Merge branch 'python-package' into awdl

* python-package:
  fix little endian labels
  libpcap fixes #15

README.md

@@ -39,6 +39,7 @@ Have a look into `nemesys.py` resp. `nemesys_fms.py` to get an impression of the
 
 ## Requirements
 * Python 3
+* libpcap for pcapy: `apt-get install libpcap-dev libpq-dev`
 * Install packages listed in requirements.txt: `pip install -r requirements.txt`
 * Manual install of Netzob from the ["fix-layer-build" branch]([email protected]:skleber/netzob.git)
   -- ~~currently NOT the official~~ [~~"next" branch~~](https://github.com/netzob/netzob/tree/next/netzob)! --

eval-nemeftr-clustering.sh

@@ -58,17 +58,6 @@ report=reports/${prefix}-${cftnpad}-clustering-${currcomm}
 mkdir ${report}
 
 
-#for fn in ${input} ; do
-#for sig in ${sigmas} ; do
-##python src/nemeftr_cluster-segments.py -s ${sig} -r PCAmoco ${fn}
-#python src/nemeftr_cluster-segments.py -s ${sig} -r PCA ${fn}
-##python src/nemeftr_cluster-segments.py -s ${sig} -r base ${fn}
-##python src/nemeftr_cluster-segments.py -s ${sig} -r original ${fn}
-#
-##python src/nemeftr_cluster-segments.py -p -s ${sig} -r PCAmoco ${fn}
-#done
-#done
-
 for seg in ${segmenters} ; do
   for ref in ${refines} ; do
       if [[ ${seg} == "zeros" ]] && [[ ! ${ref} =~ ^(none|PCA1)$ ]] ; then

src/nemeftr_cluster-segments.py

@@ -133,7 +133,7 @@ def inferredFEs4segment(segment: MessageSegment) -> List[int]:
     analysisArgs = None
     analysisTitle = analysis_method
     tokenizer = args.tokenizer
-    if tokenizer == "nemesys" and args.littleendian:
+    if tokenizer == "nemesys" and littleendian:
         tokenizer = "nemesysle"
 
     # # # # # # # # # # # # # # # # # # # # # # # #
@@ -158,13 +158,16 @@ def inferredFEs4segment(segment: MessageSegment) -> List[int]:
             fromCache.configureRefinement(nemetylRefinements)
         elif args.refinement == "PCA1":
             fromCache.configureRefinement(pcaRefinements, littleEndian=littleendian)
-            refinement = args.refinement + "le"
+            if littleendian:
+                refinement = args.refinement + "le"
         elif args.refinement == "PCAmoco":
             fromCache.configureRefinement(pcaMocoRefinements, littleEndian=littleendian)
-            refinement = args.refinement + "le"
+            if littleendian:
+                refinement = args.refinement + "le"
         elif args.refinement == "zerocharPCAmocoSF":
             fromCache.configureRefinement(zerocharPCAmocoSFrefinements, littleEndian=littleendian)
-            refinement = args.refinement + "le"
+            if littleendian:
+                refinement = args.refinement + "le"
         elif args.refinement is None or args.refinement == "none":
             print("No refinement selected. Performing raw segmentation.")
         else:
@@ -173,7 +176,8 @@ def inferredFEs4segment(segment: MessageSegment) -> List[int]:
     elif tokenizer == "zeros":
         if args.refinement == "PCA1":
             fromCache.configureRefinement(pcaRefinements, littleEndian=littleendian)
-            refinement = args.refinement + "le"
+            if littleendian:
+                refinement = args.refinement + "le"
         elif args.refinement is None or args.refinement == "none":
             print("No refinement selected. Performing zeros segmentation with CropChars.")
         else:
@@ -428,7 +432,7 @@ def inferredFEs4segment(segment: MessageSegment) -> List[int]:
     clusterStats = report.precisionRecallList
 
     # # # # # # # # # # # # # # # # # # # # # # # #
-    with open(join(reportFolder, "segmentclusters-" + filechecker.pcapstrippedname + ".csv"), "a") as segfile:
+    with open(join(filechecker.reportFullPath, "segmentclusters-" + filechecker.pcapstrippedname + ".csv"), "a") as segfile:
         segcsv = csv.writer(segfile)
         for cLabel, segments in ftclusters.items():
             segcsv.writerows([