From 4321f24b387ecce8838e8946ecc2d633e4f43006 Mon Sep 17 00:00:00 2001
From: helmutm <helmutm@fd906abe-77d9-0310-91a1-e0d9ade77398>
Date: Mon, 8 Oct 2007 11:58:24 +0000
Subject: [PATCH] more on rstat.getDataFrame()

git-svn-id: svn://svn.cy55.de/Zope3/src/cybertools/trunk@2099 fd906abe-77d9-0310-91a1-e0d9ade77398
---
 pyscript/rstat.py  | 28 ++++++++++++++++++++--------
 pyscript/rstat.txt |  3 ++-
 2 files changed, 22 insertions(+), 9 deletions(-)

diff --git a/pyscript/rstat.py b/pyscript/rstat.py
index 44a86de..c5f95d0 100644
--- a/pyscript/rstat.py
+++ b/pyscript/rstat.py
@@ -79,25 +79,37 @@ class RStat(object):
             (rowId, columnId, value). Elements with a columnId
             that is not present in all rows is omitted.
         """
-        def checkColumnId(rows, columnId):
-            for row in rows.values():
-                if columnId not in row:
+        def checkId(mapping, id):
+            for element in mapping.values():
+                if id not in element:
                     return False
             return True
         data = sorted(data)
         rows = {}
+        columns = {}
+        dataMapping = {}
         for rowId, columnId, value in data:
-            element = rows.setdefault(rowId, [])
-            element.append(columnId)
+            rows.setdefault(rowId, []).append(columnId)
+            columns.setdefault(columnId, []).append(rowId)
+            dataMapping[(rowId, columnId)] = value
         columnsToOmit = []
         for rowId, row in rows.items():
             for columnId in row:
-                if not checkColumnId(rows, columnId):
+                if not checkId(rows, columnId):
                     columnsToOmit.append(columnId)
+        rowsToOmit = []
+        for columnId, column in columns.items():
+            for rowId in column:
+                if not checkId(columns, rowId):
+                    rowsToOmit.append(rowId)
         r.library('ltm')
         result = {}
-        for rowId, columnId, value in data:
-            if columnId not in columnsToOmit:
+        #for rowId, columnId, value in data:
+            #if columnId not in columnsToOmit:
+        #    if rowId not in rowsToOmit:
+        #        result.setdefault(rowId, []).append(value)
+        for (rowId, columnId), value in sorted(dataMapping.items()):
+            if rowId not in rowsToOmit:
                 result.setdefault(rowId, []).append(value)
         self.intermediateData = result
         matrix = rpy.with_mode(rpy.NO_CONVERSION, r.data_frame)(**result)
diff --git a/pyscript/rstat.txt b/pyscript/rstat.txt
index 2b2e7d7..da6c6d9 100644
--- a/pyscript/rstat.txt
+++ b/pyscript/rstat.txt
@@ -24,9 +24,10 @@ Using the R Statistics Package with Python Scripts
   ...         ('r3', 'c1', 0),
   ...         ('r3', 'c2', 1),
   ... ]
+  ... data = [('q050', 'user1', 0), ('q044', 'user1', 1), ('q039', 'user1', 1), ('q041', 'user1', 0), ('q036', 'user1', 0), ('q053', 'user1', 0), ('q049', 'user1', 1), ('q057', 'user1', 0), ('q059', 'user1', 0), ('q060', 'user1', 0), ('q056', 'user1', 1), ('q047', 'user1', 1), ('q057', 'user1', 0), ('q041', 'user1', 1), ('q039', 'user1', 1), ('q038', 'user1', 1), ('q046', 'user1', 1), ('q040', 'user1', 1), ('q056', 'user1', 0), ('q059', 'user10', 0), ('q050', 'user10', 0), ('q058', 'user10', 0), ('q040', 'user10', 0), ('q052', 'user10', 0), ('q055', 'user10', 1), ('q039', 'user10', 1), ('q057', 'user10', 0), ('q049', 'user10', 1), ('q051', 'user10', 1), ('q041', 'user10', 1), ('q047', 'user10', 1), ('q038', 'user10', 1), ('q056', 'user10', 0), ('q037', 'user10', 1), ('q055', 'user10', 1), ('q054', 'user10', 0), ('q041', 'user10', 1), ('q044', 'user10', 1), ('q049', 'user10', 1), ('q048', 'user10', 1), ('q061', 'user10', 0), ('q056', 'user10', 1), ('q058', 'user10', 0), ('q042', 'user10', 0), ('q041', 'user10', 1), ('q036', 'user10', 1), ('q040', 'user10', 1), ('q059', 'user10', 1), ('q058', 'user10', 1), ('q057', 'user10', 0), ('q054', 'user10', 1), ('q037', 'user1', 1), ('q038', 'user1', 1), ('q047', 'user1', 1), ('q041', 'user1', 1), ('q051', 'user1', 1), ('q048', 'user1', 1), ('q058', 'user1', 1), ('q060', 'user1', 1)]
   ... matrix = rstat.getDataFrame(data)
   ... #return matrix
-  ... return rstat.intermediateData
+  ... #return rstat.intermediateData
   ... x = rpy.with_mode(rpy.BASIC_CONVERSION, r.descript)(matrix)
   ... return x
   ... """