July 2010 - Olena-patches

[Olena] #229: Have Olena's BuildBot generate GitStats information
by Olena Trac 23 Nov '11

23 Nov '11

#229: Have Olena's BuildBot generate GitStats information -------------------------+-------------------------------------------------- Reporter: levill_r | Owner: levill_r Type: enhancement | Status: new Priority: trivial | Milestone: Olena 1.1 Component: other | Version: 1.0 Keywords: | -------------------------+-------------------------------------------------- Git statistics are generated by my (Roland's) account for a limited set of branches, each night. It'd be smarter to have one of the buildslave invoke a script running `gitstats` and uploading the result (as `lrde- upload.sh` does) prior to building the project. We could even filter branches (having statistics for all branches might not be relevant). -- Ticket URL: <https://trac.lrde.org/olena/ticket/229> Olena <http://olena.lrde.epita.fr> Olena, a generic and efficient C++ image processing library.

1 4

[Olena] #225: Code not displayed in tutorial
by Olena Trac 09 Sep '11

09 Sep '11

#225: Code not displayed in tutorial ---------------------------+------------------------------------------------ Reporter: jonathan | Owner: Olena Team Type: defect | Status: new Priority: minor | Milestone: Component: Milena | Version: 1.0 Keywords: doc, tutorial | ---------------------------+------------------------------------------------ in page, [http://www.lrde.epita.fr/dload/olena/1.0/doc/milena/devel_doc/] all examples of code are not displayed (all boxes are empty) -- Ticket URL: <https://trac.lrde.org/olena/ticket/225> Olena <http://olena.lrde.epita.fr> Olena, a generic and efficient C++ image processing library.

1 1

[Olena] #220: Change the logo of the Trac
by Olena Trac 08 Sep '11

08 Sep '11

#220: Change the logo of the Trac ----------------------+----------------------------------------------------- Reporter: levill_r | Owner: Olena Team Type: task | Status: new Priority: trivial | Milestone: Component: Trac | Version: 1.0 Keywords: | ----------------------+----------------------------------------------------- Use this one instead: http://www.lrde.epita.fr/dload/olena/logos/olena_small.jpg. Unfortunately, we don't have the required permissions on the Trac's files/dirs to change it. -- Ticket URL: <https://trac.lrde.org/olena/ticket/220> Olena <http://olena.lrde.epita.fr> Olena, a generic and efficient C++ image processing library.

1 5

[Olena] #172: Experiment SCOOP 1.5/2 with C++0x concepts
by Olena 12 Jul '11

12 Jul '11

#172: Experiment SCOOP 1.5/2 with C++0x concepts -------------------------+-------------------------------------------------- Reporter: levill_r | Owner: Olena Team Type: enhancement | Status: new Priority: trivial | Milestone: Olena 2.0 Component: Olena-NG | Version: 2.0 Keywords: | -------------------------+-------------------------------------------------- What I call SCOOP 1.5 is the paradigm currently used in Milena 1.0 (as of Nov. 2008), and SCOOP 2 the paradigm described in the MPOOL'08 paper (used in Olena-NG). -- Ticket URL: <https://trac.lrde.org/olena/ticket/172> Olena <http://olena.lrde.epita.fr> Olena, a generic and efficient C++ image processing library.

2 2

last-svn-commit-137-gdb55e63 Add the binarization tool submitted to H-DIBCO 2010.
by Guillaume Lazzara 04 Nov '10

04 Nov '10

* src/Makefile.am, * src/contest/Makefile.am, * src/contest/hdibco-2010/Makefile.am: New. * src/contest/hdibco-2010/sauvola_ms_hdibco.cc: New. Source code of the submitted program for H-DIBCO 2010. --- scribo/ChangeLog | 11 ++ scribo/src/Makefile.am | 1 + .../{tests/toolchain => src/contest}/Makefile.am | 2 +- .../contest/hdibco-2010}/Makefile.am | 8 +- .../hdibco-2010/sauvola_ms_hdibco.cc} | 100 +++++++------------ 5 files changed, 56 insertions(+), 66 deletions(-) copy scribo/{tests/toolchain => src/contest}/Makefile.am (98%) copy scribo/{tests/toolchain => src/contest/hdibco-2010}/Makefile.am (87%) copy scribo/src/{binarization/pgm_sauvola_ms.cc => contest/hdibco-2010/sauvola_ms_hdibco.cc} (53%) diff --git a/scribo/ChangeLog b/scribo/ChangeLog index f2eb3c3..3888981 100644 --- a/scribo/ChangeLog +++ b/scribo/ChangeLog @@ -1,5 +1,16 @@ 2010-06-03 Guillaume Lazzara <z(a)lrde.epita.fr> + Add the binarization tool submitted to H-DIBCO 2010. + + * src/Makefile.am, + * src/contest/Makefile.am, + * src/contest/hdibco-2010/Makefile.am: New. + + * src/contest/hdibco-2010/sauvola_ms_hdibco.cc: New. Source code + of the submitted program for H-DIBCO 2010. + +2010-06-03 Guillaume Lazzara <z(a)lrde.epita.fr> + Update tools in scribo/src. * src/Makefile.am: Add postprocessing subdir. diff --git a/scribo/src/Makefile.am b/scribo/src/Makefile.am index b3a7859..774740a 100644 --- a/scribo/src/Makefile.am +++ b/scribo/src/Makefile.am @@ -22,6 +22,7 @@ include $(top_srcdir)/scribo/scribo.mk SUBDIRS = \ binarization \ + contest \ debug \ primitive \ filter \ diff --git a/scribo/tests/toolchain/Makefile.am b/scribo/src/contest/Makefile.am similarity index 98% copy from scribo/tests/toolchain/Makefile.am copy to scribo/src/contest/Makefile.am index 826fb86..81f314b 100644 --- a/scribo/tests/toolchain/Makefile.am +++ b/scribo/src/contest/Makefile.am @@ -20,4 +20,4 @@ include $(top_srcdir)/scribo/tests/tests.mk SUBDIRS = \ - nepomuk \ No newline at end of file + hdibco-2010 \ No newline at end of file diff --git a/scribo/tests/toolchain/Makefile.am b/scribo/src/contest/hdibco-2010/Makefile.am similarity index 87% copy from scribo/tests/toolchain/Makefile.am copy to scribo/src/contest/hdibco-2010/Makefile.am index 826fb86..44add64 100644 --- a/scribo/tests/toolchain/Makefile.am +++ b/scribo/src/contest/hdibco-2010/Makefile.am @@ -19,5 +19,9 @@ include $(top_srcdir)/scribo/tests/tests.mk -SUBDIRS = \ - nepomuk \ No newline at end of file +check_PROGRAMS = sauvola_ms_hdibco + +sauvola_ms_hdibco_SOURCES = sauvola_ms_hdibco.cc + + +TESTS = $(check_PROGRAMS) diff --git a/scribo/src/binarization/pgm_sauvola_ms.cc b/scribo/src/contest/hdibco-2010/sauvola_ms_hdibco.cc similarity index 53% copy from scribo/src/binarization/pgm_sauvola_ms.cc copy to scribo/src/contest/hdibco-2010/sauvola_ms_hdibco.cc index 38ff2e6..68fdaac 100644 --- a/scribo/src/binarization/pgm_sauvola_ms.cc +++ b/scribo/src/contest/hdibco-2010/sauvola_ms_hdibco.cc @@ -1,5 +1,4 @@ -// Copyright (C) 2009, 2010 EPITA Research and Development Laboratory -// (LRDE) +// Copyright (C) 2010 EPITA Research and Development Laboratory (LRDE) // // This file is part of Olena. // @@ -24,86 +23,61 @@ // exception does not however invalidate any other reasons why the // executable file might be covered by the GNU General Public License. +#include <cctype> #include <mln/core/image/image2d.hh> #include <mln/value/int_u8.hh> #include <mln/io/pgm/load.hh> +#include <mln/io/ppm/load.hh> #include <mln/io/pbm/save.hh> #include <scribo/binarization/sauvola_ms.hh> -#include <scribo/debug/usage.hh> - -bool check_args(int argc, char * argv[]) -{ - if (argc < 3 || argc > 6) - return false; - - if (argc >= 5) - { - int s = atoi(argv[4]); - - if (s < 1 || s > 3) - { - std::cout << "s must be set to 2 or 3." - << std::endl; - return false; - } - } - - return true; -} - - -const char *args_desc[][2] = -{ - { "input.pgm", "A graylevel image." }, - { "out.pbm", "A binary image." }, - { "w", "Window size at scale 1. (default: 101)" }, - { "s", "First subsampling ratio (default: 3)." }, - { "k", "Sauvola's formuale parameter (default: 0.34)" }, - {0, 0} -}; - - - +#include <scribo/postprocessing/fill_object_holes.hh> +#include <scribo/preprocessing/denoise.hh> int main(int argc, char *argv[]) { using namespace mln; using namespace scribo; - if (!check_args(argc, argv)) - return scribo::debug::usage(argv, - "Multi-Scale Binarization of a color image based on Sauvola's algorithm.", - "input.pgm output.pbm <w> <s> <k>", - args_desc); - trace::entering("main"); + if (argc != 3) + { + std::cout << "Usage: " << argv[0] << " <input.{pgm,ppm}> <output.pbm>" << std::endl; + std::cout << "--------------" << std::endl; + std::cout << "EPITA/LRDE - 2010" << std::endl; + return 1; + } - // Window size - unsigned w_1; - if (argc >= 4) - w_1 = atoi(argv[3]); // Scale 1 - else - w_1 = 101u; + std::string ext = argv[1]; + ext = ext.substr(ext.length() - 3, 3); + for (unsigned i = 0; i < ext.length(); ++i) + ext[i] = tolower(ext[i]); - // First subsampling scale. - unsigned s; - if (argc >= 5) - s = atoi(argv[4]); - else - s = 3u; + // Load + image2d<value::int_u8> input; - double k; - if (argc >= 6) - k = atof(argv[5]); - else - k = 0.34f; + if (ext == "pgm") + io::pgm::load(input, argv[1]); + else if (ext == "ppm") + { + image2d<value::rgb8> input_color; + io::ppm::load(input_color, argv[1]); - image2d<value::int_u8> input_1; - io::pgm::load(input_1, argv[1]); + // Convert to Gray level image. + input = data::transform(input_color, mln::fun::v2v::rgb_to_int_u<8>()); + } + else + { + std::cout << "Invalid input file format! Expecting PGM or PPM images." << std::endl; + return 1; + } + // Binarize image2d<bool> - output = scribo::binarization::sauvola_ms(input_1, w_1, s, k); + output = scribo::binarization::sauvola_ms(input, 101, 3, 0.4); + + output = scribo::preprocessing::denoise_bg(output, c8(), 4); + output = scribo::postprocessing::fill_object_holes(output, 0.01f); io::pbm::save(output, argv[2]); } -- 1.5.6.5

2 1

last-svn-commit-181-g565a8ac Fix Scribo's dependencies w.r.t. libraries.
by Roland Levillain 31 Jul '10

31 Jul '10

* src/Makefile.am (text_in_doc_LDFLAGS, text_in_doc_ppm_LDFLAGS): Remove -lhpdf. --- scribo/ChangeLog | 7 +++++++ scribo/src/Makefile.am | 4 ++-- 2 files changed, 9 insertions(+), 2 deletions(-) diff --git a/scribo/ChangeLog b/scribo/ChangeLog index 9506589..3f0ee54 100644 --- a/scribo/ChangeLog +++ b/scribo/ChangeLog @@ -1,3 +1,10 @@ +2010-07-31 Roland Levillain <roland(a)lrde.epita.fr> + + Fix Scribo's dependencies w.r.t. libraries. + + * src/Makefile.am (text_in_doc_LDFLAGS, text_in_doc_ppm_LDFLAGS): + Remove -lhpdf. + 2010-04-23 Roland Levillain <roland(a)lrde.epita.fr> Get C++ compiler flags for Scribo from configure. diff --git a/scribo/src/Makefile.am b/scribo/src/Makefile.am index 5f2367f..818b005 100644 --- a/scribo/src/Makefile.am +++ b/scribo/src/Makefile.am @@ -50,7 +50,7 @@ if HAVE_TIFF text_in_doc_LDFLAGS = $(AM_LDFLAGS) \ $(TESSERACT_LDFLAGS) \ $(TIFF_LDFLAGS) \ - -lpthread -lhpdf + -lpthread bin_PROGRAMS += text_in_doc_ppm text_in_doc_ppm_SOURCES = text_in_doc_ppm.cc @@ -60,7 +60,7 @@ if HAVE_TIFF text_in_doc_ppm_LDFLAGS = $(AM_LDFLAGS) \ $(TESSERACT_LDFLAGS) \ $(TIFF_LDFLAGS) \ - -lpthread -lhpdf + -lpthread endif HAVE_TIFF endif HAVE_TESSERACT -- 1.5.6.5

1 0

branch fix-scribo-lib-deps created: last-svn-commit-181-g565a8ac
by git＠lrde.epita.fr 31 Jul '10

31 Jul '10

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "Olena, a generic and efficient image processing platform". The branch fix-scribo-lib-deps has been created at 565a8acd20101cf22d58b674d4d1695d8e9926c6 (commit) - Log ----------------------------------------------------------------- 565a8ac Fix Scribo's dependencies w.r.t. libraries. ----------------------------------------------------------------------- hooks/post-receive -- Olena, a generic and efficient image processing platform

1 0

branch vtk-format updated: last-svn-commit-199-g34deb37
by git＠lrde.epita.fr 28 Jul '10

28 Jul '10

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "Olena, a generic and efficient image processing platform". The branch vtk-format has been updated via 34deb373040a6d2b7540bb7321bdb5c8615014ef (commit) from 1ff059f5bd239a640b62f51c995574dc409763d8 (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- 34deb37 Regen Milena's Makefile helpers. ----------------------------------------------------------------------- Summary of changes: milena/ChangeLog | 6 ++++++ milena/headers.mk | 2 ++ milena/tests/unit_test/unit-tests.mk | 4 ++++ 3 files changed, 12 insertions(+), 0 deletions(-) hooks/post-receive -- Olena, a generic and efficient image processing platform

1 0

last-svn-commit-199-g34deb37 Regen Milena's Makefile helpers.
by Roland Levillain 28 Jul '10

28 Jul '10

* headers.mk, tests/unit_test/unit-tests.mk: Here. --- milena/ChangeLog | 6 ++++++ milena/headers.mk | 2 ++ milena/tests/unit_test/unit-tests.mk | 4 ++++ 3 files changed, 12 insertions(+), 0 deletions(-) diff --git a/milena/ChangeLog b/milena/ChangeLog index af4885f..79e9568 100644 --- a/milena/ChangeLog +++ b/milena/ChangeLog @@ -1,5 +1,11 @@ 2010-07-28 Roland Levillain <roland(a)lrde.epita.fr> + Regen Milena's Makefile helpers. + + * headers.mk, tests/unit_test/unit-tests.mk: Here. + +2010-07-28 Roland Levillain <roland(a)lrde.epita.fr> + Add a VTK output for images of unsigned integers. * mln/io/vtk/save.hh diff --git a/milena/headers.mk b/milena/headers.mk index 110caa6..06e1d9e 100644 --- a/milena/headers.mk +++ b/milena/headers.mk @@ -686,6 +686,8 @@ mln/io/tiff/all.hh \ mln/io/tiff/load.hh \ mln/io/txt/all.hh \ mln/io/txt/save.hh \ +mln/io/vtk/all.hh \ +mln/io/vtk/save.hh \ mln/labeling/all.hh \ mln/labeling/background.hh \ mln/labeling/blobs.hh \ diff --git a/milena/tests/unit_test/unit-tests.mk b/milena/tests/unit_test/unit-tests.mk index 6a9d5d5..0cd8877 100644 --- a/milena/tests/unit_test/unit-tests.mk +++ b/milena/tests/unit_test/unit-tests.mk @@ -702,6 +702,8 @@ mln_io_ppms_all \ mln_io_ppms_load \ mln_io_txt_all \ mln_io_txt_save \ +mln_io_vtk_all \ +mln_io_vtk_save \ mln_labeling_all \ mln_labeling_background \ mln_labeling_blobs \ @@ -2000,6 +2002,8 @@ mln_io_ppms_all_SOURCES = mln_io_ppms_all.cc mln_io_ppms_load_SOURCES = mln_io_ppms_load.cc mln_io_txt_all_SOURCES = mln_io_txt_all.cc mln_io_txt_save_SOURCES = mln_io_txt_save.cc +mln_io_vtk_all_SOURCES = mln_io_vtk_all.cc +mln_io_vtk_save_SOURCES = mln_io_vtk_save.cc mln_labeling_all_SOURCES = mln_labeling_all.cc mln_labeling_background_SOURCES = mln_labeling_background.cc mln_labeling_blobs_SOURCES = mln_labeling_blobs.cc -- 1.5.6.5

1 0

last-svn-commit-195-g96628ac Add a specific (temporary) VTK output routine for binary mesh images.
by Roland Levillain 28 Jul '10

28 Jul '10

* apps/mesh-segm-skel/save_bin_alt.hh (mln::io::vtk::save_bin_alt): New function. --- milena/ChangeLog | 7 + milena/apps/mesh-segm-skel/save_bin_alt.hh | 310 +++++++++++++++++++++++++++- 2 files changed, 312 insertions(+), 5 deletions(-) diff --git a/milena/ChangeLog b/milena/ChangeLog index ab27b56..3136661 100644 --- a/milena/ChangeLog +++ b/milena/ChangeLog @@ -1,5 +1,12 @@ 2010-06-24 Roland Levillain <roland(a)lrde.epita.fr> + Add a specific (temporary) VTK output routine for binary mesh images. + + * apps/mesh-segm-skel/save_bin_alt.hh + (mln::io::vtk::save_bin_alt): New function. + +2010-06-24 Roland Levillain <roland(a)lrde.epita.fr> + Exercise mln::io::vtk::save. * tests/io/vtk/save.cc: New. diff --git a/milena/apps/mesh-segm-skel/save_bin_alt.hh b/milena/apps/mesh-segm-skel/save_bin_alt.hh index 034b840..b09fa88 100644 --- a/milena/apps/mesh-segm-skel/save_bin_alt.hh +++ b/milena/apps/mesh-segm-skel/save_bin_alt.hh @@ -1,4 +1,4 @@ -// Copyright (C) 2009 EPITA Research and Development Laboratory (LRDE) +// Copyright (C) 2009, 2010 EPITA Research and Development Laboratory (LRDE) // // This file is part of the Milena Library. This library is free // software; you can redistribute it and/or modify it under the terms @@ -28,10 +28,11 @@ #ifndef APPS_MESH_SEGM_SKEL_SAVE_BIN_ALT_HH # define APPS_MESH_SEGM_SKEL_SAVE_BIN_ALT_HH -/*--------------------------------------------------------------------. -| FIXME: Copied and adjusted (in a hurry) from mln/io/off/save.hh, | -| because fixing image_if + complex_image was much too long. Sorry. | -`--------------------------------------------------------------------*/ +/*-------------------------------------------------------------------. +| FIXME: Copied and adjusted (in a hurry) from mln/io/off/save.hh, | +| and mln/io/vtk/save.hh,because fixing image_if + complex_image was | +| much too long. Sorry. | +`-------------------------------------------------------------------*/ # include <cstdlib> @@ -51,6 +52,10 @@ namespace mln namespace io { + /*------. + | OFF. | + `------*/ + namespace off { @@ -181,6 +186,301 @@ namespace mln } // end of namespace mln::io::off + + /*------. + | VTK. | + `------*/ + + namespace vtk + { + /** FIXME: Similar to + mln::io::vtk::save(const bin_2complex_image3df&, const std::string&), + but does not save faces whose value is `false'. */ + template <typename I> + void save_bin_alt(const I& ima, + const std::string& filename) + { + const std::string me = "mln::io::vtk::save"; + + std::ofstream ostr(filename.c_str()); + if (!ostr) + { + std::cerr << me << ": `" << filename << "' invalid file." + << std::endl; + /* FIXME: Too violent. We should allow the use of + exceptions, at least to have Milena's code behave + correctly in interpreted environments (std::exit() or + std::abort() causes the termination of a Python + interpreter, for instance!). */ + std::exit(1); + } + + /*---------. + | Header. | + `---------*/ + + /* ``The legacy VTK file formats consist of five basic + parts.'' */ + + /* ``1. The first part is the file version and + identifier. This part contains the single line: + + # vtk DataFile Version x.x. + + This line must be exactly as shown with the + exception of the version number x.x, which will vary + with different releases of VTK. (Note: the current + version number is 3.0. Version 1.0 and 2.0 files are + compatible with version 3.0 files.)'' */ + ostr << "# vtk DataFile Version 2.0" << std::endl; + + /* ``2. The second part is the header. The header consists + of a character string terminated by end-of-line + character `\n'. The header is 256 characters + maximum. The header can be used to describe the data + and include any other pertinent information.'' */ + ostr << "Generated by Milena 1.0 http://olena.lrde.epita.fr" + << std::endl; + + /* ``3. The next part is the file format. The file format + describes the type of file, either ASCII or + binary. On this line the single word ASCII or BINARY + must appear.'' */ + ostr << "ASCII" << std::endl; + + /*-------. + | Data. | + `-------*/ + + /* ``4. The fourth part is the dataset structure. The + geometry part describes the geometry and topology of + the dataset. This part begins with a line containing + the keyword DATASET followed by a keyword describing + the type of dataset. Then, depending upon the type + of dataset, other keyword/data combinations define + the actual data.'' + + [...] + + Dataset Format. The Visualization Toolkit supports + five different dataset formats: structured points, + structured grid, rectilinear grid, unstructured + grid, and polygonal data.'' */ + + ostr << "DATASET POLYDATA" << std::endl << std::endl; + + // --------- // + // Complex. // + // --------- // + + static const unsigned D = I::dim; + typedef mln_geom(I) G; + + /* ``* Polygonal Data + The polygonal dataset consists of arbitrary + combinations of surface graphics primitives + vertices (and polyvertices), lines (and + polylines), polygons (of various types), and + triangle strips. Polygonal data is defined by + the POINTS, VERTICES, LINES, POLYGONS, or + TRIANGLE_STRIPS sections. The POINTS definition + is the same as we saw for structured grid + datasets.'' */ + + // ---------------------------- // + // Geometry (point locations). // + // ---------------------------- // + + ostr << "POINTS " + << ima.domain().cplx().template nfaces_of_static_dim<0>() + << " float" << std::endl; + // Iterate on 0-faces (vertices). + p_n_faces_fwd_piter<D, G> v(ima.domain(), 0); + for_all(v) + { + mln_invariant(v.to_site().size() == 1); + ostr << v.to_site().front()[0] << ' ' + << v.to_site().front()[1] << ' ' + << v.to_site().front()[2] << std::endl; + } + ostr << std::endl; + + /* ``The VERTICES, LINES, POLYGONS, or + TRIANGLE_STRIPS keywords define the polygonal + dataset topology. Each of these keywords + requires two parameters: the number of cells `n' + and the size of the cell list `size'. The cell + list size is the total number of integer values + required to represent the list (i.e., sum of + `numPoints' and connectivity indices over each + cell). None of the keywords VERTICES, LINES, + POLYGONS, or TRIANGLE_STRIPS is required.'' */ + + // ---------- // + // Vertices. // + // ---------- // + + /* FIXME: Do not create a VERTICES section if there is no + vertex. + + Likewise, only process vertices having a value attached + to them, i.e., which are part of the domain of the image + (which is different from the complex, upon which the + domain is based). */ + + /* We do not use + + ima.domain().cplx().template nfaces_of_static_dim<N>() + + to get the number of N-faces, since the image may be + masked, and exhibit less N-faces than its underlying + complex. Iterating on the N-faces is safer. */ + /* FIXME: Is there anything faster? See what the interface + of the (morphed) image can provide. */ + unsigned nvertices = 0; + for_all(v) if (ima(v)) + ++nvertices; + + ostr << "VERTICES " << nvertices << ' ' + /* Each vertex requires two numbers: the cardinal of its + ends (which is always 1) and the indices of the point + among the POINTS section. Hence the total number of + values in the VERTEX section is nvertices * 2. */ + << nvertices * 2 << std::endl; + + for_all(v) if (ima(v)) + ostr << "1 " << v.unproxy_().face().face_id() << std::endl; + ostr << std::endl; + + // ------- // + // Edges. // + // ------- // + + /* FIXME: Do not create a LINES section if there is no + edge. + + Likewise, only process edges having a value attached + to them, i.e., which are part of the domain of the image + (which is different from the complex, upon which the + domain is based). */ + + // Same comment as above about the count of N-faces. + unsigned nedges = 0; + p_n_faces_fwd_piter<D, G> e(ima.domain(), 1); + for_all (e) if (ima(e)) + ++nedges; + + ostr << "LINES " << nedges << ' ' + /* Each edge requires three numbers: the cardinal of its + ends (which is always 2) and the indices of these ends + among the POINTS section. Hence the total number of + values in the LINES section is nedges * 3. */ + << nedges * 3 << std::endl; + + // Vertices adjacent to edges. + typedef complex_lower_neighborhood<D, G> adj_vertices_nbh_t; + adj_vertices_nbh_t adj_vertices_nbh; + mln_niter(adj_vertices_nbh_t) adj_v(adj_vertices_nbh, e); + // Iterate on 1-faces (edges). + for_all (e) if (ima(e)) + { + ostr << "2"; + // Iterate on vertices (0-faces). + for_all (adj_v) + { + // FIXME: Accessing the face id is too complicated. + ostr << " " << adj_v.unproxy_().face().face_id(); + } + ostr << std::endl; + } + ostr << std::endl; + + // ---------- // + // Polygons. // + // ---------- // + + /* FIXME: Do not create a POLYGONS section if there is no + polygon. + + Likewise, only process polygons having a value attached + to them, i.e., which are part of the domain of the image + (which is different from the complex, upon which the + domain is based). */ + + // Same comment as above about the count of N-faces. + unsigned npolygons = 0; + p_n_faces_fwd_piter<D, G> p(ima.domain(), 2); + + // FIXME: Merge this loop with the next one. + + for_all (p) if (ima(p)) + ++npolygons; + + // A neighborhood where neighbors are the set of 0-faces + // transitively adjacent to the reference point. + typedef complex_m_face_neighborhood<D, G> nbh_t; + nbh_t nbh; + mln_fwd_niter(nbh_t) u(nbh, p); + /* FIXME: We should be able to pass this value (m) either at + the construction of the neighborhood or at the construction + of the iterator. */ + u.iter().set_m(0); + + /* Compute the number of values (`size') to be passed as + second parameter of the POLYGONS keyword. */ + unsigned polygons_size = 0; + // Iterate on polygons (2-face). + for_all(p) if (ima(p)) + { + unsigned nvertices = 0; + /* FIXME: There may be a faster way to do this (e.g., + the neighbordhood may provide a method returning the + number of P's neighbors. */ + // Iterate on transitively adjacent vertices (0-face). + for_all(u) + ++nvertices; + // The number of values describing this polygon P is the + // cardinal of its set of vertices (1 value) plus the + // NVERTICES indices of these vertices. + polygons_size += 1 + nvertices; + } + ostr << "POLYGONS " << npolygons << ' ' << polygons_size + << std::endl; + + /* Output polygons (one per line), with their number of + vertices and the indices of these vertices. */ + // Iterate on polygons (2-face). + for_all(p) if (ima(p)) + { + unsigned nvertices = 0; + std::ostringstream vertices; + // Iterate on transitively adjacent vertices (0-face). + for_all(u) + { + // FIXME: Likewise, this is a bit too long... + vertices << ' ' << u.unproxy_().face().face_id(); + ++nvertices; + } + ostr << nvertices << vertices.str() << std::endl; + } + + /* ``5. The final part describes the dataset + attributes. This part begins with the keywords + POINT_DATA or CELL_DATA,followed by an integer + number specifying the number of points or cells, + respectively. (It doesn't matter whether POINT_DATA + or CELL_DATA comes first.) Other keyword/data + combinations then define the actual dataset + attribute values (i.e., scalars, vectors, tensors, + normals, texture coordinates, or field data).'' */ + + // FIXME: To do. + + ostr.close(); + } + + } // end of namespace mln::io::vtk + } // end of namespace mln::io } // end of namespace mln -- 1.5.6.5

1 0