changeset 0:bd5caa8e2290 draft

Uploaded

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/semweb_tools/README.md	Wed Nov 21 12:19:18 2012 -0500
@@ -0,0 +1,20 @@
+# Semantic Web Tools for Galaxy #
+
+## SPARQL Import tool ##
+This [Galaxy (bioinformatics web portal)](http://galaxyproject.org/) toolkit currently just contains a tool that enables to specify an URL to a SPARQL endpoint, and a SPARQL query, and to save the returned data in tabular format, for further processing in Galaxy's tools.
+
+## Installation ##
+
+1. Create a folder "semweb_tools" under [galaxy_root]/tools/ and add these files.
+2. Add the following section right under the starting <toolbox> tag in [galaxy_root]/tool_conf.xml:
+
+````
+    <section name="Semantic Web Tools" id="semweb_tools">
+        <tool file="semweb_tools/sparql_import.xml" />
+    </section>
+````
+
+## Change log ##
+
+* 2012-11-16: Now works properly (at least for my simple test case)!
+* 2012-11-16: Created today, still under heavy development!

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/semweb_tools/sparql_import.py	Wed Nov 21 12:19:18 2012 -0500
@@ -0,0 +1,99 @@
+#!/usr/bin/python
+# --------------------------------------------------------
+# A Galaxy plugin for querying external SPARQL Endpoints
+# Samuel Lampa, [e-mail]
+# Created: 2012-11-16
+# --------------------------------------------------------
+
+from xml.etree import ElementTree as et
+from optparse import OptionParser
+import urllib, sys, re
+
+# -----------------------
+# Option parsing
+# -----------------------
+
+parser = OptionParser()
+parser.add_option("-u", "--url",
+	help = "The URL to the SPARQL endpoint")
+parser.add_option("-q", "--sparql_query",
+	help = "A SPARQL query to send to a SPARQL endpoint")
+parser.add_option("-o", "--output_file",
+	help = "An output file for storing the results")
+(options, args) = parser.parse_args()
+
+if not options.url:
+	sys.exit("You have to specify an URL! Use the -h flag to view command line options!")
+if not options.sparql_query:
+	sys.exit("You have to specify a SPARQL query! Use the -h flag to view command line options!")
+if not options.output_file:
+	sys.exit("You have to specify an output file! Use the -h flag to view command line options!")
+
+if len(options.sparql_query) < 9: 
+	sys.exit("Your SPARQL query is too short (printed below)!\n" + options.sparql_query)
+
+if not re.match("^http", options.url):
+	sys.exit("The URL has to start with 'http://'! Please try again!")
+
+# -----------------------
+# The main code
+# -----------------------
+
+def main():
+	# Extract command line options
+	sparql_query = options.sparql_query
+	sparql_query = sparql_query.replace("__oc__","{")
+	sparql_query = sparql_query.replace("__ob__","[")
+	sparql_query = sparql_query.replace("__cc__","}")
+	sparql_query = sparql_query.replace("__cb__","]")
+	sparql_query = sparql_query.replace("__cr__"," ")
+	sparql_query = sparql_query.replace("__cn__"," ")
+	sparql_query = urllib.quote_plus(sparql_query)
+	url = options.url
+
+	output_file = options.output_file
+
+	# Create SPARQL query URL
+	sparql_query_url = url + "?query=" + sparql_query
+
+	# Read from SPARQL Endpoint
+	sparql_endpoint = urllib.urlopen(sparql_query_url)
+	results = sparql_endpoint.read()
+	sparql_endpoint.close()
+
+        # Convert to tabular format
+        if "<sparql" in results:
+                xmldata = extract_xml( results )
+                tabular = xml_to_tabular( xmldata )
+        else:
+                sys.exit("No SPARQL content found in returned data!\nReturned data:\n" + "-"*80 + "\n" + results)
+
+	# Print to file
+	of = open(output_file, "w")
+	of.write(tabular)
+	of.close()
+
+# -----------------------
+# Helper methods
+# -----------------------
+
+def extract_xml( content ):
+	'''Extract the part of the document starting with <?xml ...'''
+	xmlcontent = re.search("<sparql.*", content, re.DOTALL).group(0)
+	return xmlcontent
+
+def xml_to_tabular( xmldata ):
+	'''Convert SPARQL result set XML format to tabular text'''
+	root = et.fromstring(xmldata)
+	tree = et.ElementTree(root)
+	tabular = ""
+
+	results = root.getchildren()[1]
+	for result in results:
+		line_bits = [binding.getchildren()[0].text for binding in result.getchildren()]
+		line = "\t".join(line_bits)
+		tabular += line + "\n"	
+	return tabular
+
+if __name__ == '__main__':
+	main()

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/semweb_tools/sparql_import.xml	Wed Nov 21 12:19:18 2012 -0500
@@ -0,0 +1,44 @@
+<tool id="sparql_import" name="SPARQL Import" version="1.0.0">
+	<description>Import data from a SPARQL Endpoint</description>
+	<command interpreter="python">sparql_import.py -u $url -q "$sparqlquery" -o $outfile</command>
+	<inputs>
+		<param name="url" type="text" format="text" size="80" label="SPARQL Endpoint URL"/>
+		<param name="sparqlquery" type="text" format="text" area="True" size="5x80" label="SPARQL Query"/>
+	</inputs>
+	<outputs>
+		<data name="outfile" format="tabular"/>
+	</outputs>
+	<help>
+**What it does**
+
+The SPARQL import executes a SPARQL query against a SPARQL endpoint and returns the results in tabular format, with one column per variable in the SPARQL query.
+
+-------
+
+**Example**
+
+You could for example use this SPARQL Endpoint URL:
+
+http://sparql.wikipathways.org/
+
+and the following SPARQL query, which will select 25 sets of "subject", "predicate" and "object" out of the database, with no filtering at all::
+
+    SELECT ?s ?p ?o
+    WHERE { ?s ?p ?o } 
+    LIMIT 10
+
+This will generate a 3-column tabular output like this::
+
+    http://identifiers.org/pubmed/12860264  http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://vocabularies.wikipathways.org/wp#PublicationReference
+    http://identifiers.org/pubmed/15782111  http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://vocabularies.wikipathways.org/wp#PublicationReference
+    http://identifiers.org/pubmed/11017945  http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://vocabularies.wikipathways.org/wp#PublicationReference
+    http://identifiers.org/pubmed/10997684  http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://vocabularies.wikipathways.org/wp#PublicationReference
+    http://identifiers.org/pubmed/19635799  http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://vocabularies.wikipathways.org/wp#PublicationReference
+    http://identifiers.org/pubmed/16480962  http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://vocabularies.wikipathways.org/wp#PublicationReference
+    http://identifiers.org/pubmed/15976321  http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://vocabularies.wikipathways.org/wp#PublicationReference
+    http://identifiers.org/pubmed/17218259  http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://vocabularies.wikipathways.org/wp#PublicationReference
+    http://identifiers.org/pubmed/17215298  http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://vocabularies.wikipathways.org/wp#PublicationReference
+    http://identifiers.org/pubmed/11177571  http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://vocabularies.wikipathways.org/wp#PublicationReference
+
+</help>
+</tool>