'''
Various generic utility functions
'''

import os
from os.path import join
import subprocess
import functools
import itertools
import io
from pkg_resources import resource_filename

# Get the folder one level above the current folder
scriptsPath = resource_filename("praatio", "praatScripts", )


def getValueAtTime(timestamp, sortedDataTupleList, fuzzyMatching=False,
                   startI=0):
    '''
    Get the value in the data list (sorted by time) that occurs at this point
    
    If fuzzyMatching is True, if there is not a value
    at the requested timestamp, the nearest feature value will be taken.
    
    The procedure assumes that all data is ordered in time.
    dataTupleList should be in the form
    [(t1, v1a, v1b, ..), (t2, v2a, v2b, ..), ..]
    
    The procedure makes one pass through dataTupleList and one
    pass through self.entryList.  If the data is not sequentially
    ordered, the incorrect response will be returned.
    
    For efficiency purposes, it takes a starting index and returns the ending
    index.
    '''
    
    i = startI
    
    # Only find exact timestamp matches
    if fuzzyMatching is False:
        while True:
            try:
                dataTuple = sortedDataTupleList[i]
            except IndexError:
                currTime = timestamp
                currVal = "--"
                break

            currTime = dataTuple[0]
            currVal = dataTuple[1]
            if timestamp == currTime:
                break
            i += 1
        retTime = currTime
        retVal = currVal
    
    # Find the closest timestamp
    else:
        bestTime = sortedDataTupleList[i][0]
        bestVal = sortedDataTupleList[i][1]
        i += 1
        while True:
            try:
                dataTuple = sortedDataTupleList[i]
            except IndexError:
                break  # Last known value is the closest one

            currTime = dataTuple[0]
            currVal = dataTuple[1]

            currDiff = abs(currTime - timestamp)
            bestDiff = abs(bestTime - timestamp)
            if currDiff < bestDiff:  # We're closer to the target val
                bestTime = currTime
                bestVal = currVal
                if currDiff == 0:
                    break  # Can't do better than a perfect match
            elif currDiff > bestDiff:
                break  # We've past the best value.
            i += 1
        
        retTime = bestTime
        retVal = bestVal

    return retTime, retVal, i


def getValuesInInterval(dataTupleList, start, stop):
    '''
    Gets the values that exist within an interval
    
    The function assumes that the data is formated as
    [(t1, v1a, v1b, ...), (t2, v2a, v2b, ...)]
    '''
            
    intervalDataList = []
    for dataTuple in dataTupleList:
        time = dataTuple[0]
        if start <= time and stop >= time:
            intervalDataList.append(dataTuple)
    
    return intervalDataList


def sign(x):
    '''Returns 1 if x is positive, 0 if x is 0, and -1 otherwise'''
    retVal = 0
    if x > 0:
        retVal = 1
    elif x < 0:
        retVal = -1
    return retVal


def invertIntervalList(inputList, maxValue=None):
    '''
    Inverts the segments of a list of intervals
    
    e.g.
    [(0,1), (4,5), (7,10)] -> [(1,4), (5,7)]
    '''
    inputList = sorted(inputList)
    
    # Special case -- empty lists
    if len(inputList) == 0 and maxValue is not None:
        invList = [(0, maxValue), ]
    else:
        # Insert in a garbage head and tail value for the purpose
        # of inverting, in the range does not start and end at the
        # smallest and largest values
        if inputList[0][0] != 0:
            inputList.insert(0, ['', 0])
        if maxValue is not None and inputList[-1][1] < maxValue:
            inputList.append((maxValue, ''))
        
        invList = [[inputList[i][1], inputList[i + 1][0]]
                   for i in range(0, len(inputList) - 1)]
    
    return invList


def makeDir(path):
    '''
    Creates a new directory

    Unlike os.mkdir, it does not throw an exception if the directory already exists.
    '''
    if not os.path.exists(path):
        os.mkdir(path)
        

def findAll(txt, subStr):
    '''
    Find the starting indicies of all instances of subStr in txt
    '''
    indexList = []
    index = 0
    while True:
        try:
            index = txt.index(subStr, index)
        except ValueError:
            break
        indexList.append(int(index))
        index += 1
    
    return indexList


class FileNotFound(Exception):
    
    def __init__(self, fullPath):
        super(FileNotFound, self).__init__()
        self.fullPath = fullPath
    
    def __str__(self):
        return ("File not found:\n%s" % self.fullPath)


class PraatExecutionFailed(Exception):
    
    def __init__(self, cmdList):
        super(PraatExecutionFailed, self).__init__()
        self.cmdList = cmdList
    
    def __str__(self):
        errorStr = ("\nPraat Execution Failed.  Please check the following:\n"
                    "- Praat exists in the location specified\n"
                    "- Praat script can execute ok outside of praat\n"
                    "- script arguments are correct\n\n"
                    "If you can't locate the problem, I recommend using "
                    "absolute paths rather than relative "
                    "paths and using paths without spaces in any folder "
                    "or file names\n\n"
                    "Here is the command that python attempted to run:\n")
        cmdTxt = " ".join(self.cmdList)
        return errorStr + cmdTxt
    
    
def runPraatScript(praatEXE, scriptFN, argList, cwd=None):
    
    # Popen gives a not-very-transparent error
    if not os.path.exists(praatEXE):
        raise FileNotFound(praatEXE)
    if not os.path.exists(scriptFN):
        raise FileNotFound(scriptFN)
    
    argList = ["%s" % arg for arg in argList]
    cmdList = [praatEXE, '--run', scriptFN] + argList
    
    myProcess = subprocess.Popen(cmdList, cwd=cwd)
    
    if myProcess.wait():
        raise PraatExecutionFailed(cmdList)


def _getMatchFunc(pattern):
    '''
    An unsophisticated pattern matching function
    '''
    
    # '#' Marks word boundaries, so if there is more than one we need to do
    #    something special to make sure we're not mis-representings them
    assert(pattern.count('#') < 2)

    def startsWith(subStr, fullStr):
        return fullStr[:len(subStr)] == subStr
            
    def endsWith(subStr, fullStr):
        return fullStr[-1 * len(subStr):] == subStr
    
    def inStr(subStr, fullStr):
        return subStr in fullStr

    # Selection of the correct function
    if pattern[0] == '#':
        pattern = pattern[1:]
        cmpFunc = startsWith
        
    elif pattern[-1] == '#':
        pattern = pattern[:-1]
        cmpFunc = endsWith
        
    else:
        cmpFunc = inStr
    
    return functools.partial(cmpFunc, pattern)


def findFiles(path, filterPaths=False, filterExt=None, filterPattern=None,
              skipIfNameInList=None, stripExt=False):
    
    fnList = os.listdir(path)
       
    if filterPaths is True:
        fnList = [folderName for folderName in fnList
                  if os.path.isdir(os.path.join(path, folderName))]

    if filterExt is not None:
        splitFNList = [[fn, ] + list(os.path.splitext(fn)) for fn in fnList]
        fnList = [fn for fn, name, ext in splitFNList if ext == filterExt]
        
    if filterPattern is not None:
        splitFNList = [[fn, ] + list(os.path.splitext(fn)) for fn in fnList]
        matchFunc = _getMatchFunc(filterPattern)
        fnList = [fn for fn, name, ext in splitFNList if matchFunc(name)]
    
    if skipIfNameInList is not None:
        targetNameList = [os.path.splitext(fn)[0] for fn in skipIfNameInList]
        fnList = [fn for fn in fnList
                  if os.path.splitext(fn)[0] not in targetNameList]
    
    if stripExt is True:
        fnList = [os.path.splitext(fn)[0] for fn in fnList]
    
    fnList.sort()
    return fnList


def openCSV(path, fn, valueIndex=None, encoding="utf-8"):
    '''
    Load a feature
    
    In many cases we only want a single value from the feature (mainly because
    the feature only contains one value).  In these situations, the user
    can indicate that rather than receiving a list of lists, they can receive
    a lists of values, where each value represents the item in the row
    indicated by valueIndex.
    '''
    
    # Load CSV file
    with io.open(join(path, fn), "r", encoding=encoding) as fd:
        featureList = fd.read().splitlines()
    featureList = [row.split(",") for row in featureList]
    
    if valueIndex is not None:
        featureList = [row[valueIndex] for row in featureList]
    
    return featureList


def safeZip(listOfLists, enforceLength):
    """
    A safe version of python's zip()

    If two sublists are of different sizes, python's zip will truncate
    the output to be the smaller of the two.

    safeZip throws an exception if the size of the any sublist is different
    from the rest.
    """
    if enforceLength is True:
        length = len(listOfLists[0])
        assert(all([length == len(subList) for subList in listOfLists]))
    
    try:
        zipFunc = itertools.izip_longest # Python 2.x
    except AttributeError:
        zipFunc = itertools.zip_longest # Python 3.x
    
    return zipFunc(*listOfLists)