Understand the failure.
Verify the repair.
Small, reproducible software failures. The broken implementation, the fix that didn’t work, and the one that passed—preserved together.
Explore the cases ↓How results are verified ↗WHAT THE ARCHIVE CONTAINS
100840 executable cases. 20168 are open.
Every case records the implementation that fails, the fix that did not work, and the repair that passed its checks—with recorded outputs and source hashes. This release adds 100840 cases across 20168 failure mechanisms and 254 domains.
The open tier gives you the failure and the unsuccessful fix for one case in every mechanism. The remaining 80672 cases, 5 variants per mechanism, are member-only: the verified repair, its recorded checks, and the full fixture suite are held in the member archive. Read the methodology ↗
A RECORD OF WHAT WENT WRONG
Browse the archive / 100840
An unknown TLV tag shifts the next record boundary · case 01
An unknown TLV tag shifts the next record boundary.
An unknown TLV tag shifts the next record boundary · case 02
An unknown TLV tag shifts the next record boundary.
An unknown TLV tag shifts the next record boundary · case 03
An unknown TLV tag shifts the next record boundary.
An unknown TLV tag shifts the next record boundary · case 04
An unknown TLV tag shifts the next record boundary.
An unknown TLV tag shifts the next record boundary · case 05
An unknown TLV tag shifts the next record boundary.
Run-length decoding treats zero count as one · case 01
Run-length decoding treats zero count as one.
Run-length decoding treats zero count as one · case 02
Run-length decoding treats zero count as one.
Run-length decoding treats zero count as one · case 03
Run-length decoding treats zero count as one.
Run-length decoding treats zero count as one · case 04
Run-length decoding treats zero count as one.
Run-length decoding treats zero count as one · case 05
Run-length decoding treats zero count as one.
Run-length encoding forgets the final run · case 01
Run-length encoding forgets the final run.
Run-length encoding forgets the final run · case 02
Run-length encoding forgets the final run.
Run-length encoding forgets the final run · case 03
Run-length encoding forgets the final run.
Run-length encoding forgets the final run · case 04
Run-length encoding forgets the final run.
Run-length encoding forgets the final run · case 05
Run-length encoding forgets the final run.
Escaped frame markers are decoded as delimiters · case 01
Escaped frame markers are decoded as delimiters.
Escaped frame markers are decoded as delimiters · case 02
Escaped frame markers are decoded as delimiters.
Escaped frame markers are decoded as delimiters · case 03
Escaped frame markers are decoded as delimiters.
Escaped frame markers are decoded as delimiters · case 04
Escaped frame markers are decoded as delimiters.
Escaped frame markers are decoded as delimiters · case 05
Escaped frame markers are decoded as delimiters.
A complemented byte mask includes unbounded high bits · case 01
A complemented byte mask includes unbounded high bits.
A complemented byte mask includes unbounded high bits · case 02
A complemented byte mask includes unbounded high bits.
A complemented byte mask includes unbounded high bits · case 03
A complemented byte mask includes unbounded high bits.
A complemented byte mask includes unbounded high bits · case 04
A complemented byte mask includes unbounded high bits.
A complemented byte mask includes unbounded high bits · case 05
A complemented byte mask includes unbounded high bits.
Delta decoding omits the first absolute sample · case 01
Delta decoding omits the first absolute sample.
Delta decoding omits the first absolute sample · case 02
Delta decoding omits the first absolute sample.
Delta decoding omits the first absolute sample · case 03
Delta decoding omits the first absolute sample.
Delta decoding omits the first absolute sample · case 04
Delta decoding omits the first absolute sample.
Delta decoding omits the first absolute sample · case 05
Delta decoding omits the first absolute sample.
Delta encoding subtracts the initial sample every time · case 01
Delta encoding subtracts the initial sample every time.
Delta encoding subtracts the initial sample every time · case 02
Delta encoding subtracts the initial sample every time.
Delta encoding subtracts the initial sample every time · case 03
Delta encoding subtracts the initial sample every time.
Delta encoding subtracts the initial sample every time · case 04
Delta encoding subtracts the initial sample every time.
Delta encoding subtracts the initial sample every time · case 05
Delta encoding subtracts the initial sample every time.
A binary boolean treats all nonzero octets as true · case 01
A binary boolean treats all nonzero octets as true.
A binary boolean treats all nonzero octets as true · case 02
A binary boolean treats all nonzero octets as true.
A binary boolean treats all nonzero octets as true · case 03
A binary boolean treats all nonzero octets as true.
A binary boolean treats all nonzero octets as true · case 04
A binary boolean treats all nonzero octets as true.
A binary boolean treats all nonzero octets as true · case 05
A binary boolean treats all nonzero octets as true.
A mixed-endian identifier reverses every field · case 01
A mixed-endian identifier reverses every field.
A mixed-endian identifier reverses every field · case 02
A mixed-endian identifier reverses every field.
A mixed-endian identifier reverses every field · case 03
A mixed-endian identifier reverses every field.
A mixed-endian identifier reverses every field · case 04
A mixed-endian identifier reverses every field.
A mixed-endian identifier reverses every field · case 05
A mixed-endian identifier reverses every field.
An all-ones unsigned sentinel becomes a valid length · case 01
An all-ones unsigned sentinel becomes a valid length.
An all-ones unsigned sentinel becomes a valid length · case 02
An all-ones unsigned sentinel becomes a valid length.
An all-ones unsigned sentinel becomes a valid length · case 03
An all-ones unsigned sentinel becomes a valid length.
An all-ones unsigned sentinel becomes a valid length · case 04
An all-ones unsigned sentinel becomes a valid length.
An all-ones unsigned sentinel becomes a valid length · case 05
An all-ones unsigned sentinel becomes a valid length.
A closing parenthesis appears before its matching opener · case 01
A closing parenthesis appears before its matching opener.
A closing parenthesis appears before its matching opener · case 02
A closing parenthesis appears before its matching opener.
A closing parenthesis appears before its matching opener · case 03
A closing parenthesis appears before its matching opener.
A closing parenthesis appears before its matching opener · case 04
A closing parenthesis appears before its matching opener.
A closing parenthesis appears before its matching opener · case 05
A closing parenthesis appears before its matching opener.
Balanced delimiters close in the wrong nesting order · case 01
Balanced delimiters close in the wrong nesting order.
Balanced delimiters close in the wrong nesting order · case 02
Balanced delimiters close in the wrong nesting order.
Balanced delimiters close in the wrong nesting order · case 03
Balanced delimiters close in the wrong nesting order.
Balanced delimiters close in the wrong nesting order · case 04
Balanced delimiters close in the wrong nesting order.
Balanced delimiters close in the wrong nesting order · case 05
Balanced delimiters close in the wrong nesting order.
A two-character operator is split into single tokens · case 01
A two-character operator is split into single tokens.
A two-character operator is split into single tokens · case 02
A two-character operator is split into single tokens.
A two-character operator is split into single tokens · case 03
A two-character operator is split into single tokens.
A two-character operator is split into single tokens · case 04
A two-character operator is split into single tokens.
A two-character operator is split into single tokens · case 05
A two-character operator is split into single tokens.
Keyword prefix matching splits an identifier · case 01
Keyword prefix matching splits an identifier.
Keyword prefix matching splits an identifier · case 02
Keyword prefix matching splits an identifier.
Keyword prefix matching splits an identifier · case 03
Keyword prefix matching splits an identifier.
Keyword prefix matching splits an identifier · case 04
Keyword prefix matching splits an identifier.
Keyword prefix matching splits an identifier · case 05
Keyword prefix matching splits an identifier.
Whitespace removal shifts token source offsets · case 01
Whitespace removal shifts token source offsets.
Whitespace removal shifts token source offsets · case 02
Whitespace removal shifts token source offsets.
Whitespace removal shifts token source offsets · case 03
Whitespace removal shifts token source offsets.
Whitespace removal shifts token source offsets · case 04
Whitespace removal shifts token source offsets.
Whitespace removal shifts token source offsets · case 05
Whitespace removal shifts token source offsets.
Character offsets are reported as UTF-8 byte offsets · case 01
Character offsets are reported as UTF-8 byte offsets.
Character offsets are reported as UTF-8 byte offsets · case 02
Character offsets are reported as UTF-8 byte offsets.
Character offsets are reported as UTF-8 byte offsets · case 03
Character offsets are reported as UTF-8 byte offsets.
Character offsets are reported as UTF-8 byte offsets · case 04
Character offsets are reported as UTF-8 byte offsets.
Character offsets are reported as UTF-8 byte offsets · case 05
Character offsets are reported as UTF-8 byte offsets.
Column tracking does not reset after a newline · case 01
Column tracking does not reset after a newline.
Column tracking does not reset after a newline · case 02
Column tracking does not reset after a newline.
Column tracking does not reset after a newline · case 03
Column tracking does not reset after a newline.
Column tracking does not reset after a newline · case 04
Column tracking does not reset after a newline.
Column tracking does not reset after a newline · case 05
Column tracking does not reset after a newline.
CRLF normalization creates duplicate newlines · case 01
CRLF normalization creates duplicate newlines.
CRLF normalization creates duplicate newlines · case 02
CRLF normalization creates duplicate newlines.
CRLF normalization creates duplicate newlines · case 03
CRLF normalization creates duplicate newlines.
CRLF normalization creates duplicate newlines · case 04
CRLF normalization creates duplicate newlines.
CRLF normalization creates duplicate newlines · case 05
CRLF normalization creates duplicate newlines.
A physical continuation inserts a semantic space · case 01
A physical continuation inserts a semantic space.
A physical continuation inserts a semantic space · case 02
A physical continuation inserts a semantic space.
A physical continuation inserts a semantic space · case 03
A physical continuation inserts a semantic space.
A physical continuation inserts a semantic space · case 04
A physical continuation inserts a semantic space.
A physical continuation inserts a semantic space · case 05
A physical continuation inserts a semantic space.
Paragraph splitting counts each blank line separately · case 01
Paragraph splitting counts each blank line separately.
Paragraph splitting counts each blank line separately · case 02
Paragraph splitting counts each blank line separately.
Paragraph splitting counts each blank line separately · case 03
Paragraph splitting counts each blank line separately.
Paragraph splitting counts each blank line separately · case 04
Paragraph splitting counts each blank line separately.
Paragraph splitting counts each blank line separately · case 05
Paragraph splitting counts each blank line separately.
INSPECTABLE BY DESIGN
Every result has a runnable source.
Runnable implementations with recorded outputs, source hashes, and explicit contracts. Related variants share a failure mechanism and belong together in evaluation splits.
Read the methodology ↗