1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178
|
#!/usr/bin/env python
#/*##########################################################################
#
# The PyMca X-Ray Fluorescence Toolkit
#
# Copyright (c) 2004-2015 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"
__doc__ = """
This module reads Renishaw Raman MAP measurements exported as .txt file.
The native .wxd file has to be converted to .txt using Renishaw tools.
The output ASCII file has no headers. It contains several columns, their meaning
being different according to the type of data. Some possibilities:
columns
wavelength value
z wavelength value
time wavelength value
y x wavelength value
Of all the previous possibilities, this module only supports the last one.
"""
import os
import sys
import re
import numpy
from PyMca5.PyMcaCore import DataObject
SOURCE_TYPE = "EdfFileStack"
class RenishawMap(DataObject.DataObject):
def __init__(self, filename, infofile=None):
DataObject.DataObject.__init__(self)
# allocate buffers greater than any spected number of channels
x = numpy.zeros((10000,), numpy.float32)
y = numpy.zeros((10000,), numpy.float32)
wl = numpy.zeros((10000,), numpy.float32)
data = numpy.zeros((10000,), numpy.float32)
#import time
#t0 = time.time()
with open(filename, 'r') as f:
nChannels = 0
for line in f:
# read the first line
y[nChannels], x[nChannels], wl[nChannels], data[nChannels] = \
[float(value) for value in line.split("\t")]
if nChannels == 0:
firstChannel = wl[0]
elif wl[nChannels] == firstChannel:
break
nChannels += 1
if y[0] == y[nChannels]:
firstChangesFirst = False
else:
firstChangesFirst = True
nLines = nChannels + 1 + sum([1 for l in f])
#print("ELAPSED 0 = ", time.time() - t0)
if nLines % nChannels:
raise IOError("Not a regular Renishaw map or a not a complete file")
nSpectra = int(nLines / nChannels)
#print("N Channels = %d" % nChannels)
#print("N Lines = %d" % nLines)
#print("N Spectra = %f" % (nLines/nChannels))
rows = numpy.zeros((nSpectra,), numpy.float32)
columns = numpy.zeros((nSpectra,), numpy.float32)
wl = numpy.array(wl[:nChannels], dtype=numpy.float32)
data = numpy.zeros((nSpectra, nChannels), numpy.float32)
nRows = 0
nColumns = 0
#t0 = time.time()
myFloat = numpy.float32
actualRows = []
actualColumns = []
indices = [None] * nSpectra
with open(filename, 'r') as f:
for i in range(nSpectra):
for j in range(nChannels):
line = f.readline()
if firstChangesFirst:
column, row, dummy1, dummy2 = \
[float(value) for value in line.split("\t")]
else:
row, column, dummy1, dummy2 = \
[float(value) for value in line.split("\t")]
#positions.append((row, column, i))
if row not in actualRows:
actualRows.append(row)
if column not in actualColumns:
actualColumns.append(column)
indices[i] = (actualRows.index(row), actualColumns.index(column))
nRows = len(actualRows)
nColumns = len(actualColumns)
#positions.sort()
data.shape = nRows, nColumns, nChannels
with open(filename, 'r') as f:
for i in range(nSpectra):
row, column = indices[i]
for j in range(nChannels):
line = f.readline()
d1, d2, d3, data[row, column, j] = \
[myFloat(value) for value in line.split("\t")]
#print "nRows = ", nRows
#print "nColumns= ", nColumns
#print columns[::nColumns]
#print rows[::nRows]
#print "product = ", nRows * nColumns
#print("ELAPSED tinal = ", time.time() - t0)
# arrange as EDF stack
self.sourceName = filename
self.data = data
self.info = {}
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"] = 1
self.info["NumberOfFiles"] = 1
self.info["FileIndex"] = 0
self.info["McaIndex"] = 2
self.info["McaCalib"] = [0.0, 1.0, 0.0]
self.info["Channel0"] = 0.0
self.x = [wl]
def isRenishawMapFile(filename):
try:
if filename.endswith(".txt"):
with open(filename, 'r') as f:
line = f.readline().split("\t")
y, x, wl, data = [float(item) for item in line]
return True
except:
# it is not a Renishaw map file
pass
return False
if __name__ == "__main__":
filename = None
if len(sys.argv) > 1:
filename = sys.argv[1]
print("is Renishaw File?", isRenishawMapFile(filename))
instance = RenishawMap(filename)
print(instance.info)
print(instance.data.size)
|