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
Excused absence disqualifies holiday pay · case 01
Workers on approved sick leave the day before a holiday lose holiday pay.
Excused absence disqualifies holiday pay · case 02
Workers on approved sick leave the day before a holiday lose holiday pay.
Excused absence disqualifies holiday pay · case 03
Workers on approved sick leave the day before a holiday lose holiday pay.
Excused absence disqualifies holiday pay · case 04
Workers on approved sick leave the day before a holiday lose holiday pay.
Excused absence disqualifies holiday pay · case 05
Workers on approved sick leave the day before a holiday lose holiday pay.
Holiday pay granted when only one surrounding shift is worked · case 01
Workers who skip the shift after a holiday still receive holiday pay.
Holiday pay granted when only one surrounding shift is worked · case 02
Workers who skip the shift after a holiday still receive holiday pay.
Holiday pay granted when only one surrounding shift is worked · case 03
Workers who skip the shift after a holiday still receive holiday pay.
Holiday pay granted when only one surrounding shift is worked · case 04
Workers who skip the shift after a holiday still receive holiday pay.
Holiday pay granted when only one surrounding shift is worked · case 05
Workers who skip the shift after a holiday still receive holiday pay.
Neighbour search treats day zero as scheduled · case 01
A holiday whose only prior days are unscheduled treats an off day at index 0 as the prior shift and is denied.
Neighbour search treats day zero as scheduled · case 02
A holiday whose only prior days are unscheduled treats an off day at index 0 as the prior shift and is denied.
Neighbour search treats day zero as scheduled · case 03
A holiday whose only prior days are unscheduled treats an off day at index 0 as the prior shift and is denied.
Neighbour search treats day zero as scheduled · case 04
A holiday whose only prior days are unscheduled treats an off day at index 0 as the prior shift and is denied.
Neighbour search treats day zero as scheduled · case 05
A holiday whose only prior days are unscheduled treats an off day at index 0 as the prior shift and is denied.
Shift running past the window end counted in full · case 01
A long shift starting late in the window pushes the total over the limit.
Shift running past the window end counted in full · case 02
A long shift starting late in the window pushes the total over the limit.
Shift running past the window end counted in full · case 03
A long shift starting late in the window pushes the total over the limit.
Shift running past the window end counted in full · case 04
A long shift starting late in the window pushes the total over the limit.
Shift running past the window end counted in full · case 05
A long shift starting late in the window pushes the total over the limit.
Rolling 24-hour limit evaluated on calendar days · case 01
Two long shifts straddling midnight never exceed a calendar-day total.
Rolling 24-hour limit evaluated on calendar days · case 02
Two long shifts straddling midnight never exceed a calendar-day total.
Rolling 24-hour limit evaluated on calendar days · case 03
Two long shifts straddling midnight never exceed a calendar-day total.
Rolling 24-hour limit evaluated on calendar days · case 04
Two long shifts straddling midnight never exceed a calendar-day total.
Rolling 24-hour limit evaluated on calendar days · case 05
Two long shifts straddling midnight never exceed a calendar-day total.
Exactly sixteen hours in 24 flagged as fatigue breach · case 01
Double shifts totalling 16 hours are rejected.
Exactly sixteen hours in 24 flagged as fatigue breach · case 02
Double shifts totalling 16 hours are rejected.
Exactly sixteen hours in 24 flagged as fatigue breach · case 03
Double shifts totalling 16 hours are rejected.
Exactly sixteen hours in 24 flagged as fatigue breach · case 04
Double shifts totalling 16 hours are rejected.
Exactly sixteen hours in 24 flagged as fatigue breach · case 05
Double shifts totalling 16 hours are rejected.
Overlapping callouts paid twice · case 01
Two overlapping callout records pay the shared minutes at full rate twice.
Overlapping callouts paid twice · case 02
Two overlapping callout records pay the shared minutes at full rate twice.
Overlapping callouts paid twice · case 03
Two overlapping callout records pay the shared minutes at full rate twice.
Overlapping callouts paid twice · case 04
Two overlapping callout records pay the shared minutes at full rate twice.
Overlapping callouts paid twice · case 05
Two overlapping callout records pay the shared minutes at full rate twice.
Callout minutes also earn the standby allowance · case 01
Minutes worked on a callout are paid both full rate and standby allowance.
Callout minutes also earn the standby allowance · case 02
Minutes worked on a callout are paid both full rate and standby allowance.
Callout minutes also earn the standby allowance · case 03
Minutes worked on a callout are paid both full rate and standby allowance.
Callout minutes also earn the standby allowance · case 04
Minutes worked on a callout are paid both full rate and standby allowance.
Callout minutes also earn the standby allowance · case 05
Minutes worked on a callout are paid both full rate and standby allowance.
Weekend standby rate uses a Sunday-first week · case 01
Monday standby is paid at the weekend rate and Saturday at the weekday rate.
Weekend standby rate uses a Sunday-first week · case 02
Monday standby is paid at the weekend rate and Saturday at the weekday rate.
Weekend standby rate uses a Sunday-first week · case 03
Monday standby is paid at the weekend rate and Saturday at the weekday rate.
Weekend standby rate uses a Sunday-first week · case 04
Monday standby is paid at the weekend rate and Saturday at the weekday rate.
Weekend standby rate uses a Sunday-first week · case 05
Monday standby is paid at the weekend rate and Saturday at the weekday rate.
Standby allowance truncated to the cent · case 01
Allowances with half a cent or more lose a cent.
Standby allowance truncated to the cent · case 02
Allowances with half a cent or more lose a cent.
Standby allowance truncated to the cent · case 03
Allowances with half a cent or more lose a cent.
Standby allowance truncated to the cent · case 04
Allowances with half a cent or more lose a cent.
Standby allowance truncated to the cent · case 05
Allowances with half a cent or more lose a cent.
Leave weeks averaged in as zero-hour weeks · case 01
Taking leave lowers the average, hiding excessive hours in other weeks.
Leave weeks averaged in as zero-hour weeks · case 02
Taking leave lowers the average, hiding excessive hours in other weeks.
Leave weeks averaged in as zero-hour weeks · case 03
Taking leave lowers the average, hiding excessive hours in other weeks.
Leave weeks averaged in as zero-hour weeks · case 04
Taking leave lowers the average, hiding excessive hours in other weeks.
Leave weeks averaged in as zero-hour weeks · case 05
Taking leave lowers the average, hiding excessive hours in other weeks.
Reference period not extended past skipped weeks · case 01
After a leave week the average covers fewer weeks than required.
Reference period not extended past skipped weeks · case 02
After a leave week the average covers fewer weeks than required.
Reference period not extended past skipped weeks · case 03
After a leave week the average covers fewer weeks than required.
Reference period not extended past skipped weeks · case 04
After a leave week the average covers fewer weeks than required.
Reference period not extended past skipped weeks · case 05
After a leave week the average covers fewer weeks than required.
Average divides by the nominal reference length · case 01
Short records understate the average when fewer weeks are available.
Average divides by the nominal reference length · case 02
Short records understate the average when fewer weeks are available.
Average divides by the nominal reference length · case 03
Short records understate the average when fewer weeks are available.
Average divides by the nominal reference length · case 04
Short records understate the average when fewer weeks are available.
Average divides by the nominal reference length · case 05
Short records understate the average when fewer weeks are available.
Average exactly at the limit flagged · case 01
A worker averaging exactly 48 hours is treated as in breach.
Average exactly at the limit flagged · case 02
A worker averaging exactly 48 hours is treated as in breach.
Average exactly at the limit flagged · case 03
A worker averaging exactly 48 hours is treated as in breach.
Average exactly at the limit flagged · case 04
A worker averaging exactly 48 hours is treated as in breach.
Average exactly at the limit flagged · case 05
A worker averaging exactly 48 hours is treated as in breach.
Weeks with partial leave kept in the reference period · case 01
A week with three leave days drags the average down.
Weeks with partial leave kept in the reference period · case 02
A week with three leave days drags the average down.
Weeks with partial leave kept in the reference period · case 03
A week with three leave days drags the average down.
Weeks with partial leave kept in the reference period · case 04
A week with three leave days drags the average down.
Weeks with partial leave kept in the reference period · case 05
A week with three leave days drags the average down.
Junior workers bid before senior workers · case 01
Newest hires take the most preferred shifts.
Junior workers bid before senior workers · case 02
Newest hires take the most preferred shifts.
Junior workers bid before senior workers · case 03
Newest hires take the most preferred shifts.
Junior workers bid before senior workers · case 04
Newest hires take the most preferred shifts.
Junior workers bid before senior workers · case 05
Newest hires take the most preferred shifts.
Bid cap compared with all shifts taken by anyone · case 01
Workers stop bidding once the whole team has taken as many shifts as their personal cap.
Bid cap compared with all shifts taken by anyone · case 02
Workers stop bidding once the whole team has taken as many shifts as their personal cap.
Bid cap compared with all shifts taken by anyone · case 03
Workers stop bidding once the whole team has taken as many shifts as their personal cap.
Bid cap compared with all shifts taken by anyone · case 04
Workers stop bidding once the whole team has taken as many shifts as their personal cap.
Bid cap compared with all shifts taken by anyone · case 05
Workers stop bidding once the whole team has taken as many shifts as their personal cap.
Senior worker takes several shifts in one bidding turn · case 01
A senior bidder claims all compatible preferences before juniors pick once.
Senior worker takes several shifts in one bidding turn · case 02
A senior bidder claims all compatible preferences before juniors pick once.
Senior worker takes several shifts in one bidding turn · case 03
A senior bidder claims all compatible preferences before juniors pick once.
Senior worker takes several shifts in one bidding turn · case 04
A senior bidder claims all compatible preferences before juniors pick once.
Senior worker takes several shifts in one bidding turn · case 05
A senior bidder claims all compatible preferences before juniors pick once.
Back-to-back shifts rejected as overlapping during bidding · case 01
Workers cannot bid for a shift that starts when their held shift ends.
Back-to-back shifts rejected as overlapping during bidding · case 02
Workers cannot bid for a shift that starts when their held shift ends.
Back-to-back shifts rejected as overlapping during bidding · case 03
Workers cannot bid for a shift that starts when their held shift ends.
Back-to-back shifts rejected as overlapping during bidding · case 04
Workers cannot bid for a shift that starts when their held shift ends.
Back-to-back shifts rejected as overlapping during bidding · case 05
Workers cannot bid for a shift that starts when their held shift ends.
Bidding stops after a single round · case 01
Workers with capacity left never pick a second shift.
Bidding stops after a single round · case 02
Workers with capacity left never pick a second shift.
Bidding stops after a single round · case 03
Workers with capacity left never pick a second shift.
Bidding stops after a single round · case 04
Workers with capacity left never pick a second shift.
Bidding stops after a single round · case 05
Workers with capacity left never pick a second shift.
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 ↗