From 87ba0e2fae57764863af2c978a8878f92f043ab2 Mon Sep 17 00:00:00 2001 From: pi coding agent Date: Thu, 15 Oct 2026 20:44:47 +0000 Subject: Sort collected genotype symbols for deterministic output The genotype symbols were converted to a tuple straight from a set, so their order depended on the set's iteration order. That made the output (and the JSON emitted by the CLI) non-deterministic, and caused the multi-processed result to differ from the single-threaded result once symbols were merged across chunks. Sort the symbols so the output is stable and identical for both implementations. The set of symbols, which is what consumers care about, is unchanged. --- scripts/genotypes/preprocess_csv_collect_info.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'scripts') diff --git a/scripts/genotypes/preprocess_csv_collect_info.py b/scripts/genotypes/preprocess_csv_collect_info.py index 0fb1b96..b153279 100644 --- a/scripts/genotypes/preprocess_csv_collect_info.py +++ b/scripts/genotypes/preprocess_csv_collect_info.py @@ -350,7 +350,7 @@ def _as_file_stats( dot_geno_lines=stats.dot_geno_lines, dot_geno_fields=(DotGenoFields(**stats.dot_geno_fields) if bool(stats.dot_geno_fields) else None), - geno_symbols=tuple(stats.geno_symbols), + geno_symbols=tuple(sorted(stats.geno_symbols)), samples_list=tuple(samples), markers_list=stats.markers, errors=tuple(header_errors) + tuple(stats.errors)) -- cgit 1.4.1