Compliance cases

These are the 7,246 inputs a submission must answer. A conforming implementation returns true for the 2,791 valid ones and false for the 4,455 invalid ones. You can run the suite without reading it, or open any case to see its input and where its first ill-formed byte is.

Well-formed means Table 3-7 of the Unicode Standard, as RFC 3629 restates it: no overlong forms, surrogates, code points above U+10FFFF, stray continuation bytes or sequences cut short. The compliance suite combines Markus Kuhn's UTF-8 stress test, WPT's fatal TextDecoder tests, every lead byte against the edges of its allowed ranges, errors at every offset around 16-, 32- and 64-byte blocks and at the end of every length up to 130 bytes, large inputs, and seeded random text. simdutf, Python and a Table 3-7 reference agree on every answer.

Run the full compliance suite locally. Replace your-component.wasm with your compiled component. Download the suite

curl -fLo validate_utf8.comply.wasm https://calculated.world/comply/validate_utf8.wasm
npx qipok your-component.wasm --suite validate_utf8.comply.wasm
Clear

1,451 of 7,246 cases · 334 true · 1,117 false

Case Name Expected Note Size
000961 regression/lead DE second 90 Focused regression true DE 90: lead byte DE then 90. 2 B
000962 regression/lead DE second 9F Focused regression true DE 9F: lead byte DE then 9F. 2 B
000963 regression/lead DE second A0 Focused regression true DE A0: lead byte DE then A0. 2 B
000964 regression/lead DE second BF Focused regression true DE BF: lead byte DE then BF. 2 B
000965 regression/lead DE second C0 Focused regression false DE C0: lead byte DE then C0. 2 B
000966 regression/lead DE second C1 Focused regression false DE C1: lead byte DE then C1. 2 B
000967 regression/lead DE second C2 Focused regression false DE C2: lead byte DE then C2. 2 B
000968 regression/lead DE second DF Focused regression false DE DF: lead byte DE then DF. 2 B
000969 regression/lead DE second E0 Focused regression false DE E0: lead byte DE then E0. 2 B
000970 regression/lead DE second ED Focused regression false DE ED: lead byte DE then ED. 2 B
000971 regression/lead DE second EF Focused regression false DE EF: lead byte DE then EF. 2 B
000972 regression/lead DE second F0 Focused regression false DE F0: lead byte DE then F0. 2 B
000973 regression/lead DE second F4 Focused regression false DE F4: lead byte DE then F4. 2 B
000974 regression/lead DE second F5 Focused regression false DE F5: lead byte DE then F5. 2 B
000975 regression/lead DE second FF Focused regression false DE FF: lead byte DE then FF. 2 B
000976 regression/lead DF second 00 Focused regression false DF 00: lead byte DF then 00. 2 B
000977 regression/lead DF second 41 Focused regression false DF 41: lead byte DF then 41. 2 B
000978 regression/lead DF second 7F Focused regression false DF 7F: lead byte DF then 7F. 2 B
000979 regression/lead DF second 80 Focused regression true DF 80: lead byte DF then 80. 2 B
000980 regression/lead DF second 8F Focused regression true DF 8F: lead byte DF then 8F. 2 B
000981 regression/lead DF second 90 Focused regression true DF 90: lead byte DF then 90. 2 B
000982 regression/lead DF second 9F Focused regression true DF 9F: lead byte DF then 9F. 2 B
000983 regression/lead DF second A0 Focused regression true DF A0: lead byte DF then A0. 2 B
000984 regression/lead DF second C0 Focused regression false DF C0: lead byte DF then C0. 2 B
000985 regression/lead DF second C1 Focused regression false DF C1: lead byte DF then C1. 2 B
000986 regression/lead DF second C2 Focused regression false DF C2: lead byte DF then C2. 2 B
000987 regression/lead DF second DF Focused regression false DF DF: lead byte DF then DF. 2 B
000988 regression/lead DF second E0 Focused regression false DF E0: lead byte DF then E0. 2 B
000989 regression/lead DF second ED Focused regression false DF ED: lead byte DF then ED. 2 B
000990 regression/lead DF second EF Focused regression false DF EF: lead byte DF then EF. 2 B
000991 regression/lead DF second F0 Focused regression false DF F0: lead byte DF then F0. 2 B
000992 regression/lead DF second F4 Focused regression false DF F4: lead byte DF then F4. 2 B
000993 regression/lead DF second F5 Focused regression false DF F5: lead byte DF then F5. 2 B
000994 regression/lead DF second FF Focused regression false DF FF: lead byte DF then FF. 2 B
000995 regression/lead E0 second 00 Focused regression false E0 00 80: lead byte E0 then 00. 3 B
000996 regression/lead E0 second 41 Focused regression false E0 41 80: lead byte E0 then 41. 3 B
000997 regression/lead E0 second 7F Focused regression false E0 7F 80: lead byte E0 then 7F. 3 B
000998 regression/lead E0 second 8F Focused regression false E0 8F 80: lead byte E0 then 8F. 3 B
000999 regression/lead E0 second 90 Focused regression false E0 90 80: lead byte E0 then 90. 3 B
001000 regression/lead E0 second 9F Focused regression false E0 9F 80: lead byte E0 then 9F. 3 B
001001 regression/lead E0 second BF Focused regression true E0 BF 80: lead byte E0 then BF. 3 B
001002 regression/lead E0 second C0 Focused regression false E0 C0 80: lead byte E0 then C0. 3 B
001003 regression/lead E0 second C1 Focused regression false E0 C1 80: lead byte E0 then C1. 3 B
001004 regression/lead E0 second C2 Focused regression false E0 C2 80: lead byte E0 then C2. 3 B
001005 regression/lead E0 second DF Focused regression false E0 DF 80: lead byte E0 then DF. 3 B
001006 regression/lead E0 second E0 Focused regression false E0 E0 80: lead byte E0 then E0. 3 B
001007 regression/lead E0 second ED Focused regression false E0 ED 80: lead byte E0 then ED. 3 B
001008 regression/lead E0 second EF Focused regression false E0 EF 80: lead byte E0 then EF. 3 B
001009 regression/lead E0 second F0 Focused regression false E0 F0 80: lead byte E0 then F0. 3 B
001010 regression/lead E0 second F4 Focused regression false E0 F4 80: lead byte E0 then F4. 3 B

Where these rules come from

Well-formed UTF-8 is defined by Table 3-7 of the Unicode Standard, chapter 3 , which RFC 3629 restates as ABNF. Every answer is agreed by simdutf 9.4.0, Python’s strict codec and a direct reading of the table; the suite’s own tests add Node.js’s buffer.isUtf8 and a fatal TextDecoder.