From 7039ac81914edd84b50792cb340050d7c4f46a6d Mon Sep 17 00:00:00 2001
From: Julian Dolby <juliandolby@gmail.com>
Date: Mon, 9 Jul 2018 22:13:23 -0400
Subject: [PATCH] trivial start of pandas model

---
 .../cast/python/ml/test/TestPandasModel.java  | 19 +++++++++++
 com.ibm.wala.cast.python.ml/data/pandas.xml   | 34 +++++++++++++++++++
 .../ml/client/PythonTensorAnalysisEngine.java |  9 +++--
 com.ibm.wala.cast.python.test/.project        |  6 ++++
 com.ibm.wala.cast.python.test/data/pandas1.py |  8 +++++
 5 files changed, 73 insertions(+), 3 deletions(-)
 create mode 100644 com.ibm.wala.cast.python.ml.test/source/com/ibm/wala/cast/python/ml/test/TestPandasModel.java
 create mode 100644 com.ibm.wala.cast.python.ml/data/pandas.xml
 create mode 100644 com.ibm.wala.cast.python.test/data/pandas1.py

diff --git a/com.ibm.wala.cast.python.ml.test/source/com/ibm/wala/cast/python/ml/test/TestPandasModel.java b/com.ibm.wala.cast.python.ml.test/source/com/ibm/wala/cast/python/ml/test/TestPandasModel.java
new file mode 100644
index 000000000..dc9e4f994
--- /dev/null
+++ b/com.ibm.wala.cast.python.ml.test/source/com/ibm/wala/cast/python/ml/test/TestPandasModel.java
@@ -0,0 +1,19 @@
+package com.ibm.wala.cast.python.ml.test;
+
+import java.io.IOException;
+
+import org.junit.Test;
+
+import com.ibm.wala.ipa.callgraph.CallGraph;
+import com.ibm.wala.ipa.cha.ClassHierarchyException;
+import com.ibm.wala.util.CancelException;
+
+public class TestPandasModel extends TestPythonMLCallGraphShape {
+
+	@Test
+	public void testPandas1() throws ClassHierarchyException, IllegalArgumentException, CancelException, IOException {
+		CallGraph CG = process("pandas1.py");
+		System.err.println(CG);
+	}
+
+}
diff --git a/com.ibm.wala.cast.python.ml/data/pandas.xml b/com.ibm.wala.cast.python.ml/data/pandas.xml
new file mode 100644
index 000000000..2028c4988
--- /dev/null
+++ b/com.ibm.wala.cast.python.ml/data/pandas.xml
@@ -0,0 +1,34 @@
+<?xml version="1.0" ?>
+<!DOCTYPE summary-spec>
+<!-- Pandas model -->
+<summary-spec>
+  <classloader name="PythonLoader">
+    <class name="pandas" allocatable="true">
+      <method name="import"
+	      static="true"
+	      descriptor="()Lpandas;">
+        <new def="x" class="Lpandas"/>
+
+	<new def="read_excel" class="Lpandas/functions/read_excel"/>
+	<putfield class="LRoot"
+                  field="read_excel"
+                  fieldType="LRoot"
+                  ref="x"
+                  value="read_excel"/>
+
+	<return value="x"/>
+      </method>
+    </class>
+
+    <package name="pandas/functions">
+
+      <class name="read_excel" allocatable="true">
+	<method name="do" descriptor="()LRoot;" numArgs="3" paramNames="self file sheet">
+	  <new def="v" class="Lobject"/>
+	  <return vaue="v"/>
+	</method>
+      </class>
+      
+    </package>
+  </classloader>
+</summary-spec>
diff --git a/com.ibm.wala.cast.python.ml/source/com/ibm/wala/cast/python/ml/client/PythonTensorAnalysisEngine.java b/com.ibm.wala.cast.python.ml/source/com/ibm/wala/cast/python/ml/client/PythonTensorAnalysisEngine.java
index bcfd5cf67..c25268ce1 100644
--- a/com.ibm.wala.cast.python.ml/source/com/ibm/wala/cast/python/ml/client/PythonTensorAnalysisEngine.java
+++ b/com.ibm.wala.cast.python.ml/source/com/ibm/wala/cast/python/ml/client/PythonTensorAnalysisEngine.java
@@ -186,10 +186,13 @@ public Map<PointerKey, AnalysisError> getErrors() {
 	
 	protected void addBypassLogic(AnalysisOptions options) {
 		super.addBypassLogic(options);
-		
+		addSummaryBypassLogic(options, "tensorflow.xml");
+		addSummaryBypassLogic(options, "pandas.xml");
+	}
+
+	private void addSummaryBypassLogic(AnalysisOptions options, String summary) {
 		IClassHierarchy cha = getClassHierarchy();
-		
-		XMLMethodSummaryReader xml = new XMLMethodSummaryReader(getClass().getClassLoader().getResourceAsStream("tensorflow.xml"), scope);
+		XMLMethodSummaryReader xml = new XMLMethodSummaryReader(getClass().getClassLoader().getResourceAsStream(summary), scope);
 		for(TypeReference t : xml.getAllocatableClasses()) {
 			BypassSyntheticClassLoader ldr = (BypassSyntheticClassLoader) cha.getLoader(scope.getSyntheticLoader());
 			ldr.registerClass(t.getName(), new SyntheticClass(t, cha) {
diff --git a/com.ibm.wala.cast.python.test/.project b/com.ibm.wala.cast.python.test/.project
index 9ab89339d..125dd967c 100644
--- a/com.ibm.wala.cast.python.test/.project
+++ b/com.ibm.wala.cast.python.test/.project
@@ -5,6 +5,11 @@
 	<projects>
 	</projects>
 	<buildSpec>
+		<buildCommand>
+			<name>org.python.pydev.PyDevBuilder</name>
+			<arguments>
+			</arguments>
+		</buildCommand>
 		<buildCommand>
 			<name>org.eclipse.jdt.core.javabuilder</name>
 			<arguments>
@@ -30,5 +35,6 @@
 		<nature>org.eclipse.m2e.core.maven2Nature</nature>
 		<nature>org.eclipse.pde.PluginNature</nature>
 		<nature>org.eclipse.jdt.core.javanature</nature>
+		<nature>org.python.pydev.pythonNature</nature>
 	</natures>
 </projectDescription>
diff --git a/com.ibm.wala.cast.python.test/data/pandas1.py b/com.ibm.wala.cast.python.test/data/pandas1.py
new file mode 100644
index 000000000..ce24d2463
--- /dev/null
+++ b/com.ibm.wala.cast.python.test/data/pandas1.py
@@ -0,0 +1,8 @@
+import pandas as pd
+
+dfqol = pd.read_excel("Master_Data_Sheet_06252018.xlsx", sheetname="EQ5D")
+dfdemog = pd.read_excel("Master_Data_Sheet_06252018.xlsx", sheetname="Patient_Info_Demographics", skiprows = 1)
+
+# format participant IDs uniformly across each dataframe
+dfqol['PID'] = dfqol['PID'].str.replace("'",'')
+dfdemog['PID'] = dfdemog['Patient_ID'].str.replace('-','')