DAMASK_EICMD/python/damask/table.py

190 lines
5.7 KiB
Python

import re
import pandas as pd
import numpy as np
class Table():
"""Store spreadsheet-like data."""
def __init__(self,array,headings,comments=None):
"""
New spreadsheet data.
Parameters
----------
array : numpy.ndarray
Data.
headings : dict
Column headings. Labels as keys and shape as tuple. Example 'F':(3,3) for a deformation gradient.
comments : iterable of str, optional
Additional, human-readable information
"""
self.data = pd.DataFrame(data=array)
d = {}
i = 0
for label in headings:
for components in range(np.prod(headings[label])):
d[i] = label
i+=1
if i != self.data.shape[1]:
raise IndexError('Mismatch between array shape and headings')
self.data.rename(columns=d,inplace=True)
if comments is None:
self.comments = []
else:
self.comments = [c for c in comments]
self.headings = headings
@staticmethod
def from_ASCII(fname):
"""
Create table from ASCII file.
The first line needs to indicate the number of subsequent header lines as 'n header'.
Vector data labels are indicated by '1_v, 2_v, ..., n_v'.
Tensor data labels are indicated by '3x3:1_T, 3x3:2_T, ..., 3x3:9_T'.
Parameters
----------
fname : file, str, or pathlib.Path
Filename or file for reading.
"""
try:
f = open(fname)
except TypeError:
f = fname
header,keyword = f.readline().split()
if keyword == 'header':
header = int(header)
else:
raise Exception
comments = [f.readline()[:-1] for i in range(header-1)]
labels = f.readline().split()
headings = {}
for label in labels:
tensor_column = re.search(r'[0-9,x]*?:[0-9]*?_',label)
if tensor_column:
my_shape = tensor_column.group().split(':',1)[0].split('x')
headings[label.split('_',1)[1]] = tuple([int(d) for d in my_shape])
else:
vector_column = re.match(r'[0-9]*?_',label)
if vector_column:
headings[label.split('_',1)[1]] = (int(label.split('_',1)[0]),)
else:
headings[label]=(1,)
return Table(np.loadtxt(f),headings,comments)
def get_array(self,label):
"""
Return data as array.
Parameters
----------
label : str
Label of the array.
"""
if re.match(r'[0-9]*?_',label):
idx,key = label.split('_',1)
return self.data[key].to_numpy()[:,int(idx)-1]
else:
return self.data[label].to_numpy().reshape((-1,)+self.headings[label])
def set_array(self,label,array,info):
"""
Modify data in the spreadsheet.
Parameters
----------
label : str
Label for the new data.
array : np.ndarray
New data.
info : str
Human-readable information about the new data.
"""
if np.prod(array.shape[1:],dtype=int) == 1:
self.comments.append('{}: {}'.format(label,info))
else:
self.comments.append('{} {}: {}'.format(label,array.shape[1:],info))
if re.match(r'[0-9]*?_',label):
idx,key = label.split('_',1)
iloc = self.data.columns.get_loc(key).tolist().index(True) + int(idx) -1
self.data.iloc[:,iloc] = array
else:
self.data[label] = array.reshape(self.data[label].shape)
def get_labels(self):
"""Return the labels of all columns."""
return [label for label in self.headings]
def add_array(self,label,array,info):
"""
Add data to the spreadsheet.
Parameters
----------
label : str
Label for the new data.
array : np.ndarray
New data.
info : str
Human-readable information about the new data.
"""
if np.prod(array.shape[1:],dtype=int) == 1:
self.comments.append('{}: {}'.format(label,info))
else:
self.comments.append('{} {}: {}'.format(label,array.shape[1:],info))
self.headings[label] = array.shape[1:] if len(array.shape) > 1 else (1,)
size = np.prod(array.shape[1:],dtype=int)
new_data = pd.DataFrame(data=array.reshape(-1,size),
columns=[label for l in range(size)])
self.data = pd.concat([self.data,new_data],axis=1)
def to_ASCII(self,fname):
"""
Store as plain text file.
Parameters
----------
fname : file, str, or pathlib.Path
Filename or file for reading.
"""
labels = []
for l in self.headings:
if(self.headings[l] == (1,)):
labels.append('{}'.format(l))
elif(len(self.headings[l]) == 1):
labels+=['{}_{}'.format(i+1,l)\
for i in range(self.headings[l][0])]
else:
labels+=['{}:{}_{}'.format('x'.join([str(d) for d in self.headings[l]]),i+1,l)\
for i in range(np.prod(self.headings[l],dtype=int))]
header = ['{} header'.format(len(self.comments)+1)]\
+ self.comments\
+ [' '.join(labels)]
try:
f = open(fname,'w')
except TypeError:
f = fname
for line in header: f.write(line+'\n')
self.data.to_csv(f,sep=' ',index=False,header=False)