³ã¢ã ãã§çµè«ãåºãã®ã§ã¯ãªããè€æ°ã®èŠ³æž¬çµæãããããã¹ã¯ãšããŠè¡šçŸããããšã§ãã©ã®ç¹åŸŽãäžèŽããã©ã®ç¹åŸŽãç°ãªã£ãŠããã®ãã远跡ã§ããä»çµã¿ãç®æããŠããŸãã
æ¬èšäºã§ã¯ãSieve-Scopeã®èšèšææ³ã7ã€ã®ä»®èª¬ã«ããè©äŸ¡ã¢ãã«ãã¯ã©ã¹ã¿å®å šæ¡ä»¶ã鿥é£éã®æ€åºããããŠæ±ºå®è«çãªæ¯èŒãæã€éçã«ã€ããŠè§£èª¬ããŸãã
Sieve ãšã³ã·ã¹ãã ã«ãããŠãsieve-coreïŒæ§é éçŽïŒã sieve-refereeïŒæ€èšŒã»ã¬ããã³ã¹ïŒãããŒã¿ã®è©äŸ¡ãæ€èšŒãæ
ãäžæ¹ããããã®è©äŸ¡å¯Ÿè±¡ãšãªãé åãå¢çãæ£ç¢ºã«åãåºãã®ã sieve-scope ã§ãã
æ¬ãããžã§ã¯ãã®èµ·ç¹ã¯ããšã©ãã¹ããã¹ã®ç¯©ãã«çæ³ãåŸãŠèšèšãããéåè«ã®ç¯©ã Sieve-AI ã®èšèšææ³ããããæ±çšçãªããã¹ãã»ããŒã¿åŠçãã¡ã€ã³ã«é©çšããã³ã¢ãšã³ãžã³ã Sieve-Core ã§ãã
次ã«ããã® Sieve-Core ã«ãçµã¿åããè«ã®ç¯©ãã远å èšèšãããSieve Referee ã¯ãè€æ°ã®ç¬ç«ããæ
å ±æºã«ããè£ä»ãã«åºã¥ãããã€ãºãã確ããªæ
å ±ãæœåºãããŒããã£ãã³ãã³ã·ãŒã®Pythonã©ã€ãã©ãªã§ãã
ãããŠãSieve Scope ã¯ããã«2軞è©äŸ¡ãå ããŠåèšèšããããšã§ãã³ãŒãã®ã衚é¢çãªãã©ãŒãããïŒå€æ°åã»ã³ã¡ã³ãã»ç©ºçœïŒããšãæ¬è³ªçãªããžãã¯ïŒã¢ã«ãŽãªãºã æ§é ïŒããåé¢ããŠèŠ³æž¬ããã説æå¯èœãªã³ãŒãççšæ€ç¥ãšã³ãžã³ã§ãã
Sieveãšã³ã·ã¹ãã ã«ãããSieve-Scopeã®åœ¹å²
Sieve-Scopeã¯ãSieve-CoreãSieve-Refereeãšå ±éããèšèšææ³ãæã¡ãªãããäž»ã«ãœãŒã¹ã³ãŒãã®æ¯èŒãšè©äŸ¡å¯Ÿè±¡ã®å¢ç管çã«çŠç¹ãåœãŠããããžã§ã¯ãã§ãã
ããããã®åœ¹å²ãæŽçãããšã次ã®ããã«ãªããŸãã
Sieve-CoreïŒé¡äŒŒããããŒã¿ã®é¢ä¿ãæ§é åããéçŽããã
Sieve-RefereeïŒããã¹ãã®é¡äŒŒæ§ãè€æ°ã®ä»®èª¬ã«åè§£ããå€å®æ ¹æ ãåé¡ããã
Sieve-ScopeïŒãœãŒã¹ã³ãŒããè€æ°ã®æ§é çç¹åŸŽãžåè§£ããã³ãŒãéã®é¢ä¿ã芳枬ããã
ãããã«å ±éããã®ã¯ãè©äŸ¡ããã©ãã¯ããã¯ã¹ã«å§ããã®ã§ã¯ãªããåŠçæ¡ä»¶ãšå€å®æ ¹æ ãæç€ºçã«å®çŸ©ããããšã§ãã
Sieve-Scopeã§ã¯ããã®èãæ¹ãããã°ã©ã ã®æ§é ãžé©çšããŸãã
ãœãŒã¹ã³ãŒãã¯ãåãªãæååã§ã¯ãããŸãããæ§ææšãèå¥åã®äžŠã³ã宿°ãæŒç®åãå¶åŸ¡æ§é ãªã©ãè€æ°ã®ç°ãªãåŽé¢ãã芳枬ã§ããŸãã
ãã®ãããã³ãŒãã®é¡äŒŒæ§ãäžã€ã®å°ºåºŠã«æŒã蟌ããã®ã§ã¯ãªãã芳枬察象ãåè§£ããŠæ¯èŒããããšãéèŠã ãšèããŠããŸãã
åŸæ¥ã®æå忝èŒãåçŽãªé¡äŒŒåºŠå€å®ã§ã¯ã
倿°åãå€ããã ãã®ççšïŒå€æ°åãã³ããªã³ã°ïŒ
å ±éãã³ãã¬ãŒãã«ããå¶ç¶ã®äžèŽïŒãã€ã©ãŒãã¬ãŒãïŒ
ãåºå¥ããããšãå°é£ã§ããã
Sieve Scope ã¯ã
ASTïŒæœè±¡æ§ææšïŒ ããèå¥åãæšãŠããããžãã¯ã®éªšæ Œã
tokenize ããæœåºãããèå¥åã®èªåœã»é åºã
ã®2軞ãç¬ç«ã«èŠ³æž¬ããSieve-Referee ããç¶æ¿ãã仮説ãã¹ã¯ïŒH1ãH7ïŒãšLUTã«ãã£ãŠããªããã®å€å®ã«ãªã£ãã®ããå®å šã«èª¬æããŸãã
ãSieve Scopeããšããåç§°ã¯ãã³ãŒãã®è¡šé¢ïŒãã©ãŒãããïŒãééããé¡åŸ®é¡ïŒMicroscopeïŒã®ããã«ã¢ã«ãŽãªãºã ã®çŽç²ãªããžãã¯ã ããèŠã蟌ãã§æ¯èŒããããšããèšèšææ³ã«ç±æ¥ããŸãã
æ¬èšäºã§ã¯ãå€éšã©ã€ãã©ãªã«äžåäŸåããªã Python æšæºã©ã€ãã©ãªã®ã¿ã®èšèšææ³ã®ããšãå€§èŠæš¡ãªäŸåã°ã©ããããã¥ã¡ã³ã矀ããç¹å®ã®ã³ã³ããã¹ããå®å šãã€æ±ºå®è«çã«æœåºã»ç®¡çãã sieve-scope ã®ä»çµã¿ã«ã€ããŠè§£èª¬ããŸãã
1. ã¹ã³ãŒã管çã®å¿ èŠæ§
è€éãªã·ã¹ãã ã巚倧ãªããŒã¿çŸ€ãè©äŸ¡ããéãå šé åãäžåºŠã«åŠçããããšã¯èšç®ã³ã¹ãã®å¢å€§ãæãã ãã§ãªããæå³ããªãã³ã³ããã¹ãã®æ··å ¥ïŒãã€ãºïŒãåŒãèµ·ãããŸãã
決å®è«çå¢çã®èšå®: sieve-scope ã¯ãå ¥åããŒã¿ãäŸåã°ã©ãã«å¯ŸããŠæç¢ºãªå¢çïŒã¹ã³ãŒãïŒãå®çŸ©ããè©äŸ¡ãšã³ãžã³ãåŠçãã¹ãéšåéåãæ£ç¢ºã«åãåºããŸãã
ãŒãäŸåã«ããåçŸæ§: ãµãŒãããŒãã£è£œã©ã€ãã©ãªãæé€ããçŽç²ãª Python å®è£ ã«ãããã©ã®å®è¡ç°å¢ã§ãã£ãŠãå®å šã«åäžã®ã¹ã³ãŒãå¢çãå°åºãããŸãã
2. ã¹ã³ãŒãæœåºãšãã£ã«ã¿ãªã³ã°ã®å éšæ§é
sieve-scope ã®ã³ã¢ããžãã¯ã¯ãã°ã©ãæ§é ããã³ããã¹ã矀ã«å¯Ÿããéå±€çã»æ¡ä»¶çãªãã£ã«ã¿ãªã³ã°ã決å®è«çã«å®è¡ããŸãã
éå±€çã³ã³ããã¹ãã®è§£æ±º
äŸåé¢ä¿ã®äžäžé¢ä¿ãã¢ãžã¥ãŒã«éã®æå±ãèµ°æ»ããéšåã°ã©ãïŒSub-graphïŒãç¹å®ã®éšåæååã¬ã³ãžãæœåºããŸãã
æ¢çŽ¢é åºãè©äŸ¡åªå 床ã¯ã¢ã«ãŽãªãºã çã«å³å¯ã«åºå®ãããŠãããå ¥åããŒã¿ã®äžŠã³é ã«äŸåããªãå®å®ããçµæãè¿ããŸãã
å·®åã¹ã³ãŒããšå¹çå
åçã«å€åããã°ã©ããããã¥ã¡ã³ã矀ã«å¯Ÿãã倿Žã®ãã£ãé åïŒå·®åïŒåšèŸºã®ã¿ãå²ãã¹ã³ãŒããè¿ éã«æ§ç¯ããåŸæ®µã® sieve-core ã sieve-referee ãžå¹ççã«åŒãæž¡ããŸãã
3. ãã€ãã©ã€ã³å šäœã«ãããã€ã³ãã°ã¬ãŒã·ã§ã³
ãããŸã§ã«è§£èª¬ããã¢ãžã¥ãŒã«çŸ€ãš sieve-scope ã¯ã次ã®ãããªäžé£ã®æ±ºå®è«çãã€ãã©ã€ã³ã圢æããŸãã
Sieve-Lens / Redact: 7-bit 芳枬空éã§ã®ããŒã¿ååŸãšãä»ãã€ãå®å šæ§ãç¶æãããã¹ãã³ã°ã
Sieve-Scope: åŠç察象ãšãªãã³ã³ããã¹ãã®å¢çïŒã¹ã³ãŒãïŒãå®çŸ©ããäžå¿ èŠãªãã€ãºãæé€ã
Sieve-Referee: ã¹ã³ãŒãå ã§åãåºãããããŒã¿ã«å¯Ÿãããã§ãŒã³æ€åºãæ§é çã¬ããã³ã¹ãé©çšã
Sieve-Core: æçµçãªäŸåã°ã©ãã®é路瞮çŽãšã·ã°ã¢ã€ã確信床è©äŸ¡ãå®è¡ã
ãã®é£æºã«ãããã·ã¹ãã å šäœã§ç¢ºççæšæž¬ãææ§ããäžåæé€ãããå ç¢ãªããŒã«ã«ã»ã¬ããã³ã¹ãå®çŸãããŸãã
4. 決å®è«çã¢ãããŒãã®å¶çŽãšéç
峿 Œãªå¢çå®çŸ©ãéèŠããèšèšäžãsieve-scope ã«ã¯ä»¥äžã®ãããªãã¬ãŒããªããååšããŸãã
åçã³ã³ããã¹ãå€åã«å¯Ÿãã远åŸã®ã³ã¹ã
å¢çã峿 Œã«åºå®ããããããæèã®ããããæé»çãªã¯ãã¹ãªãã¡ã¬ã³ã¹ã«å¯ŸããŠã¯ãæç€ºçãªã¹ã³ãŒãæ¡åŒµã«ãŒã«ã远å ããå¿ èŠããããŸãã
éå°ãªåæã®ãªã¹ã¯
ã¹ã³ãŒãã®ç²åºŠã现ãããããšãåºåã«ããã埪ç°äŸåïŒTarjan SCC ã§è§£æ±ºãããéè·¯ãªã©ïŒã®å šäœåãèŠãã¥ãããªããããé©åãªèšèšç²åºŠãæ±ããããŸãã
ãªãæå忝èŒã ãã§ã¯äžååãªã®ã
ã³ãŒãã®é¡äŒŒæ§ã調ã¹ãæ¹æ³ã«ã¯ãããããç°ãªãåŸæåéããããŸãã
æå忝èŒ
æå忝èŒã¯ãå®å šäžèŽãéšåäžèŽã確èªããããã§ç°¡åã§é«éãªæ¹æ³ã§ãã
ãããã空çœãã³ã¡ã³ããèå¥åã®å€æŽãªã©ã«ãã£ãŠæååã倧ããå€åãããšãå®éã®æ§é ã䌌ãŠããŠãé¡äŒŒåºŠãäžããå ŽåããããŸãã
ASTã«ããæ§é æ¯èŒ
ASTã¯ããœãŒã¹ã³ãŒããæ§æäžã®ããŒããšé¢ä¿ãšããŠè¡šçŸããŸãã
ããšãã°ã颿°å®çŸ©ãã«ãŒããä»£å ¥ã颿°åŒã³åºããæŒç®ãªã©ããæååãšã¯ç°ãªãæ§é ãšããŠæ±ããŸãã
ããã«ããããœãŒã¹ã³ãŒãã®ç©ºçœãã³ã¡ã³ãã«å·Šå³ããã«ããæ¯èŒãå¯èœã«ãªããŸãã
ãã ããASTã®ããŒãæ§é ãäžèŽããããšã¯ãããã°ã©ã ã®æå³ãäžèŽããããšãä¿èšŒããŸããããŸããASTã®æ§é ã ãã§ã¯ã倿°åãèå¥åã®äœ¿ããæ¹ã«é¢ããæ å ±ãååã«è©äŸ¡ã§ããªãå ŽåããããŸãã
èå¥åã®æ¯èŒ
èå¥åã¯ã倿°åã颿°åãã¯ã©ã¹åãªã©ãã³ãŒãå ã§äœ¿çšãããååã§ãã
èå¥åã®äžèŽãé åºãæ¯èŒããããšã§ãæ§é æ¯èŒã ãã§ã¯èŠãã«ããå ±éæ§ã芳枬ã§ããŸãã
äžæ¹ãèå¥åãç°ãªã£ãŠããŠãåãåŠçãå®çŸã§ãããããèå¥åã ãã§ã³ãŒãã®é¡äŒŒæ§ãå€å®ããããšãé©åã§ã¯ãããŸããã
Sieve-Scopeã¯ããããã®æ¹æ³ãçµã¿åãããæ§é ãšèå¥åã®åæ¹ã芳枬ããèšèšãæ¡çšããŠããŸãã
7ã€ã®ä»®èª¬ã«ããå€è»žè©äŸ¡ã¢ãã«
Sieve-Scope v1.3ã§ã¯ãã³ãŒããã¢ãH1ããH7ãŸã§ã®7ã€ã®äºå€ç¹åŸŽã«ãã£ãŠè¡šçŸããŸãã
åç¹åŸŽã¯ãå®çŸ©ãããæ¡ä»¶ãæºããå Žåã«1ãæºãããªãå Žåã«0ãšãªããŸãã
[
f(C_1,C_2)=(H_1,H_2,H_3,H_4,H_5,H_6,H_7)
]
ããã§ã(C_1) ãš (C_2) ã¯æ¯èŒå¯Ÿè±¡ã®ã³ãŒãã§ãã
ãã®ã¢ãã«ã§ã¯ã7ã€ã®ç¹åŸŽãäžã€ã®ã¹ã³ã¢ã«ãŸãšããã®ã§ã¯ãªããåå¥ã®èŠ³æž¬çµæãšããŠä¿æããŸãã
H1ïŒæ§æè§£æå¯èœæ§ïŒParseabilityïŒ
H1ã¯ãäž¡æ¹ã®ã³ãŒããPythonã® ast.parse() ã§æ§æè§£æã§ãããã衚ããŸãã
äž¡æ¹ã®è§£æã«æåããã°H1ã¯1ãšãªãããããããæ§æãšã©ãŒã«ãªãå Žåã¯0ãšãªããŸãã
æ§æè§£æã«å€±æããã³ãŒãã«ã€ããŠã¯ãåŸç¶ã®æ¯èŒåŠçãå®è¡ããã0000-000 ãè¿ããŸãã
ãã®æ®µéã¯ãã³ãŒãã®åè³ªãæ£ãããè©äŸ¡ãããã®ã§ã¯ãããŸããããããŸã§ãæ¯èŒã«å¿ èŠãªASTãæ§ç¯ã§ãããã©ããã確èªããŸãã
H2ïŒASTéªšæ Œã®äžèŽïŒAST SkeletonïŒ
H2ã¯ãASTãæ·±ãåªå ã§èµ°æ»ãã蚪åããããŒãåã®åãæ¯èŒããŸãã
ããŒãåã«ã¯ã颿°å®çŸ©ãä»£å ¥ãã«ãŒããªã©ã«å ããŠãLoadãStoreãAddãSub ãšãã£ãã³ã³ããã¹ããæŒç®åã«é¢ããããŒããå«ãŸããŸãã
çæãããããŒãåã®åãäžèŽããã°H2ã¯1ãšãªããŸãã
ãã®æ¯èŒã¯ããœãŒã¹ã³ãŒãã®æååãã®ãã®ã§ã¯ãªããæ§ææ§é ã®äžŠã³æ¹ã芳枬ãããã®ã§ãã
ãã ããH2ã¯ASTãæå³çã«åçŽåããå®å šãªæœè±¡è¡šçŸã§ã¯ãããŸããããã¹ãŠã®ããŒãåã察象ãšãããããæŒç®åãããŒãã»ã¹ãã¢ã®éããçµæã«åœ±é¿ããŸãã
ãŸããæ§é ãäžèŽããŠããèå¥åã宿°ã®éãã«ãã£ãŠããã°ã©ã ã®æå³ãç°ãªãå ŽåããããŸãã
H3ïŒèå¥åã®é¡äŒŒæ§ïŒPure Identifier SimilarityïŒ
H3ã¯ãPythonã® tokenize ã䜿ã£ãŠèå¥åãæœåºãããã®éåçãªäžèŽåºŠãšé åºã®äžèŽåºŠãçµã¿åãããŠè©äŸ¡ããŸãã
Pythonã®ããŒã¯ãŒãã¯é€å€ãããprintãlenãrangeãsumãintãstrãselfã__init__ ãªã©ãäºææ§ã®ããã«å®çŸ©ãããèå¥åãé€å€ãããŸãã
èå¥åã®è©äŸ¡å€ã¯ã次ã®2ã€ã®ææšãçµã¿åãããŠç®åºããŸãã
[
S_{\mathrm{H3}}=0.6J_m+0.4R_s
]
(J_m)ïŒèå¥åã®åºçŸåæ°ãèæ ®ãããã«ãã»ããJaccardé¡äŒŒåºŠ
(R_s)ïŒèå¥ååã®é åºãæ¯èŒããSequenceMatcherã®é¡äŒŒåºŠ
éŸå€ã¯ããã©ã«ãã§0.50ã§ããå®è£ ã§ã¯ãè©äŸ¡å€ãéŸå€ãå³å¯ã«è¶ ããå Žåã«H3ã1ãšããŸãã
ãã®æ¹æ³ã§ã¯ãå ±éããèå¥åãå€ããã©ããã ãã§ãªããããããã©ã®ãããªé åºã§çŸããããè©äŸ¡ã§ããŸãã
ãã ããèå¥åã®äžèŽã¯ãåãã¢ã«ãŽãªãºã ã§ããããšã®èšŒæã§ã¯ãããŸãããäžè¬çãªååãäžèŽããã ãã®å Žåãããã°ãååãå šé¢çã«å€æŽããããšã§èå¥åé¡äŒŒåºŠãäžããå ŽåããããŸãã
H4ïŒã¯ã©ã¹ã¿å®å šæ¡ä»¶ïŒCluster Safety InvariantïŒ
H4ã¯ãåäžã®ã³ãŒããã¢ã ãã§ã¯æ±ºå®ããŸããã
ã³ãŒãéåå šäœãè§£æããåãæå¹ã¯ã©ã¹ã¿ã«å±ããŠããããšãã¯ã©ã¹ã¿ãµã€ãºãèŠå®å€ä»¥äžã§ããããšããããŠå¯Ÿè±¡ãã¢èªèº«ã«çŽæ¥çãªé¡äŒŒæ ¹æ ãããããšã確èªããŠããèšå®ãããŸãã
æ¡ä»¶ã¯æ¬¡ã®ãšããã§ãã
[
H_4=1
\iff
|C|\geq 3
\land
(H_2=1\lor H_3=1)
]
ããã§ã(C) ã¯å¯Ÿè±¡ãã¢ãå±ããæå¹ã¯ã©ã¹ã¿ã§ãã
éèŠãªã®ã¯ãåãã¯ã©ã¹ã¿ã«å±ããŠãããšããçç±ã ãã§H4ã1ã«ããªãããšã§ããã¯ã©ã¹ã¿å ã®å¥ã®ã³ãŒããä»ãã鿥çãªã€ãªããã ãã§ã¯ã察象ãã¢èªèº«ã®çŽæ¥çãªé¡äŒŒæ ¹æ ãäœãåºããŸããã
ãã®å¶çŽã¯ã鿥é£éã«ãã£ãŠèª€ã£ã匷ãå€å®ãçãŸããããšãé²ãããã®å®å šæ¡ä»¶ã§ãã
H5ïŒäœçœ®çASTæ§é ïŒg_POSïŒ
H5ã§ã¯ãASTã®ãã£ãŒã«ãæ§é ãååž°çãªã¿ãã«ãžå€æãããã®è¡šçŸãSHA-256ã§ããã·ã¥åããŠæ¯èŒããŸãã
H2ãããŒãåã®åãæ¯èŒããã®ã«å¯ŸããH5ã¯ASTãã£ãŒã«ãã®éå±€çãªæ§é ã察象ã«ããŸãã
å®è£ ã§ã¯ãvalueãidãattr ãªã©ã®ãã£ãŒã«ãã®ãã¡ãASTããŒãã§ã¯ãªãå€ãé€å€ããŸãããã ãããã¹ãŠã®èå¥åæ å ±ãé€å»ããããã§ã¯ãªãã颿°åŒæ°åã®ãããªäžéšã®ã¹ã«ã©ãŒå€ã¯ä¿æãããŸãã
ãããã£ãŠãH5ã¯ãèå¥åãå®å šã«ç¡èŠããæå³æ§é ã®æ¯èŒãã§ã¯ãããŸãããäœçœ®ããã£ãŒã«ãã®æ§é ããå®çŸ©ãããé€å€èŠåã®ããšã§æ¯èŒããææšã§ãã
ããã·ã¥å€ãäžèŽããã°H5ã¯1ãšãªããŸãã
ãªããããã·ã¥å€ã®äžèŽã¯ãçŸåšã®è¡šçŸæ¹æ³ã«ãããæ§é æçŽã®äžèŽãæå³ããŸããããã°ã©ã ã®æå³çãªç䟡æ§ã蚌æãããã®ã§ã¯ãããŸããã
H6ïŒå®æ°ã·ã°ããã£ïŒg_CONSTïŒ
H6ã§ã¯ãASTãã ast.Constant ã®å€ãåéãããã®å€ãæåååããŠãœãŒããããªã¹ããæ¯èŒããŸãã
äž¡æ¹ã®ãªã¹ããäžèŽããã°H6ã¯1ãšãªããŸãã
ãã®ææšã¯ãæ°å€ãæååãçåœå€ãªã©ã®ãªãã©ã«ããã©ã®çšåºŠå ±éããŠãããã芳枬ããããã®ãã®ã§ãã
ããšãã°ãã«ãŒãã颿°åŒã³åºãã®æ§é ã䌌ãŠããŠãã䜿çšããéŸå€ãæååãªãã©ã«ãç°ãªãã°ãH6ã§å·®ç°ã芳枬ã§ããå¯èœæ§ããããŸãã
ãã ãã宿°ã®é åºã¯ãœãŒãã«ãã£ãŠå€±ãããŸãããŸããåã宿°ã䜿ãããŠãããããšãã£ãŠãåŠçã®æå³ãåäžã§ãããšã¯éããŸããã
H7ïŒèå¥åã®é£ç¶æ§ïŒg_FREQïŒ
H7ã§ã¯ãèå¥ååãã飿¥ãã2ã€ã®èå¥åãçµã¿åãããbi-gramãçæããåºçŸåæ°ã«åºã¥ãDiceä¿æ°ãèšç®ããŸãã
[
D(A,B)=
\frac{2\sum_g\min(c_A(g),c_B(g))}
{\sum_g c_A(g)+\sum_g c_B(g)}
]
ããã§ã(c_A(g)) ãš (c_B(g)) ã¯ãããããã®ã³ãŒãã«ãããbi-gram (g) ã®åºçŸåæ°ã§ãã
ããã©ã«ãã®éŸå€ã¯0.60ã§ãããDiceä¿æ°ãéŸå€ä»¥äžã®å Žåã«H7ã1ãšããŸãã
H3ãèå¥åå šäœã®ååžãšé åºãçµã¿åãããŠæ¯èŒããã®ã«å¯ŸããH7ã¯èå¥åãé£ç¶ããŠçŸãããã¿ãŒã³ã«çç®ããŸãã
ãã®éãã«ãã£ãŠãåãèå¥åã䜿ã£ãŠããŠãäžŠã³æ¹ãç°ãªãã±ãŒã¹ããå¥ã®èŠ³æž¬è»žããè©äŸ¡ã§ããŸãã
ãã ããèå¥åã2ã€æªæºã§bi-gramãçæãããªãå Žåã¯ãå®è£ äžã®ã¹ã³ã¢ã¯0ãšãªããŸããçãã³ãŒãã§ã¯ããã®ææšã ãã§ææãªæ¯èŒãè¡ããªãããšããããŸãã
ããããã¹ã¯ â å€å®æ ¹æ ãæ§é ãšããŠæ®ã
H1ããH7ãŸã§ã®çµæã¯ã7ãããã®ãã¹ã¯ãšããŠè¡šçŸãããŸãã
H1H2H3H4-H5H6H7ããšãã°ã1110-111 ãšãããã¹ã¯ã¯ã次ã®ç¶æ ã衚ããŸãã
H1ïŒäž¡æ¹ã®ã³ãŒããæ§æè§£æå¯èœ
H2ïŒASTéªšæ ŒãäžèŽ
H3ïŒèå¥åã®é¡äŒŒåºŠãéŸå€ãè¶ é
H4ïŒåç¬ã®ãã¢è©äŸ¡æç¹ã§ã¯æªç¢ºå®
H5ïŒäœçœ®çASTæ§é ãäžèŽ
H6ïŒå®æ°ã·ã°ããã£ãäžèŽ
H7ïŒèå¥åbi-gramã®é¡äŒŒåºŠãéŸå€ä»¥äž
ããã§æ³šæãããã®ã¯ãH4ãã¯ã©ã¹ã¿è§£æãéããŠèšå®ãããç¹åŸŽã ãšããããšã§ããåäžã®ãã¢ãè©äŸ¡ãã evaluate_pair() ã§ã¯H4ã¯åžžã«0ãšãªããã³ãŒãã¹å šäœã®è§£æã§å®å šæ¡ä»¶ãæºãããå Žåã«ã®ã¿1ãžææ ŒããŸãã
ãŸããH2ãšH3ã®äž¡æ¹ã0ã§ããã°ãH5ãH7ã®èšç®ãçç¥ãã1000-000 ãè¿ãã·ã§ãŒããµãŒããããé©çšãããŸãã
ããã¯ãäž»èŠãªæ§é ã»èå¥åã®æ¯èŒã§çŽæ¥çãªé¡äŒŒæ ¹æ ãèŠã€ãããªãå Žåã«ã远å ã®èšç®ãçãããã®æé©åã§ãã
ãã ããã·ã§ãŒããµãŒãããã«ãã£ãŠæªèšç®ãšãªã£ãH5ãH7ã0ã§ããããšã¯ããããã®ç¹åŸŽãç¬ç«ã«æ€èšŒããããšãæå³ããŸããããã¹ã¯ãèªãéã«ã¯ãshort_circuited ã®ç¶æ ã䜵ããŠç¢ºèªããå¿ èŠããããŸãã
ã¯ã©ã¹ã¿è§£æãšéæ¥é£éã®æ€åº
ã³ãŒãéåãæ¯èŒããå Žåãåã ã®ãã¢ã®çµæã ãã§ãªããã³ãŒãå士ãã©ã®ããã«ã€ãªãã£ãŠããããéèŠã§ãã
Sieve-Scopeã§ã¯ãH2ãŸãã¯H3ã«ãã£ãŠçŽæ¥çãªé¡äŒŒæ ¹æ ã確èªããããã¢ã蟺ãšããŠãã°ã©ããæ§ç¯ããŸãã
ãã®ã°ã©ãã®é£çµæåã調ã¹ãèŠå®ã®ãµã€ãºä»¥äžã®æåãæå¹ã¯ã©ã¹ã¿ãšããŠæ±ããŸãã
çŸåšã®ããã©ã«ãã§ã¯ãã¯ã©ã¹ã¿ãµã€ãºã®æå°å€ã¯3ã§ãã
ãªã鿥é£éã«æ³šæããã®ã
次ã®ãããªé¢ä¿ãèããŠã¿ãŸãã
A âââ B âââ CAãšBã®éã«ã¯çŽæ¥çãªé¡äŒŒæ ¹æ ããããBãšCã®éã«ãçŽæ¥çãªé¡äŒŒæ ¹æ ããããšããŸãã
ããããAãšCã®éã«çŽæ¥çãªé¡äŒŒæ ¹æ ããããšã¯éããŸããã
é¡äŒŒæ§ã®é¢ä¿ã¯ãå¿ ãããæšç§»çã§ã¯ãããŸãããAãšBã䌌ãŠããŠãBãšCã䌌ãŠãããããšãã£ãŠãAãšCãåãçšåºŠã«äŒŒãŠãããšã¯éããªãã®ã§ãã
Sieve-Scopeã§ã¯ãå¹ åªå æ¢çŽ¢ïŒBFSïŒã䜿ã£ãŠæå¹ã¯ã©ã¹ã¿å ã®éæ¥çµè·¯ãæ€åºããçŽæ¥çãªèŸºãååšããªããã¢ã«ã€ããŠèŠåãçæããŸãã
èŠåã«ã¯ã察象ãã¢ãçµç±ããçµè·¯ããããæ°ãªã©ãå«ãŸããŸãã
ãã®ä»çµã¿ã«ãã£ãŠãçŽæ¥æ¯èŒã®çµæãšãã¯ã©ã¹ã¿å ã§éæ¥çã«ã€ãªãã£ãŠãããšããäºå®ãåºå¥ã§ããŸãã
ãã ãã鿥çµè·¯ãååšããããšèªäœã¯ãäžæ£è¡çºãççšã®èšŒæ ã§ã¯ãããŸãããããã¯ãã³ãŒãéã®é¢ä¿ã確èªããããã®èŠåã§ãã
Sieve-Scopeãç®æãã®ã¯ããã¹ãŠã®é¢ä¿ãäžåŸã«åŒ·ãäžèŽãšæ±ãããšã§ã¯ãªããé¢ä¿ã®çš®é¡ãåããŠæç€ºããããšã§ãã
決å®è«æ§ãšåçŸæ§
Sieve-Scopeã®ã³ã¢ãšã³ãžã³ã¯ãPythonæšæºã©ã€ãã©ãªãäžå¿ã«æ§æãããŠããŸãã
ASTãtokenizeãCounterãSequenceMatcherãSHA-256ãProcessPoolExecutorãªã©ãå©çšããå€éšã®æ©æ¢°åŠç¿ãµãŒãã¹ããµãŒãããŒãã£è£œã©ã€ãã©ãªã«äŸåããã«ã³ãŒããã¢ãè©äŸ¡ããŸãã
ãã®èšèšã«ã¯ãããã€ãã®å©ç¹ããããŸãã
åŠçæ¡ä»¶ãæç€ºããã
å仮説ã®è©äŸ¡æ¹æ³ãéŸå€ãé€å€å¯Ÿè±¡ãã¯ã©ã¹ã¿ã®æ¡ä»¶ãã³ãŒããšä»æ§æžã«å®çŸ©ãããŠããŸãã
å®è¡çµæãæ¯èŒãããã
åãå ¥åãšèšå®ã§è©äŸ¡ããçµæãæ¯èŒãã倿Žã«ããå·®åãæ€èšŒã§ããŸãã
ããŒã«ã«ç°å¢ã§å©çšã§ãã
ãœãŒã¹ã³ãŒããå€éšãµãŒãã¹ãžéä¿¡ããã«è©äŸ¡ã§ãããããããŒã«ã«ã§ã®èª¿æ»ãéçºã¯ãŒã¯ãããŒãžã®çµã¿èŸŒã¿ã«é©ããŠããŸãã
ãŸããè€æ°ã®ãã¢ã䞊ååŠçããå Žåã«ããçµæãããŒã§ãœãŒãããŠããè¿ãããšã§ãåŠçå®äºé åºã«å·Šå³ãããªãåºåé ã確ä¿ããŠããŸãã
ãã ããæ±ºå®è«æ§ã¯ããã¹ãŠã®PythonããŒãžã§ã³ãå®è¡ç°å¢ã§ç¡æ¡ä»¶ã«åãçµæãåŸãããããšãæå³ããŸãããASTã®ä»æ§ãããŒãµãŒã®æåãå®è£ ã®ããŒãžã§ã³ãªã©ãå€ããã°ãèŠ³æž¬çµæã«åœ±é¿ããå¯èœæ§ããããŸãã
ãã®ãããåçŸæ§ãéèŠããéçšã§ã¯ãPythonã®ããŒãžã§ã³ãSieve-Scopeã®ããŒãžã§ã³ãè©äŸ¡ãã©ã¡ãŒã¿ãèšé²ããããšãéèŠã§ãã
ãã¹ããšéŸå€ã®æ€èšŒ
決å®è«çãªãšã³ãžã³ã§ãã£ãŠããã«ãŒã«ãéŸå€ãé©åã§ãããšã¯éããŸããã
Sieve-Scopeã§ã¯ãååž°ãã¹ãããšããžã±ãŒã¹ã®ãã¹ããè€æ°ã®PythonããŒãžã§ã³ã§ã®æ€èšŒãéŸå€ã®èª¿æŽãªã©ãéããŠãå®è£ ã®æåã確èªããŠããŸãã
v1.3ã§ã¯ãåŸæ¥ã®4ããããã¹ã¯ãšã®äºææ§ãç¶æããªãããH5ãH7ã®è¿œå 芳枬軞ãå°å ¥ããŠããŸãã
ãã®äºææ§ã¯ãæ¢åã®è©äŸ¡çµæããã¹ããç¶æããªããã芳枬ã¢ãã«ã段éçã«æ¡åŒµããããã®ãã®ã§ãã
ãŸããéçºæã®åæããŒã¿ã»ããã§ã¯ãå°æ°ã®æ£äŸã»è² äŸãçšããŠéŸå€ã調æŽããŠããŸãã仿§æžã«ã¯ã6ä»¶ã®æ£äŸãš3ä»¶ã®è² äŸãããªãåæããŒã¿ã»ããã§F1ã¹ã³ã¢1.0ãåŸãçµæãèšé²ãããŠããŸãã
ãã ãããã®æ°å€ã¯å°èŠæš¡ãªåæããŒã¿ã»ããã«å¯Ÿããçµæã§ããå®éã®ã³ãŒãããŒã¹ããç°ãªãã¢ã«ãŽãªãºã ãå«ãå€§èŠæš¡ãªããŒã¿ã»ããã«ãããæ€åºæ§èœãä¿èšŒãããã®ã§ã¯ãããŸããã
ä»åŸã¯ããã倿§ãªå®ããŒã¿ã䜿ãã誀æ€åºãšèŠéãã®äž¡æ¹ã枬å®ããå¿ èŠããããŸãã
Sieve-Scopeã®éç
Sieve-Scopeã¯ã決å®è«çãªã³ãŒãæ¯èŒãè¡ãããã®ãšã³ãžã³ã§ãããããã°ã©ã ã®æå³ãå®å šã«çè§£ãããã®ã§ã¯ãããŸããã
æ§é ãäžèŽããŠãæå³ã¯äžèŽããªã
éåžžã«çã颿°ãå®åçãªã³ãŒãã§ã¯ãç°ãªãåŠçã§ãåãASTããŒãåã®åãåŸãããå ŽåããããŸããH5ãH7ãªã©ã®è¿œå ç¹åŸŽã«ãã£ãŠåºå¥ã§ããã±ãŒã¹ã¯ãããŸãããæå³çãªç䟡æ§ãä¿èšŒãããã®ã§ã¯ãããŸããã
èå¥åã®å€æŽã«ã¯éçããã
倿°åã颿°åã倿ŽãããšãH3ãH7ã®çµæãå€ããå¯èœæ§ããããŸããäžæ¹ãH2ãH5ã®äžéšã¯ãèå¥ååãã®ãã®ã«äŸåãã«ããæ¯èŒãæäŸããŸãã
ãã ãããã¹ãŠã®ç¹åŸŽãèå¥åããå®å šã«ç¬ç«ããŠããããã§ã¯ãããŸãããããšãã°ãH5ã®æ§é 衚çŸã«ã¯äžéšã®ã¹ã«ã©ãŒå€ãæ®ãåŸãŸãã
宿°ã®äžèŽã¯ã¢ã«ãŽãªãºã ã®äžèŽãæå³ããªã
åã宿°ã䜿çšããŠããã³ãŒãã§ããåŠçã®ç®çãçµæã¯ç°ãªãå¯èœæ§ããããŸãããŸããH6ã§ã¯å®æ°ã®é åºã倱ãããããã宿°ãçŸããäœçœ®ã®éãã¯çŽæ¥è©äŸ¡ãããŸããã
æ§æè§£æå¯èœãªã³ãŒãã ããéåžžã®æ¯èŒå¯Ÿè±¡ã«ãªã
æ§æè§£æã«å€±æããã³ãŒãã¯ãéåžžã®ASTæ¯èŒãè¡ããã«çµäºããŸãããã®ãããäžå®å šãªã³ãŒããéäžã®ç·šéç¶æ ã«ããã³ãŒããå¥èšèªã®ã³ãŒããªã©ãæ±ãå Žåã«ã¯ãäºåã®å€æãå¥ã®è©äŸ¡æ¹æ³ãå¿ èŠã«ãªããŸãã
ã³ãŒãã®é¡äŒŒæ§ã¯ççšã®èšŒæã§ã¯ãªã
å ±éã®ã¢ã«ãŽãªãºã ãå ¬éãããŠãããµã³ãã«ã課é¡ãã³ãã¬ãŒããäžè¬çãªå®è£ ãã¿ãŒã³ãªã©ã«ãã£ãŠãæ£åœãªçç±ããã³ãŒãã䌌ãããšããããŸãã
Sieve-Scopeã®çµæã¯ãã³ãŒãã®é¡äŒŒæ§ã調æ»ããããã®æ ¹æ ã§ãããèäœæš©äŸµå®³ãççšã®æå³ãã©ã€ã»ã³ã¹éåãªã©ãåç¬ã§ç¢ºå®ãããã®ã§ã¯ãããŸããã
ä»åŸã®å±æ â 芳枬ã§ããç¯å²ãæç¢ºã«ãã
ä»åŸã®éçºã§ã¯ãåçŽã«ä»®èª¬ã®æ°ãå¢ããããšã ãã§ãªããå仮説ãã©ã®ãããªæ¡ä»¶ã§æå¹ãªã®ãããããå³å¯ã«æ€èšŒããŠãããããšèããŠããŸãã
å ·äœçã«ã¯ã次ã®ãããªæ¹åæ§ããããŸãã
å®ããŒã¿ã«ããæ§èœè©äŸ¡ïŒå€æ§ãªã¢ã«ãŽãªãºã ãå®è£ ã¹ã¿ã€ã«ãå«ãããŒã¿ã»ããã§ã誀æ€åºçãšèŠéãçãæž¬å®ããã
å€å®æ ¹æ ã®å¯èŠåïŒäžèŽããASTããŒããèå¥ååã宿°ãªã©ãã¬ããŒãäžã§æ¯èŒã§ããããã«ããã
PythonããŒãžã§ã³éã®æ€èšŒïŒAST仿§ã®å·®ç°ãèŠ³æž¬çµæã«äžãã圱é¿ã調ã¹ãã
å€§èŠæš¡ã³ãŒãã¹ãžã®å¯Ÿå¿ïŒãã¢æ¯èŒã®ã³ã¹ããæž¬å®ããå¿ èŠã«å¿ããŠåè£æœåºã䞊ååŠçãæ¹åããã
ã¹ã³ãŒãå¢çã®æç¢ºåïŒã©ã®ãã¡ã€ã«ãã³ãŒãé åãæ¯èŒå¯Ÿè±¡ãšããã©ããŸã§ã®é¢ä¿ãè©äŸ¡ããã®ãããããæç€ºçã«ç®¡çããã
ãããã¯ãä»åŸã®æ¹å課é¡ã§ããçŸåšã®å®è£ ã§ä¿èšŒãããŠããæ§è³ªãšãããããæ€èšŒããŠããæ§æ³ãåºå¥ããªããããããžã§ã¯ããçºå±ãããŠãããŸãã
ãããã« â 芳枬è ã§ãã£ãŠã仲è£è ã§ã¯ãªã
Sieve-Scopeã®äžå¿ã«ããã®ã¯ãã³ãŒãã®é¡äŒŒæ§ãäžã€ã®æ°åã ãã§æ±ºããªããšããèãæ¹ã§ãã
ASTéªšæ Œãèå¥åã®é¡äŒŒæ§ãäœçœ®çãªæ§é ã宿°ãèå¥åã®é£ç¶æ§ãããããã®ç¹åŸŽãç¬ç«ãã芳枬軞ãšããŠæ±ããæ¡ä»¶ã«åºã¥ããŠçµæãçµã¿åãããããšã§ãå€å®ã®æ ¹æ ã远跡ã§ããæ§é ãç®æããŠããŸãã
ãŸããã¯ã©ã¹ã¿å ã®éæ¥çãªã€ãªãããã察象ãã¢èªèº«ã®çŽæ¥çãªé¡äŒŒæ ¹æ ãšæ··åããªãããšãéèŠã§ãã
Observer, not arbiter. â 芳枬è ã§ãã£ãŠã仲è£è ã§ã¯ãªãã
é¡äŒŒããŠãããšããçµæã¯ãççšã®èšŒæ ãã®ãã®ã§ã¯ãããŸãããé¡äŒŒæ§ãæ€åºãããªãã£ãããšããæå³çãªé¢ä¿ãååšããªãããšã®èšŒæã§ã¯ãããŸããã
Sieve-Scopeã¯ãã³ãŒãã«ä»£ãã£ãŠçµè«ãäžãããã®ã·ã¹ãã ã§ã¯ãªãã人éãã³ãŒãã®é¢ä¿ãæ€èšŒããããã®èŠ³æž¬åºç€ã§ãã
決å®è«æ§ã説æå¯èœæ§ããããŠéçã®æç€ºããã®3ã€ã軞ã«ãã³ãŒãã®æ§é ãããæ£ç¢ºã«èŠ³æž¬ã§ãããšã³ãžã³ãç®æããŠéçºãç¶ããŠãããŸãã
â»æ¬èšäºã¯Sieve-Scope v1.3ã®ä»æ§ããã³å®è£ ã«åºã¥ã解説ã§ããé¡äŒŒæ§ã®è©äŸ¡çµæã¯ãå®çŸ©ãããç¹åŸŽéãšéŸå€ã«ãããã®ã§ãããããã°ã©ã ã®æå³çãªç䟡æ§ãççšã®äºå®ãåç¬ã§èšŒæãããã®ã§ã¯ãããŸããã
é¢é£ãªã³ã¯
Sieve ãšã³ã·ã¹ãã
Sieveã·ãªãŒãºå šäœã®èšèšææ³ãåãšã³ãžã³ã®åœ¹å²ã芳枬空éã®é¢ä¿ã«ã€ããŠã¯ãã¢ãŒããã¯ãã£ã€ã³ããã¯ã¹ãåç §ããŠãã ããã
Sieve ãšã³ã·ã¹ãã ã¢ãŒããã¯ãã£ã€ã³ããã¯ã¹ïŒnoteïŒ
PyPI â https://pypi.org/user/neguseatama/
GitHub ãªããžããª
Sieve-Core â æ±ºå®è«çãªæ§é éçŽã»ç¢ºä¿¡åºŠè©äŸ¡ã®åºç€ãšã³ãžã³
Sieve-Referee â ããã¹ãã®æ§é ã»æµçšã»èšãæãã芳枬ãããšã³ãžã³
Sieve-Scope â ãœãŒã¹ã³ãŒãã®æ§é çé¡äŒŒæ§ã芳枬ãããšã³ãžã³
Sieve-Lens â äžå¯èŠã³ã³ãã³ããé èœèŠçŽ ã芳枬ãããšã³ãžã³
Sieve-Redact â æå®é åã®æ±ºå®è«çãã¹ãã³ã°ãšå®å šæ§æ€èšŒ
Sieve-Order â ããã³ããã®è¡šå±€æ§é ã6-bit芳枬空éã§èŠ³æž¬ãããšã³ãžã³
Sieve-Tray â åãšã³ãžã³ãçµ±åããããŒã«ã«ããã³ããšã³ã
note æè¡èšäº
Sieve-ScopeïŒæ±ºå®è«çã¹ã³ãŒã管çãšã³ã³ããã¹ãå¢çã®å®çŸ©
Sieve-RedactïŒæ±ºå®è«çãã¹ãã³ã°ãšä»ãã€ãå®å šæ§æ€èšŒã«ããããŒã¿ä¿å š
Sieve-TrayïŒè€æ°ã®èŠ³æž¬ç©ºéãçµ±åããããŒã«ã«æ€èšŒããã³ããšã³ã
èšèšã®èšé²(ææèšäº)
Sieve ã·ãªãŒãºã®æ°çã¢ãã« â å°åºæ³ãšå®å®æ§
ã·ãªãŒãºã®ååŒããå£ãæ¹ã®å®æž¬ãããã©ãå°ãããããã
æèšç®ã§è¿œããç²åºŠã§è§£èª¬ãå°åºã® 3 段æé ãšèšèšãã§ãã¯ãªã¹ãä»ããSieve ã·ãªãŒãºã®æ§é èšèš â äŸåæ¹åãã¢ãŒããã¯ãã£ã決ãã
åäžãã¡ã€ã«ã»ãã©ã°ã€ã³ã»ã«ã¿ãã°çé¢ã»byte pinã5 ã€ã®æ§é 倿ã®
çç±ãšåäŸããããŠãã»ãŠã¹ã®è¹ããå§ãŸã£ãéçºçµç·¯ã