add checks for different accession types with browser API

python3/assemblyGet.py

@@ -39,6 +39,9 @@ def check_format(output_format):
         )
         sys.exit(1)
 
+def get_default_format():
+    return utils.EMBL_FORMAT
+
 def get_sequence_report_url(record):
     for link in record.iter('URL_LINK'):
         label = (link.find('LABEL').text)
@@ -137,8 +140,6 @@ def extract_wgs_scaffolds(assembly_dir, wgs_scaffolds, wgs_set, output_format, q
     target_file.close()
 
 def download_assembly(dest_dir, accession, output_format, fetch_wgs, extract_wgs, expanded, quiet=False):
-    if output_format is None:
-        output_format = utils.EMBL_FORMAT
     assembly_dir = os.path.join(dest_dir, accession)
     utils.create_dir(assembly_dir)
     # download xml

python3/enaDataGet.py

@@ -58,6 +58,11 @@ def set_parser():
     parser.add_argument('-v', '--version', action='version', version='%(prog)s 1.5.3')
     return parser
 
+def check_availability(accession, output_format):
+    if not utils.is_available(accession, output_format):
+        sys.stderr.write(
+        'ERROR: Record does not exist or is not available for accession provided\n')
+        sys.exit(1)
 
 if __name__ == '__main__':
     parser = set_parser()
@@ -81,26 +86,34 @@ def set_parser():
         if utils.is_wgs_set(accession):
             if output_format is not None:
                 sequenceGet.check_format(output_format)
+            else:
+                output_format = sequenceGet.get_default_format()
+            check_availability(accession, output_format)
             sequenceGet.download_wgs(dest_dir, accession, output_format)
-        elif not utils.is_available(accession, output_format):
-            sys.stderr.write('ERROR: Record does not exist or is not available for accession provided\n')
-            sys.exit(1)
+        elif utils.is_assembly(accession):
+            if output_format is not None:
+                assemblyGet.check_format(output_format)
+            else:
+                output_format = assemblyGet.get_default_format()
+            check_availability(accession, output_format)
+            assemblyGet.download_assembly(dest_dir, accession, output_format, fetch_wgs, extract_wgs, expanded)
         elif utils.is_sequence(accession):
             if output_format is not None:
                 sequenceGet.check_format(output_format)
+            else:
+                output_format = sequenceGet.get_default_format()
+            check_availability(accession, output_format)
             sequenceGet.download_sequence(dest_dir, accession, output_format, expanded)
         elif utils.is_analysis(accession):
             if output_format is not None:
                 readGet.check_read_format(output_format)
+            check_availability(accession, output_format)
             readGet.download_files(accession, output_format, dest_dir, fetch_index, fetch_meta, aspera)
         elif utils.is_run(accession) or utils.is_experiment(accession):
             if output_format is not None:
                 readGet.check_read_format(output_format)
+            check_availability(accession, output_format)
             readGet.download_files(accession, output_format, dest_dir, fetch_index, fetch_meta, aspera)
-        elif utils.is_assembly(accession):
-            if output_format is not None:
-                assemblyGet.check_format(output_format)
-            assemblyGet.download_assembly(dest_dir, accession, output_format, fetch_wgs, extract_wgs, expanded)
         else:
             sys.stderr.write('ERROR: Invalid accession provided\n')
             sys.exit(1)

python3/sequenceGet.py

@@ -31,8 +31,6 @@ def write_record(dest_file, accession, output_format, expanded=False):
     return utils.write_record(url, dest_file)
 
 def download_sequence(dest_dir, accession, output_format, expanded):
-    if output_format is None:
-        output_format = utils.EMBL_FORMAT
     success = utils.download_record(dest_dir, accession, output_format, expanded)
     if not success:
         print ('Unable to fetch file for {0}, format {1}'.format(accession, output_format))
@@ -45,8 +43,6 @@ def download_wgs(dest_dir, accession, output_format):
 
 def download_versioned_wgs(dest_dir, accession, output_format):
     prefix = accession[:6]
-    if output_format is None:
-        output_format = utils.EMBL_FORMAT
     public_set_url = utils.get_wgs_ftp_url(prefix, utils.PUBLIC, output_format)
     supp_set_url = utils.get_wgs_ftp_url(prefix, utils.SUPPRESSED, output_format)
     success = utils.get_ftp_file(public_set_url, dest_dir)
@@ -58,8 +54,6 @@ def download_versioned_wgs(dest_dir, accession, output_format):
 
 def download_unversioned_wgs(dest_dir, accession, output_format):
     prefix = accession[:4]
-    if output_format is None:
-        output_format = utils.EMBL_FORMAT
     public_set_url = utils.get_nonversioned_wgs_ftp_url(prefix, utils.PUBLIC, output_format)
     if public_set_url is not None:
         utils.get_ftp_file(public_set_url, dest_dir)
@@ -76,3 +70,6 @@ def check_format(output_format):
     if output_format not in allowed_formats:
         sys.stderr.write('Please select a valid format for this accession: {0}\n'.format(allowed_formats))
         sys.exit(1)
+
+def get_default_format():
+    return utils.EMBL_FORMAT

python3/utils.py

@@ -236,7 +236,7 @@ def get_record_url(accession, output_format):
 def is_available(accession, output_format):
     if is_taxid(accession):
         url = get_record_url('Taxon:{0}'.format(accession), XML_FORMAT)
-    elif is_study(accession) or is_sample(accession):
+    elif is_study(accession) or is_sample(accession) or is_assembly(accession):
         url = get_record_url(accession, XML_FORMAT)
     else:
         url = get_record_url(accession, output_format)
@@ -245,7 +245,7 @@ def is_available(accession, output_format):
     try:
         print('Checking availability of ' + url)
         response = requests.get(url)
-        return response.status_code == 200
+        return response.status_code == 200 and len(response.content) != 0
     except urlerror.URLError as e:
         if 'CERTIFICATE_VERIFY_FAILED' in str(e):
             print ('Error verifying SSL certificate. Have you run "Install Certificates" as part of your Python3 installation?')