From 9322da0f79dfa4c3f9f899f5a861ce302ce21e9c Mon Sep 17 00:00:00 2001 From: Frederick Muriuki Muriithi Date: Wed, 10 Jan 2024 04:44:53 +0300 Subject: Make identifier column name explicit Since the R/qtl2 bundle generator could name the identifier column anything, this commit converts the incoming identifier column name into something explicit that we know and can use. --- r_qtl/r_qtl2.py | 30 +++++++++++++++++++++++++----- 1 file changed, 25 insertions(+), 5 deletions(-) (limited to 'r_qtl') diff --git a/r_qtl/r_qtl2.py b/r_qtl/r_qtl2.py index d3a3805..d8231bb 100644 --- a/r_qtl/r_qtl2.py +++ b/r_qtl/r_qtl2.py @@ -48,12 +48,26 @@ def with_non_transposed(zfile: ZipFile, def not_comment_line(line): return not line.startswith(cdata.get("comment.char", "#")) + sep = cdata.get("sep", ",") with zfile.open(cdata[member_key]) as innerfile: - reader = csv.DictReader( - filter(not_comment_line, io.TextIOWrapper(innerfile)), - delimiter=cdata.get("sep", ",")) + wrapped_file = io.TextIOWrapper(innerfile) + firstrow = tuple( + field.strip() for field in + next(filter(not_comment_line, wrapped_file)).strip().split(sep)) + id_key = firstrow[0] + wrapped_file.seek(0) + reader = csv.DictReader(filter(not_comment_line, wrapped_file), + delimiter=sep) for row in reader: - yield process_value(row) + processed = process_value(row) + yield { + "id": processed[id_key], + **{ + key: value + for key, value in processed.items() + if key != id_key + } + } def __make_organise_by_id__(id_key): """Return a function to use with `reduce` to organise values by some @@ -101,7 +115,13 @@ def with_transposed(zfile: ZipFile, for line in batch for row in process_value(id_key, headers, line)), {}).items(): - yield row + yield { + "id": row[id_key], + **{ + key: value + for key, value in row.items() + if key != id_key + }} except StopIteration: pass -- cgit v1.2.3