Related
I'm using prince to perform Factor Analysis of Mixed Data (FAMD). During training everything goes as planned, but when I try to transform new data, I get an error (shapes not aligned:(dim 1) != (dim 0)).
I've seen people claiming to have this problem before and they claim the problem is that they're trying to transform data from a test set, but I tried different approaches and I still get this error. Let me describe my situation:
My dataset has some 500k rows and some 20 columns.
Originally, I separated the dataset into a train(80%) and test(20%) and everything worked fine
The problem appears when I try to transform a small subset of the data
When I try transforming less than 100 rows, I get the aforementioned error
Between 100 and 2000 rows, there's no error, but the output is all nan
Over 2000 rows it works properly
Does anyone know how to fix this? I think it may be some issue with matrix rank, but right now I'm clueless
Here's the error message
ValueError Traceback (most recent call
last) in
----> 7 famd.transform(c)
\site-packages\prince\mfa.py in
transform(self, X)
135 def transform(self, X):
136 """Returns the row principal coordinates of a dataset."""
--> 137 return self.row_coordinates(X)
138
139 def _row_coordinates_from_global(self, X_global):
\site-packages\prince\mfa.py in
row_coordinates(self, X)
152 X = self._prepare_input(X)
153
--> 154 return self._row_coordinates_from_global(self._build_X_global(X))
155
156 def row_contributions(self, X):
\site-packages\prince\mfa.py in
_row_coordinates_from_global(self, X_global)
139 def _row_coordinates_from_global(self, X_global):
140 """Returns the row principal coordinates."""
--> 141 return len(X_global) ** 0.5 * super().row_coordinates(X_global)
142
143 def row_coordinates(self, X):
\site-packages\prince\pca.py in
row_coordinates(self, X)
136 X = self.scaler_.transform(X)
137
--> 138 return pd.DataFrame(data=X.dot(self.V_.T), index=index, dtype=np.float64)
139
140 def row_standard_coordinates(self, X):
ValueError: shapes (100,41) and (43,10) not aligned: 41 (dim 1) != 43
(dim 0)
You'll have to pardon the rather large-ish block of code, but I believe this is a near-minimal reproduction of my problem. The problem is not isolated to example.com but persists across many other sites.
If I have 4 threads actively making network requests, curl works 100% fine.
If I add one more thread, that thread takes ~10x the time to execute. I feel like I must be missing something obvious, but it escapes me right now.
UPDATE with more information: These tests are in a virtual machine. Independent of the number of cores available to the machine, four of the requests take ~100ms, and the remainder take ~5500ms.
UPDATE 2: Actually, I was wrong in one aspect, it isn't always 4 / n-4 distribution -- when I've changed to 4 cores, sometimes I get a different result distribution (running on 1 core at least seemed relatively consistent) - here's a snippet of the results when the threads return their latency (ms) instead of their http code when running on a 4 core VM:
191 191
198 198 167
209 208 202 208
215 207 214 209 209
5650 213 5649 222 193 207
206 201 164 205 201 201 205
5679 5678 5666 5678 216 173 205 175
5691 212 179 206 5685 5688 211 5691 5680
5681 199 210 5678 5663 213 5679 212 5666 428
UPDATE 3: I built curl and openssl from scratch, removed the locking (as openssl 1.1.0g doesn't require it) and the problem persists. ( Sanity check / verified by the following):
std::cout << "CURL:\n " << curl_version_info(CURLVERSION_NOW)->ssl_version
<< "\n";
std::cout << "SSLEAY:\n " << SSLeay_version(SSLEAY_VERSION) << "\n";
outputting:
CURL:
OpenSSL/1.1.0g
SSLEAY:
OpenSSL 1.1.0g 2 Nov 2017
With example latencies:
191 191
197 197 196
210 210 201 210
212 212 199 200 165
5656 5654 181 214 181 212
5653 5651 5647 211 206 205 162
5681 5674 5669 165 201 204 201 5681
5880 5878 5657 5662 197 209 5664 173 174
5906 5653 5664 5905 5663 173 5666 173 165 204
UPDATE 4: Setting CURLOPT_CONNECTTIMEOUT_MS equal to x makes x an upper limit on the time it takes to return.
UPDATE 5, MOST IMPORTANT:
Running the program under strace -T ./a.out 2>&1 | vim - with 5 threads, when the program only had 1 slow request, yielded two very slow lines. It was two calls to the same futex, one took way longer than the second, but both took longer than all other futex calls (most were 0.000011 ms, these two calls took 5.4 and 0.2 seconds to unlock).
Additionally, I verified that the slow-ness was entirely in curl_easy_perform.
futex(0x7efcb66439d0, FUTEX_WAIT, 3932, NULL) = 0 <5.390086>
futex(0x7efcb76459d0, FUTEX_WAIT, 3930, NULL) = 0 <0.204908>
Finally, after some looking around in source code, I found that the bug is somewhere in the DNS lookup. Replacing hostnames with IP addresses is a bandaid over the problem, wherever or whatever it is.
-----------
Below is my minimal reproduction / distillation of the issue, compiled with g++ -lpthread -lcurl -lcrypto main.cc, linked to the versions of openssl and libcurl built from source.
#include <chrono>
#include <iomanip>
#include <iostream>
#include <thread>
#include <vector>
#include <curl/curl.h>
#include <openssl/crypto.h>
size_t NoopWriteFunction(void *buffer, size_t size, size_t nmemb, void *userp) {
return size * nmemb;
};
int GetUrl() {
CURL *hnd = curl_easy_init();
curl_easy_setopt(hnd, CURLOPT_URL, "https://www.example.com/");
curl_easy_setopt(hnd, CURLOPT_HEADERFUNCTION, NoopWriteFunction);
curl_easy_setopt(hnd, CURLOPT_WRITEFUNCTION, NoopWriteFunction);
curl_easy_setopt(hnd, CURLOPT_SSH_KNOWNHOSTS, "/home/web/.ssh/known_hosts");
CURLcode ret = curl_easy_perform(hnd);
long http_code = 0;
curl_easy_getinfo(hnd, CURLINFO_RESPONSE_CODE, &http_code);
curl_easy_cleanup(hnd);
hnd = NULL;
if (ret != CURLE_OK) {
return -ret;
}
return http_code;
}
int main() {
curl_global_init(CURL_GLOBAL_ALL);
for (int i = 1; i < 10; i++) {
std::vector<std::thread> threads;
int response_code[10]{};
auto clock = std::chrono::high_resolution_clock();
auto start = clock.now();
threads.resize(i);
for (int j = 0; j < i; j++) {
threads.emplace_back(std::thread(
[&response_code](int x) { response_code[x] = GetUrl(); }, j));
}
for (auto &t : threads) {
if (t.joinable()) {
t.join();
}
}
auto end = clock.now();
int time_to_execute =
std::chrono::duration_cast<std::chrono::milliseconds>(end - start)
.count();
std::cout << std::setw(10) << time_to_execute;
for (int j = 0; j < i; j++) {
std::cout << std::setw(5) << response_code[j];
}
std::cout << "\n";
}
}
And when I run the program on my machine, I get the following result (I can change the domain to whatever, the results are ~ the same):
123 200
99 200 200
113 200 200 200
119 200 200 200 200
5577 200 200 200 200 200
5600 200 200 200 200 200 200
5598 200 200 200 200 200 200 200
5603 200 200 200 200 200 200 200 200
5606 200 200 200 200 200 200 200 200 200
And here is my curl version and openssl version:
$curl --version
curl 7.52.1 (x86_64-pc-linux-gnu) libcurl/7.52.1 OpenSSL/1.0.2l zlib/1.2.8 libidn2/0.16 libpsl/0.17.0 (+libidn2/0.16) libssh2/1.7.0 nghttp2/1.18.1 librtmp/2.3
Protocols: dict file ftp ftps gopher http https imap imaps ldap ldaps pop3 pop3s rtmp rtsp scp sftp smb smbs smtp smtps telnet tftp
Features: AsynchDNS IDN IPv6 Largefile GSS-API Kerberos SPNEGO NTLM NTLM_WB SSL libz TLS-SRP HTTP2 UnixSockets HTTPS-proxy PSL
$ openssl version
OpenSSL 1.1.0f 25 May 2017
The bug is somewhere in the DNS resolution, as indicated by my UPDATE 5.
It's something to do with the lookup for IPV6, somewhere in getaddrinfo.
Searching around indicates this is usually an ISP issue, or an over-aggressive-packet-filtering issue, combined with something else (what, I don't know) that makes this a really really odd edge case.
Following the instructions on this page leads to the following workaround / solution:
curl_easy_setopt(curl, CURLOPT_IPRESOLVE, CURL_IPRESOLVE_V4);
Which eliminates the issue as I've perceived it. IPV6 is hard. :(
If the http service is based on mongoose or CivetWeb, look at this answer
libcurl delays for 1 second before uploading data, command-line curl does not
The problem is that curl sends out Expect:100-continue in the header, but mongoose/civetweb does not respond to it. Curl times-out after 1000 ms and continues.
The answer above shows how to fix either curl or CivetWeb.
I'm writing an application in Python that applies a watermark to a font file. The glyphs are non-convex polygons with holes and consist of bezier splines defined in PostScript. The watermark needs to merge with the glyphs, not overlap. I was unable to find a library to do this in Python so I'm using CGAL & C++. I have it working nicely on most glyphs, but it's mysteriously failing on others.
Initially, I was very impressed with CGAL. It looked very comprehensive and sophisticated and seemed to provide all the functionality I needed, but it suffers one fatal flaw - and I almost can't believe it - the library contains no error handling. No error codes, no exceptions, just assertions - but only in the debug build. In the release build you get a nice segmentation fault instead. Furthermore, the assertions reveal nothing about the nature of the problem.
The program fails only on certain glyphs. The letter 'e' works fine, but it crashes on 'a'. The crash occurs when I attempt to compute the union of the glyph with the watermark. There is nothing visibly wrong with the 'a' glyph. My application parses the PostScript correctly and renders it fine on a QGraphicsView.
This program needs to run on a server and its output sent directly to the customer so it must be reliable. I can't recover from an assertion failure or a segfault, so what can I do?
Even if I get it working reliably, how can I trust that it will never fail? If there was some kind of error handling in place, I could just skip the few glyphs that it fails on, leaving them unwatermarked - not ideal, but acceptable. I just don't understand what the authors of this library were thinking; they went to such tremendous effort to make the most comprehensive geometry library available only to ensure that it's completely unfit for purpose.
Currently, it's looking like I'm going to have to modify the code myself to handle the error in a sensible way, but this just seems so ridiculous.
I'm sorry if I come off as impatient, but I'm way past my deadline and my client isn't going to care about or understand these excuses.
The assertion failure is occurring on line 2141 of multiset.h:
CGAL_multiset_precondition (comp_f(object, nodeP->object) != LARGER);
It happens when I call join() on a BezierPolygonSet. My types are as follows:
typedef CGAL::CORE_algebraic_number_traits NtTraits;
typedef NtTraits::Rational Rational;
typedef NtTraits::Algebraic Algebraic;
typedef CGAL::Cartesian<Rational> RatKernel;
typedef CGAL::Cartesian<Algebraic> AlgKernel;
typedef RatKernel::Point_2 BezierRatPoint;
typedef CGAL::Arr_Bezier_curve_traits_2<RatKernel, AlgKernel, NtTraits> Traits;
typedef Traits::Point_2 BezierPoint;
typedef Traits::Curve_2 BezierCurve;
typedef CGAL::Gps_traits_2<Traits> BezierTraits;
typedef BezierTraits::X_monotone_curve_2 BezierXMonotoneCurve;
typedef BezierTraits::General_polygon_2 BezierPolygon;
typedef BezierTraits::General_polygon_with_holes_2 BezierPolygonWithHoles;
typedef CGAL::Gps_default_dcel<BezierTraits> BezierDcelTraits;
typedef CGAL::General_polygon_set_2<BezierTraits, BezierDcelTraits> BezierPolygonSet;
Any help would be much appreciated. Thanks.
EDIT:
I have a module called Geometry which wraps the CGAL code and exposes a bunch of geometric primitives (Point, Curve, LineSegment, CubicBezier, Path) and the functions:
PathList toPathList(const PolyList& polyList);
PolyList toPolyList(const PathList& paths);
The Path class has a method called computeUnion, which looks like this:
PathList Path::computeUnion(const PathList& paths1, const PathList& paths2) {
PolyList polyList1 = toPolyList(paths1);
PolyList polyList2 = toPolyList(paths2);
cgal_wrap::BezierPolygonSet polySet;
for (auto i : polyList1) {
polySet.join(i);
}
for (auto i : polyList2) {
polySet.join(i);
}
PolyList polyList;
polySet.polygons_with_holes(std::back_inserter(polyList));
return toPathList(polyList);
}
The error occurs when I call join(). The polygons are created from paths like so:
PolyList toPolyList(const PathList& paths) {
cgal_wrap::Traits traits;
cgal_wrap::Traits::Make_x_monotone_2 fnMakeXMonotone = traits.make_x_monotone_2_object();
cgal_wrap::RatKernel ratKernel;
cgal_wrap::RatKernel::Equal_2 fnEqual = ratKernel.equal_2_object();
PolyList polyList; // The final polygons with holes
cgal_wrap::BezierPolygon outerPoly;
std::list<cgal_wrap::BezierPolygon> holes;
std::list<cgal_wrap::BezierXMonotoneCurve> monoCurves;
bool first = true;
cgal_wrap::BezierRatPoint firstPoint;
// For each path in the list
for (auto i = paths.begin(); i != paths.end(); ++i) {
const Path& path = *i;
cgal_wrap::BezierRatPoint prevEndPoint;
// For each curve in the path
for (auto j = path.begin(); j != path.end(); ++j) {
const Curve& curve = **j;
std::list<cgal_wrap::BezierRatPoint> points;
if (curve.type() == LineSegment::type) {
const LineSegment& lseg = dynamic_cast<const LineSegment&>(curve);
cgal_wrap::BezierRatPoint A = lseg.A();
if (j != path.begin()) {
if (A != prevEndPoint) {
// TODO
assert(false);
}
A = prevEndPoint;
}
points.push_back(cgal_wrap::BezierRatPoint(A));
points.push_back(cgal_wrap::BezierRatPoint(lseg.B()));
}
else if (curve.type() == CubicBezier::type) {
const CubicBezier& bezier = dynamic_cast<const CubicBezier&>(curve);
cgal_wrap::BezierRatPoint A = bezier.A();
if (j != path.begin()) {
if (A != prevEndPoint) {
// TODO
assert(false);
}
A = prevEndPoint;
}
points.push_back(cgal_wrap::BezierRatPoint(A));
points.push_back(cgal_wrap::BezierRatPoint(bezier.B()));
points.push_back(cgal_wrap::BezierRatPoint(bezier.C()));
points.push_back(cgal_wrap::BezierRatPoint(bezier.D()));
}
bool bClosesCurve = false;
if (!first && Point(points.back()) == Point(firstPoint)) {
points.pop_back();
points.push_back(firstPoint);
bClosesCurve = true;
}
prevEndPoint = points.back();
cgal_wrap::BezierCurve cgalCurve(points.begin(), points.end());
std::list<CGAL::Object> monoObjs;
fnMakeXMonotone(cgalCurve, std::back_inserter(monoObjs));
// Append the x-monotone curves to the list
cgal_wrap::BezierXMonotoneCurve monoCurve;
for (auto o = monoObjs.begin(); o != monoObjs.end(); ++o) {
if (CGAL::assign(monoCurve, *o)) {
monoCurves.push_back(monoCurve);
}
}
if (!first) {
// If this curve closes the current chain, thereby creating a new polygon
if (bClosesCurve) {
// Add the new polygon to the list
cgal_wrap::BezierPolygon subPoly(monoCurves.begin(), monoCurves.end());
if (subPoly.orientation() == CGAL::COUNTERCLOCKWISE) {
if (!outerPoly.is_empty()) {
polyList.push_back(cgal_wrap::BezierPolygonWithHoles(outerPoly, holes.begin(), holes.end()));
holes.clear();
}
outerPoly = subPoly;
}
else {
holes.push_back(subPoly);
}
monoCurves.clear();
first = true;
}
}
else {
// This is the first curve in the chain - store its source point
firstPoint = cgalCurve.control_point(0);
first = false;
}
}
}
polyList.push_back(cgal_wrap::BezierPolygonWithHoles(outerPoly, holes.begin(), holes.end()));
return polyList;
}
Notice that I'm careful to ensure that the polygon boundaries have no gaps by setting the first point of curve n+1 to the last point of curve n in case they were slightly different. I was hoping this would solve the problem, but it didn't. I can't think of any other things that might make the shapes invalid.
Here is the successful merging of the 'e' glyph with the watermark (an X).
Here is what the 'a' glyph looks like. The merging fails on this glyph.
EDIT 2:
Here are the curves that make up the 'a' glyph after parsing it from PostScript. There doesn't appear to be anything wrong with it. As I said, it looks okay when rendered. The error probably occurs during the translation from this data into the CGAL types. The line segments get translated into BezierCurves with 2 control points. I will investigate further.
LineSegment[(344, 0), (409, 0)]
CubicBezier[(409, 0), (403, 24), (400, 68), (400, 161)]
LineSegment[(400, 161), (400, 324)]
CubicBezier[(400, 324), (400, 437), (330, 485), (232, 485)]
CubicBezier[(232, 485), (180, 485), (121, 472), (66, 437)]
LineSegment[(66, 437), (94, 385)]
CubicBezier[(94, 385), (127, 405), (167, 424), (224, 424)]
CubicBezier[(224, 424), (283, 424), (326, 392), (326, 320)]
LineSegment[(326, 320), (326, 290)]
LineSegment[(326, 290), (236, 287)]
CubicBezier[(236, 287), (188, 285), (150, 280), (118, 264)]
CubicBezier[(118, 264), (70, 242), (38, 199), (38, 136)]
CubicBezier[(38, 136), (38, 45), (102, -10), (188, -10)]
CubicBezier[(188, -10), (247, -10), (293, 18), (330, 53)]
LineSegment[(330, 53), (344, 0)]
LineSegment[(326, 234), (326, 114)]
CubicBezier[(326, 114), (304, 91), (260, 52), (201, 52)]
CubicBezier[(201, 52), (147, 52), (113, 88), (113, 140)]
CubicBezier[(113, 140), (113, 171), (127, 198), (154, 213)]
CubicBezier[(154, 213), (175, 224), (202, 230), (243, 231)]
LineSegment[(243, 231), (326, 234)]
EDIT 3:
Here are the 'a' glyph curves after translation into CGAL curves. Notice that they exactly match the curves before translation implying that none of them had to be split into X-monotone subcurves; they must have all been X-monotone already.
Outer boundary:
2 344 0 409 0 [1] | 344 0 --> 409 0
4 409 0 403 24 400 68 400 161 [1] | 409 0 --> 400 161
2 400 161 400 324 [1] | 400 161 --> 400 324
4 400 324 400 437 330 485 232 485 [1] | 400 324 --> 232 485
4 232 485 180 485 121 472 66 437 [1] | 232 485 --> 66 437
2 66 437 94 385 [1] | 66 437 --> 94 385
4 94 385 127 405 167 424 224 424 [1] | 94 385 --> 224 424
4 224 424 283 424 326 392 326 320 [1] | 224 424 --> 326 320
2 326 320 326 290 [1] | 326 320 --> 326 290
2 326 290 236 287 [1] | 326 290 --> 236 287
4 236 287 188 285 150 280 118 264 [1] | 236 287 --> 118 264
4 118 264 70 242 38 199 38 136 [1] | 118 264 --> 38 136
4 38 136 38 45 102 -10 188 -10 [1] | 38 136 --> 188 -10
4 188 -10 247 -10 293 18 330 53 [1] | 188 -10 --> 330 53
2 330 53 344 0 [1] | 330 53 --> 344 0
Holes:
Hole:
2 326 234 326 114 [1] | 326 234 --> 326 114
4 326 114 304 91 260 52 201 52 [1] | 326 114 --> 201 52
4 201 52 147 52 113 88 113 140 [1] | 201 52 --> 113 140
4 113 140 113 171 127 198 154 213 [1] | 113 140 --> 154 213
4 154 213 175 224 202 230 243 231 [1] | 154 213 --> 243 231
2 243 231 326 234 [1] | 243 231 --> 326 234
This polygon causes the assertion failure when added to a BezierPolygonSet. Any ideas?
You can customize error handling; see the manual.
Please attach a complete standalone program. Nothing looks wrong to me with statements you listed.
You can approximate the Bezier curves with polylines and process the whole things using polylines. If the problem is with the handling of Bezier curves, then this would solve it. If this is acceptable, it will also be more efficient.
We have fixed a bug in the CGAL component that handles Bezier curves, namely, Arr_Bezier_curve_traits_2.h.
I am building a small application. in this i have saved some data in a txt file. i need to edit a value in particular row and column. i wrote a code to go to a particular line and fetch the values but i have tried almost everything to got a particular column and edit that value.
1000 400 120 110 800 110 150 500 0 1000
1000 400 90 150 800 120 150 600 0 1000
1000 400 80 60 **800** 132 150 700 0 1000
1000 400 120 60 800 123 150 200 0 1000
1000 400 111 80 800 143 150 700 0 1000
1000 400 30 90 800 155 150 500 0 1000
for example i have edit the highlighted value , which is the best way to do. i cannot paste my whole code as it is very long.
this is the one where i can go to a particular line
fstream& GotoLine(int num) {
infile.seekg(std::ios::beg);
for (int i = 0; i < num - 1; ++i) {
infile.ignore(std::numeric_limits<std::streamsize>::max(), '\n');
}
return infile;
}
i would appreciate any help over this.
Assuming that your file is formatted as fixed column size, my advice would be:
prepare an array or a vector of struct { size_t pos; size_t width} to define the fields
open the file in ios::in | ios::out mode
read the file one line at a time with your GotoLine function up to the line you want to process
note the index in the fstream with tellg
read the interesting line in a char array with a size greater than the line - here I would use 64 to be large enough - with istream::getline
as you have read the line as a plain array, you can rewrite the characters of one single field
go back to beginning of line with seekp
write the line back to the file
This is not a general method of editing a text file. It only works here because as you have fields of fixed size the edited line has exactly same size than original one, so it can be re-written in place. But never use it in a general case.
This question is unlikely to help any future visitors; it is only relevant to a small geographic area, a specific moment in time, or an extraordinarily narrow situation that is not generally applicable to the worldwide audience of the internet. For help making this question more broadly applicable, visit the help center.
Closed 9 years ago.
I am currently trying to understand how the following code (http://pastebin.com/zTHUrmyx) works, my approach is currently compiling the software in debug and using gdb to step through the code.
However, I'm running into the problem that 'step' does not always tell me what is going on. Particularly unclear to me is the EXECUTE {...} which I cannot step into.
How do I go about learning what the code is doing?
1 /*
2 Copyright 2008 Brain Research Institute, Melbourne, Australia
3
4 Written by J-Donald Tournier, 27/06/08.
5
6 This file is part of MRtrix.
7
8 MRtrix is free software: you can redistribute it and/or modify
9 it under the terms of the GNU General Public License as published by
10 the Free Software Foundation, either version 3 of the License, or
11 (at your option) any later version.
12
13 MRtrix is distributed in the hope that it will be useful,
14 but WITHOUT ANY WARRANTY; without even the implied warranty of
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 GNU General Public License for more details.
17
18 You should have received a copy of the GNU General Public License
19 along with MRtrix. If not, see <http://www.gnu.org/licenses/>.
20
21
22 15-10-2008 J-Donald Tournier <d.tournier#brain.org.au>
23 * fix -prs option handling
24 * remove MR::DICOM_DW_gradients_PRS flag
25
26 15-10-2008 J-Donald Tournier <d.tournier#brain.org.au>
27 * add -layout option to manipulate data ordering within the image file
28
29 14-02-2010 J-Donald Tournier <d.tournier#brain.org.au>
30 * fix -coord option so that the "end" keyword can be used
31
32
33 */
34
35 #include "app.h"
36 #include "image/position.h"
37 #include "image/axis.h"
38 #include "math/linalg.h"
39
40 using namespace std;
41 using namespace MR;
42
43 SET_VERSION_DEFAULT;
44
45 DESCRIPTION = {
46 "perform conversion between different file types and optionally extract a subset of the input image.",
47 "If used correctly, this program can be a very useful workhorse. In addition to converting images between different formats, it can be used to extract specific studies from a data set, extract a specific region of interest, flip the images, or to scale the intensity of the images.",
48 NULL
49 };
50
51 ARGUMENTS = {
52 Argument ("input", "input image", "the input image.").type_image_in (),
53 Argument ("ouput", "output image", "the output image.").type_image_out (),
54 Argument::End
55 };
56
57
58 const gchar* type_choices[] = { "REAL", "IMAG", "MAG", "PHASE", "COMPLEX", NULL };
59 const gchar* data_type_choices[] = { "FLOAT32", "FLOAT32LE", "FLOAT32BE", "FLOAT64", "FLOAT64LE", "FLOAT64BE",
60 "INT32", "UINT32", "INT32LE", "UINT32LE", "INT32BE", "UINT32BE",
61 "INT16", "UINT16", "INT16LE", "UINT16LE", "INT16BE", "UINT16BE",
62 "CFLOAT32", "CFLOAT32LE", "CFLOAT32BE", "CFLOAT64", "CFLOAT64LE", "CFLOAT64BE",
63 "INT8", "UINT8", "BIT", NULL };
64
65 OPTIONS = {
66 Option ("coord", "select coordinates", "extract data only at the coordinates specified.", false, true)
67 .append (Argument ("axis", "axis", "the axis of interest").type_integer (0, INT_MAX, 0))
68 .append (Argument ("coord", "coordinates", "the coordinates of interest").type_sequence_int()),
69
70 Option ("vox", "voxel size", "change the voxel dimensions.")
71 .append (Argument ("sizes", "new dimensions", "A comma-separated list of values. Only those values specified will be changed. For example: 1,,3.5 will change the voxel size along the x & z axes, and leave the y-axis voxel size unchanged.")
72 .type_sequence_float ()),
73
74 Option ("datatype", "data type", "specify output image data type.")
75 .append (Argument ("spec", "specifier", "the data type specifier.").type_choice (data_type_choices)),
76
77 Option ("scale", "scaling factor", "apply scaling to the intensity values.")
78 .append (Argument ("factor", "factor", "the factor by which to multiply the intensities.").type_float (NAN, NAN, 1.0)),
79
80 Option ("offset", "offset", "apply offset to the intensity values.")
81 .append (Argument ("bias", "bias", "the value of the offset.").type_float (NAN, NAN, 0.0)),
82
83 Option ("zero", "replace NaN by zero", "replace all NaN values with zero."),
84
85 Option ("output", "output type", "specify type of output")
86 .append (Argument ("type", "type", "type of output.")
87 .type_choice (type_choices)),
88
89 Option ("layout", "data layout", "specify the layout of the data in memory. The actual layout produced will depend on whether the output image format can support it.")
90 .append (Argument ("spec", "specifier", "the data layout specifier.").type_string ()),
91
92 Option ("prs", "DW gradient specified as PRS", "assume that the DW gradients are specified in the PRS frame (Siemens DICOM only)."),
93
94 Option::End
95 };
96
97
98
99 inline bool next (Image::Position& ref, Image::Position& other, const std::vector<int>* pos)
100 {
101 int axis = 0;
102 do {
103 ref.inc (axis);
104 if (ref[axis] < ref.dim(axis)) {
105 other.set (axis, pos[axis][ref[axis]]);
106 return (true);
107 }
108 ref.set (axis, 0);
109 other.set (axis, pos[axis][0]);
110 axis++;
111 } while (axis < ref.ndim());
112 return (false);
113 }
114
115
116
117
118
119 EXECUTE {
120 std::vector<OptBase> opt = get_options (1); // vox
121 std::vector<float> vox;
122 if (opt.size())
123 vox = parse_floats (opt[0][0].get_string());
124
125
126 opt = get_options (3); // scale
127 float scale = 1.0;
128 if (opt.size()) scale = opt[0][0].get_float();
129
130 opt = get_options (4); // offset
131 float offset = 0.0;
132 if (opt.size()) offset = opt[0][0].get_float();
133
134 opt = get_options (5); // zero
135 bool replace_NaN = opt.size();
136
137 opt = get_options (6); // output
138 Image::OutputType output_type = Image::Default;
139 if (opt.size()) {
140 switch (opt[0][0].get_int()) {
141 case 0: output_type = Image::Real; break;
142 case 1: output_type = Image::Imaginary; break;
143 case 2: output_type = Image::Magnitude; break;
144 case 3: output_type = Image::Phase; break;
145 case 4: output_type = Image::RealImag; break;
146 }
147 }
148
149
150
151
152 Image::Object &in_obj (*argument[0].get_image());
153
154 Image::Header header (in_obj);
155
156 if (output_type == 0) {
157 if (in_obj.is_complex()) output_type = Image::RealImag;
158 else output_type = Image::Default;
159 }
160
161 if (output_type == Image::RealImag) header.data_type = DataType::CFloat32;
162 else if (output_type == Image::Phase) header.data_type = DataType::Float32;
163 else header.data_type.unset_flag (DataType::ComplexNumber);
164
165
166 opt = get_options (2); // datatype
167 if (opt.size()) header.data_type.parse (data_type_choices[opt[0][0].get_int()]);
168
169 for (guint n = 0; n < vox.size(); n++)
170 if (isfinite (vox[n])) header.axes.vox[n] = vox[n];
171
172 opt = get_options (7); // layout
173 if (opt.size()) {
174 std::vector<Image::Axis> ax = parse_axes_specifier (header.axes, opt[0][0].get_string());
175 if (ax.size() != (guint) header.axes.ndim())
176 throw Exception (String("specified layout \"") + opt[0][0].get_string() + "\" does not match image dimensions");
177
178 for (guint i = 0; i < ax.size(); i++) {
179 header.axes.axis[i] = ax[i].axis;
180 header.axes.forward[i] = ax[i].forward;
181 }
182 }
183
184
185 opt = get_options (8); // prs
186 if (opt.size() && header.DW_scheme.rows() && header.DW_scheme.columns()) {
187 for (guint row = 0; row < header.DW_scheme.rows(); row++) {
188 double tmp = header.DW_scheme(row, 0);
189 header.DW_scheme(row, 0) = header.DW_scheme(row, 1);
190 header.DW_scheme(row, 1) = tmp;
191 header.DW_scheme(row, 2) = -header.DW_scheme(row, 2);
192 }
193 }
194
195 std::vector<int> pos[in_obj.ndim()];
196
197 opt = get_options (0); // coord
198 for (guint n = 0; n < opt.size(); n++) {
199 int axis = opt[n][0].get_int();
200 if (pos[axis].size()) throw Exception ("\"coord\" option specified twice for axis " + str (axis));
201 pos[axis] = parse_ints (opt[n][1].get_string(), header.dim(axis)-1);
202 header.axes.dim[axis] = pos[axis].size();
203 }
204
205 for (int n = 0; n < in_obj.ndim(); n++) {
206 if (pos[n].empty()) {
207 pos[n].resize (in_obj.dim(n));
208 for (guint i = 0; i < pos[n].size(); i++) pos[n][i] = i;
209 }
210 }
211
212
213 in_obj.apply_scaling (scale, offset);
214
215
216
217
218
219
220 Image::Position in (in_obj);
221 Image::Position out (*argument[1].get_image (header));
222
223 for (int n = 0; n < in.ndim(); n++) in.set (n, pos[n][0]);
224
225 ProgressBar::init (out.voxel_count(), "copying data...");
226
227 do {
228
229 float re, im = 0.0;
230 in.get (output_type, re, im);
231 if (replace_NaN) if (gsl_isnan (re)) re = 0.0;
232 out.re (re);
233
234 if (output_type == Image::RealImag) {
235 if (replace_NaN) if (gsl_isnan (im)) im = 0.0;
236 out.im (im);
237 }
238
239 ProgressBar::inc();
240 } while (next (out, in, pos));
241
242 ProgressBar::done();
243 }
As was noted in the comments, EXECUTE seems to be a macro, apparent from the context a function header (and maybe a bit more, e.g. some global variables and functions), so the part in curly braces is the function body.
To get to the definition of EXECUTE, you will have to examine the headers.
However, if you can reach some part of the code during debugging, you could insert a string or char[] at that point, giving it the stringified version of EXECUTE, so you get whatever the preprocessor will emit for EXECUTE at that position in the code.
#define STR(x) #x
#define STRINGIFY(x) STR(x)
char c[] = STRINGIFY(EXECUTE);
the two macros are a known little macro trick to get the content of any macro as a string literal. Try it out and inspect the char array in your debugger to get the content of execute.
My wild guess here: EXECUTE is the main function or a replacement for it, the OPTIONS and ARGUMENTS describe what arguments the program expects and what command line options you can pass to it. Those macros and some of the used functions and variables (get_options, argument) are part of a little framework that should facilitate the usage, evaluation and user information about command line options.