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
Address label line wrapper: overlong words are split only once · case 01
A word more than twice the width leaves a line longer than the label.
Address label line wrapper: overlong words are split only once · case 02
A word more than twice the width leaves a line longer than the label.
Address label line wrapper: overlong words are split only once · case 03
A word more than twice the width leaves a line longer than the label.
Address label line wrapper: overlong words are split only once · case 04
A word more than twice the width leaves a line longer than the label.
Address label line wrapper: overlong words are split only once · case 05
A word more than twice the width leaves a line longer than the label.
Address label line wrapper: lines beyond the budget are returned · case 01
A two-line label receives three lines.
Address label line wrapper: lines beyond the budget are returned · case 02
A two-line label receives three lines.
Address label line wrapper: lines beyond the budget are returned · case 03
A two-line label receives three lines.
Address label line wrapper: lines beyond the budget are returned · case 04
A two-line label receives three lines.
Address label line wrapper: lines beyond the budget are returned · case 05
A two-line label receives three lines.
Address label line wrapper: the last line is never emitted · case 01
The final words of every label disappear.
Address label line wrapper: the last line is never emitted · case 02
The final words of every label disappear.
Address label line wrapper: the last line is never emitted · case 03
The final words of every label disappear.
Address label line wrapper: the last line is never emitted · case 04
The final words of every label disappear.
Address label line wrapper: the last line is never emitted · case 05
The final words of every label disappear.
Address label line wrapper: repeated spaces become empty words · case 01
"1 FAKE RD" keeps its runs of spaces on the label.
Address label line wrapper: repeated spaces become empty words · case 02
"1 FAKE RD" keeps its runs of spaces on the label.
Address label line wrapper: repeated spaces become empty words · case 03
"1 FAKE RD" keeps its runs of spaces on the label.
Address label line wrapper: repeated spaces become empty words · case 04
"1 FAKE RD" keeps its runs of spaces on the label.
Address label line wrapper: repeated spaces become empty words · case 05
"1 FAKE RD" keeps its runs of spaces on the label.
House number parser: hyphenated suffix is rejected · case 01
"13-B FAKE ST" returns None instead of suffix B.
House number parser: hyphenated suffix is rejected · case 02
"13-B FAKE ST" returns None instead of suffix B.
House number parser: hyphenated suffix is rejected · case 03
"13-B FAKE ST" returns None instead of suffix B.
House number parser: hyphenated suffix is rejected · case 04
"13-B FAKE ST" returns None instead of suffix B.
House number parser: hyphenated suffix is rejected · case 05
"13-B FAKE ST" returns None instead of suffix B.
House number parser: two-letter suffixes are accepted · case 01
"13AB" parses with suffix "AB".
House number parser: two-letter suffixes are accepted · case 02
"13AB" parses with suffix "AB".
House number parser: two-letter suffixes are accepted · case 03
"13AB" parses with suffix "AB".
House number parser: two-letter suffixes are accepted · case 04
"13AB" parses with suffix "AB".
House number parser: two-letter suffixes are accepted · case 05
"13AB" parses with suffix "AB".
House number parser: any denominator is accepted as a house fraction · case 01
"13 5/8 FAKE ST" consumes 5/8 as a house fraction.
House number parser: any denominator is accepted as a house fraction · case 02
"13 5/8 FAKE ST" consumes 5/8 as a house fraction.
House number parser: any denominator is accepted as a house fraction · case 03
"13 5/8 FAKE ST" consumes 5/8 as a house fraction.
House number parser: any denominator is accepted as a house fraction · case 04
"13 5/8 FAKE ST" consumes 5/8 as a house fraction.
House number parser: any denominator is accepted as a house fraction · case 05
"13 5/8 FAKE ST" consumes 5/8 as a house fraction.
House number parser: improper fractions are consumed · case 01
"13 2/2 FAKE ST" consumes "2/2" as a fraction.
House number parser: improper fractions are consumed · case 02
"13 2/2 FAKE ST" consumes "2/2" as a fraction.
House number parser: improper fractions are consumed · case 03
"13 2/2 FAKE ST" consumes "2/2" as a fraction.
House number parser: improper fractions are consumed · case 04
"13 2/2 FAKE ST" consumes "2/2" as a fraction.
House number parser: improper fractions are consumed · case 05
"13 2/2 FAKE ST" consumes "2/2" as a fraction.
House number parser: the fraction is duplicated in the street remainder · case 01
A fraction is reported and also left at the start of the street.
House number parser: the fraction is duplicated in the street remainder · case 02
A fraction is reported and also left at the start of the street.
House number parser: the fraction is duplicated in the street remainder · case 03
A fraction is reported and also left at the start of the street.
House number parser: the fraction is duplicated in the street remainder · case 04
A fraction is reported and also left at the start of the street.
House number parser: the fraction is duplicated in the street remainder · case 05
A fraction is reported and also left at the start of the street.
House number parser: house number zero is accepted · case 01
"0 FAKE ST" parses as house number 0.
House number parser: house number zero is accepted · case 02
"0 FAKE ST" parses as house number 0.
House number parser: house number zero is accepted · case 03
"0 FAKE ST" parses as house number 0.
House number parser: house number zero is accepted · case 04
"0 FAKE ST" parses as house number 0.
House number parser: house number zero is accepted · case 05
"0 FAKE ST" parses as house number 0.
Post office box canonicalizer: periods are deleted instead of becoming spaces · case 01
"P.O.BOX 101" is classified as a street line "POBOX 101".
Post office box canonicalizer: periods are deleted instead of becoming spaces · case 02
"P.O.BOX 101" is classified as a street line "POBOX 101".
Post office box canonicalizer: periods are deleted instead of becoming spaces · case 03
"P.O.BOX 101" is classified as a street line "POBOX 101".
Post office box canonicalizer: periods are deleted instead of becoming spaces · case 04
"P.O.BOX 101" is classified as a street line "POBOX 101".
Post office box canonicalizer: periods are deleted instead of becoming spaces · case 05
"P.O.BOX 101" is classified as a street line "POBOX 101".
Post office box canonicalizer: street names starting with BOX are treated as boxes · case 01
"BOXWOOD LANE" is reported as an invalid box line.
Post office box canonicalizer: street names starting with BOX are treated as boxes · case 02
"BOXWOOD LANE" is reported as an invalid box line.
Post office box canonicalizer: street names starting with BOX are treated as boxes · case 03
"BOXWOOD LANE" is reported as an invalid box line.
Post office box canonicalizer: street names starting with BOX are treated as boxes · case 04
"BOXWOOD LANE" is reported as an invalid box line.
Post office box canonicalizer: street names starting with BOX are treated as boxes · case 05
"BOXWOOD LANE" is reported as an invalid box line.
Post office box canonicalizer: text after the box identifier is ignored · case 01
"PO BOX 101 FAKETOWN" is accepted as "PO BOX 101".
Post office box canonicalizer: text after the box identifier is ignored · case 02
"PO BOX 101 FAKETOWN" is accepted as "PO BOX 101".
Post office box canonicalizer: text after the box identifier is ignored · case 03
"PO BOX 101 FAKETOWN" is accepted as "PO BOX 101".
Post office box canonicalizer: text after the box identifier is ignored · case 04
"PO BOX 101 FAKETOWN" is accepted as "PO BOX 101".
Post office box canonicalizer: text after the box identifier is ignored · case 05
"PO BOX 101 FAKETOWN" is accepted as "PO BOX 101".
Post office box canonicalizer: alphanumeric box identifiers are rejected · case 01
"Box 1A" is reported invalid.
Post office box canonicalizer: alphanumeric box identifiers are rejected · case 02
"Box 1A" is reported invalid.
Post office box canonicalizer: alphanumeric box identifiers are rejected · case 03
"Box 1A" is reported invalid.
Post office box canonicalizer: alphanumeric box identifiers are rejected · case 04
"Box 1A" is reported invalid.
Post office box canonicalizer: alphanumeric box identifiers are rejected · case 05
"Box 1A" is reported invalid.
Post office box canonicalizer: a prefix without a number falls through to street · case 01
"POST OFFICE BOX" alone is classified as a street line.
Post office box canonicalizer: a prefix without a number falls through to street · case 02
"POST OFFICE BOX" alone is classified as a street line.
Post office box canonicalizer: a prefix without a number falls through to street · case 03
"POST OFFICE BOX" alone is classified as a street line.
Post office box canonicalizer: a prefix without a number falls through to street · case 04
"POST OFFICE BOX" alone is classified as a street line.
Post office box canonicalizer: a prefix without a number falls through to street · case 05
"POST OFFICE BOX" alone is classified as a street line.
Post office box canonicalizer: street lines keep their runs of spaces · case 01
"1 FAKE RD" is returned with its original spacing.
Post office box canonicalizer: street lines keep their runs of spaces · case 02
"1 FAKE RD" is returned with its original spacing.
Post office box canonicalizer: street lines keep their runs of spaces · case 03
"1 FAKE RD" is returned with its original spacing.
Post office box canonicalizer: street lines keep their runs of spaces · case 04
"1 FAKE RD" is returned with its original spacing.
Post office box canonicalizer: street lines keep their runs of spaces · case 05
"1 FAKE RD" is returned with its original spacing.
Post office box canonicalizer: the spelled-out prefix is missing · case 01
"post office box 101" is a street line.
Post office box canonicalizer: the spelled-out prefix is missing · case 02
"post office box 101" is a street line.
Post office box canonicalizer: the spelled-out prefix is missing · case 03
"post office box 101" is a street line.
Post office box canonicalizer: the spelled-out prefix is missing · case 04
"post office box 101" is a street line.
Post office box canonicalizer: the spelled-out prefix is missing · case 05
"post office box 101" is a street line.
City-state-code last line parser: commas stay attached to city tokens · case 01
"FAKETOWN,ZZ 11101" is rejected and "HEIGHTS," keeps its comma.
City-state-code last line parser: commas stay attached to city tokens · case 02
"FAKETOWN,ZZ 11101" is rejected and "HEIGHTS," keeps its comma.
City-state-code last line parser: commas stay attached to city tokens · case 03
"FAKETOWN,ZZ 11101" is rejected and "HEIGHTS," keeps its comma.
City-state-code last line parser: commas stay attached to city tokens · case 04
"FAKETOWN,ZZ 11101" is rejected and "HEIGHTS," keeps its comma.
City-state-code last line parser: commas stay attached to city tokens · case 05
"FAKETOWN,ZZ 11101" is rejected and "HEIGHTS," keeps its comma.
City-state-code last line parser: any ten-digit token is taken as a code · case 01
"1110112345" is accepted as a postal code.
City-state-code last line parser: any ten-digit token is taken as a code · case 02
"1110112345" is accepted as a postal code.
City-state-code last line parser: any ten-digit token is taken as a code · case 03
"1110112345" is accepted as a postal code.
City-state-code last line parser: any ten-digit token is taken as a code · case 04
"1110112345" is accepted as a postal code.
City-state-code last line parser: any ten-digit token is taken as a code · case 05
"1110112345" is accepted as a postal code.
City-state-code last line parser: a last line without a city is accepted · case 01
"ZZ 11101" parses with an empty city.
City-state-code last line parser: a last line without a city is accepted · case 02
"ZZ 11101" parses with an empty city.
City-state-code last line parser: a last line without a city is accepted · case 03
"ZZ 11101" parses with an empty city.
City-state-code last line parser: a last line without a city is accepted · case 04
"ZZ 11101" parses with an empty city.
City-state-code last line parser: a last line without a city is accepted · case 05
"ZZ 11101" parses with an empty city.
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 ↗