def getFileInfo(self, filename):
'''
return general information of the given [filename], like it's size
'''
f = QtCore.QFileInfo(filename)
size = f.size()
size_type = ['Byte', 'kB', 'MB', 'GB', 'TB', 'PB']
size_index = 0
while size > 1024:
size /= 1024.0
size_index += 1
size = "%.3g %s" % (round(size, 3), size_type[size_index])
n = f.fileName()
return '''File name:\t%s
Folder:\t%s
File type:\t%s
File size:\t%s
Last changed:\t%s''' % (n, f.filePath()[:-len(n)], f.suffix(), size,
f.lastModified().toString())
#TODO: not used at the moment
#but useful
# def getMetaData(self, filename):
# '''
# Find and format the meta data of [filename]
# '''
# text = ""
# if filename.filetype().lower() not in ('tif', 'tiff'):
# #for all files except TIF
# filename, realname = unicodeFilename(filename), filename
# parser = createParser(filename, realname)
# if not parser:
# print "Unable to parse file"
# try:
# metadata = extractMetadata(parser)
# except HachoirError, err:
# #AttributeError because hachoir uses
# print "Metadata extraction error: %s" % unicode(err)
# metadata = None
# if not metadata:
# print "Unable to extract metadata"
# else:
# text = "\n".join(metadata.exportPlaintext())
# else:
# #for TIF images
# import exifread
# # Open image file for reading (binary mode)
# f = open(filename, 'rb')
# # Return Exif tags
# tag_dict = exifread.process_file(f)
# for key, value in tag_dict.iteritems():
# text += "%s: %s\n" %(key, value)
# return text
def open(self, filename):
prefs = self.preferences
#VARIABLES:
buff = prefs.pBuffer.value()
step = prefs.pReadEveryNLine.value()
stop_at_line = prefs.pHasStopLine.value()
stop_n = prefs.pStopLine.value()
x_col = prefs.pXColumn.value()
has_x_col = prefs.pHasXColumn.value()
self.n_line = 0 #number of current line
step_len = 0 #number of lines in the read file_part
n_defective_lines = 0 # number of corrupted lines
i = 0 # current index in data array
at_end = False # is file the the end
labels = filename
#COUNT LINES:
if stop_at_line:
self.nLines = stop_n
else:
modified = QtCore.QFileInfo(filename).lastModified().toString()
#ONLY OF THIS IS A NEW FILE OR THE FILE WAS MODIFIED:
if not (self._lastFilename == filename
and self._lastModified == modified):
self.nLines = countLines(filename, prefs.pBuffer.value())
self._lastnLines = self.nLines
self._lastFilename = filename
self._lastModified = modified
else:
self.nLines = self._lastnLines
#GET NUMBER OF COLUMNS:
with open(filename, 'r') as f:
#FIND SEPARATOR:
if prefs.pFindSeparator.value():
prefs.pSeparator.setValue(self._findSeparator(f))
separator = prefs.separators[prefs.pSeparator.value()]
#DEFINE NUMBER OF COLUMNS:
n_col = f.readline().count(separator) + 1
f.seek(0) #back to begin
#FILTER COLUMNS:
filter_columns = prefs.pFilterColumns.value()
if filter_columns:
c = [ch.value()
for ch in prefs.pFilterColumns.children()[1:]][:n_col]
#ADD X COLUMN TO THE BEGINNING:
if has_x_col:
if x_col not in c:
c.insert(0, x_col)
n_col = len(c)
col_filter = itemgetter(*c)
elif n_col > 1:
col_filter = lambda l: l[:n_col]
else:
col_filter = lambda l: l[0]
#GET FIRST LINE NAMES:
fline = prefs.pFirstLine.value()
if fline != '-':
names = col_filter(f.readline().split(separator))
if fline == 'axes names':
for n, ax in enumerate(self.display.axes):
ax.p.setValue(names[n])
else: #plot names:
labels = names
if has_x_col:
labels.pop(x_col)
#JUMP TO START POSITION:
startline = prefs.pStartLine.value()
if startline:
f.readline(startline)
self.n_line = startline
#PRINT FIRST 10 LINES:
if prefs.pPrintFirstLines.value():
startPos = f.tell()
print '=========='
for line in f.readlines(min(self.nLines, 10)):
#TODO: do the check for \n only once
# ??? doesn't have every line \n at the end??
#print line[-2:]
#if line[-2:] == '\n':
# line = line[:-2]
print '[%s]: %s' % (self.n_line, line)
print '--> %s\n' % str(col_filter(line.split(separator)))
f.seek(startPos)
print '=========='
#CREATE DATA ARRAY:
shape = self.nLines / step
if n_col == 0:
raise Exception('no columns given')
elif n_col > 1:
shape = (shape, n_col)
data = np.empty(shape=shape,
dtype=prefs.dtypes[prefs.pDType.value()])
#MAIN LOOP:
while not self.canceled:
#READ PART OF THE FILE:
file_piece = f.readlines(buff)
l = len(file_piece)
if not l:
break
for n, line in enumerate(file_piece[::step]):
#line = line[:-1]
#FILTER COLUMNS:
line = col_filter(line.split(separator))
self.n_line = n + step_len
#ADD LINE TO ARRAY:
try:
data[i] = line
i += 1
except ValueError:
n_defective_lines += 1
#CHECK BREAK CRITERIA:
if stop_at_line and self.n_line >= stop_n:
at_end = True
break
if at_end:
break
step_len += l
data = data[:i]
print '%s lines were corrupted' % n_defective_lines
#SPLIT ARRAY IF NEEDED:
if (has_x_col and n_col > 2) or (not has_x_col and n_col > 1):
if has_x_col:
x = data[:, x_col]
#GET Y COLUMNS THROUGH REMOVING THE X COLUMN:
y_cols = np.c_[data[:, :x_col], data[:, x_col + 1:]]
else:
y_cols = data
l = []
#CREATE TUPLE OF [y_n] OR [x, y_n] arrays:
for n in range(y_cols.shape[1]):
y = y_cols[:, n]
if has_x_col:
y = np.c_[x, y]
l.append(y)
return tuple(l), labels
return [data], labels