def codeMetricsTable(nameLabel, dataFrame, sourceFilesDirectory, storageConnection):
import storage
import scmData as scm
import dataUtilities
tableName = nameLabel + '_codeMetrics'
if storage.tableExists(tableName, storageConnection):
return storage.readTable(tableName, storageConnection)
scm.gitlog.switchToRevision(sourceFilesDirectory,nameLabel)
metricsData = collectMetrics(os.path.join(os.getcwd(),"..","ck"), sourceFilesDirectory)
#if not metricsData:
#return None
#metricsData = pd.concat(metricsData)
metricsData = dropUnusedMetrics(metricsData)
#metricsDataFrame = metricsData.rename(columns={"file":"entity"})
#metricsData = dataUtilities.formatEntityNames(metricsData, sourceFilesDirectory)
storage.writeTable(tableName, storageConnection, metricsData)
return metricsData
def makeCommitDateMapping(sourceDirectory,
filesToInspect,
storageConnection,
branchname='master'):
import storage
tableName = branchname + '_commitDates'
if storage.tableExists(tableName, storageConnection):
return storage.readTable(tableName, storageConnection)
switchToRevision(sourceDirectory, branchname)
command = "cd %s; git log --pretty=format:'%%h,%%ad' --date=short -- %s" % (
sourceDirectory, filesToInspect)
try:
dates = pd.read_csv(StringIO(
subprocess.check_output(command, shell=True)),
names=['sha', 'date'],
dtype={
'sha': str,
'date': str
})
storage.writeTable(tableName, storageConnection, dates)
except subprocess.CalledProcessError:
dates = "no commit date mapping available"
return dates
def indentMetricsTable(nameLabel,
dataFrame,
sourceFilesDirectory,
storageConnection,
tabsize=4):
import storage
import scmData as scm
import dataUtilities
tableName = nameLabel + '_indentMetrics'
if storage.tableExists(tableName, storageConnection):
return storage.readTable(tableName, storageConnection)
scm.gitlog.switchToRevision(sourceFilesDirectory, nameLabel)
indentMetrics = pd.DataFrame()
for entityName in dataFrame['entity']:
nextFrame = indentCountStats(sourceFilesDirectory + entityName,
tabsize)
nextFrame['entity'] = sourceFilesDirectory + entityName
indentMetrics = pd.concat([indentMetrics, nextFrame])
if indentMetrics.empty:
return None
indentMetrics = dataUtilities.formatEntityNames(indentMetrics,
sourceFilesDirectory)
storage.writeTable(tableName, storageConnection, indentMetrics)
return indentMetrics
def codeMetricsTable(nameLabel, dataFrame, sourceFilesDirectory, storageConnection): import storage import scmData as scm import dataUtilities tableName = nameLabel + '_codeMetrics' if storage.tableExists(tableName, storageConnection): return storage.readTable(tableName, storageConnection) scm.gitlog.switchToRevision(sourceFilesDirectory,nameLabel) metricsData = [] for entityName in dataFrame['entity']: analysis = runAnalysisOnFile(sourceFilesDirectory+entityName) metricsData.append(analysis) if not metricsData: return None metricsData = pd.concat(metricsData) metricsData = dropUnusedCMetrics(metricsData) metricsData = dataUtilities.formatEntityNames(metricsData, sourceFilesDirectory) storage.writeTable(tableName, storageConnection, metricsData) return metricsData
def makeGitLog(sourceDirectory, filesToInspect, lastTimestamp,
previousTimestamp, storageConnection):
import storage
tableName = 'gitLog_' + lastTimestamp + '_' + previousTimestamp
if storage.tableExists(tableName, storageConnection):
return
command = "cd %s; git log %s...%s --pretty=format:'[%%h] %%aN %%ad %%s' --date=short --date-order --numstat -- %s" % (
sourceDirectory, lastTimestamp, previousTimestamp, filesToInspect)
gitlog = subprocess.check_output(command, shell=True)
storage.writeFile(tableName, storageConnection, gitlog)
def changeMetricsTable(lastTime, previousTime, logFileType, storageConnection):
import storage
tableNameChange = lastTime + '_changeMetrics'
if storage.tableExists(tableNameChange, storageConnection):
return storage.readTable(tableNameChange, storageConnection)
tableNameLog = 'gitLog_' + lastTime + '_' + previousTime
scmLogPath = storage.readFile(tableNameLog,
storageConnection,
toTemporaryFile=True)
results = runAnalysisList(scmLogPath, logFileType, entityAnalysisTypes)
changeMetrics = pd.read_csv(results[0])
for result in results[1:]:
changeMetrics = changeMetrics.merge(pd.read_csv(result), on='entity')
changeMetrics.drop('n-revs_y', axis=1, inplace=True)
changeMetrics = changeMetrics.rename(columns={'n-revs_x': 'n-revs'})
storage.writeTable(tableNameChange, storageConnection, changeMetrics)
return changeMetrics