ADG Git Change Log

Age Message
7 days 13 hours doc: annotate AgdGtkArea:canvas-changed parameters

For some unknown reason, gobject-introspection is unable to get the
proper C type for the old_canvas argument of the "canvas-changed" signal
of AdgGtkArea object. Explicitely specifying its C type solved the

Closes issue #102:
7 days 20 hours i18n: update italian translations

Also, corrected some minor mistakes in cpml-demo localization.
7 days 21 hours adg: bare export API adg_canvas_export()

Added also the helper function adg_type_from_filename() that tries to
guess the surface type from the file name. Modified adg-demo to use the
new APIs accordingly.

WARNING: one argument of adg_canvas_export() is a cairo_surface_type_t.
Actually cairo does not export such type via gobject-introspection so
that API will not be available to language bindings. I made a pull
request to add support to that (and more) types:
The API will be available only when that PR will be merged upstream.

Closes issue #106:
13 days 3 hours doc: correct bad typo in NEWS
13 days 3 hours build: bump version to 0.7.5
13 days 3 hours doc: update NEWS
13 days 4 hours adg: use another diameter symbol by default

The UTF-8 symbol "\xE2\x8C\x80" is more semantically correct but not
backed up by the fonts (e.g. CentOS 5.4 and Windows XP). Use the
"\xC3\xB8" phi instead.
1 month 2 days doc: add Travis badge to README
1 month 3 days adg: correct adg_canvas_autoscale() bug

The above method manually call the arrange() virtual method instead of
emitting the "arrange" signal (it cannot, because it must skip the paper
entities). As a consecuence, the global matrix was not picked up.

Forced a global-changed signal before the arrange() calls.
1 month 9 days adg: merge AdgGtkLayout gtk2/gtk3 versions

The codebase differences between the two versions were too few to
justify a split in the source code.
1 month 10 days doc: unfinished SAIOT documentation

A new algorithm that tries to address the shortcomings of BAIOCA.
1 month 10 days adg: merge AdgGtkArea gtk2/gtk3 versions

The codebase differences between the two versions were too few to
justify a split in the source code.
1 month 10 days build: correct dependencies
1 month 22 days doc: correct hodograph in BAIOCA

This does not change anything in the algorithm because the error was not
1 month 22 days cpml: correct handcraft description
1 month 22 days doc: add some missing vector notation to BAIOCA
1 month 22 days cpml: clean up offset code and doc

Correct the description of the handcraft algorithm and cleaned up some
1 month 23 days cpml: split the geometrical offset algorithm

This algorithm was previously embedded in handcraft. Made it a separate
function that can be used as standard offset algorithm. Also, provided
overview documentation of the various algorithm types.
1 month 23 days cpml: avoid division by 0 in baioca

When the discriminant is 0, fallbacks in using the handcraft algorithm.
1 month 23 days demo: avoid GtkAlignment in cpml-demo

A similar result can be achieved by leveraging the padding property of
the children.
2 months 3 days cpml: add BAIOCA offset implementation

Implemented BAIOCA as described in nodist/baioca.tex. The result is
still suboptimal, hence BAIOCA is not the default offsetting algorithm.
2 months 3 days doc: remove duplication formula in baioca

Further simplification by removing some redundant math: the equation for
calculating Q_0 and Q_3 was bogus because Q_0 = C_0 and Q_3 = C_n.
2 months 3 days doc: correct and simplify baioca algorithm

There was a sign error somewhere: I changed and simplified the formulae
anyway and the error went away.
2 months 6 days doc: cosmetic improvements on baioca.tex

Using more idiomatic tags for math.
2 months 6 days doc: improve t_i methods section in BAIOCA

Corrected formatting, renamed variables to avoid clashes and corrected
some documentation bug in the section "Choosing t_i".