def processDiffs(self,proj,path):
#self.got_some = {'java':True, 'cxx':True, 'hxx':True}
self.got_some = {'java':False, 'cxx':True, 'hxx':False}
self.pb = PathBuilder(self.tmpPath, force_clean = True)
if os.path.isdir(path) is True:
self.filterDiffProj(proj)
elif os.path.isfile(path) is True:
self.filterDiffFile(proj)
else:
return ('Invalid path : ' + path, False)
# Second, change each diff into ccFinder input format
converter = CCFXInputConverter()
progress = (self.operations_so_far.incr() / float(self.num_operations))*100
callback = lambda: self.progress('Converting to ccfx input format')
converter.convert(proj, self.pb, callback)
self.num_operations = 3 * 2
self.operations_so_far = IntegerWrapper(0)
return ("Converting diffs to ccFinder compatible format is done",True)
def processDiffs(self, proj, path):
#self.got_some = {'java':True, 'cxx':True, 'hxx':True}
self.got_some = {'java': False, 'cxx': True, 'hxx': False}
self.pb = PathBuilder(self.tmpPath, force_clean=True)
if os.path.isdir(path) is True:
self.filterDiffProj(proj)
elif os.path.isfile(path) is True:
self.filterDiffFile(proj)
else:
return ('Invalid path : ' + path, False)
# Second, change each diff into ccFinder input format
converter = CCFXInputConverter()
progress = (self.operations_so_far.incr() /
float(self.num_operations)) * 100
callback = lambda: self.progress('Converting to ccfx input format')
converter.convert(proj, self.pb, callback)
self.num_operations = 3 * 2
self.operations_so_far = IntegerWrapper(0)
return ("Converting diffs to ccFinder compatible format is done", True)
def processImpl(self, model):
proj0 = model.getProj(PathBuilder.Proj0)
proj1 = model.getProj(PathBuilder.Proj1)
path_builder = model.getPathBuilder()
converter = CCFXInputConverter()
ccfx = CCFXEntryPoint(path_builder, model.getCcfxPath(), model.getCcfxTokenSize())
step = 0
total_steps = 20.0
final_status = False
while not self.sync.stopRequested():
if step == 0:
self.progress("Loading version histories for first project",
step / total_steps)
step += 1
proj0.load()
elif step == 1:
self.progress("Loading version histories for second project",
step / total_steps)
step += 1
proj1.load()
elif step == 2:
self.progress("Dumping commits for first project",
step / total_steps)
step += 1
proj0.dumpCommits()
elif step == 3:
self.progress("Dumping commits for second project",
step / total_steps)
step += 1
proj1.dumpCommits()
elif step == 4:
self.progress("Converting diffs to ccfx compatible format for first project",
step / total_steps)
step += 1
converter.convert(path_builder)
elif step == 5:
self.progress("Converting diffs to ccfx compatible format for second project",
step / total_steps)
step += 1
elif step == 6:
self.progress("Running ccFinder for old C, this will take quite some time...",
step / total_steps)
step += 1
have_old_c = ccfx.processPairs(LangDecider.CXX, False)
elif step == 7:
self.progress("Running ccFinder for new C, this will take quite some time...",
step / total_steps)
step += 1
have_new_c = ccfx.processPairs(LangDecider.CXX, True)
elif step == 8:
self.progress("Running ccFinder for old headers, this will take quite some time...",
step / total_steps)
step += 1
have_old_h = ccfx.processPairs(LangDecider.HXX, False)
elif step == 9:
self.progress("Running ccFinder for new headers, this will take quite some time...",
step / total_steps)
step += 1
have_new_h = ccfx.processPairs(LangDecider.HXX, True)
elif step == 10:
self.progress("Running ccFinder for old Java, this will take quite some time...",
step / total_steps)
step += 1
have_old_j = ccfx.processPairs(LangDecider.JAVA, False)
elif step == 11:
self.progress("Running ccFinder for new Java, this will take quite some time...",
step / total_steps)
step += 1
have_new_j = ccfx.processPairs(LangDecider.JAVA, True)
elif step == 12:
self.progress("Filtering ccFinder old C output based on operation...",
step / total_steps)
step += 1
if not have_old_c:
continue
is_new = False
lang = LangDecider.CXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
elif step == 13:
self.progress("Filtering ccFinder new C output based on operation...",
step / total_steps)
step += 1
if not have_new_c:
continue
is_new = True
lang = LangDecider.CXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
elif step == 14:
self.progress("Filtering ccFinder old header output based on operation...",
step / total_steps)
step += 1
if not have_old_h:
continue
is_new = False
lang = LangDecider.HXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
elif step == 15:
self.progress("Filtering ccFinder new header output based on operation...",
step / total_steps)
step += 1
if not have_new_h:
continue
is_new = True
lang = LangDecider.HXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
elif step == 16:
self.progress("Filtering ccFinder old java output based on operation...",
step / total_steps)
step += 1
if not have_old_j:
continue
is_new = False
lang = LangDecider.JAVA
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
elif step == 17:
self.progress("Filtering ccFinder new java output based on operation...",
step / total_steps)
step += 1
if not have_new_j:
continue
is_new = True
lang = LangDecider.JAVA
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
elif step == 18:
self.progress(
"Combining ccFinder output into a unified database...",
step / total_steps)
step += 1
pickle.dump(model, open(path_builder.getModelPathAndName(), 'w'))
rep_populator = RepDBPopulator(path_builder)
db = rep_populator.generateDB(proj0, proj1)
db_file = open(path_builder.getDBPathAndName(), 'w')
pickle.dump(db, db_file)
db_file.close()
else:
final_status = True
break
if final_status:
return 'Success!', final_status
return 'Aborting', final_status
def filterDiffs(self, interface):
self.got_some = {'java':True, 'cxx':True, 'hxx':True}
# self.haveJava = haveC = haveH = False
self.pb = PathBuilder(self.tmpPath, force_clean = True)
# First, filter the input diffs by file type, so that all c diffs
# are in one set of files, and similarly for java/headers
if self.processDirectory is True:
self.filterDiffProjs(interface)
else:
self.filterDiffFiles(interface)
# Second, change each diff into ccFinder input format
converter = CCFXInputConverter()
callback = lambda: interface.progress(
'Converting to ccfx input format',
self.operations_so_far.incr() / float(self.num_operations))
converter.convert(self.pb, callback)
#new and old for 3 langs
self.num_operations = 3 * 2
self.operations_so_far = IntegerWrapper(0)
clone_path = self.pb.getCCFXOutputPath()
# Third, call ccfx for each directory
worked = True
for lang in ['java', 'cxx', 'hxx']:
if not self.got_some[lang]:
interface.progress('ccFinderX executing',
self.operations_so_far.incr() / float(self.num_operations))
continue
old_path0 = self.pb.getCCFXInputPath(PathBuilder.PROJ0, lang, False)
old_path1 = self.pb.getCCFXInputPath(PathBuilder.PROJ1, lang, False)
new_path0 = self.pb.getCCFXInputPath(PathBuilder.PROJ0, lang, True)
new_path1 = self.pb.getCCFXInputPath(PathBuilder.PROJ1, lang, True)
tmp_old_out = clone_path + self.pb.getCCFXOutputFileName(
lang, is_new = False, is_tmp = True)
tmp_new_out = clone_path + self.pb.getCCFXOutputFileName(
lang, is_new = True, is_tmp = True)
old_out = clone_path + self.pb.getCCFXOutputFileName(
lang, is_new = False, is_tmp = False)
new_out = clone_path + self.pb.getCCFXOutputFileName(
lang, is_new = True, is_tmp = False)
if self.paths['proj0'] == self.paths['proj1']:
old_path1 = old_path0
new_path1 = new_path0
worked = worked and self.ccfx.processPair(
old_path0, old_path1, tmp_old_out, old_out, lang)
interface.progress('ccFinderX executing',
self.operations_so_far.incr() / float(self.num_operations))
worked = worked and self.ccfx.processPair(
new_path0, new_path1, tmp_new_out, new_out, lang)
interface.progress('ccFinderX executing',
self.operations_so_far.incr() / float(self.num_operations))
if not worked:
return ('ccFinderX execution failed', False)
# Fourth, build up our database of clones
print "Repertoire filtering...."
#new and old for 3 langs
self.num_operations = 3 * 2
self.operations_so_far = IntegerWrapper(0)
for lang in ['java', 'cxx', 'hxx']:
if not self.got_some[lang]:
interface.progress('Repertoire filtering based on operation',
self.operations_so_far.incr() / float(self.num_operations))
continue
for is_new in [True, False]:
output = convert_ccfx_output(self.pb, lang, is_new)
rep_out_path = self.pb.getRepertoireOutputPath(lang, is_new)
suffix = '_old.txt'
if is_new:
suffix = '_new.txt'
output.writeToFile(rep_out_path + lang + suffix)
interface.progress('Repertoire filtering based on operation',
self.operations_so_far.incr() / float(self.num_operations))
print "Processing successful!!"
return ('Processing successful', True)
def processImpl(self, model):
proj0 = model.getProj(PathBuilder.Proj0)
proj1 = model.getProj(PathBuilder.Proj1)
path_builder = model.getPathBuilder()
converter = CCFXInputConverter()
ccfx = CCFXEntryPoint(path_builder, model.getCcfxPath(), model.getCcfxTokenSize())
step = 0
total_steps = 20.0
final_status = False
# step == 0:
print "Loading version histories for first project" + str(step / total_steps)
step += 1
proj0.load()
print "Loading version histories for second project" + str(step / total_steps)
step += 1
proj1.load()
print "Dumping commits for first project" + str(step / total_steps)
step += 1
proj0.dumpCommits()
print "Dumping commits for second project" + str(step / total_steps)
step += 1
proj1.dumpCommits()
print "Converting diffs to ccfx compatible format" + str(step / total_steps)
step += 1
converter.convert(path_builder)
print "Running ccFinder for old C, this will take quite some time..." + str(step / total_steps)
step += 1
have_old_c = ccfx.processPairs(LangDecider.CXX, False)
print "Running ccFinder for new C, this will take quite some time..." + str(step / total_steps)
step += 1
have_new_c = ccfx.processPairs(LangDecider.CXX, True)
print "Running ccFinder for old headers, this will take quite some time..." + str(step / total_steps)
step += 1
have_old_h = ccfx.processPairs(LangDecider.HXX, False)
print "Running ccFinder for new headers, this will take quite some time..." + str(step / total_steps)
step += 1
have_new_h = ccfx.processPairs(LangDecider.HXX, True)
print "Running ccFinder for old Java, this will take quite some time..." + str(step / total_steps)
step += 1
have_old_j = ccfx.processPairs(LangDecider.JAVA, False)
print "Running ccFinder for new Java, this will take quite some time..." + str(step / total_steps)
step += 1
have_new_j = ccfx.processPairs(LangDecider.JAVA, True)
print "Filtering ccFinder old C output based on operation..." + str(step / total_steps)
step += 1
if have_old_c:
is_new = False
lang = LangDecider.CXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Filtering ccFinder new C output based on operation..." + str(step / total_steps)
step += 1
if have_new_c:
is_new = True
lang = LangDecider.CXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Filtering ccFinder old header output based on operation..." + str(step / total_steps)
step += 1
if have_old_h:
is_new = False
lang = LangDecider.HXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Filtering ccFinder new header output based on operation..." + str(step / total_steps)
step += 1
if have_new_h:
is_new = True
lang = LangDecider.HXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Filtering ccFinder old java output based on operation..." + str(step / total_steps)
step += 1
if not have_old_j:
is_new = False
lang = LangDecider.JAVA
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Filtering ccFinder new java output based on operation..." + str(step / total_steps)
step += 1
if not have_new_j:
is_new = True
lang = LangDecider.JAVA
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Combining ccFinder output into a unified database..." + str(step / total_steps)
step += 1
pickle.dump(model, open(path_builder.getModelPathAndName(), "w"))
rep_populator = RepDBPopulator(path_builder)
db = rep_populator.generateDB(proj0, proj1)
db_file = open(path_builder.getDBPathAndName(), "w")
pickle.dump(db, db_file)
db_file.close()
final_status = True
if final_status:
return "Success!", final_status
return "Aborting", final_status
def __init__(self, model):
self.proj0 = model.getProj(PathBuilder.Proj0)
self.proj1 = model.getProj(PathBuilder.Proj1)
self.path_builder = model.getPathBuilder()
self.converter = CCFXInputConverter()
self.ccfx = CCFXEntryPoint(self.path_builder, model.getCcfxPath(), model.getCcfxTokenSize())
class RepDriver:
def __init__(self, model):
self.proj0 = model.getProj(PathBuilder.Proj0)
self.proj1 = model.getProj(PathBuilder.Proj1)
self.path_builder = model.getPathBuilder()
self.converter = CCFXInputConverter()
self.ccfx = CCFXEntryPoint(self.path_builder, model.getCcfxPath(), model.getCcfxTokenSize())
def ccfxConvert(self):
print "Converting diffs to ccfx compatible format"
if self.proj0:
proj0_repo = self.proj0.getRepoRoot() + os.sep
print proj0_repo
self.path_builder.setExtDiffPath(0, proj0_repo)
if self.proj1:
proj1_repo = self.proj1.getRepoRoot() + os.sep
print proj1_repo
self.path_builder.setExtDiffPath(1, proj1_repo)
self.converter.convertExtDiffs(self.path_builder)
def runCCFX_old(self, lang):
print "Running ccFinder for old files, this will take quite some time..."
have_old_lang = self.ccfx.processPairs(lang, False)
print "Filtering ccFinder old output based on operation..."
if have_old_lang:
is_new = False
output = convert_ccfx_output(self.path_builder, lang, is_new)
rep_out_path = self.path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = self.path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
def runCCFX_new(self, lang):
print "Running ccFinder for new files, this will take quite some time..."
have_new_lang = self.ccfx.processPairs(lang, True)
print "Filtering ccFinder new output based on operation..."
if have_new_lang:
is_new = True
# lang = have_new_lang
output = convert_ccfx_output(self.path_builder, lang, is_new)
rep_out_path = self.path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = self.path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
def runCCFX(self):
# self.runCCFX_old()
for lang in [LangDecider.CXX, LangDecider.HXX, LangDecider.JAVA]:
self.runCCFX_new(lang)
self.runCCFX_old(lang)
def process(self, rep_model):
msg, success = self.processImpl(rep_model)
def processImpl(self, model):
proj0 = model.getProj(PathBuilder.Proj0)
proj1 = model.getProj(PathBuilder.Proj1)
path_builder = model.getPathBuilder()
converter = CCFXInputConverter()
ccfx = CCFXEntryPoint(path_builder, model.getCcfxPath(), model.getCcfxTokenSize())
step = 0
total_steps = 20.0
final_status = False
# step == 0:
print "Loading version histories for first project" + str(step / total_steps)
step += 1
proj0.load()
print "Loading version histories for second project" + str(step / total_steps)
step += 1
proj1.load()
print "Dumping commits for first project" + str(step / total_steps)
step += 1
proj0.dumpCommits()
print "Dumping commits for second project" + str(step / total_steps)
step += 1
proj1.dumpCommits()
print "Converting diffs to ccfx compatible format" + str(step / total_steps)
step += 1
converter.convert(path_builder)
print "Running ccFinder for old C, this will take quite some time..." + str(step / total_steps)
step += 1
have_old_c = ccfx.processPairs(LangDecider.CXX, False)
print "Running ccFinder for new C, this will take quite some time..." + str(step / total_steps)
step += 1
have_new_c = ccfx.processPairs(LangDecider.CXX, True)
print "Running ccFinder for old headers, this will take quite some time..." + str(step / total_steps)
step += 1
have_old_h = ccfx.processPairs(LangDecider.HXX, False)
print "Running ccFinder for new headers, this will take quite some time..." + str(step / total_steps)
step += 1
have_new_h = ccfx.processPairs(LangDecider.HXX, True)
print "Running ccFinder for old Java, this will take quite some time..." + str(step / total_steps)
step += 1
have_old_j = ccfx.processPairs(LangDecider.JAVA, False)
print "Running ccFinder for new Java, this will take quite some time..." + str(step / total_steps)
step += 1
have_new_j = ccfx.processPairs(LangDecider.JAVA, True)
print "Filtering ccFinder old C output based on operation..." + str(step / total_steps)
step += 1
if have_old_c:
is_new = False
lang = LangDecider.CXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Filtering ccFinder new C output based on operation..." + str(step / total_steps)
step += 1
if have_new_c:
is_new = True
lang = LangDecider.CXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Filtering ccFinder old header output based on operation..." + str(step / total_steps)
step += 1
if have_old_h:
is_new = False
lang = LangDecider.HXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Filtering ccFinder new header output based on operation..." + str(step / total_steps)
step += 1
if have_new_h:
is_new = True
lang = LangDecider.HXX
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Filtering ccFinder old java output based on operation..." + str(step / total_steps)
step += 1
if not have_old_j:
is_new = False
lang = LangDecider.JAVA
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Filtering ccFinder new java output based on operation..." + str(step / total_steps)
step += 1
if not have_new_j:
is_new = True
lang = LangDecider.JAVA
output = convert_ccfx_output(path_builder, lang, is_new)
rep_out_path = path_builder.getRepertoireOutputPath(lang, is_new)
rep_out_file = path_builder.getRepertoireOutputFileName(lang, is_new)
output.writeToFile(rep_out_path + rep_out_file)
print "Combining ccFinder output into a unified database..." + str(step / total_steps)
step += 1
pickle.dump(model, open(path_builder.getModelPathAndName(), "w"))
rep_populator = RepDBPopulator(path_builder)
db = rep_populator.generateDB(proj0, proj1)
db_file = open(path_builder.getDBPathAndName(), "w")
pickle.dump(db, db_file)
db_file.close()
final_status = True
if final_status:
return "Success!", final_status
return "Aborting", final_status