Skip to content

Restore ISO-8859-1 encoding for accented country names - #10

Merged
raunaqmorarka merged 1 commit into
masterfrom
fix-country-encoding-iso-8859-1
Jun 23, 2026
Merged

Restore ISO-8859-1 encoding for accented country names#10
raunaqmorarka merged 1 commit into
masterfrom
fix-country-encoding-iso-8859-1

Conversation

@raunaqmorarka

@raunaqmorarka raunaqmorarka commented Jun 23, 2026

Copy link
Copy Markdown
Member

The official TPC-DS 4.0.0 answer sets emit accented country names in ISO-8859-1, and the library reads distribution files as ISO-8859-1 (DistributionUtils) and writes generated output as ISO-8859-1 (TableGenerator). Commit a1e81d5 re-encoded countries.dst to UTF-8, which the ISO-8859-1 reader then corrupted: CÔTE D'IVOIRE became CÃ"TE D'IVOIRE both in generated files and through the Trino connector path.

This reverts a1e81d5, restoring countries.dst to ISO-8859-1 (CÔTE D'IVOIRE = 0xD4, RÉUNION = 0xC9) and the corresponding TestCustomerGenerator MD5s. The bytes now match the official answer_sets/30.ans, which is the fidelity target.

@cla-bot cla-bot Bot added the cla-signed label Jun 23, 2026
The official TPC-DS 4.0.0 answer sets emit these names in ISO-8859-1 and
the library reads distribution files as ISO-8859-1, so a UTF-8 countries.dst
produced corrupted values (both in generated files and via the Trino connector).
@raunaqmorarka
raunaqmorarka force-pushed the fix-country-encoding-iso-8859-1 branch from 3b0cfee to 82ccf68 Compare June 23, 2026 06:44
@raunaqmorarka
raunaqmorarka requested a review from wendigo June 23, 2026 06:45
@raunaqmorarka
raunaqmorarka merged commit e41ef08 into master Jun 23, 2026
2 checks passed
@raunaqmorarka
raunaqmorarka deleted the fix-country-encoding-iso-8859-1 branch June 23, 2026 07:05
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Development

Successfully merging this pull request may close these issues.

2 participants