2013-03-30  Potapov Ilya  <elias.potapov@gmail.com>

	* configure.ac: Reversed the order of the library checking. Also
	check for gsl_odeiv2_step_alloc function as it comes from the most
	recent version of GSL.

2013-03-29  Potapov Ilya  <elias.potapov@gmail.com>

	* src/input_interpreter.c (graphics_interp): HAVE_GNUPLOT gives
	the right to change graph_flag. The same in the read_ode.c where
	the .ode input file is read.

	* configure.ac: Extensive Gnuplot check in the configure.

2013-03-28  Potapov Ilya  <elias.potapov@gmail.com>

	* configure.ac: The first seeming plausible configure and
	corresponding makefiles are created for the package. configure.ac
	uses some trick to include the path for the libdin.a for Dinamica,
	since the package requires compilation during the functioning,
	which is not usual situation.
	HAVE_GNUPLOT definition, if the program is found during the
	configure check.

2013-03-22  Potapov Ilya  <elias.potapov@gmail.com>

	* src/integrator.c (run): MA analysis of the trajectory if the
	ma_span is specified.

2013-03-21  Potapov Ilya  <elias.potapov@gmail.com>

	* src/integrator.c (run): New moving average(MA) possibility for
	approximation of the trajectory.

2013-03-14  Potapov Ilya  <elias.potapov@gmail.com>

	* src/read_ode.c (process_inter_par): Print error if not found the
	option name.

2013-03-13  Potapov Ilya  <elias.potapov@gmail.com>

	* src/trajectory.c (get_dynamics_1d): If there are no peak/trough
	found during the exploration of the trajectory, there must be
	either not enough data (too little time of simulation) OR the real
	SS occured.

2013-03-12  Potapov Ilya  <elias.potapov@gmail.com>

	* src/graph.c (send_to_eps): Just plotting commands, not taking
	the file name. The latter is up to the corresponding menu
	functions in input_interpreter().

	* src/cross.c (period_cross): nper is updated inside the function
	call.
	(compute_hist_per): Wrong center of the histogram was fixed.

	* src/input_interpreter.c (periods_interp): per_thresh is the
	period threshold for the period_cross() function. Global variable.

	* src/cross.c (period_cross): New cross period algorithm
	accounting for the pre-defined threshold. The summing of
	subsequent periods is carried out to get the threshold.

2013-03-01  Potapov Ilya  <elias.potapov@gmail.com>

	* src/integrator.c (analyze_traj): The function takes an
	additional argument -- the output file name modifier.

	* src/input_interpreter.c (file_interp): Output file name modifier
	is changeable through the file menu.

	* src/init.c (init_command_line): out_name is used to keep the
	affix name for the output file. out_name is appended to the
	data_name to form output filename.

2013-02-25  Potapov Ilya  <elias.potapov@gmail.com>

	* src/continue.c (run_extend): It is possible to fit the
	simulation time to the desired number of periods. The function
	runs an addtional simulation in order to estimate the mean period
	and multiplies this estimate to the number of periods that user
	wants. This value is set for the total time of the simulation.
	(run_extend): Estimation of the period is carried out by the
	deterministic period, if one has it. However, mean of the
	stochastic periods is the bad estimate, so we use max as an
	estimator.

2013-02-23  Potapov Ilya  <elias.potapov@gmail.com>

	* src/input_interpreter.c (cont_interp): Run on data. If one has
	obtained the actual trajectories and wants to reanalyze them for
	different technical parameters, one can use this option.

2013-02-21  Potapov Ilya  <elias.potapov@gmail.com>

	* src/integrator.c (analyze_traj): Change the output file name,
	now with .csv extenstion and comma separated data.

	* src/input_interpreter.c (cont_interp): Range and values for the
	parameter to take in the continuation menu.

2013-02-20  Potapov Ilya  <elias.potapov@gmail.com>

	* src/continue.c (run_extend): Multiple runs for different
	parameter sets. Extended continuation menu. Rennovation of the
	menu has started.

	* src/file.c (set_info_data): Set the info array of the trajectory.

	* src/integrator.c (analyze_traj): Introduce a new file generated
	by the function. This contains the common entities computed from
	the data file. It has an extension to the <data_name> ".out".

2013-02-18  Potapov Ilya  <elias.potapov@gmail.com>

	* src/main.c (main): Mark the obsolete command line arguments in
	the function for the future remove.

	* src/trajectory.c (check_slope_ampl): The function has the sane
	amplitudes check: it checks with the certain level of error every
	pair that is subject to the comparison(according to the 95/5%
	criterion. In the pair, at lease one value must be larger than the
	error level.
	(slope): Increase in the slope amplitudes indicate the OS.
	(check_slope_ampl): 1% of the max amplitudes is the error level
	for the function to determine booms and drops of the amplitudes.
	(slope): Increase is measured in %, that is larger than 100% if it
	is real increase.
	(slope): We do not need 50% criterion for the equal amplitudes for
	the OS, since if only two last amplitudes are equal we are at OS.
	(slope): Debugging info on the increase of the amplitudes:
	percentage of increase on the each comparison test.

	* src/din_main.c (main): LDFLAGS can be /usr/local/lib as well as
	/opt/local/lib satisfying both Linux and Mac so far. At max, the
	gcc/ld will give a warning about non-existing directory for the
	libraries.

	* src/cross.c (peak_trough2): Remove the error level to compare
	with from the peak/trough determining function.

2013-02-17  Potapov Ilya  <elias.potapov@gmail.com>

	* src/graph.c (send_to_eps): Max file name length is 100 symbols.

	* src/integrator.c (write_data_file): Chenge the format of
	printing.

2013-02-16  Potapov Ilya  <elias.potapov@gmail.com>

	* src/trajectory.c (check_slope_ampl): Checking the amplitude is
	improved. Seems to work on all considered cases.

	* src/cross.c (peak_trough): Improved peak/trough computing
	function, taking into account for plateau of equal(!) values.

2013-02-13  Potapov Ilya  <elias.potapov@gmail.com>

	* src/random.c (generate_seed): generate seed for the random
	generator as clocks defined by the standard library function
	clock().

2013-02-11  Potapov Ilya  <elias.potapov@gmail.com>

	* src/integrator.c (analyze_traj): Plotting within this function.

	* src/read_ode.c (read_source): Correct position of the trailing
	null character for the var_name. Before caused a wrong var names
	read from the ode file.

2013-02-04  Potapov Ilya  <elias.potapov@gmail.com>

	* everywhere printf Qualifiers giving warning messages on Mac were replaced
	with more correct ones.

	* src/din_main.c (main): Changed compilation LDFLAGS from
	-L/usr/local/lib(Linux) to -L/opt/local/lib(Mac). This should be
	arranged automatically in the future.

	* src/read_ode.c (read_source): All realloc's on the NULL pointers
	must be substituted with malloc: dinamica fails to run on Mac
	without this.

	* Makefile (LDFLAGS+CPPFLAGS): Started maintainance of the program
	in Mac OS X. There are some issues on the Mac I have never
	encountered in Linux. There is a strong need in Autoconf scripts
	for the program.

2013-01-30  Ilya Potapov  <elias.potapov@gmail.com>

	* Makefile (SUBVER): Removed sub-sub-version of the program. The
	current release is 0.9.

	* src/trajectory.c (slope): Introduced the slope amplitude array
	size since it can be changed by check_slope_ampl(...).

2013-01-29  Ilya Potapov  <elias.potapov@gmail.com>

	* src/input_interpreter.c (numerics_interp): BSIMP method in the
	menu.

	* src/integrator.c (run): Bulirsch-Stoer implicit method for the
	stiff systems.

2013-01-28  Ilya Potapov  <elias.potapov@gmail.com>

	* src/trajectory.c (slope): Criteria for the oscillatory
	regime. Compare the last slope amplitude to the rest one by
	one. We are on the oscillatory regime if the last amplitude equal
	to any previous one.

2013-01-27  Ilya Potapov  <elias.potapov@gmail.com>

	* src/trajectory.c (slope): Detect the steady state by the
	computing the decrease in the slope amplitudes along the
	trajectory. This happens if the trajectory is not within the
	tolerance boundaries for being at the steady state, but tends to
	it.
	(slope): 50% of slopes demonstrating the monotonical decrease is
	the criterion for the steady state.
	(slope): Compare monotonical decrease with
	D(abs_am,rel_am,sl_ampl[i]) rather than with the absolute
	tolerance level(like in case of the very small amplitudes). The
	same criterion but of the opposite sing is applied to find
	oscillations.
	(slope): >50% slopes demonstrate monotonical decrease in amplitude
	AND >50% of the amplitude drop allows to claim the SS was observed.

2013-01-26  Ilya Potapov  <elias.potapov@gmail.com>

	* src/trajectory.c (slope): Detect the steady state by comparing
	with the system-wide tolerance level = eps_abs_am, in other word,
	this could be also interpreted as the computational error of the
	integration.

	* src/input_interpreter.c (periods_interp): Recompute common
	entities within the periodics menu.

2013-01-21  Ilya Potapov  <elias.potapov@gmail.com>

	* src/trajectory.c (slope): This new function gets the info about
	the slopes of the trajectory. Slope is the part of the trajectory
	between the peak and the trough or vice versa.
	(slope_ampl): Given the struct trajPeak of the trajectory the
	function computes the amplitudes of the slopes. This info can be
	used to assert the steady state and oscillatory attractors
	convergence.

	* src/cross.c (peak_trough): The function determines the peaks and
	troughs of the deterministic time series. This info is kept in the
	special struct trajPeak defined in the trajectory.h.

2013-01-14  Ilya Potapov  <elias.potapov@gmail.com>

	* src/graph.c (graph_set_labels): Graph: X11 first window for the
	kinetics/phase portraits, the second window is for the histograms
	"thist" and "mdist" types.

2013-01-12  Ilya Potapov  <elias.potapov@gmail.com>

	* src/graph.c (init_graph): Enhanced mode for X11 terminal of gnuplot.

	* src/input_interpreter.c (periods_interp): Do not print variable
	for periods calculation when no runs were performed.
	(periods_interp): perVarInd --> per_var.

	* src/init.h: We cannot remove per_var!!! It is a flag, while
	perVarInd is a real index.

2013-01-11  Ilya Potapov  <elias.potapov@gmail.com>

	* src/trajectory.c (mol_dist): Automatic detection of the variable
	during the creation of a molecular histogram.

	* src/input_interpreter.c (traj_interp): Molecular distribution in
	trajectory menu.

	* src/trajectory.c (mol_dist): free_frame was killing frame before
	it is used.

	* src/input_interpreter.c (traj_interp): Trajectory menu.

	* src/trajectory.c (mol_dist): New function to compute molecular
	distributions. First simple implementation.

	* src/cross.c (compute_period): perVarInd used everywhere instead
	of per_var.

	* src/input_interpreter.c (load_initials): New function for
	loading the initial conditions. Simple fscanf() implementation.
	(traj_interp): New trajectory menu is introduced.

2012-12-31  Ilya Potapov  <elias.potapov@gmail.com>

	NEW YEAR STARTS!

	* Makefile (SUBVER): 0.9.4a release.

