-
Notifications
You must be signed in to change notification settings - Fork 122
/
LoadAscii.cpp
439 lines (398 loc) · 15.9 KB
/
LoadAscii.cpp
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
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
/*WIKI*
The LoadAscii algorithm reads in spectra data from a text file and stores it in a [[Workspace2D]] as data points. The data in the file must be organized in columns separated by commas, tabs, spaces, colons or semicolons. Only one separator type can be used throughout the file; use the "Separator" property to tell the algorithm which to use. The algorithm [[SaveAscii]] is normally able to produce such a file.
By default the algorithm attempts to guess which lines are header lines by trying to see where a contiguous block of numbers starts. This can be turned off by specifying the "SkipNumLines" property, which will then tell the algorithm to simply use that as the the number of header lines.
The format can be one of:
* Two columns: 1st column=X, 2nd column=Y, E=0
* For a workspace of ''n'' spectra, 2''n''+1 columns: 1''st'' column=X, 2i''th'' column=Y, 2i+1''th'' column =E
* Four columns: 1st column=X, 2nd column=Y, 3rd column=E, 4th column=DX (X error)
The number of bins is defined by the number of rows.
The resulting workspace will have common X binning for all spectra.
This algorithm cannot load a file created by [[SaveAscii]] if it has X errors written and several spectra.
*WIKI*/
//----------------------------------------------------------------------
// Includes
//----------------------------------------------------------------------
#include "MantidDataHandling/LoadAscii.h"
#include "MantidDataObjects/Workspace2D.h"
#include "MantidKernel/UnitFactory.h"
#include "MantidAPI/FileProperty.h"
#include "MantidAPI/RegisterFileLoader.h"
#include "MantidKernel/BoundedValidator.h"
#include "MantidKernel/ListValidator.h"
#include <fstream>
#include <boost/tokenizer.hpp>
#include <Poco/StringTokenizer.h>
// String utilities
#include <boost/algorithm/string.hpp>
namespace Mantid
{
namespace DataHandling
{
DECLARE_FILELOADER_ALGORITHM(LoadAscii);
/// Sets documentation strings for this algorithm
void LoadAscii::initDocs()
{
this->setWikiSummary("Loads data from a text file and stores it in a 2D [[workspace]] ([[Workspace2D]] class). ");
this->setOptionalMessage("Loads data from a text file and stores it in a 2D workspace (Workspace2D class).");
}
using namespace Kernel;
using namespace API;
/// Empty constructor
LoadAscii::LoadAscii() : m_columnSep(), m_separatorIndex()
{
}
/**
* Return the confidence with with this algorithm can load the file
* @param descriptor A descriptor for the file
* @returns An integer specifying the confidence level. 0 indicates it will not be used
*/
int LoadAscii::confidence(Kernel::FileDescriptor & descriptor) const
{
const std::string & filePath = descriptor.filename();
const size_t filenameLength = filePath.size();
// Avoid some known file types that have different loaders
int confidence(0);
if( filePath.compare(filenameLength - 12,12,"_runinfo.xml") == 0 ||
filePath.compare(filenameLength - 6,6,".peaks") == 0 ||
filePath.compare(filenameLength - 10,10,".integrate") == 0 )
{
confidence = 0;
}
else if(descriptor.isAscii())
{
confidence = 9; // Low so that others may try but not stopping version 2
}
return confidence;
}
/**
* Check if a file is a text file
* @param file :: The file pointer
* @returns true if the file an ascii text file, false otherwise
*/
bool LoadAscii::isAscii(FILE *file)
{
char data[256];
char *pend = &data[fread(data, 1, sizeof(data), file)];
fseek(file,0,SEEK_SET);
/*
* Call it a binary file if we find a non-ascii character in the
* first 256 bytes of the file.
*/
for( char *p = data; p < pend; ++p )
{
unsigned long ch = (unsigned long)*p;
if( !(ch <= 0x7F) )
{
return false;
}
}
return true;
}
//--------------------------------------------------------------------------
// Protected methods
//--------------------------------------------------------------------------
/**
* Process the header information. This implementation just skips it entirely.
* @param file :: A reference to the file stream
*/
void LoadAscii::processHeader(std::ifstream & file) const
{
// Most files will have some sort of header. If we've haven't been told how many lines to
// skip then try and guess
int numToSkip = getProperty("SkipNumLines");
if( numToSkip == EMPTY_INT() )
{
const int rowsToMatch(5);
// Have a guess where the data starts. Basically say, when we have say "rowsToMatch" lines of pure numbers
// in a row then the line that started block is the top of the data
int numCols(-1), matchingRows(0), row(0);
std::string line;
std::vector<double> values;
while( getline(file,line) )
{
++row;
//int nchars = (int)line.length(); TODO dead code?
boost::trim(line);
if( this->skipLine(line) )
{
continue;
}
std::list<std::string> columns;
int lineCols = this->splitIntoColumns(columns, line);
try
{
fillInputValues(values, columns);
}
catch(boost::bad_lexical_cast&)
{
continue;
}
if( numCols < 0 ) numCols = lineCols;
if( lineCols == numCols )
{
++matchingRows;
if( matchingRows == rowsToMatch ) break;
}
else
{
numCols = lineCols;
matchingRows = 1;
}
}
// if the file does not have more than rowsToMatch + skipped lines, it will stop
// and raise the EndOfFile, this may cause problems for small workspaces.
// In this case clear the flag
if (file.eof()){
file.clear(file.eofbit);
}
// Seek the file pointer back to the start.
// NOTE: Originally had this as finding the stream position of the data and then moving the file pointer
// back to the start of the data. This worked when a file was read on the same platform it was written
// but failed when read on a different one due to underlying differences in the stream translation.
file.seekg(0,std::ios::beg);
// We've read the header plus the number of rowsToMatch
numToSkip = row - rowsToMatch;
}
int i(0);
std::string line;
while( i < numToSkip && getline(file, line) )
{
++i;
}
g_log.information() << "Skipped " << numToSkip << " line(s) of header information()\n";
}
/**
* Reads the data from the file. It is assumed that the provided file stream has its position
* set such that the first call to getline will be give the first line of data
* @param file :: A reference to a file stream
* @returns A pointer to a new workspace
*/
API::Workspace_sptr LoadAscii::readData(std::ifstream & file) const
{
// Get the first line and find the number of spectra from the number of columns
std::string line;
getline(file,line);
boost::trim(line);
std::list<std::string> columns;
const int numCols = splitIntoColumns(columns, line);
if( numCols < 2 )
{
g_log.error() << "Invalid data format found in file \"" << getPropertyValue("Filename") << "\"\n";
throw std::runtime_error("Invalid data format. Fewer than 2 columns found.");
}
size_t numSpectra(0);
bool haveErrors(false);
bool haveXErrors(false);
// Assume single data set with no errors
if( numCols == 2 )
{
numSpectra = numCols/2;
}
// Data with errors
else if( (numCols-1) % 2 == 0 )
{
numSpectra = (numCols - 1)/2;
haveErrors = true;
}
// Data with errors on both X and Y (4-column file)
else if( numCols == 4 )
{
numSpectra = 1;
haveErrors = true;
haveXErrors = true;
}
else
{
g_log.error() << "Invalid data format found in file \"" << getPropertyValue("Filename") << "\"\n";
g_log.error() << "LoadAscii requires the number of columns to be an even multiple of either 2 or 3.";
throw std::runtime_error("Invalid data format.");
}
// A quick check at the number of lines won't be accurate enough as potentially there
// could be blank lines and comment lines
int numBins(0), lineNo(0);
std::vector<DataObjects::Histogram1D> spectra(numSpectra);
std::vector<double> values(numCols, 0.);
do
{
++lineNo;
boost::trim(line);
if( this->skipLine(line) ) continue;
columns.clear();
int lineCols = this->splitIntoColumns(columns, line);
if( lineCols != numCols )
{
std::ostringstream ostr;
ostr << "Number of columns changed at line " << lineNo;
throw std::runtime_error(ostr.str());
}
try
{
fillInputValues(values, columns); //ignores nans and replaces them with 0
}
catch(boost::bad_lexical_cast&)
{
g_log.error() << "Invalid value on line " << lineNo << " of \""
<< getPropertyValue("Filename") << "\"\n";
throw std::runtime_error("Invalid value encountered.");
}
for (size_t i = 0; i < numSpectra; ++i)
{
spectra[i].dataX().push_back(values[0]);
spectra[i].dataY().push_back(values[i*2+1]);
if( haveErrors )
{
spectra[i].dataE().push_back(values[i*2+2]);
}
if( haveXErrors )
{
// Note: we only have X errors with 4-column files.
// We are only here when i=0.
spectra[i].dataDx().push_back(values[3]);
}
}
++numBins;
}
while(getline(file,line));
MatrixWorkspace_sptr localWorkspace = boost::dynamic_pointer_cast<MatrixWorkspace>
(WorkspaceFactory::Instance().create("Workspace2D",numSpectra,numBins,numBins));
try
{
localWorkspace->getAxis(0)->unit() = UnitFactory::Instance().create(getProperty("Unit"));
}
catch (Exception::NotFoundError&)
{
// Asked for dimensionless workspace (obviously not in unit factory)
}
for (size_t i = 0; i < numSpectra; ++i)
{
localWorkspace->dataX(i) = spectra[i].dataX();
localWorkspace->dataY(i) = spectra[i].dataY();
/* If Y or E errors are not there, DON'T copy across as the 'spectra' vectors
have not been filled above. The workspace will by default have vectors of
the right length filled with zeroes. */
if ( haveErrors ) localWorkspace->dataE(i) = spectra[i].dataE();
if ( haveXErrors ) localWorkspace->dataDx(i) = spectra[i].dataDx();
// Just have spectrum number start at 1 and count up
localWorkspace->getSpectrum(i)->setSpectrumNo(static_cast<specid_t>(i)+1);
}
return localWorkspace;
}
/**
* Peek at a line without extracting it from the stream
*/
void LoadAscii::peekLine(std::ifstream & is, std::string & str) const
{
getline(is, str);
is.seekg(-(int)str.length(),std::ios::cur);
boost::trim(str);
}
/**
* Return true if the line is to be skipped.
* @param line :: The line to be checked
* @return True if the line should be skipped
*/
bool LoadAscii::skipLine(const std::string & line) const
{
// Empty or comment
return ( line.empty() || boost::starts_with(line, "#") );
}
/**
* Split the data into columns based on the input separator
* @param[out] columns :: A reference to a list to store the column data
* @param[in] str :: The input string
* @returns The number of columns
*/
int LoadAscii::splitIntoColumns(std::list<std::string> & columns, const std::string & str) const
{
boost::split(columns, str, boost::is_any_of(m_columnSep), boost::token_compress_on);
return static_cast<int>(columns.size());
}
/**
* Fill the given vector with the data values. Its size is assumed to be correct
* @param[out] values :: The data vector fill
* @param columns :: The list of strings denoting columns
*/
void LoadAscii::fillInputValues(std::vector<double> &values,
const std::list<std::string>& columns) const
{
values.resize(columns.size());
std::list<std::string>::const_iterator iend = columns.end();
int i = 0;
for( std::list<std::string>::const_iterator itr = columns.begin();
itr != iend; ++itr )
{
std::string value = *itr;
boost::trim(value);
boost::to_lower(value);
if (value == "nan"|| value == "1.#qnan") //ignores nans (not a number) and replaces them with a nan
{
double nan = std::numeric_limits<double>::quiet_NaN();//(0.0/0.0);
values[i] = nan;
}
else
{
values[i] = boost::lexical_cast<double>(value);
}
++i;
}
}
//--------------------------------------------------------------------------
// Private methods
//--------------------------------------------------------------------------
/// Initialisation method.
void LoadAscii::init()
{
std::vector<std::string> exts;
exts.push_back(".dat");
exts.push_back(".txt");
exts.push_back(".csv");
exts.push_back("");
declareProperty(new FileProperty("Filename", "", FileProperty::Load, exts),
"The name of the text file to read, including its full or relative path. The file extension must be .tst, .dat, or .csv");
declareProperty(new WorkspaceProperty<Workspace>("OutputWorkspace",
"",Direction::Output), "The name of the workspace that will be created, filled with the read-in data and stored in the [[Analysis Data Service]].");
std::string spacers[6][6] = { {"Automatic", ",\t:; "}, {"CSV", ","},
{"Tab", "\t"}, {"Space", " "}, {"Colon", ":"}, {"SemiColon", ";"} };
// For the ListValidator
std::vector<std::string> sepOptions;
for( size_t i = 0; i < 5; ++i )
{
std::string option = spacers[i][0];
m_separatorIndex.insert(std::pair<std::string,std::string>(option, spacers[i][1]));
sepOptions.push_back(option);
}
declareProperty("Separator", "Automatic", boost::make_shared<StringListValidator>(sepOptions),
"The separator between data columns in the data file. The possible values are \"CSV\", \"Tab\", "
"\"Space\", \"SemiColon\", or \"Colon\" (default: Automatic selection).");
std::vector<std::string> units = UnitFactory::Instance().getKeys();
units.insert(units.begin(),"Dimensionless");
declareProperty("Unit","Energy", boost::make_shared<StringListValidator>(units),
"The unit to assign to the X axis (anything known to the [[Unit Factory]] or \"Dimensionless\")");
auto mustBePosInt = boost::make_shared<BoundedValidator<int> >();
mustBePosInt->setLower(0);
declareProperty("SkipNumLines", EMPTY_INT(), mustBePosInt,
"If given, skip this number of lines at the start of the file.");
}
/**
* Executes the algorithm.
*/
void LoadAscii::exec()
{
std::string filename = getProperty("Filename");
std::ifstream file(filename.c_str());
if (!file)
{
g_log.error("Unable to open file: " + filename);
throw Exception::FileError("Unable to open file: " , filename);
}
std::string sepOption = getProperty("Separator");
m_columnSep = m_separatorIndex[sepOption];
// Process the header information.
processHeader(file);
// Read the data
MatrixWorkspace_sptr outputWS = boost::dynamic_pointer_cast<MatrixWorkspace>(readData(file));
outputWS->mutableRun().addProperty("Filename",filename);
setProperty("OutputWorkspace", outputWS);
}
} // namespace DataHandling
} // namespace Mantid