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
Objects from another native window win semantic hit testing · case 01
Objects from another native window win semantic hit testing.
Objects from another native window win semantic hit testing · case 02
Objects from another native window win semantic hit testing.
Objects from another native window win semantic hit testing · case 03
Objects from another native window win semantic hit testing.
Objects from another native window win semantic hit testing · case 04
Objects from another native window win semantic hit testing.
Objects from another native window win semantic hit testing · case 05
Objects from another native window win semantic hit testing.
Unexposed native records participate in semantic hit testing · case 01
Unexposed native records participate in semantic hit testing.
Unexposed native records participate in semantic hit testing · case 02
Unexposed native records participate in semantic hit testing.
Unexposed native records participate in semantic hit testing · case 03
Unexposed native records participate in semantic hit testing.
Unexposed native records participate in semantic hit testing · case 04
Unexposed native records participate in semantic hit testing.
Unexposed native records participate in semantic hit testing · case 05
Unexposed native records participate in semantic hit testing.
Transparent semantic overlays mask inspectable objects · case 01
Transparent semantic overlays mask inspectable objects.
Transparent semantic overlays mask inspectable objects · case 02
Transparent semantic overlays mask inspectable objects.
Transparent semantic overlays mask inspectable objects · case 03
Transparent semantic overlays mask inspectable objects.
Transparent semantic overlays mask inspectable objects · case 04
Transparent semantic overlays mask inspectable objects.
Transparent semantic overlays mask inspectable objects · case 05
Transparent semantic overlays mask inspectable objects.
Touching an exclusive object boundary counts as a semantic hit · case 01
Touching an exclusive object boundary counts as a semantic hit.
Touching an exclusive object boundary counts as a semantic hit · case 02
Touching an exclusive object boundary counts as a semantic hit.
Touching an exclusive object boundary counts as a semantic hit · case 03
Touching an exclusive object boundary counts as a semantic hit.
Touching an exclusive object boundary counts as a semantic hit · case 04
Touching an exclusive object boundary counts as a semantic hit.
Touching an exclusive object boundary counts as a semantic hit · case 05
Touching an exclusive object boundary counts as a semantic hit.
Native clipped regions remain semantically hit-testable · case 01
Native clipped regions remain semantically hit-testable.
Native clipped regions remain semantically hit-testable · case 02
Native clipped regions remain semantically hit-testable.
Native clipped regions remain semantically hit-testable · case 03
Native clipped regions remain semantically hit-testable.
Native clipped regions remain semantically hit-testable · case 04
Native clipped regions remain semantically hit-testable.
Native clipped regions remain semantically hit-testable · case 05
Native clipped regions remain semantically hit-testable.
Semantic hit testing selects background objects before foreground objects · case 01
Semantic hit testing selects background objects before foreground objects.
Semantic hit testing selects background objects before foreground objects · case 02
Semantic hit testing selects background objects before foreground objects.
Semantic hit testing selects background objects before foreground objects · case 03
Semantic hit testing selects background objects before foreground objects.
Semantic hit testing selects background objects before foreground objects · case 04
Semantic hit testing selects background objects before foreground objects.
Semantic hit testing selects background objects before foreground objects · case 05
Semantic hit testing selects background objects before foreground objects.
Equal-stack semantic objects reverse native encounter precedence · case 01
Equal-stack semantic objects reverse native encounter precedence.
Equal-stack semantic objects reverse native encounter precedence · case 02
Equal-stack semantic objects reverse native encounter precedence.
Equal-stack semantic objects reverse native encounter precedence · case 03
Equal-stack semantic objects reverse native encounter precedence.
Equal-stack semantic objects reverse native encounter precedence · case 04
Equal-stack semantic objects reverse native encounter precedence.
Equal-stack semantic objects reverse native encounter precedence · case 05
Equal-stack semantic objects reverse native encounter precedence.
Semantic hit results return anonymous native records instead of semantic proxies · case 01
Semantic hit results return anonymous native records instead of semantic proxies.
Semantic hit results return anonymous native records instead of semantic proxies · case 02
Semantic hit results return anonymous native records instead of semantic proxies.
Semantic hit results return anonymous native records instead of semantic proxies · case 03
Semantic hit results return anonymous native records instead of semantic proxies.
Semantic hit results return anonymous native records instead of semantic proxies · case 04
Semantic hit results return anonymous native records instead of semantic proxies.
Semantic hit results return anonymous native records instead of semantic proxies · case 05
Semantic hit results return anonymous native records instead of semantic proxies.
Detector cutout reference: Padded cutout reference pixel omits the left halo · case 01
The adapter reports an incorrect crpix x while other fields remain valid.
Detector cutout reference: Padded cutout reference pixel omits the left halo · case 02
The adapter reports an incorrect crpix x while other fields remain valid.
Detector cutout reference: Padded cutout reference pixel omits the left halo · case 03
The adapter reports an incorrect crpix x while other fields remain valid.
Detector cutout reference: Padded cutout reference pixel omits the left halo · case 04
The adapter reports an incorrect crpix x while other fields remain valid.
Detector cutout reference: Padded cutout reference pixel omits the left halo · case 05
The adapter reports an incorrect crpix x while other fields remain valid.
Detector cutout reference: Vertical reference pixel reads the horizontal cutout metadata · case 01
The adapter reports an incorrect crpix y while other fields remain valid.
Detector cutout reference: Vertical reference pixel reads the horizontal cutout metadata · case 02
The adapter reports an incorrect crpix y while other fields remain valid.
Detector cutout reference: Vertical reference pixel reads the horizontal cutout metadata · case 03
The adapter reports an incorrect crpix y while other fields remain valid.
Detector cutout reference: Vertical reference pixel reads the horizontal cutout metadata · case 04
The adapter reports an incorrect crpix y while other fields remain valid.
Detector cutout reference: Vertical reference pixel reads the horizontal cutout metadata · case 05
The adapter reports an incorrect crpix y while other fields remain valid.
Detector cutout reference: FITS data section begins on the halo rather than first science pixel · case 01
The adapter reports an incorrect science first x while other fields remain valid.
Detector cutout reference: FITS data section begins on the halo rather than first science pixel · case 02
The adapter reports an incorrect science first x while other fields remain valid.
Detector cutout reference: FITS data section begins on the halo rather than first science pixel · case 03
The adapter reports an incorrect science first x while other fields remain valid.
Detector cutout reference: FITS data section begins on the halo rather than first science pixel · case 04
The adapter reports an incorrect science first x while other fields remain valid.
Detector cutout reference: FITS data section begins on the halo rather than first science pixel · case 05
The adapter reports an incorrect science first x while other fields remain valid.
Detector cutout reference: Inclusive data section endpoint is confused with science width · case 01
The adapter reports an incorrect science last x while other fields remain valid.
Detector cutout reference: Inclusive data section endpoint is confused with science width · case 02
The adapter reports an incorrect science last x while other fields remain valid.
Detector cutout reference: Inclusive data section endpoint is confused with science width · case 03
The adapter reports an incorrect science last x while other fields remain valid.
Detector cutout reference: Inclusive data section endpoint is confused with science width · case 04
The adapter reports an incorrect science last x while other fields remain valid.
Detector cutout reference: Inclusive data section endpoint is confused with science width · case 05
The adapter reports an incorrect science last x while other fields remain valid.
Detector cutout reference: Parent detector section origin includes padding from the child array · case 01
The adapter reports an incorrect parent first y while other fields remain valid.
Detector cutout reference: Parent detector section origin includes padding from the child array · case 02
The adapter reports an incorrect parent first y while other fields remain valid.
Detector cutout reference: Parent detector section origin includes padding from the child array · case 03
The adapter reports an incorrect parent first y while other fields remain valid.
Detector cutout reference: Parent detector section origin includes padding from the child array · case 04
The adapter reports an incorrect parent first y while other fields remain valid.
Detector cutout reference: Parent detector section origin includes padding from the child array · case 05
The adapter reports an incorrect parent first y while other fields remain valid.
Detector cutout reference: Parent inclusive detector endpoint is translated using padded height · case 01
The adapter reports an incorrect parent last y while other fields remain valid.
Detector cutout reference: Parent inclusive detector endpoint is translated using padded height · case 02
The adapter reports an incorrect parent last y while other fields remain valid.
Detector cutout reference: Parent inclusive detector endpoint is translated using padded height · case 03
The adapter reports an incorrect parent last y while other fields remain valid.
Detector cutout reference: Parent inclusive detector endpoint is translated using padded height · case 04
The adapter reports an incorrect parent last y while other fields remain valid.
Detector cutout reference: Parent inclusive detector endpoint is translated using padded height · case 05
The adapter reports an incorrect parent last y while other fields remain valid.
Binned detector centers: Binning treats detector corners as bin centers · case 01
The adapter reports an incorrect reference x while other fields remain valid.
Binned detector centers: Binning treats detector corners as bin centers · case 02
The adapter reports an incorrect reference x while other fields remain valid.
Binned detector centers: Binning treats detector corners as bin centers · case 03
The adapter reports an incorrect reference x while other fields remain valid.
Binned detector centers: Binning treats detector corners as bin centers · case 04
The adapter reports an incorrect reference x while other fields remain valid.
Binned detector centers: Binning treats detector corners as bin centers · case 05
The adapter reports an incorrect reference x while other fields remain valid.
Binned detector centers: Anisotropic binning applies the other detector axis geometry · case 01
The adapter reports an incorrect reference y while other fields remain valid.
Binned detector centers: Anisotropic binning applies the other detector axis geometry · case 02
The adapter reports an incorrect reference y while other fields remain valid.
Binned detector centers: Anisotropic binning applies the other detector axis geometry · case 03
The adapter reports an incorrect reference y while other fields remain valid.
Binned detector centers: Anisotropic binning applies the other detector axis geometry · case 04
The adapter reports an incorrect reference y while other fields remain valid.
Binned detector centers: Anisotropic binning applies the other detector axis geometry · case 05
The adapter reports an incorrect reference y while other fields remain valid.
Binned detector centers: Binned angular pixel scale uses the inverse bin factor · case 01
The adapter reports an incorrect plate scale while other fields remain valid.
Binned detector centers: Binned angular pixel scale uses the inverse bin factor · case 02
The adapter reports an incorrect plate scale while other fields remain valid.
Binned detector centers: Binned angular pixel scale uses the inverse bin factor · case 03
The adapter reports an incorrect plate scale while other fields remain valid.
Binned detector centers: Binned angular pixel scale uses the inverse bin factor · case 04
The adapter reports an incorrect plate scale while other fields remain valid.
Binned detector centers: Binned angular pixel scale uses the inverse bin factor · case 05
The adapter reports an incorrect plate scale while other fields remain valid.
Binned detector centers: Dispersion follows spatial rather than spectral binning · case 01
The adapter reports an incorrect spectral step while other fields remain valid.
Binned detector centers: Dispersion follows spatial rather than spectral binning · case 02
The adapter reports an incorrect spectral step while other fields remain valid.
Binned detector centers: Dispersion follows spatial rather than spectral binning · case 03
The adapter reports an incorrect spectral step while other fields remain valid.
Binned detector centers: Dispersion follows spatial rather than spectral binning · case 04
The adapter reports an incorrect spectral step while other fields remain valid.
Binned detector centers: Dispersion follows spatial rather than spectral binning · case 05
The adapter reports an incorrect spectral step while other fields remain valid.
Binned detector centers: A binned sample footprint assumes square detector bins · case 01
The adapter reports an incorrect footprint pixels while other fields remain valid.
Binned detector centers: A binned sample footprint assumes square detector bins · case 02
The adapter reports an incorrect footprint pixels while other fields remain valid.
Binned detector centers: A binned sample footprint assumes square detector bins · case 03
The adapter reports an incorrect footprint pixels while other fields remain valid.
Binned detector centers: A binned sample footprint assumes square detector bins · case 04
The adapter reports an incorrect footprint pixels while other fields remain valid.
Binned detector centers: A binned sample footprint assumes square detector bins · case 05
The adapter reports an incorrect footprint pixels while other fields remain valid.
Binned detector centers: Last binned detector center is replaced by the outer edge · case 01
The adapter reports an incorrect last center x while other fields remain valid.
Binned detector centers: Last binned detector center is replaced by the outer edge · case 02
The adapter reports an incorrect last center x while other fields remain valid.
Binned detector centers: Last binned detector center is replaced by the outer edge · case 03
The adapter reports an incorrect last center x while other fields remain valid.
Binned detector centers: Last binned detector center is replaced by the outer edge · case 04
The adapter reports an incorrect last center x while other fields remain valid.
Binned detector centers: Last binned detector center is replaced by the outer edge · case 05
The adapter reports an incorrect last center x while other fields remain valid.
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 ↗