From 05901b6b30bf41d0d7064800622297bce5c729fe Mon Sep 17 00:00:00 2001
From: boasvdp <boasvdp@gmail.com>
Date: Wed, 23 Feb 2022 19:38:29 +0100
Subject: [PATCH] Add handling of empty files

---
 .github/workflows/main.yml |  2 ++
 extract_genes_abricate.py  | 13 ++++++++-----
 test/input/test_empty.tsv  |  1 +
 3 files changed, 11 insertions(+), 5 deletions(-)
 create mode 100644 test/input/test_empty.tsv

diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml
index b1b5063..de462c8 100644
--- a/.github/workflows/main.yml
+++ b/.github/workflows/main.yml
@@ -56,6 +56,8 @@ jobs:
          # Testing with flanking sequences which run off the contig end
          python extract_genes_abricate.py -a test/input/flanking.tsv -g test/genomes -o test/out_flanking_15bp --flanking --flanking-bp 15
          cmp test/correct/correct_flanking_15bp.out test/out_flanking_15bp/flanking_blaCTX-M-15.out
+         # Test with empty file
+         python extract_genes_abricate.py -a test/input/test_empty.tsv -g test/genomes -o test/out_empty
       - name: Tests with combined ABRicate files from multiple genomes
         run: |
          # Normal gene extraction should work for files with multiple genomes
diff --git a/extract_genes_abricate.py b/extract_genes_abricate.py
index 7a730c5..5b19467 100644
--- a/extract_genes_abricate.py
+++ b/extract_genes_abricate.py
@@ -220,12 +220,15 @@ def main(args):
   logging.debug(f"Calling function main")
   check_outdir(args.outdir)
   df = read_abricatefile(args.abricatefile, args.csv)
-  if args.genecluster == False:
-    logging.debug(f"Executing main_genes for extraction of single genes")
-    main_genes(df, args)
+  if df.shape[0] > 0:
+    if args.genecluster == False:
+      logging.debug(f"Executing main_genes for extraction of single genes")
+      main_genes(df, args)
+    else:
+      logging.debug(f"Executing main_genecluster for extraction of gene clusters")
+      main_genecluster(df, args)
   else:
-    logging.debug(f"Executing main_genecluster for extraction of gene clusters")
-    main_genecluster(df, args)
+    logging.warning(f"No hits found in {args.abricatefile}, outputting empty directory {args.outdir}")
 
 if __name__ == '__main__':
   parser = argparse.ArgumentParser(description='Extract genes from genes based on ABRicate output.')
diff --git a/test/input/test_empty.tsv b/test/input/test_empty.tsv
new file mode 100644
index 0000000..9423b57
--- /dev/null
+++ b/test/input/test_empty.tsv
@@ -0,0 +1 @@
+#FILE	SEQUENCE	START	END	STRAND	GENE	COVERAGE	COVERAGE_MAP	GAPS	%COVERAGE	%IDENTITY	DATABASE	ACCESSION	PRODUCT	RESISTANCE