Merge pull request #327 from bioinfo-chru-strasbourg/fix_output_empty

Fix empty file output #324

howard/objects/database.py

@@ -1809,11 +1809,21 @@ def get_columns(
                     "bed",
                     "json",
                 ]:
+                    # Query
                     sql_from = self.get_sql_from(
                         database=database, header_file=header_file
                     )
                     sql_query = f"SELECT * FROM {sql_from} LIMIT 0"
-                    return list(self.conn.query(sql_query).columns)
+
+                    # Get columns
+                    result_description = self.conn.execute(sql_query).description
+
+                    # Extract columns' names
+                    columns = [desc[0] for desc in result_description]
+
+                    # Return columns as list
+                    return columns
+
         except ValueError:
             return []
 
@@ -2557,8 +2567,10 @@ def export(
                     query_empty = False
                     break
                 if query_empty:
-                    log.error("Export failed: Empty")
-                    raise ValueError("Export failed: Empty")
+                    log.warning("Export warning: Empty")
+                    remove_header_line = False
+                else:
+                    remove_header_line = True
 
                 # Schema names
                 schema_names = None
@@ -2597,7 +2609,7 @@ def export(
                             query_output_header_tmp = os.path.join(tmp_dir, "header")
                             self.get_header_file(
                                 header_file=query_output_header_tmp,
-                                remove_header_line=True,
+                                remove_header_line=remove_header_line,
                                 sql_query=query,
                             )
 
@@ -2834,7 +2846,7 @@ def export(
                         query_output_header_tmp = os.path.join(tmp_dir, "header")
                         tmp_files.append(query_output_header_tmp)
                         self.get_header_file(
-                            header_file=query_output_header_tmp, remove_header_line=True
+                            header_file=query_output_header_tmp, remove_header_line=remove_header_line
                         )
 
                         # Add tmp header file for concat and compress

tests/data/example.empty.vcf

@@ -0,0 +1,55 @@
+##fileformat=VCFv4.1
+##FILTER=<ID=PASS,Description="All filters passed">
+##fileDate=20140624
+##source=./export.pl release 1.5
+##reference=IRC
+##phasing=unknown
+##INFO=<ID=NS,Number=1,Type=Integer,Description="Number of Samples With Data">
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">
+##INFO=<ID=AA,Number=1,Type=String,Description="Ancestral Allele">
+##FILTER=<ID=FSFilter,Description="FS > 200.0">
+##FILTER=<ID=LowQual,Description="Low quality">
+##FILTER=<ID=QDFilter,Description="QD < 2.0">
+##FILTER=<ID=ReadPosFilter,Description="ReadPosRankSum < -20.0">
+##FILTER=<ID=TruthSensitivityTranche99.00to99.90,Description="Truth sensitivity tranche level at VSQ Lod: -3.9813 <= x < 3.448">
+##FILTER=<ID=TruthSensitivityTranche99.90to100.00+,Description="Truth sensitivity tranche level at VQS Lod < -29586.8217">
+##FILTER=<ID=TruthSensitivityTranche99.90to100.00,Description="Truth sensitivity tranche level at VSQ Lod: -29586.8217 <= x < -3.9813">
+##FORMAT=<ID=HQ,Number=2,Type=Integer,Description="Haplotype Quality">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Float,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+##FORMAT=<ID=AD,Number=2,Type=Integer,Description="Allelic Depth">
+##FORMAT=<ID=FILTER,Number=.,Type=String,Description="Filter quality">
+##FORMAT=<ID=QUAL,Number=1,Type=Integer,Description="Filter quality value">
+##INFO=<ID=CLNSIG,Number=1,Type=String,Description="CLNSIG">
+##contig=<ID=chr1,length=249250621,assembly=hg19>
+##contig=<ID=chr7,length=159138663,assembly=hg19>
+##contig=<ID=1,length=249250621,assembly=hg19>
+##contig=<ID=10,length=135534747,assembly=hg19>
+##contig=<ID=11,length=135006516,assembly=hg19>
+##contig=<ID=12,length=133851895,assembly=hg19>
+##contig=<ID=13,length=115169878,assembly=hg19>
+##contig=<ID=14,length=107349540,assembly=hg19>
+##contig=<ID=15,length=102531392,assembly=hg19>
+##contig=<ID=16,length=90354753,assembly=hg19>
+##contig=<ID=17,length=81195210,assembly=hg19>
+##contig=<ID=18,length=78077248,assembly=hg19>
+##contig=<ID=19,length=59128983,assembly=hg19>
+##contig=<ID=2,length=243199373,assembly=hg19>
+##contig=<ID=20,length=63025520,assembly=hg19>
+##contig=<ID=21,length=48129895,assembly=hg19>
+##contig=<ID=22,length=51304566,assembly=hg19>
+##contig=<ID=3,length=198022430,assembly=hg19>
+##contig=<ID=4,length=191154276,assembly=hg19>
+##contig=<ID=5,length=180915260,assembly=hg19>
+##contig=<ID=6,length=171115067,assembly=hg19>
+##contig=<ID=7,length=159138663,assembly=hg19>
+##contig=<ID=8,length=146364022,assembly=hg19>
+##contig=<ID=9,length=141213431,assembly=hg19>
+##contig=<ID=M,length=16571,assembly=hg19>
+##contig=<ID=X,length=155270560,assembly=hg19>
+##contig=<ID=Y,length=59373566,assembly=hg19>
+##INFO=<ID=SIFT,Number=.,Type=String,Description="Annotation 'SIFT'">
+##bcftools_viewVersion=1.15.1+htslib-1.15.1
+##bcftools_viewCommand=view tests/data/example.vcf.gz; Date=Fri Mar 10 21:25:44 2023
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	sample1	sample2	sample3	sample4

tests/test_needed.py

@@ -106,6 +106,7 @@
     "example_vcf": tests_data_folder + "/example.vcf",
     "example_vcf_gz": tests_data_folder + "/example.vcf.gz",
     "example_vcf_gzip": tests_data_folder + "/example.vcf.gzip",
+    "example_empty_vcf": tests_data_folder + "/example.empty.vcf",
 }
 
 

tests/test_objects_database.py

@@ -20,6 +20,31 @@
 from test_needed import *
 
 
+def test_export_empty():
+    """
+    The function tests the export functionality of an empty database.
+    """
+
+    with TemporaryDirectory(dir=tests_folder) as tmp_dir:
+
+        tmp_dir = "/tmp"
+
+        # No database input
+        database = Database()
+        input_database = database_files.get("example_empty_vcf")
+        output_database = f"{tmp_dir}/output.tsv"
+
+        # Export database
+        database = Database(database=input_database)
+        database.export(output_database)
+        assert os.path.exists(output_database)
+
+        # Check if exported database is empty
+        database = Database(database=output_database)
+        results = database.query(query=f"""SELECT * FROM variants""")
+        assert len(results) == 0
+
+
 @pytest.mark.parametrize(
     "order_by, first_pos, first_qual, first_alt",
     [

tests/test_tools_query.py

@@ -53,18 +53,14 @@ def test_query_empty():
         log.debug(result)
         assert len(result) == 0
 
-        # Check if VCF is in correct format with pyVCF
-        with pytest.raises(ValueError) as e:
-            variants.export_output(query=query)
-        assert str(e.value) == "Export failed: Empty"
-
-        # Set output
-        variants.set_output = output_parquet
-
-        # Check if VCF is in correct format with pyVCF
-        with pytest.raises(ValueError) as e:
-            variants.export_output(query=query)
-        assert str(e.value) == "Export failed: Empty"
+        # Export file
+        variants.export_output(query=query)
+        assert os.path.exists(output_vcf)
+
+        # Check if exported file is empty
+        variants_output_vcf = Variants(conn=None, input=output_vcf, load=True)
+        results = variants_output_vcf.get_query_to_df(query=f"""SELECT * FROM variants""")
+        assert len(results) == 0
 
 
 def test_query():