GAMGI Changelogs: GAMGI 0.10

Completed the conversion from chemical to crystallographic point groups of symmetry, thus ending the 0.10 series. Major achievements in this series: implemented the 230 space groups of symmetry, 3D extruded Truetype fonts, reciprocal lattice, stereographic projection, radical Voronoi tesselation, conversion from chemical to crystallographic point symmetry.

What is new:

Cleaned the m,C2 sort functions in gamgi_chem_symmetry.c.
When the point group is Undefined, in Periodic (restricted) mode, the elements with impossible rotation orders (not 6, 4, 3, 2) are now discarded. All other elements are shown (even if they are notoriously incompatible with periodicity, as for example too many mirror planes).
Updated Molecule->Measure and Molecule->Measure->Symmetry Help documentation, regarding chemical and crystallographic point groups.
Converted chemical Ih,I to crystallographic Th,T point groups: there are 5 sets of axes/planes to choose from when converting icosahedral to cubic groups.
Cleaned the code to convert chemical to crystallographic point groups, except I,Ih groups.
Removed an error when converting Dnd chemical to crystallographic point groups: C2h exists when n is odd and D2 exists when n is even.
Converted Dnd chemical groups to Crystallographic Dnd, C6v, C4v, C3v, D6, D4, D3, S6, S4, Ci, Cs and Ci.
Redesigned the chemical to crystallographic conversion code. It is much simpler, solid and easier to maintain, now.
Added Reduced International notation to identify point groups when applying crystallographic restrictions.
Added code to sort m, C2 elements oriented along the main direction.
Added T, Th, Oh building routines (to convert I, Ih, Kh chemical groups).
Chemical or Crystallographic point symmetry can now be selected in the Molecule->Measure dialog.
More cleaning in the code converting chemical to crystallographic point groups.
Converted chemical Dnh and Cnh to crystallographic (Dnh, C2v) and (Cnh, Cs) point groups, respectively. Tested against structures with higher rotation orders, written automatically, such as D30h and C30h. Added the XML files and the building utility to dat/.
We need to guarantee that, when they exist, the first plane is the horizontal plane and the first C2 axis is the vertical C2 axis, so the conversion between chemical and crystallographic point symmetry produces proper geometry for all symmetry elements.

What is next:

Add Plane->Modify and Direction->Modify functionality.
Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

Implemented the so-called Radical Voronoi Tesselation, where different polyhedra have different weights (given by the atomic radius) and added anisotropy tensor results. Improved the way area/node is handled in non-Miller planes. Improved the point group symmetry code, and started converting unrestricted (chemical) to restricted (crystallographic) point group information in non-periodic structures.

What is new:

Wrote utilities to automatically create molecule files with arbitrarily high orders, for Cn, Dn, Cnv, Sn groups.
Tested for chemical and crystalographic groups, for S10, S12, S14, S16, S18, S20, S22, S24, S30.
Conversion from chemical Cnv and Sn to crystallographic Cnv, Sn, Cn and Ci point groups is done.
Tested symmetry operations for C12, C24 and D12, D24 groups. Added C12.xml, C24.xml and D12.xml, D24.xml examples to dat/molecule/symmetry.
When 6 and 4 axes are both present in the chemical group, the crystallographic group can have as main order 4 or 6. Done for Cn and Dn groups.
Conversion from chemical Cnv and Dn to crystallographic Cnv, Dn, Cs and C2 point groups is done.
Conversion from chemical to crystallographic Cn point group is done.
Conversion from chemical Coov, Dooh and Kh to crystallographic C4v/C6v and D4h/D6h point groups is done: elegant and solid.
Improved the point symmetry code, in order to implement restricted (crystallographic) point group symmetry.
Simplified primitive reciprocal lattice colors: (0.5, 0.5, 0.5). It's simple, reliable, and almost always a good choice.
When creating (Cell->Measure) the reciprocal of a centered lattice, with a conventional volume, users can now choose to represent also the reciprocal of the corresponding primitive lattice.
Added screenshots to www.gamgi.org, showing stereographic projections and reciprocal lattices.
Updated cell_create.html, cell_modify.html and plane_measure.html documentation.
Cleaned the tutorial code: the names of the tutorial functions are now independent of their order, so this can be changed, and new tutorials added in any ranking position, without conflicts.
Added a new tutorial showing how to build a crystalline cell of Orthorhombic Polyethylene, using space group and Wyckoff positions data.
Updated the tutorial on Directional,Positional,Spot Lights.
Voronoi tesselation: added a variancy parameter in the Cluster->Measure dialog to control the partition weight scaling.
Voronoi tesselation: the radical tesselation, with different weigths for different polyhedra (defined by the atomic radius) seems to be working fine.
Cleaned Voronoi code and added atomic radius, in preparation for Radical Tesselation.
For planes with indices higher than Miller, the area/node now gives an error message. For planes with Miller indices or below, the area/node for Miller planes is shown.
Cleaned Voronoi code: lists of face vertices can be made cyclic now. Simplified static_face_orientate. Removed redundant static_angle.
Voronoi tesselation: the system tensor anisotropy and the polyhedron scalar anisotropy are now calculated. Cleaned code.

What is next:

Add Plane->Modify and Direction->Modify functionality.
Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

Improved the Voronoi Tesselation code, the Reciprocal Lattice code, the Stereographic Projection code, simplified the paths to get documentation, data and source files.

What is new:

Redesigned the reciprocal lattice code, to handle simultaneously the primitive lattice in centered lattices, and completed the special cases: hR and tI.
Cleaned gamgi_mesa.h.
Increased slightly the point size used to draw cell nodes, so these are still visible when another cell is present.
Found and removed a bug in Cell->Measure: the Gtk error dialog did not show up when an error was found. Some code automatically removed the task AND the message dialog without checking for errors.
When transforming a direct into a reciprocal lattice, when the direct lattice is centered, the corresponding primitive lattice can now be also transformed.
Updated documentation and data in http://www.gamgi.org/ and ftp://ftp.gamgi.org/gamgi/.
Improved name handling for planes and directions in gamgi_expat_export_object.c.
Corrected a little bug in gamgi_expat_export_object.c: plane->pole and plane->trace were compared to FALSE, not TRUE.
Improved name and error handling for Planes and Directions in gamgi_expat_import_object.c.
Stereographic Projection: added new pole symbols (masks) for planes and directions, so when they are in the same spot, planes and directions can still be seen (and measured), even if they use the same color.
Voronoi tesselation: face angles are now measured.
Removed a bug in Direction->Measure and Plane->Measure: in some cases (Plane-Direction and Direction-Plane angles) X flushing/synchronization was not done properly: now an expose event is triggered immediately after scanning for an object in the back buffer.
Changed the Stereographic Projection, so now only the planes and directions that are above can be seen.
Improved website documentation about Gamgi installation.
Improved error detection in Atom->Measure, Direction->Measure, Plane->measure dialogs.
Sections of the GAMGI website which are not related with documentation (like screenshot images), are not included anymore in the doc/ tree, resulting in much smaller documentation tarballs.
The paths to documentation files are much shorter and simpler now. This is valid for local files, that come with the distribution, and files available from http://www.gamgi.org/ and ftp://ftp.gamgi.org/.
Rotate,Move,Scale now also works when the selected object is Window, updating all layers in the window, both in Axes and Objects mode.

What is next:

Add Plane->Modify and Direction->Modify functionality.
Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

Created new dialogs Atom->Measure, Plane->Measure, Direction->Measure. Several new properties involving angles, lengths, areas can now be measured. Crystallographic planes,directions can now be imported,exported to XML files.

What is new:

Distances between an atom and a 2-atom direction, and an atom and a 3-atom plane can now be measured.
Distances between an atom and a crystallographic direction, and an atom and a crystallographic plane can now be measured.
Distances between nodes in crystallographic directions, linear and planar node densities can now be measured.
Angles between crystallographic planes, directions and planes with directions can now be measured.
Implemented the Atom->Measure dialog with the new GUI rules and removed the Group->Measure dialog, replaced by Atom->Measure.
Improved and cleaned gamgi_gtk_plane_measure.c and gamgi_gtk_direction_measure.c, so now the dialogs Plane->Measure and Direction->Measure are much more user-friendly.
Implemented the new Plane->Measure dialog.
Designed and implemented the Direction->Measure dialog.
Improved the whole GUI strategy to handle measuring properties.
Plane, Direction objects can now be imported from XML files.
Updated documentation for Cell->Measure and Plane->Measure.
Plane, Direction objects can now be exported to XML files.

What is next:

Add Plane->Modify and Direction->Modify functionality.
Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

Added Wigner-Seitz cells for hR rhombohedral lattices. Improved stereographic projection for planes and directions. Improved reciprocal lattice functionality.

What is new:

Filtered inner faces for hR high and low Wigner-Seitz cells.
The high c/a Wigner-Seitz cell for the hR lattice is working.
Found and removed a bug in gamgi_expat_import_object.c: Wigner-Seitz cells for monoclinic and triclinic lattices (not implemented yet) were not flagged down as an error.
The low c/a Wigner-Seitz cell for the hR lattice is working.
Calculated the coordinates for both Wigner-Seitz hR cells, with low and high c/a.
Cleaned the plane,direction code, including gamgi_engine_start.c, gamgi_engine_copy_out.c, gamgi_phys_projection.c and gamgi_math_plane_measure.c.
Improved the interface handling the reciprocal lattice constant.
The reciprocal lattice constant can be chosen now. The new default is the so-called Bravais polar lattice constant, where the volume per node is equal for direct and reciprocal lattices.
Improved Stereographic Projection: i) when a pole points to the half-sphere below, the inverted direction is now used; ii) only the trace in the half-sphere above is now represented;
Corrected Layer->Create and Light->Select HTML Help pages.
When calculating the area per node, multiple planes are taken into account: distance between adjacent planes is 1/2 or 1/3.
Added dialog Plane->Measure. Currently only measures the distance between adjacent planes, and the area per node.

What is next:

Add XML input/output functionality for plane/direction objects.
Add Plane->Modify and Direction->Modify functionality.
Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

Removed a couple of serious bugs, in Space groups, Reciprocal Lattice and Voronoi Tesselation. Plane->Select, Plane->Remove, Direction->Select, Direction->Remove dialogs are implemented and working. Plane->Create and Direction->Create have more features, are more robust and user-friendly.

What is new:

Found and removed a fresh bug in Planes,Directions: origin was not initialized, sometimes producing erratic values.
Disabled Scale,Move,Rotate for Planes,Directions, when they are the selected objects. Initialized Plane and Direction origin to Cell origin. All this needs to be improved in the future.
Implemented Plane->Remove and Direction->Remove dialogs, so single Planes and Directions can now be easily removed.
Improved Plane->Select and Direction->Select dialogs, which now recognize and select planes and directions, respectively.
Improved Plane->Create and Direction->Create dialogs, which now support both conventional and primitive vectors. Moreover, both dialogs are now initialized and reset.
Cleaned gamgi_gtk_cell_measure.c. In particular, current layer parameters are now copied to the new layer, when determining the reciprocal lattice, instead of using the default layer parameters as before.
Found and removed a bug in the reciprocal lattice code: parameters for R cells were still determined with the pre-GAMGI 0.10.5 convention of using primitive rather than conventional hexagonal parameters to define R cells.
In stereographic projection, added code to get directions normal to planes, using reciprocal lattice vectors. Added code to handle plane and direction indices based in conventional and primitive vectors.
Found and removed a bug in the reciprocal lattice code, related with the cell center: center was calculated too late.
Corrected edge faces and edge lengths, for topologies not well formed, in the Voronoi code. Cleaned the statistics part.
Found and removed a bug in the Voronoi code: when the topology is not well formed (aka C60), an infinite loop could appear.
Updated documentation about Euler angles, Text objects, the Freetype library and the GAMGI_TEXT environment variable.
Found a removed a bug in the Cell Link code: the inside_triangles routine was receiving always the first coordinates.

What is next:

Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

Started working on crystallographic plane and direction objects: stereographic projection is working, including Wulff (mostly used in Materials Science) and Schmidt (mostly used in Geological Sciences) projections. A bug was fixed in the Euler angles code.

What is new:

Updated the pathnames in the gtk/*/* files.
Wrote gamgi_gtk_plane_create.c: the Plane>Create dialog is working, for both Wulff and Schmidt projections, for all planes, colors, poles and traces.
Found and removed a rather old bug in the function gamgi_math_euler_from_axis: sometimes the Euler angles were wrong: 90 instead of 270. Cleaned the Euler angles code.
Andre made new rpms for GAMGI 0.10.6 and Suse9.x. Cleaned the HTTP and FTP download sections, at www.gamgi.org.
Wulff and Schmidt stereographic projections are now rotating fine, in interactive mode (with the mouse or the ruler).
Updated gamgi_mesa_draw and gamgi_mesa_scan, to work with planes and directions.
Updated gamgi_engine_create, gamgi_engine_start, gamgi_engine_copy_out, gamgi_engine_remove, to work with planes and directions.
Adapted and cleaned gamgi_expat_import_object and gamgi_expat_export_object, in preparation for the stereographic projection cell representation.
Adapted and cleaned gamgi_gtk_cell_create and gamgi_gtk_cell_modify, in preparation for the stereographic projection cell representation.
We have now a simple and flexible model to handle everything that is needed with stereographic projections: we will add a new tupe of cell representation, a projection, specifically to contain and handle planes and directions. Multiple cells and relations between them will be handled by Assembies (including epitaxial relations).
Pole and trace representations seem to be working fine, for both Wulff and Schmidt projections, for all types of lattices, for all possible rotations.
Started working on Plane and Direction objects.
The mathematical constructions needed to handle Stereographic (Wullf) and Schmith projections are worked out.

What is next:

Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

This is a cleaning release: reviewed the Voronoi code and removed a serious bug, improved the way Light shininess is handled, added an environment variable so solid fonts can now be easily found, clarified that code exporting to raster .ps files is fine and usable.

What is new:

Found and removed a bug in the Voronoi code: the number of faces per edge was not properly calculated. Edges are now are linked for each polyhedron AND for each face.
Added code to find equivalent vertices and edges, when using periodic boundary conditions, in the Voronoi code.
Cleaned the Voronoi code. Searching for redundant vertices now requires looking only into 8 rather than 27 cells.
Added a new environment variable, GAMGI_TEXT, to indicate the path to Vera truetype fonts.
Cleaned the postscript exporting code.
Clarified why the postscript code was not working properly: in Gimp we need to use a resolution of 72dpi (the default is 100). In practise this means that exporting to .ps files is working fine. We tested this issue also on Windows, on 3 programs, and all of them worked fine with our .ps files.
Cleaned again the text font code. Now when a text object is modified, the font glyphs must be available (even if they were previously compiled), otherwise an error is triggered.
Special XML characters must be translated to predefined entities when exporting, otherwise they cannot be reimported. Fixed.
Improved the way we handled Light shininess: Shininess is disabled for 0.0, is minimum close to 0.0 and is maximum for 1.0.
Only wired text objects need line width control. Fixed.
Added a new screenshot about crystallographic space groups.

What is next:

Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

Updated/added documentation for Text and Light formats, and Text and Cell dialogs. Improved solid fonts code. Rhombohedral cells now are defined with conventional (hexagonal) cell parameters.

What is new:

Added a new screenshot, showing how the Cell->Link dialog is used to create a crystal using space group information.
Updated help documentation for Cell->Create, Cell->Modify and Cell->Link.
Found two bugs in the code handling solid fonts, resulting in segmentation faults when font files are modified or removed AFTER removing the font files. Removed one, by checking first if glyphs must be compiled. Removed the other by checking if the glyphs are available, when calculating the text center.
When solid Truetype fonts are not found, the error message produced now shows the name of the file that is missing, both in Gtk Text->Create, Text->Modify dialogs and from XML files.
Rhombohedral cells are now defined using a,c hexagonal cell parameters rather than a,bc rhombohedral parameters as before. This affects only the lattices for the 7 R space groups. Essentially crystallographers tend to define cells belonging to these groups using hexagonal, not rhombohedral, parameters. Moreover, this case was an exception, from now on parameters are always for the conventional cell, which is simpler for users and coders alike.
Added a screenshot to gamgi.org with stroked and extruded fonts, illuminated with directional and spot lights.
Updated format documentation for <light></light> (Create) and <gamgi><light></light></gamgi> (Config) blocks in XML files.
Updated format documentation for <text></text> (Create) and <gamgi><text></text></gamgi> (Config) blocks in XML files.
Updated documentation for Text->Create and Text->Modify dialogs.
Scaled up very slightly the wired fonts, so wired and solid and fonts now look essentially the same size.
Changed the expected names for the 10 Vera Truetype files, back to the original names in Suse 9.0, hopefully this wil help finding these files in other Linux distributions.
Added documentation for Cell->Measure, for the reciprocal lattice transformation.
Added the relevant original Allen Barnett's comments to the extruded fonts code, plus metrics information taken from the Freetype documentation.
Cleaned the code in gamgi_mesa_text.c; defined the scaling procedure for solid fonts, so wired and scaled fonts look the same size; calculated the center for solid text objects.

What is next:

Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

Text and Light properties can now be configured, both from GUI dialogs and XML files. Solid (extruded) Truetype fonts can now be created/modified from GUI dialogs and imported/exported to XML files.

What is new:

When objects and lights have null overlapping color components, the light cannot iluminate the object and the object becomes invisible. This more testing, and perhaps an option for Light Materials functionality.
Found and removed a bug in the code initialising text objects: font was started, so a core dump would be produced when a parsing error was found as the font engine would try wrongly to decrease the glyph count.
Cleaned all the text font engine code and improved the text font API, to acommodate all the GUI and XML code needs. The dialogs Text->Create and Tex->Modify and the XML import/export files are now be handling wired and solid fonts just fine (famous last words!).
Solid fonts are now working fine. The code to center the text objects, for both wired and solid fonts has been redesigned and is working. Removed gamgi_mesa_center_out.
Wired fonts are working again, in all cases, with the new font engine. Cleaned the text code. Solid fonts are working, but still need some work.
Changed completely the text font engine. Before, when a glyph was required, the whole font was compiled. Now only the glyphs actually needed are compiled. This is important mainly because the new solid fonts require much more resources.
Refined the GUI handling Systems,Lattices,Groups in Cell->Create and Cell->Modify: the toggle buttons handling the Systems,Lattices,Groups Lists can now be used to create AND remove the dialogs. Previous methods to close these dialogs (Cancel button and WM decorations) remain available, of course.
Removed a bug in the the Cluster Voronoi code: the statictics for global edge lengths were calculated using redundant edges, not just the real ones.
Updated the website Links and Contributors pages, to include all the contributions for the extruded fonts code.
Cleaned and integrated further the text code: Removed files gamgi_gtk_text and gamgi_mesa_extruded and added the code to gamgi_gtk_text_config, gamgi_gtk_text_modify and gamgi_mesa_text.
Wired (tested) and Solid (not integrated yet) fonts can now be imported and exported to GAMGI XML files.
Found a bug in the XML export code: the XML space offset is not correct for text configuration data: Fixed. We were sending garbage, instead of the right argument, to a function. Surprising that this was not found before...
Light and Text configuration parameters can now be exported to XML files. Corrected a minor bug in the XML layer export code: the decimal precision was wrong for some position components.
Modified Text->Create and Text->Modify dialogs so they are now ready to work with both Wired and Solid fonts.
Default Text Style, Color, and Solid and Wired fonts, can now be configured, selecting Text->Config or importing config data from XML files.
Materials Light Shininess and default Light components can now be configured, selecting Light->Config or importing config data from XML files.

What is next:

Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

Solid extruded fonts are now working on GAMGI, thanks to Allen Barnett and Linas Vepstas, of OGLFT and GLE libraries, respectively, which code we used, and the team developing the FreeType library, which is now required. A screenshot is available here. The interface with the user is not built yet, so this is just a working release, not for final users.

What is new:

Removed the global variables introduced earlier in the extruded fonts code.
After sorting out a few glitches, the extruded fonts are now running great on GAMGI!. Wait for the screenshots...
GAMGI compiles with the solid font code: the FreeType library is now required to compile and run GAMGI.
Cleaned the solid fonts code a bit more. There is a bug with lights and Scalef probably due to unormalized normal vectors, which was already in the original code.
The solid fonts code has now less than 800 lines and the extrusion code is fully understood.
Cleaned and merged the OGLFT and GLE C codes into a single file, with just 844 lines.
Introduced C-based dynamic arrays, a new type of data object inspired in C++ <vector>, that should help to reduce the number of reallocs directly hard-coded in GAMGI, making the code more elegant and simpler to maintain.
Removed the last C++ header in the OGLFT code, <vector>, and converted the last C++ bits to C. The whole code is now ansi C and works fine.
Removed C++ standard libraries, <map> and <list>, in the OGLFT code, and replaced by GAMGI lists.
Simplified the GLE code to 122 lines only, and removed all the macros.
Reduced all the code to render solid fonts to 3 files (.c and .cpp and .h), plus a bunch of macros.
The GLE code to render solid fonts is now down to 420 lines in .c files and 633 lines in .h files, and the GLE and OGLFT codes are working fine together, producing exactly the same output as the pristine GLE and OGLFT libraries combined. From now on, it should be just a matter of time until we have solid fonts in GAMGI, linking against the FreeType library! Of course this is only possible due to the hard work of Linas Vepstas (GLE), Allen Barnett (OGLFT), and all the people working on the FreeType library, many, many thanks to all!
The whole OGLFT code to render solid fonts is now down to less than 800 lines, including header and demo code. Started working on the GLE library.
Stripped down solid.cpp to 766 and solid.h to 82 lines only. Essentially we have now to replace the STL functionality by our own to have pure C code. Moreover, the code doesn't look too difficult to understand.
We have a working example showing a message with a solid font, stripped down from the OGLFT C++ library, with two files, solid.cpp (1048 lines) and solid.h (320 lines) only.
We have a working model for 3D solid (extruded) fonts, which requires three libraries: Freetype, OGLFT and GLE. Our goal is to adapt the OGLFT and GLE code to GAMGI, to decrease the number of dependencies, keeping just Freetype, which is a larger lib and is one of the dependencies of Gtk2.0, so we will need it anyway if we move to Gtk2.0.

What is next:

Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

Reciprocal lattices can now be determined for all 14 Bravais lattices. A bug was removed in the code converting primitive to conventional axes in Rombohedral lattices.

What is new:

Cleaned the reciprocal lattice code.
Tested the OpenGL Light and Materials properties. Apparently we are following the best solution on this.
The orientation of reciprocal Triclinic and Monoclinic lattices is not correct: a change in orientation occurs in the transformation that we are not taking into account. Solved: get the rotation matrix that takes the conventional vectors in the standard orientation to the final orientation.
Decreased the minimum and maximum number of slices used to draw atomic spheres to 4 and 64, respectively.
The reciprocal lattice stuff seems to be finished, apparently is working correctly in all cases, even the tricky ones involving multiple rotations around different centers.
Reciprocal lattice for Tetragonal I is tricky because we do not work with the F reciprocal lattice but with the equivalent I lattice, so we need to scale and rotate 45 degrees. Reciprocal lattice for Hexagonal P is tricky because a rotation of 30 degrees occurs during the transformation.
Found and removed a bug in gamgi_phys_cell_create: the hexagonal height in rombohedric cells was wrong. As a result of this, the reciprocal lattice now works fine for this case.
Reciprocal lattices seem to be working fine, except for Hexagonal P, Hexagonal R and Tetragonal I.
Using Freetype, OGLFT and GLE, we have demos with solid fonts rotating fine. The problem now is how to adapt this to GAMGI.
Thanks to Andre, we have new packages available, for Gamgi 0.10.1: rpms for Suse 9.1, and for the first time, deb packages as well, for Woody 3.0 r1.
Renewed gamgi.org, which is now valid until 2010.

What is next:

Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.

The 230 crystallographic groups are now fully working. The structures linked to Wigner-Seitz cells are greatly improved. Distributed the gtk code by several sub-directories.

What is new:

The methodology to convert direct to reciprocal Bravais lattices seems to be clarified, hopefully this will be working soon.
The algorithm to link Wigner-Seitz cells to atoms showing all the relevant atoms and the proper symmetry is working fine.
Cleaned the gamgi_phys_cell_link.c code that handles the container cell, including static_cell and the if statement to call it.
Refined the cell link interface: users are not allowed anymore to add nodes to conventional volumes or to remove nodes from the other volumes.
Removed a bug in gamgi_phys_space.c code that prevented cell link to work correctly when only a motif was entered.
The code to handle space group information is completed and seems to be working well, though it is not properly tested yet. It should be able to build crystals for all 230 space groups, for all Wyckoff sites, for all types of cell volumes. Moreover, users can combine space group information with node selection and occupancy rules.
Merged and updated gamgi_phys_space.c, gamgi_phys_cell_link.c and gamgi_gtk_cell_link.c (found and removed a typo bug in gamgi_phys_cell_link.c affecting the size of the outside cell when the volume is a sphere).
Updated the HTTP and FTP website sections.
Prepared the dialog and layer infra-structure to determine and show reciprocal lattices for cell objects.
Completed and added the 230 arrays describing Wyckoff site symmetry.
Writing the Wyckoff site symmetry in arrays (one for each group, as for positions and drivers): more than half is done.
More cleaning on gamgi_gtk_cell_link.c, which is now ready to handle all the space group information.
The Cell->Link dialog is automatically creating and removing menuitems for all Wyckoff positions, for all the 230 space groups.
Redesigned the Cell->Link dialog: added Wyckoff positions and removed Directions.
When importing files from the command line, the code printing the current layer and object in the statusbar was not correct: removed the bug, removed the variables windows_top and layer_top, and cleaned the code.
Created sub-directories in src/gtk, for each object classe, to avoid hundreds of files in a single directory. Changed the makefiles accordingly (it is still possible to compile Gamgi from any of the directories containing source code).
The standalone code to generate Wyckoff positions for all the sites of the 230 space groups seems to be working fine. When repeated positions are found, it also tries to find a Wyckoff site with lower multiplicity (and higher symmetry) that generates the same positions.

What is next:

Replace set/get_data functions by structures defined in gtk/gamgi_object.h headers.
Add Object->Copy dialogs with basic copy functionality.
Add Contents and Container options in Object->Remove dialogs.
Improve Move and Rotate interactive usability.
Implement dynamic text objects, to measure distances, angles, etc. on real time.
Implement global selection, including global picking algorithms and selection dialogs that include all the selection methods.
Implement multiple object picking, including a dynamic frame with moving ants and dragging anchors, to select multiple objects and images from the screen.
Review the entire GAMGI code, file by file.