Add gemini.sqlite datatype.

config/datatypes_conf.xml.sample

@@ -181,7 +181,8 @@
     <datatype extension="taxonomy" type="galaxy.datatypes.tabular:Taxonomy" display_in_upload="true"/>
     <datatype extension="tabular" type="galaxy.datatypes.tabular:Tabular" display_in_upload="true" description="Any data in tab delimited format (tabular)." description_url="https://wiki.galaxyproject.org/Learn/Datatypes#Tabular_.28tab_delimited.29"/>
     <datatype extension="twobit" type="galaxy.datatypes.binary:TwoBit" mimetype="application/octet-stream" display_in_upload="true"/>
-    <datatype extension="sqlite" type="galaxy.datatypes.binary:SQlite" mimetype="application/octet-stream" display_in_upload="true"/>    
+    <datatype extension="sqlite" type="galaxy.datatypes.binary:SQlite" mimetype="application/octet-stream" display_in_upload="true"/>
+    <datatype extension="gemini.sqlite" type="galaxy.datatypes.binary:GeminiSQLite" mimetype="application/octet-stream" display_in_upload="True" />
     <datatype extension="txt" type="galaxy.datatypes.data:Text" display_in_upload="true" description="Any text file." description_url="https://wiki.galaxyproject.org/Learn/Datatypes#Plain_text"/>
     <datatype extension="linecount" type="galaxy.datatypes.data:LineCount" display_in_upload="false"/>
     <datatype extension="memexml" type="galaxy.datatypes.xml:MEMEXml" mimetype="application/xml" display_in_upload="true"/>
@@ -271,6 +272,7 @@
     -->
     <sniffer type="galaxy.datatypes.tabular:Vcf"/>
     <sniffer type="galaxy.datatypes.binary:TwoBit"/>
+    <sniffer type="galaxy.datatypes.binary:GeminiSQLite"/>
     <sniffer type="galaxy.datatypes.binary:SQlite"/>
     <sniffer type="galaxy.datatypes.binary:Bam"/>
     <sniffer type="galaxy.datatypes.binary:Sff"/>

lib/galaxy/datatypes/binary.py

@@ -20,7 +20,7 @@ eggs.require( "bx-python" )
 from bx.seq.twobit import TWOBIT_MAGIC_NUMBER, TWOBIT_MAGIC_NUMBER_SWAP, TWOBIT_MAGIC_SIZE
 
 from galaxy.util import sqlite
-from galaxy.datatypes.metadata import MetadataElement,ListParameter,DictParameter
+from galaxy.datatypes.metadata import MetadataElement, MetadataParameter, ListParameter, DictParameter
 from galaxy.datatypes import metadata
 import dataproviders
 
@@ -640,8 +640,66 @@ class SQlite ( Binary ):
         return dataproviders.dataset.SQliteDataDictProvider( dataset_source, **settings )
 
 
+#Binary.register_sniffable_binary_format("sqlite", "sqlite", SQlite)
+
+
+class GeminiSQLite( SQlite ):
+    """Class describing a Gemini Sqlite database """
+    MetadataElement( name="gemini_version", default='0.10.0' , param=MetadataParameter, desc="Gemini Version", 
+                     readonly=True, visible=True, no_value='0.10.0' )
+    file_ext = "gemini.sqlite"
+
+    def set_meta( self, dataset, overwrite = True, **kwd ):
+        super( GeminiSQLite, self ).set_meta( dataset, overwrite = overwrite, **kwd )
+        try:
+            conn = sqlite.connect( dataset.file_name )
+            c = conn.cursor()
+            tables_query = "SELECT version FROM version"
+            result = c.execute( tables_query ).fetchall()
+            for version, in result:
+                dataset.metadata.gemini_version = version
+            # TODO: Can/should we detect even more attributes, such as use of PED file, what was input annotation type, etc.
+        except Exception, e:
+            log.warn( '%s, set_meta Exception: %s', self, e )
+
+    def sniff( self, filename ):
+        if super( GeminiSQLite, self ).sniff( filename ):
+            gemini_table_names = [ "gene_detailed", "gene_summary", "resources", "sample_genotype_counts", "sample_genotypes", "samples", 
+                                  "variant_impacts", "variants", "version" ]
+            try:
+                conn = sqlite.connect( filename )
+                c = conn.cursor()
+                tables_query = "SELECT name FROM sqlite_master WHERE type='table' ORDER BY name"
+                result = c.execute( tables_query ).fetchall()
+                result = map( lambda x: x[0], result )
+                for table_name in gemini_table_names:
+                    if table_name not in result:
+                        return False
+                return True
+            except Exception, e:
+                log.warn( '%s, sniff Exception: %s', self, e )
+        return False
+
+    def set_peek( self, dataset, is_multi_byte=False ):
+        if not dataset.dataset.purged:
+            dataset.peek  = "Gemini SQLite Database, version %s" % ( dataset.metadata.gemini_version or 'unknown' )
+            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 "Gemini SQLite Database, version %s" % ( dataset.metadata.gemini_version or 'unknown' )
+
+Binary.register_sniffable_binary_format( "gemini.sqlite", "gemini.sqlite", GeminiSQLite )
+# FIXME: We need to register gemini.sqlite before sqlite, since register_sniffable_binary_format and is_sniffable_binary called in upload.py
+# ignores sniff order declared in datatypes_conf.xml
 Binary.register_sniffable_binary_format("sqlite", "sqlite", SQlite)
 
+
 class Xlsx(Binary):
     """Class for Excel 2007 (xlsx) files"""
     file_ext="xlsx"

tools/data_source/upload.py

@@ -113,6 +113,8 @@ def add_file( dataset, registry, json_file, output_path ):
         ext = sniff.guess_ext( dataset.path, is_multi_byte=True )
     # Is dataset content supported sniffable binary?
     else:
+        # FIXME: This ignores the declared sniff order in datatype_conf.xml
+        # resulting in improper behavior
         type_info = Binary.is_sniffable_binary( dataset.path )
         if type_info:
             data_type = type_info[0]