Package: rdkit / 201203-3

big-endian_support.patch Patch series | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
Description: Upstream changes to support big-endian architectures
 This patch consists of upstream revisions 2071, 2079, 2080 and 2140.
Author: Greg Landrum
Origin: upstream
Bug-Debian: http://bugs.debian.org/670054
Last-Update: 2012-07-28

--- rdkit-201203.orig/CMakeLists.txt
+++ rdkit-201203/CMakeLists.txt
@@ -26,6 +26,9 @@ option(RDK_INSTALL_STATIC_LIBS "install
 option(RDK_BUILD_SLN_SUPPORT "include support for the SLN format" ON )
 option(RDK_TEST_MULTITHREADED "run some tests of multithreading" OFF )
 
+include(TestBigEndian)
+TEST_BIG_ENDIAN(RDK_BIG_ENDIAN)
+
 # At build time put runtime binaries in the bin subdirectory
 set(RDK_RUNTIME_OUTPUT_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}/bin")
 set(RDK_LIBRARY_OUTPUT_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}/lib")
--- rdkit-201203.orig/Code/CMakeLists.txt
+++ rdkit-201203/Code/CMakeLists.txt
@@ -15,7 +15,11 @@ add_subdirectory(GraphMol)
 add_subdirectory(Query)
 
 add_subdirectory(DataManip)
-add_subdirectory(SimDivPickers)
+if(RDK_BIG_ENDIAN)
+  message("Skipping build of SimDivPickers on big endian system")
+else(RDK_BIG_ENDIAN)
+  add_subdirectory(SimDivPickers)
+endif(RDK_BIG_ENDIAN)
 
 add_subdirectory(ML)
 add_subdirectory(ChemicalFeatures)
--- rdkit-201203.orig/Code/DataStructs/BitVect.cpp
+++ rdkit-201203/Code/DataStructs/BitVect.cpp
@@ -40,7 +40,7 @@ void BitVect::initFromText(const char *d
   
   // earlier versions of the code did not have the version number encoded, so
   //  we'll use that to distinguish version 0
-  ss.read((char *)&size,sizeof(size));
+  RDKit::streamRead(ss,size);
   if(size<0){
     version = -1*size;
     if (version == 16) {
@@ -52,12 +52,12 @@ void BitVect::initFromText(const char *d
     else {
       throw ValueErrorException("bad version in BitVect pickle");
     }
-
-    ss.read((char *)&size,sizeof(size));
+    RDKit::streamRead(ss,size);
   } else if( !allowOldFormat ) {
     throw ValueErrorException("invalid BitVect pickle");
   }
-  ss.read((char *)&nOn,sizeof(nOn));
+
+  RDKit::streamRead(ss,nOn);
   _initForSize(static_cast<int>(size));
 
   // if the either have older version or or version 16 with ints for on bits
@@ -65,13 +65,13 @@ void BitVect::initFromText(const char *d
       ( (format == 1) && (size >= std::numeric_limits<unsigned short>::max()) ) ) {
     boost::uint32_t tmp;
     for(unsigned int i=0; i<nOn; i++){
-      ss.read((char *)&tmp,sizeof(tmp));
+      RDKit::streamRead(ss,tmp);
       setBit(tmp);
     }
   } else if (format == 1) { // version 16 and on bits stored as short ints
     boost::uint16_t tmp;
     for(unsigned int i=0; i<nOn; i++){
-      ss.read((char *)&tmp,sizeof(tmp));
+      RDKit::streamRead(ss,tmp);
       setBit(tmp);
     }
   } else if (format == 2) { // run length encoded format
--- rdkit-201203.orig/Code/DataStructs/BitOps.cpp
+++ rdkit-201203/Code/DataStructs/BitOps.cpp
@@ -21,18 +21,20 @@
 
 #include <boost/lexical_cast.hpp>
 
+using namespace RDKit;
+
 int getBitId(const char *&text,int format,int size,int curr){
   PRECONDITION(text,"no text");
   int res=-1;
   if( (format==0) || 
       ( (format == 1) && (size >= std::numeric_limits<unsigned short>::max()) ) ) {
     int tmp;
-    tmp = *(int *)text;
+    tmp = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(*(int *)text);
     text += sizeof(tmp);
     res=tmp;
   } else if (format == 1) { // version 16 and on bits sotred as short ints
     unsigned short tmp;
-    tmp = *(unsigned short *)text;
+    tmp = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(*(unsigned short *)text);
     text += sizeof(tmp);
     res=tmp;
   } else if (format == 2) { // run length encoded format
@@ -54,7 +56,7 @@ bool AllProbeBitsMatch(const char *probe
   int refFormat=0;
   int version=0;
 
-  int probeSize = *(int *)probe;
+  int probeSize = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(*(int *)probe);
   probe+=sizeof(probeSize);
   if(probeSize<0){
     version = -1*probeSize;
@@ -67,11 +69,11 @@ bool AllProbeBitsMatch(const char *probe
     else {
       throw("Unknown version type for the encode bit vect");
     }
-    probeSize = *(int *)probe;
+    probeSize = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(*(int *)probe);
     probe+=sizeof(probeSize);
   }
 
-  int refSize = *(int *)ref;
+  int refSize = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(*(int *)ref);
   ref+=sizeof(refSize);
   if(refSize<0){
     version = -1*refSize;
@@ -84,14 +86,14 @@ bool AllProbeBitsMatch(const char *probe
     else {
       throw("Unknown version type for the encode bit vect");
     }
-    refSize = *(int *)ref;
+    refSize = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(*(int *)ref);
     ref+=sizeof(refSize);
   }
 
 
-  int nProbeOn = *(int *)probe;
+  int nProbeOn = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(*(int *)probe);
   probe+=sizeof(nProbeOn);
-  int nRefOn = *(int *)ref;
+  int nRefOn = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(*(int *)ref);
   ref+=sizeof(nRefOn);
 
   int currProbeBit=0;
@@ -121,7 +123,7 @@ bool AllProbeBitsMatch(const T1 &probe,c
   const char *text=pkl.c_str();
   int format=0;
   int nOn=0,size,version=0;
-  size = *(int *)text;
+  size = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(*(int *)text);
   text+=sizeof(size);
   if(size<0){
     version = -1*size;
@@ -134,10 +136,10 @@ bool AllProbeBitsMatch(const T1 &probe,c
     else {
       throw("Unknown version type for the encode bit vect");
     }
-    size = *(int *)text;
+    size = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(*(int *)text);
     text+=sizeof(size);
   }
-  nOn = *(int *)text;
+  nOn = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(*(int *)text);
   text+=sizeof(nOn);
 
   int currBit=0;
--- rdkit-201203.orig/Code/DataStructs/DiscreteValueVect.cpp
+++ rdkit-201203/Code/DataStructs/DiscreteValueVect.cpp
@@ -1,6 +1,6 @@
 // $Id: DiscreteValueVect.cpp 1528 2010-09-26 17:04:37Z glandrum $
 //
-//  Copyright (C) 2004-2008 Greg Landrum and Rational Discovery LLC
+//  Copyright (C) 2004-2012 Greg Landrum and Rational Discovery LLC
 //
 //  @@ All Rights Reserved @@
 //  This file is part of the RDKit.
@@ -143,8 +143,15 @@ namespace RDKit {
     streamWrite(ss,tInt);
     tInt=d_numInts;
     streamWrite(ss,tInt);
+
+#if defined(BOOST_BIG_ENDIAN)
+    boost::uint32_t *td = new boost::uint32_t[d_numInts];
+    for(unsigned int i=0;i<d_numInts;++i) td[i]=EndianSwapBytes<HOST_ENDIAN_ORDER,LITTLE_ENDIAN_ORDER>(d_data.get()[i]);
+    ss.write((const char *)td,d_numInts*sizeof(tInt));
+    delete [] td;
+#else    
     ss.write((const char *)d_data.get(),d_numInts*sizeof(tInt));
-    
+#endif    
     std::string res(ss.str());
     return res;
   };
@@ -175,7 +182,15 @@ namespace RDKit {
     d_numInts=tInt;
     boost::uint32_t *data = new boost::uint32_t[d_numInts];
     ss.read((char *)data,d_numInts*sizeof(boost::uint32_t));
+
+#if defined(BOOST_BIG_ENDIAN)
+    boost::uint32_t *td = new boost::uint32_t[d_numInts];
+    for(unsigned int i=0;i<d_numInts;++i) td[i]=EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(data[i]);
+    d_data.reset(td);
+    delete [] data;
+#else    
     d_data.reset(data);
+#endif    
 
   };
 
--- rdkit-201203.orig/Code/DataStructs/testDatastructs.cpp
+++ rdkit-201203/Code/DataStructs/testDatastructs.cpp
@@ -125,9 +125,13 @@ template<typename T> void Test(T arg){
   delete t5;
 
   std::string pkl=t1.toString();
+  BOOST_LOG(rdInfoLog) << " a" << std::endl;
   const char *pkl64=Base64Encode(pkl.c_str(),pkl.size());
+  BOOST_LOG(rdInfoLog) << " b" << std::endl;
   T t6(t1.getNumBits());
+  BOOST_LOG(rdInfoLog) << " c" << std::endl;
   t6.initFromText(pkl64,strlen(pkl64),true);
+  BOOST_LOG(rdInfoLog) << " d" << std::endl;
   delete [] pkl64;
   TEST_ASSERT(t6==t1);
 }
@@ -149,10 +153,14 @@ template<typename T> void TaniTest(T &ar
   int idx=0;
   for(int i=0;i<4;i++){
     T v1(256);
+    BOOST_LOG(rdInfoLog) << " a "<<i << std::endl;
     FromDaylightString(v1,fps[i]);
+    BOOST_LOG(rdInfoLog) << " b"<<i << std::endl;
     for(int j=i;j<4;j++){
       T v2(256);
+      BOOST_LOG(rdInfoLog) << "     c " << j << std::endl;
       FromDaylightString(v2,fps[j]);
+      BOOST_LOG(rdInfoLog) << "     d " << j << std::endl;
       double tani=TanimotoSimilarity(v1,v2);
       TEST_ASSERT(feq(tani,dists[idx]));
       tani = TverskySimilarity(v1,v2,1.,1.);
@@ -1174,15 +1182,9 @@ int main(){
 
   ss.write((const char *)&v1,sizeof(v1));
   ss.write((const char *)&v2,sizeof(v2));
-#if 0
-  ss.close();
-  fstream ss2("blah.bin",ios_base::binary|ios_base::in);
-  ss2.read((char *)&v3,sizeof(v3));
-  ss2.read((char *)&v4,sizeof(v4));
-#endif
   ss.seekp(0,ios_base::beg);
-  ss.read((char *)&v3,sizeof(v3));
-  ss.read((char *)&v4,sizeof(v4));
+  RDKit::streamRead(ss,v3);
+  RDKit::streamRead(ss,v4);
   
   TXTMSG("v3",v3);
   TXTMSG("v4",v4);
@@ -1190,12 +1192,16 @@ int main(){
   BOOST_LOG(rdInfoLog) << " SPARSE -----------------------------------" << std::endl;
   SparseBitVect sparseFoo(10);
   Test(sparseFoo);
+  BOOST_LOG(rdInfoLog) << " 1" << std::endl;
   TaniTest(sparseFoo);
+  BOOST_LOG(rdInfoLog) << " 2" << std::endl;
   ProbeTest(sparseFoo);
   BOOST_LOG(rdInfoLog) << " Explicit ----------------------------------" << std::endl;
   ExplicitBitVect explicitFoo(10);
   Test(explicitFoo);
+  BOOST_LOG(rdInfoLog) << " 1" << std::endl;
   TaniTest(explicitFoo);
+  BOOST_LOG(rdInfoLog) << " 2" << std::endl;
   BOOST_LOG(rdInfoLog) << " Done" << std::endl;
   
   BOOST_LOG(rdInfoLog) << " Test DiscreteValue Vectors 1 ----------------------------" << endl;
--- rdkit-201203.orig/Code/DataStructs/SparseBitVect.cpp
+++ rdkit-201203/Code/DataStructs/SparseBitVect.cpp
@@ -317,10 +317,15 @@ std::string SparseBitVect::toString() co
 
   boost::int32_t tInt = ci_BITVECT_VERSION*-1;
   RDKit::streamWrite(ss,tInt);
+std::cerr<<"version: "<<tInt<<std::endl;
   tInt=d_size;
   RDKit::streamWrite(ss,tInt);
+std::cerr<<"size: "<<tInt<<std::endl;
+
   tInt=getNumOnBits();
   RDKit::streamWrite(ss,tInt);
+std::cerr<<"nOn: "<<tInt<<std::endl;
+
 
   int prev = -1;
   unsigned int zeroes;
--- rdkit-201203.orig/Code/RDGeneral/StreamOps.h
+++ rdkit-201203/Code/RDGeneral/StreamOps.h
@@ -16,10 +16,72 @@
 #include <sstream>
 #include <iostream>
 #include <boost/cstdint.hpp>
-
+#include <boost/detail/endian.hpp>
 
 namespace RDKit{
-    
+  // this code block for handling endian problems is from :
+  // http://stackoverflow.com/questions/105252/how-do-i-convert-between-big-endian-and-little-endian-values-in-c
+  enum EEndian
+    {
+      LITTLE_ENDIAN_ORDER,
+      BIG_ENDIAN_ORDER,
+#if defined(BOOST_LITTLE_ENDIAN)
+      HOST_ENDIAN_ORDER = LITTLE_ENDIAN_ORDER
+#elif defined(BOOST_BIG_ENDIAN)
+      HOST_ENDIAN_ORDER = BIG_ENDIAN_ORDER
+#else
+#error "Failed to determine the system endian value"
+#endif
+    };
+
+  // this function swap the bytes of values given it's size as a template
+  // parameter (could sizeof be used?).
+  template <class T, unsigned int size>
+  inline T SwapBytes(T value)
+  {
+    union
+    {
+      T value;
+      char bytes[size];
+    } in, out;
+
+    in.value = value;
+
+    for (unsigned int i = 0; i < size / 2; ++i)
+      {
+        out.bytes[i] = in.bytes[size - 1 - i];
+        out.bytes[size - 1 - i] = in.bytes[i];
+      }
+
+    return out.value;
+  }
+
+  // Here is the function you will use. Again there is two compile-time assertion
+  // that use the boost librarie. You could probably comment them out, but if you
+  // do be cautious not to use this function for anything else than integers
+  // types. This function need to be calles like this :
+  //
+  //     int x = someValue;
+  //     int i = EndianSwapBytes<HOST_ENDIAN_ORDER, BIG_ENDIAN_ORDER>(x);
+  //
+  template<EEndian from, EEndian to, class T>
+  inline T EndianSwapBytes(T value)
+  {
+    // A : La donnée à swapper à une taille de 2, 4 ou 8 octets
+    BOOST_STATIC_ASSERT(sizeof(T)==1 || sizeof(T) == 2 || sizeof(T) == 4 || sizeof(T) == 8);
+    if(sizeof(T)==1) return value;
+
+    // A : La donnée à swapper est d'un type arithmetic
+    //BOOST_STATIC_ASSERT(boost::is_arithmetic<T>::value);
+
+    // Si from et to sont du même type on ne swap pas.
+    if (from == to)
+      return value;
+
+    return SwapBytes<T, sizeof(T)>(value);
+  }
+  // --------------------------------------
+  
   //! Packs an integer and outputs it to a stream
   inline void appendPackedIntToStream(std::stringstream &ss, boost::uint32_t num) {
     int nbytes, bix;
@@ -27,7 +89,7 @@ namespace RDKit{
     char tc;
     
     CHECK_INVARIANT(num >= 0, "");
-    res = num;
+    res=num;
     while (1) {
       if (res < (1<<7)) {
         val = (res<<1);
@@ -53,9 +115,10 @@ namespace RDKit{
         break;
       }
       else {
-        CHECK_INVARIANT(0, "ERROR: Integer to big to pack\n");
+        CHECK_INVARIANT(0, "ERROR: Integer too big to pack\n");
       }
     }
+    //val = EndianSwapBytes<HOST_ENDIAN_ORDER,LITTLE_ENDIAN_ORDER>(val);
     
     for (bix = 0; bix < nbytes; bix++) {
       tc = (char) (val & 255);
@@ -100,6 +163,7 @@ namespace RDKit{
       offset = (1<<7) + (1<<14) + (1<<21);
     }
     num = (val >> shift) + offset;
+    //num = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(num);
     return num;
   }
 
@@ -147,18 +211,22 @@ namespace RDKit{
       offset = (1<<7) + (1<<14) + (1<<21);
     }
     num = (val >> shift) + offset;
+    //num = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(num);
     return num;
   }
   
   //! does a binary write of an object to a stream
   template <typename T>
     void streamWrite(std::ostream &ss,const T &val){
-    ss.write((const char *)&val,sizeof(T));
+    T tval=EndianSwapBytes<HOST_ENDIAN_ORDER,LITTLE_ENDIAN_ORDER>(val);
+    ss.write((const char *)&tval,sizeof(T));
   }
   //! does a binary read of an object from a stream
   template <typename T>
     void streamRead(std::istream &ss,T &loc){
-    ss.read((char *)&loc,sizeof(T));
+    T tloc;
+    ss.read((char *)&tloc,sizeof(T));
+    loc = EndianSwapBytes<LITTLE_ENDIAN_ORDER,HOST_ENDIAN_ORDER>(tloc);
   }
  
   //! grabs the next line from an instream and returns it.
--- rdkit-201203.orig/Code/ML/CMakeLists.txt
+++ rdkit-201203/Code/ML/CMakeLists.txt
@@ -1,4 +1,8 @@
-add_subdirectory(Cluster)
-add_subdirectory(FeatureSelect)
+if(RDK_BIG_ENDIAN)
+  message("Skipping build of clustering and cmim libraries on big endian system")
+else(RDK_BIG_ENDIAN)
+  add_subdirectory(Cluster)
+  add_subdirectory(FeatureSelect)
+endif(RDK_BIG_ENDIAN)
 add_subdirectory(InfoTheory)
 add_subdirectory(Data)
--- rdkit-201203.orig/Code/ML/FeatureSelect/Wrap/testCMIM.py
+++ rdkit-201203/Code/ML/FeatureSelect/Wrap/testCMIM.py
@@ -31,13 +31,13 @@ class TestCase(unittest.TestCase):
      examples.append([0,bv,1])
 
      r = FS.selectCMIM(examples,2)
-     self.failUnless(r==(2,4))
+     self.failUnlessEqual(r,(2,4))
 
      r = FS.selectCMIM(examples,1)
-     self.failUnless(r==(2,))
+     self.failUnlessEqual(r,(2,))
 
      r = FS.selectCMIM(examples,3)
-     self.failUnless(r==(2,4,-1))
+     self.failUnlessEqual(r,(2,4,-1))
 
    def test1FromList(self) :
      examples = []
@@ -63,13 +63,13 @@ class TestCase(unittest.TestCase):
      examples.append([0,bv,1])
 
      r = FS.selectCMIM(examples,2)
-     self.failUnless(r==(2,4))
+     self.failUnlessEqual(r,(2,4))
 
      r = FS.selectCMIM(examples,1)
-     self.failUnless(r==(2,))
+     self.failUnlessEqual(r,(2,))
 
      r = FS.selectCMIM(examples,3)
-     self.failUnless(r==(2,4,-1))
+     self.failUnlessEqual(r,(2,4,-1))
 
 
 if __name__ == '__main__':
--- rdkit-201203.orig/rdkit/DataStructs/VectCollection.py
+++ rdkit-201203/rdkit/DataStructs/VectCollection.py
@@ -230,12 +230,13 @@ class VectCollection(object):
   # set up our support for pickling:
   #
   def __getstate__(self):
-    pkl = struct.pack('I',len(self.__vects))
+    pkl = struct.pack('<I',len(self.__vects))
     for k,v in self.__vects.iteritems():
-      pkl += struct.pack('I',k)
+      pkl += struct.pack('<I',k)
       p = v.ToBinary()
       l = len(p)
-      pkl += struct.pack('I%ds'%(l),l,p)
+      pkl += struct.pack('<I',l)
+      pkl += struct.pack('%ds'%(l),p)
     return pkl
 
   def __setstate__(self,pkl):
@@ -245,12 +246,12 @@ class VectCollection(object):
     self.__needReset=True
     szI = struct.calcsize('I')
     offset = 0
-    nToRead = struct.unpack('I',pkl[offset:offset+szI])[0]
+    nToRead = struct.unpack('<I',pkl[offset:offset+szI])[0]
     offset += szI
     for i in range(nToRead):
-      k = struct.unpack('I',pkl[offset:offset+szI])[0]
+      k = struct.unpack('<I',pkl[offset:offset+szI])[0]
       offset += szI
-      l = struct.unpack('I',pkl[offset:offset+szI])[0]
+      l = struct.unpack('<I',pkl[offset:offset+szI])[0]
       offset += szI
       sz = struct.calcsize('%ds'%l)
       bv = DataStructs.ExplicitBitVect(struct.unpack('%ds'%l,pkl[offset:offset+sz])[0])
--- rdkit-201203.orig/rdkit/SimDivFilters/__init__.py
+++ rdkit-201203/rdkit/SimDivFilters/__init__.py
@@ -3,4 +3,8 @@
 #    All Rights Reserved
 #
 from rdkit import rdBase
-from rdSimDivPickers import *
+try:
+  import rdSimDivPickers
+  from rdSimDivPickers import *
+except ImportError:
+  rdSimDivPickers=None
--- rdkit-201203.orig/rdkit/ML/test_list.py
+++ rdkit-201203/rdkit/ML/test_list.py
@@ -5,7 +5,7 @@ tests=[
   ("python","UnitTestAnalyzeComposite.py",{}),
   ]
 
-for dir in ['Cluster','Composite','Data','DecTree','Descriptors','FeatureSelect','InfoTheory','KNN','ModelPackage','NaiveBayes','Neural','SLT']:
+for dir in ['Cluster','Composite','Data','DecTree','Descriptors','InfoTheory','KNN','ModelPackage','NaiveBayes','Neural','SLT']:
     tests.append(('python','test_list.py',{'dir':dir}))
 
 longTests=[
--- rdkit-201203.orig/rdkit/ML/InfoTheory/BitClusterer.py
+++ rdkit-201203/rdkit/ML/InfoTheory/BitClusterer.py
@@ -3,6 +3,8 @@
 #
 
 from rdkit.SimDivFilters import rdSimDivPickers as rdsimdiv
+if rdsimdiv is None:
+   raise ImportError,'rdSimDivPickers not built'
 from rdkit import DataStructs
 import numpy
 
--- rdkit-201203.orig/rdkit/ML/InfoTheory/testCorrMatGen.py
+++ rdkit-201203/rdkit/ML/InfoTheory/testCorrMatGen.py
@@ -2,7 +2,11 @@
 
 from rdkit import RDConfig
 import unittest
-from rdkit.ML.InfoTheory import rdInfoTheory, BitClusterer
+from rdkit.ML.InfoTheory import rdInfoTheory
+try:
+  from rdkit.ML.InfoTheory import BitClusterer
+except ImportError:
+  BitClusterer=None
 from rdkit.ML.Data import DataUtils
 from rdkit import DataStructs
 import random
@@ -64,6 +68,8 @@ class TestCase(unittest.TestCase):
         assert 2*navr/self.nbits == 158.3,2*navr/self.nbits
 
     def test1Cluster(self) :
+        if BitClusterer is None:
+            return
         cmg = rdInfoTheory.BitCorrMatGenerator()
         cmg.SetBitList(self.blist)
         for fp in self.fps:
--- rdkit-201203.orig/rdkit/ML/Cluster/UnitTestCluster.py
+++ rdkit-201203/rdkit/ML/Cluster/UnitTestCluster.py
@@ -14,7 +14,11 @@
 from rdkit import RDConfig
 import unittest
 from rdkit.ML.Cluster import Standardize,ClusterUtils
-from rdkit.ML.Cluster import Clusters,Murtagh
+from rdkit.ML.Cluster import Clusters
+try:
+  from rdkit.ML.Cluster import Murtagh
+except ImportError:
+  Murtagh=None
 import numpy
 import cPickle
 
@@ -57,6 +61,7 @@ class TestCase(unittest.TestCase):
     assert indices==[8,7,5,6],'bad index order'
 
   def testMurtaghUPGMA(self):
+    if Murtagh is None: return
     nPts = 5
     sz = 5
     dataP = numpy.random.random((nPts,sz))
--- rdkit-201203.orig/rdkit/ML/FeatureSelect/UnitTestCMIM.py
+++ rdkit-201203/rdkit/ML/FeatureSelect/UnitTestCMIM.py
@@ -1,4 +1,7 @@
-from rdkit.ML.FeatureSelect import CMIM
+try:
+  from rdkit.ML.FeatureSelect import CMIM
+except ImportError:
+  CMIM = None
 from rdkit import DataStructs as DS
 from rdkit import RDConfig
 import unittest
@@ -8,6 +11,8 @@ class TestCase(unittest.TestCase):
       pass
 
    def test0FromList(self) :
+     if CMIM is None:
+       return
      examples = []
 
      bv = DS.ExplicitBitVect(5)
--- rdkit-201203.orig/rdkit/ML/DecTree/BuildSigTree.py
+++ rdkit-201203/rdkit/ML/DecTree/BuildSigTree.py
@@ -12,7 +12,10 @@
 import numpy
 from rdkit.ML.DecTree import SigTree
 from rdkit.ML import InfoTheory
-from rdkit.ML.FeatureSelect import CMIM
+try:
+  from rdkit.ML.FeatureSelect import CMIM
+except ImportError:
+  CMIM=None
 from rdkit.DataStructs.VectCollection import VectCollection
 import copy
 import random
@@ -136,7 +139,7 @@ def BuildSigTree(examples,nPossibleRes,e
     nBits = fp.GetNumBits()
     ranker = InfoTheory.InfoBitRanker(nBits,nPossibleRes,metric)
     if biasList: ranker.SetBiasList(biasList)
-    if useCMIM > 0 and not ensemble:
+    if CMIM is not None and useCMIM > 0 and not ensemble:
       ensemble = CMIM.SelectFeatures(examples,useCMIM,bvCol=1)
     if random:
       if ensemble:
--- rdkit-201203.orig/rdkit/ML/NaiveBayes/CrossValidate.py
+++ rdkit-201203/rdkit/ML/NaiveBayes/CrossValidate.py
@@ -9,14 +9,17 @@ and evaluation of individual models
 """
 from rdkit.ML.NaiveBayes.ClassificationModel import NaiveBayesClassifier
 from rdkit.ML.Data import SplitData
-from rdkit.ML.FeatureSelect import CMIM
+try:
+  from rdkit.ML.FeatureSelect import CMIM
+except ImportError:
+  CMIM=None
 
 def makeNBClassificationModel(trainExamples, attrs, nPossibleValues, nQuantBounds,
                               mEstimateVal=-1.0,
                               useSigs=False,
                               ensemble=None,useCMIM=0,
                               **kwargs) :
-  if useCMIM > 0 and useSigs and not ensemble:
+  if CMIM is not None and useCMIM > 0 and useSigs and not ensemble:
     ensemble = CMIM.SelectFeatures(trainExamples,useCMIM,bvCol=1)
   if ensemble:
     attrs = ensemble