#!/usr/bin/env python
#/*##########################################################################
#
# The PyMca X-Ray Fluorescence Toolkit
#
# Copyright (c) 2004-2014 European Synchrotron Radiation Facility
#
# This file is part of the PyMca X-ray Fluorescence Toolkit developed at
# the ESRF by the Software group.
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE.
#
#############################################################################*/
from __future__ import with_statement
__author__ = "V.A. Sole - ESRF Data Analysis"
__contact__ = "sole@esrf.fr"
__license__ = "MIT"
__copyright__ = "European Synchrotron Radiation Facility, Grenoble, France"
import os
import sys
import re
import numpy
from PyMca5.PyMcaCore import DataObject

SOURCE_TYPE = "EdfFileStack"


class LuciaMap(DataObject.DataObject):
    def __init__(self, filename, infofile=None):
        DataObject.DataObject.__init__(self)

        with open(filename, 'r') as f:
            data = f.read()

        data.replace("\r\n", "\n")
        self.sourceName = [filename]
        firstByte = data.index("\n\n")

        header = data[0:firstByte]
        #get rid of the date
        data = data[firstByte:]

        #leave only the '----' as separator
        data.replace("\r", "")
        data.replace("\n", "")
        sep = '-'
        while sep in data:
            sep = sep + '-'
        sep = sep[1:]
        data = data.split(sep)
        if len(data[0]) != len(data[-1]):
            if len(data[0]) > 1:
                del data[-1]
            else:
                del data[0]

        #get the number of channels
        exp = re.compile('(-?[0-9]+\.?[0-9]*)')
        spectrum = [float(x) for x in exp.findall(data[0])]
        self.nChannels = len(spectrum)
        self.nSpectra = len(data)
        self.nRows = self.nSpectra

        #try to get the information
        if infofile is None:
            infofile = ""
            split = filename.split('_')
            if len(split) > 1:
                for i in range(len(split) - 1):
                    if i == 0:
                        infofile = split[i]
                    else:
                        infofile += "_" + split[i]
                infofile = infofile + "_Infos_" +\
                            split[-1].replace('.mca', '.dat')

        if os.path.exists(infofile):
            info = self._getInfo(infofile)
            if ('vwidth' in info) and ('vstep' in info):
                vwidth = info['vwidth']
                vstep = info['vstep']
                if abs(vstep) > 0:
                    self.nRows = int((vwidth / vstep) + 1)

        #fill the header
        self.header = header

        #arrange as an EDF Stack
        self.info = {}
        self.__nFiles = 1

        self.__nImagesPerFile = 1

        #self.nRows = 41
        self.nCols = self.nSpectra / self.nRows

        self.data = numpy.zeros((self.nRows,
                                 self.nCols,
                                 self.nChannels),
                                 numpy.float32)
        n = 0
        for i in range(self.nRows):
            for j in range(self.nCols):
                s = data[n]
                spectrum = numpy.array([float(x) for x in exp.findall(s)])
                self.data[i, j, :] = spectrum[:]
                n = n + 1

        shape = self.data.shape
        for i in range(len(shape)):
            key = 'Dim_%d' % (i + 1,)
            self.info[key] = shape[i]

        self.info["SourceType"] = SOURCE_TYPE
        self.info["SourceName"] = self.sourceName
        self.info["Size"]       = self.__nFiles * self.__nImagesPerFile
        self.info["NumberOfFiles"] = self.__nFiles * 1
        self.info["FileIndex"] = 0
        self.info["McaCalib"] = [0.0, 1.0, 0.0]
        self.info["Channel0"] = 0.0

    def _getInfo(self, filename):
        '''
        This dictionnary is to be internally normalized for the time
        being no I0 nor dead time
        '''
        exp = re.compile('(-?[0-9]+\.?[0-9]*)')
        #read the file in one go to minimize access to disk
        with open(filename) as f:
            data = f.readlines()
        ddict = {}
        for line in data:
            if line.startswith("# Horizontal center position"):
                ddict['center'] = [float(x) for x in exp.findall(line)][0]
            elif line.startswith("# Horizontal width"):
                ddict['hwidth'] = [float(x) for x in exp.findall(line)][0]
            elif line.startswith("# Horizontal step"):
                ddict['hstep'] = [float(x) for x in exp.findall(line)][0]
            elif line.startswith("# Vertical width"):
                ddict['vwidth'] = [float(x) for x in exp.findall(line)][0]
            elif line.startswith("# Vertical step"):
                ddict['vstep'] = [float(x) for x in exp.findall(line)][0]
        return ddict


def main():
    filename = None
    if len(sys.argv) > 1:
        filename = sys.argv[1]
    elif os.path.exists("S10S_6_01.mca"):
        filename = "S10S_6_01.mca"
    if filename is not None:
        w = LuciaMap(filename)
        print(w.info)
    else:
        print("Please supply input filename")

if __name__ == "__main__":
    main()
