use unidecode for clinvar

docs/content/history.rst

@@ -11,6 +11,7 @@ Release History
 #. Use SQLAlchemy for table definitions to support different RDBMS backends.
 #. Several optimizations to loading.
 #. X-linked recessive and dominant and de novo tools.
+#. Raise Exceptions rathern than sys.exit() to facilitate use as library. (thanks @brainstorm .
 
 0.18.3
 ======

gemini/annotations.py

@@ -604,7 +604,7 @@ def get_clinvar_info(var):
 
  raw_dbsource = info_map['CLNSRC'] or None
  #interpret 8-bit strings and convert to plain text
- clinvar.clinvar_dbsource = raw_dbsource.decode('utf8', 'ignore').encode('ascii', 'ignore')
+ clinvar.clinvar_dbsource = unidecode(raw_dbsource.decode('utf8'))
  clinvar.clinvar_dbsource_id = info_map['CLNSRCID'] or None
  clinvar.clinvar_origin = \
  clinvar.lookup_clinvar_origin(info_map['CLNORIGIN'])
@@ -614,7 +614,7 @@ def get_clinvar_info(var):
  clinvar.clinvar_dsdbid = info_map['CLNDSDBID'] or None
  # Remap all unicode characters into plain text string replacements
  raw_disease_name = info_map['CLNDBN'] or None
- clinvar.clinvar_disease_name = raw_disease_name.decode('utf8', 'ignore').encode('ascii', 'ignore')
+ clinvar.clinvar_disease_name = unidecode(raw_disease_name.decode('utf8')).decode('string_escape')
  # Clinvar represents commas as \x2c. Make them commas.
 
  clinvar.clinvar_disease_acc = info_map['CLNACC'] or None

master-test.sh

@@ -19,7 +19,7 @@ fi
 echo "Using gemini found at: $SCRIPT_PATH" 1>&2
 
 cd test
-#rm ./*.db
+rm ./*.db
 
 # setup the testing databases from the testing VCF files
 set -e