Merge pull request #4410 from jmchilton/excel97

Add Excel97 Datatype

config/datatypes_conf.xml.sample

@@ -384,6 +384,7 @@
     <datatype extension="rdf" type="galaxy.datatypes.triples:Rdf" display_in_upload="true"/>
     <datatype extension="jsonld" type="galaxy.datatypes.triples:Jsonld" display_in_upload="true"/>
     <!-- Excel datatypes -->
+    <datatype extension="excel.xls" type="galaxy.datatypes.binary:ExcelXls" display_in_upload="true"/>
     <datatype extension="xlsx" type="galaxy.datatypes.binary:Xlsx" display_in_upload="true" />
     <datatype extension="btwisted" type="galaxy.datatypes.data:Text" subclass="true"/>
     <datatype extension="cai" type="galaxy.datatypes.data:Text" subclass="true"/>

lib/galaxy/datatypes/binary.py

@@ -1174,6 +1174,40 @@ def sniff( self, filename ):
 Binary.register_sniffable_binary_format("xlsx", "xlsx", Xlsx)
 
 
+class ExcelXls( Binary ):
+    """Class describing an Excel (xls) file"""
+    file_ext = "excel.xls"
+    edam_format = "format_3468"
+
+    def sniff( self, filename ):
+        mime_type = subprocess.check_output("file --mime-type '{}'".format(filename), shell=True).rstrip()
+        if mime_type.find("application/vnd.ms-excel") != -1:
+            return True
+        else:
+            return False
+
+    def get_mime( self ):
+        """Returns the mime type of the datatype"""
+        return 'application/vnd.ms-excel'
+
+    def set_peek( self, dataset, is_multi_byte=False ):
+        if not dataset.dataset.purged:
+            dataset.peek = "Microsoft Excel XLS file"
+            dataset.blurb = data.nice_size( dataset.get_size() )
+        else:
+            dataset.peek = 'file does not exist'
+            dataset.blurb = 'file purged from disk'
+
+    def display_peek( self, dataset ):
+        try:
+            return dataset.peek
+        except:
+            return "Microsoft Excel XLS file (%s)" % ( data.nice_size( dataset.get_size() ) )
+
+
+Binary.register_sniffable_binary_format("excel.xls", "excel.xls", ExcelXls)
+
+
 class Sra( Binary ):
     """ Sequence Read Archive (SRA) datatype originally from mdshw5/sra-tools-galaxy"""
     file_ext = 'sra'

lib/galaxy/datatypes/sniff.py

@@ -383,6 +383,9 @@ def guess_ext( fname, sniff_order, is_multi_byte=False ):
     >>> fname = get_test_fname('diamond_db.dmnd')
     >>> guess_ext(fname, sniff_order)
     'dmnd'
+    >>> fname = get_test_fname('1.xls')
+    >>> guess_ext(fname, sniff_order)
+    'excel.xls'
     """
     file_ext = None
     for datatype in sniff_order: