TODO

WISHLIST:
	Relative waist position
	Lens selection
	Sensitivity graph
	Lens movement magnification -> multiple views
	Telescope lens lock
	Print
	Forbidden area
	Diaphragme
	Relative distances on graph
	Export graph & tables to text data
	Optical cavities, eigen mode and stability.
	Waist relative to selection
	Wavefront on graph
	Tune the number of significative digits in editors and delegates
	grid
	when moving an optics, display the optics position
	make unit tests for saving/loading
	undo framework / file modified
	Thick lens, thermal lens
	Camera interfaces
	Set view limits to bench limits button
	TEMmn modes
	more intrication between opticsbench bounds and view bounds
	Scripting
	Unittest
	explore the q/n = (Aq/n + B)/(Cq/n + D) expression

BUGS:
	When adding an optics, the validity of the positioning (optics ovelap, opticsbench boundaries, exlcusion areas is not checked)
	When moving an optics to an invalid area, the optics does not stick to the closest possible point to the invalid area
	Save properties for H and V views
	Even when the table value did not changed, modified is triggered
	Update all data when chaging the delegate
	Save all views information (save beam scale, optics height and lock view)
	Different graphism for H/V lenses -> for instance more alpha ?
	Disappearing beam in test.xml
	Lock to create beam does not work : changing the position of createBeam should change the waist position i.e. the beam stored in create beam should be relative to the create beam position
	On the same account, the sensitivity does not work for createBeam
	After implementing the signal mechanism for optics, check that the signals are reset after cloning, so that clones don't emit messages
	Grammar : change optics properties order (common first, specific after)
	Grammar : Change orientation data type
	Grammar : add orientation attribute to genericABCD

MINOR BUGS
	Check that curved mirrors are flipped when loading and old files

FUTURE RELEASE BUG
	Problem with n > 1 and target beam
	Really find local optimum, not 100% local optimum
	Changing the waist in an intermediate beam might scramble the optics ordering with angle (e.g. linear cavity)

CONSTANT VIGILENCE:
	Change all parameters when changing the wavelength
	Check initial values
	Check that the save command saves all parameters & optics
	Translations
	Pedantic
	Valgrind
	API documentation
	check that .c_str() stl strings are enclosed in a QString::fromUtf8
	check const functions
	check that the program starts without settings

TESTS:
	check grammar conformity : xmllint --dtdvalid gui/xslt/gaussianbeam_1_2.dtd file.xml
	apply xsl stylesheet     : xmlpatterns

Problems with id :
	- in 0.4 (file format 1.1), there is a bug when loading an old file : all the id are loaded and set to 0, since the id was not existing...
	- in 0.4 (file format 1.1), the id counter starts at 0 when loading a file. For this reason, several optics might have the same id.