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
Dawn and dusk parking positions: night boundary · case 01
At the moment of sunrise the tracker stays flat instead of parking toward the sun.
Dawn and dusk parking positions: night boundary · case 02
At the moment of sunrise the tracker stays flat instead of parking toward the sun.
Dawn and dusk parking positions: night boundary · case 03
At the moment of sunrise the tracker stays flat instead of parking toward the sun.
Dawn and dusk parking positions: night boundary · case 04
At the moment of sunrise the tracker stays flat instead of parking toward the sun.
Dawn and dusk parking positions: night boundary · case 05
At the moment of sunrise the tracker stays flat instead of parking toward the sun.
Dawn and dusk parking positions: wake elevation boundary · case 01
Tracking starts one elevation step later than configured.
Dawn and dusk parking positions: wake elevation boundary · case 02
Tracking starts one elevation step later than configured.
Dawn and dusk parking positions: wake elevation boundary · case 03
Tracking starts one elevation step later than configured.
Dawn and dusk parking positions: wake elevation boundary · case 04
Tracking starts one elevation step later than configured.
Dawn and dusk parking positions: wake elevation boundary · case 05
Tracking starts one elevation step later than configured.
Dawn and dusk parking positions: dawn parking side · case 01
At dawn the rows face west, away from the rising sun.
Dawn and dusk parking positions: dawn parking side · case 02
At dawn the rows face west, away from the rising sun.
Dawn and dusk parking positions: dawn parking side · case 03
At dawn the rows face west, away from the rising sun.
Dawn and dusk parking positions: dawn parking side · case 04
At dawn the rows face west, away from the rising sun.
Dawn and dusk parking positions: dawn parking side · case 05
At dawn the rows face west, away from the rising sun.
Dawn and dusk parking positions: dusk parking side · case 01
At dusk the rows face east, away from the setting sun.
Dawn and dusk parking positions: dusk parking side · case 02
At dusk the rows face east, away from the setting sun.
Dawn and dusk parking positions: dusk parking side · case 03
At dusk the rows face east, away from the setting sun.
Dawn and dusk parking positions: dusk parking side · case 04
At dusk the rows face east, away from the setting sun.
Dawn and dusk parking positions: dusk parking side · case 05
At dusk the rows face east, away from the setting sun.
Dawn and dusk parking positions: asymmetric tracking limits · case 01
Trackers with unequal stops are clamped with the limits swapped.
Dawn and dusk parking positions: asymmetric tracking limits · case 02
Trackers with unequal stops are clamped with the limits swapped.
Dawn and dusk parking positions: asymmetric tracking limits · case 03
Trackers with unequal stops are clamped with the limits swapped.
Dawn and dusk parking positions: asymmetric tracking limits · case 04
Trackers with unequal stops are clamped with the limits swapped.
Dawn and dusk parking positions: asymmetric tracking limits · case 05
Trackers with unequal stops are clamped with the limits swapped.
Dual-axis pointing error: elevation trig terms · case 01
Azimuth errors near the zenith are reported as huge.
Dual-axis pointing error: elevation trig terms · case 02
Azimuth errors near the zenith are reported as huge.
Dual-axis pointing error: elevation trig terms · case 03
Azimuth errors near the zenith are reported as huge.
Dual-axis pointing error: elevation trig terms · case 04
Azimuth errors near the zenith are reported as huge.
Dual-axis pointing error: elevation trig terms · case 05
Azimuth errors near the zenith are reported as huge.
Dual-axis pointing error: angular distance metric · case 01
Small commanded offsets across north read as nearly 360 degrees of error.
Dual-axis pointing error: angular distance metric · case 02
Small commanded offsets across north read as nearly 360 degrees of error.
Dual-axis pointing error: angular distance metric · case 03
Small commanded offsets across north read as nearly 360 degrees of error.
Dual-axis pointing error: angular distance metric · case 04
Small commanded offsets across north read as nearly 360 degrees of error.
Dual-axis pointing error: angular distance metric · case 05
Small commanded offsets across north read as nearly 360 degrees of error.
Dual-axis pointing error: azimuth difference units · case 01
Pointing errors swing wildly for small azimuth offsets.
Dual-axis pointing error: azimuth difference units · case 02
Pointing errors swing wildly for small azimuth offsets.
Dual-axis pointing error: azimuth difference units · case 03
Pointing errors swing wildly for small azimuth offsets.
Dual-axis pointing error: azimuth difference units · case 04
Pointing errors swing wildly for small azimuth offsets.
Dual-axis pointing error: azimuth difference units · case 05
Pointing errors swing wildly for small azimuth offsets.
Rest validation walks shifts in submission order · case 01
Rosters entered out of chronological order report spurious overlaps and miss real short rests.
Rest validation walks shifts in submission order · case 02
Rosters entered out of chronological order report spurious overlaps and miss real short rests.
Rest validation walks shifts in submission order · case 03
Rosters entered out of chronological order report spurious overlaps and miss real short rests.
Rest validation walks shifts in submission order · case 04
Rosters entered out of chronological order report spurious overlaps and miss real short rests.
Rest validation walks shifts in submission order · case 05
Rosters entered out of chronological order report spurious overlaps and miss real short rests.
Rest gap computed as wall-clock difference modulo one day · case 01
Two days off between shifts looks like a short rest, and overlapping shifts disappear.
Rest gap computed as wall-clock difference modulo one day · case 02
Two days off between shifts looks like a short rest, and overlapping shifts disappear.
Rest gap computed as wall-clock difference modulo one day · case 03
Two days off between shifts looks like a short rest, and overlapping shifts disappear.
Rest gap computed as wall-clock difference modulo one day · case 04
Two days off between shifts looks like a short rest, and overlapping shifts disappear.
Rest gap computed as wall-clock difference modulo one day · case 05
Two days off between shifts looks like a short rest, and overlapping shifts disappear.
Exactly-minimum rest flagged as a short rest violation · case 01
A roster giving precisely the required rest is rejected.
Exactly-minimum rest flagged as a short rest violation · case 02
A roster giving precisely the required rest is rejected.
Exactly-minimum rest flagged as a short rest violation · case 03
A roster giving precisely the required rest is rejected.
Exactly-minimum rest flagged as a short rest violation · case 04
A roster giving precisely the required rest is rejected.
Exactly-minimum rest flagged as a short rest violation · case 05
A roster giving precisely the required rest is rejected.
Back-to-back shifts classified as overlapping · case 01
A shift that starts the minute the previous one ends is reported as an overlap instead of a short rest.
Back-to-back shifts classified as overlapping · case 02
A shift that starts the minute the previous one ends is reported as an overlap instead of a short rest.
Back-to-back shifts classified as overlapping · case 03
A shift that starts the minute the previous one ends is reported as an overlap instead of a short rest.
Back-to-back shifts classified as overlapping · case 04
A shift that starts the minute the previous one ends is reported as an overlap instead of a short rest.
Back-to-back shifts classified as overlapping · case 05
A shift that starts the minute the previous one ends is reported as an overlap instead of a short rest.
Rest reference forgets a long enclosing shift · case 01
After a short shift nested inside a long one, later shifts are measured from the short shift's end.
Rest reference forgets a long enclosing shift · case 02
After a short shift nested inside a long one, later shifts are measured from the short shift's end.
Rest reference forgets a long enclosing shift · case 03
After a short shift nested inside a long one, later shifts are measured from the short shift's end.
Rest reference forgets a long enclosing shift · case 04
After a short shift nested inside a long one, later shifts are measured from the short shift's end.
Rest reference forgets a long enclosing shift · case 05
After a short shift nested inside a long one, later shifts are measured from the short shift's end.
Minimum rest parser drops the minutes component · case 01
A 10:30 minimum rest behaves like a 10-hour rest, accepting 10h15 gaps.
Minimum rest parser drops the minutes component · case 02
A 10:30 minimum rest behaves like a 10-hour rest, accepting 10h15 gaps.
Minimum rest parser drops the minutes component · case 03
A 10:30 minimum rest behaves like a 10-hour rest, accepting 10h15 gaps.
Minimum rest parser drops the minutes component · case 04
A 10:30 minimum rest behaves like a 10-hour rest, accepting 10h15 gaps.
Minimum rest parser drops the minutes component · case 05
A 10:30 minimum rest behaves like a 10-hour rest, accepting 10h15 gaps.
Seventh-day premium triggered by any Sunday shift · case 01
A worker with a normal Sunday shift after days off receives seventh-consecutive-day premiums.
Seventh-day premium triggered by any Sunday shift · case 02
A worker with a normal Sunday shift after days off receives seventh-consecutive-day premiums.
Seventh-day premium triggered by any Sunday shift · case 03
A worker with a normal Sunday shift after days off receives seventh-consecutive-day premiums.
Seventh-day premium triggered by any Sunday shift · case 04
A worker with a normal Sunday shift after days off receives seventh-consecutive-day premiums.
Seventh-day premium triggered by any Sunday shift · case 05
A worker with a normal Sunday shift after days off receives seventh-consecutive-day premiums.
Double-time minutes also counted in the time-and-a-half band · case 01
Long days pay minutes past 12 hours at both 1.5x and 2x.
Double-time minutes also counted in the time-and-a-half band · case 02
Long days pay minutes past 12 hours at both 1.5x and 2x.
Double-time minutes also counted in the time-and-a-half band · case 03
Long days pay minutes past 12 hours at both 1.5x and 2x.
Double-time minutes also counted in the time-and-a-half band · case 04
Long days pay minutes past 12 hours at both 1.5x and 2x.
Double-time minutes also counted in the time-and-a-half band · case 05
Long days pay minutes past 12 hours at both 1.5x and 2x.
Weekly overtime pyramids on top of daily overtime minutes · case 01
Weeks with long days move extra regular minutes into overtime, even making regular minutes negative.
Weekly overtime pyramids on top of daily overtime minutes · case 02
Weeks with long days move extra regular minutes into overtime, even making regular minutes negative.
Weekly overtime pyramids on top of daily overtime minutes · case 03
Weeks with long days move extra regular minutes into overtime, even making regular minutes negative.
Weekly overtime pyramids on top of daily overtime minutes · case 04
Weeks with long days move extra regular minutes into overtime, even making regular minutes negative.
Weekly overtime pyramids on top of daily overtime minutes · case 05
Weeks with long days move extra regular minutes into overtime, even making regular minutes negative.
Seventh-day double time starts at the ordinary 12-hour mark · case 01
Minutes 8 to 12 on the seventh consecutive day vanish from pay entirely.
Seventh-day double time starts at the ordinary 12-hour mark · case 02
Minutes 8 to 12 on the seventh consecutive day vanish from pay entirely.
Seventh-day double time starts at the ordinary 12-hour mark · case 03
Minutes 8 to 12 on the seventh consecutive day vanish from pay entirely.
Seventh-day double time starts at the ordinary 12-hour mark · case 04
Minutes 8 to 12 on the seventh consecutive day vanish from pay entirely.
Seventh-day double time starts at the ordinary 12-hour mark · case 05
Minutes 8 to 12 on the seventh consecutive day vanish from pay entirely.
Gross pay truncates half cents instead of rounding half up · case 01
Pay lines ending in exactly half a cent (or more) lose a cent.
Gross pay truncates half cents instead of rounding half up · case 02
Pay lines ending in exactly half a cent (or more) lose a cent.
Gross pay truncates half cents instead of rounding half up · case 03
Pay lines ending in exactly half a cent (or more) lose a cent.
Gross pay truncates half cents instead of rounding half up · case 04
Pay lines ending in exactly half a cent (or more) lose a cent.
Gross pay truncates half cents instead of rounding half up · case 05
Pay lines ending in exactly half a cent (or more) lose a cent.
Spread of hours computed from worked minutes instead of first-to-last span · case 01
Split days with long unpaid gaps never reach the spread premium.
Spread of hours computed from worked minutes instead of first-to-last span · case 02
Split days with long unpaid gaps never reach the spread premium.
Spread of hours computed from worked minutes instead of first-to-last span · case 03
Split days with long unpaid gaps never reach the spread premium.
Spread of hours computed from worked minutes instead of first-to-last span · case 04
Split days with long unpaid gaps never reach the spread premium.
Spread of hours computed from worked minutes instead of first-to-last span · case 05
Split days with long unpaid gaps never reach the spread premium.
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 ↗