2006-07-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.9.7

	* Added polyxmass-bin.desktop entry (slightly modified from the
	one by Phil Bull (thanks to Barry deFreese <bddebian@comcast.net>
	(Ubuntu project) for providing a patch, although based on the
	debian/polyxmass-bin.desktop infrastructure).

2006-07-07  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass/(any relevant function): changed the copy/cut/paste
	and selection stuff from the previous selections mechanisms to the
	GtkClipboard mechanisms. The functionality is the same. No
	user-side change. Compiles under MS-Windows, but does not work,
	like the previous version: still have the OpenClipboard failed
	with invalid window handle (gdk_selection-win32.c:1068) message.

	* polyxmass/polyxmass-ui-seqed-widget-clipboard-import.c: renamed
	this file and contained function to a more suitable name :
	polyxmass-ui-sequence-purification.c.

2006-06-26  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.9.6 (that compiles under MS-Windows).

	* polyxedit: wrote code to allow importing a sequence from a raw
	text file.

 	* Whole_project: since a number of months the development has
	brought a new organization of the code. The porting of GNU
	polyxmass to Microsoft Windows has promted to remove the
	dependency between polyxmass-bin and libpolyxmass. The
	libpolyxmass code tree is now part of polyxmass-bin (libpolyxmass
	subdir) and is compiled as a convenience library. The source code
	for the polyxmass-bin package is now located not in src but in the
	polyxmass subdir. Also dependency over popt.h was removed, as
	libglib provides options stuff.

2006-04-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.9.5 (that now depends on the libpolyxmass11 library
	package version 0.9.1).

2006-04-09  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in: added AM_MAINTAINER_MODE to configure.in.

2006-04-05  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-polchemdefctxt.c (polyxmass_polchemdefctxt_unref):
	commented out the "about to close last polymer chem. def context"
	g_message call.

2006-03-21  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-ui-console.c (polyxmass_console_wnd_focus_in_event):
	added this function to remove the urgency hint to the console
	window once the focus has been set to that window.

	* configure.in: had to bump the gtk+ requirement to 2.8.13 in
	order to be able to use the gtk_window_set_urgency_hint()
	function.

	* polyxmass-globals.c (polyxmass_globals_log_* functions): removed
	the call to force the console window to show up and take
	focus. The users are now invited to put the console window in a
	place clearly visible... Now these function use the
	gtk_window_set_urgency_hint() brought by Gtk+ 2.8.

2006-03-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.9.4 (no change regarding the libpolyxmass dependency
	nor the other polyxmass packages).

	* polyxmass-ui-main.c (on_polyxedit_new_sequence_file_activate):
	same as below.

	* polyxmass-ui-filechoosers.c
	(polyxmass_filechoosers_polseq_init_cancel_button_clicked):
	changed code so as to remove the critical messages that are
	displayed if one cancels (or closes the window) without going
	through all the setting up of the new sequence.

	* polyxmass-globals.c (polyxmass_globals_log_warning_msg): the
	console window does not present itself to the user anymore. Same
	for message msg. Critical messages do retain that feature,
	however.

	* polyxcalc-ui-recorder.c
	(polyxcalc_ui_recorder_wnd_setup_window): changed the creation of
	the window procedure so that the recorder window and the chempad
	window start living with a reasonable size.

	* configure.in (CONFDIR): I have reversed the changes done in
	configure.in about testing the host during compilation so as to
	set CONFDIR to an appropriate value. CONFDIR now always is set as
	${sysconfdir}/polyxmass. Since ${sysconfdir} equiparates to
	${prefix}/etc if no value is given to ./configure, then it should
	be OK.

2006-03-19  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* all_project (glade_files+[ch]_source_code): changed a lot of
	Polymer Sequence Properties expander to show not expanded at
	first. This is more intuitive as in a big number of cases a single
	polymer sequence is open at any given time, thus there is no need
	to have unambiguous identity data available. The code was adapted
	to the glade changes accordingly. Thus dependency on the new glade
	files has to be enforced (Debian GNU/Linux-only).

	* polyxmass-ui-filechoosers.c
	(polyxmass_filechoosers_polseq_init_treeview_row_activated): added
	this function to trigger activation through activation of an item
	in the treeview. This is also true for the polchem_def and
	atom_def variants of the same source file.

	* polyxedit-ui-cleave-options.c
	(polyxedit_cleave_opt_wnd_cls_treeview_row_activated): added this
	function to activate the cleavage only by double-clicking onto a
	cleavage specification row in the treeview.

	* configure.in (PKG_CONFIG_PATH): the CONFDIR pathname is now set
	in a case statement, where the host system name is tested. If
	"darwin" is found, CONFDIR=${prefix}/etc/polyxmass. Otherwise it
	is ${sysconfdir}/polyxmass.

2006-03-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-ui-filechoosers.c
	(polyxmass_filechoosers_setup_polchemdef_wnd): fixed bug due to
	calling gtk_widget_show_all (window) while in fact some widget
	therein had to be hidden (and actually were hidden). This resulted
	in the re-appearance of widgets which looked misleading to the
	user and caused bad gtk boxing behaviour. Thanks to Lionel Elie
	Mamane for pointing that bug to me.

2006-03-11  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.9.3 (no change regarding the libpolyxmass dependency
	nor the other polyxmass packages).

	* Updated the translation (French).

	* Whole_project (configure.in): updated the dependency on
	libgtk+-2 because of use of the GtkAboutDialog functionalities
	(2.6.10; latest of the 2.6.x branch). Removed the libglib-2
	dependency because it is automatically dealt with by libgtk+-2.

	* glade/polyxmass.glade: Put the "console window visible" menu in
	the main "View" menu instead of the "File" menu where it did not
	belong. Removed the old About window object.
	
	* configure.in: changed the dependencies due to using the
	GtkAboudDialog functionalities.

	* polyxmass-ui-main.c (on_polyxmass_about_activate):
	re-implemented the About dialog the GtkAboutDialog programmatic
	way (with gtk_show_about_dialog()). This fixed a bug that
	occasionally would show up when displaying that dialog
	window. Further, this ensures that no more than one dialog will
	open at any given time. Much better.

2006-03-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_update_monomer_flag): changed code so that
	only NOTE and MODIF properties are taken into account when
	updating the flags for monomer characteristics display in the
	sequence editor. Removed the drawing area and related code that
	used to show the presence of mere prop objects in monomers, as now
	NOTE and MODIF property objects are the sole prop objects avaible
	to annotate/modify monomers. Thanks to Mark Tracy for pointing these
	problems to me.

2006-02-21  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.9.2 (no change regarding the libpolyxmass dependency).

2006-02-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-searchmass-options.c and
	polyxedit-ui-searchmass-results.c: The problem was that the Search
	Mass results window did not get registered correctly by the
	function in the Search Mass Options window. Thus, when the results
	window was opened with the masses that were found, and when the
	user would want to perform reporting for these results, this would
	fail, as the results window would not be listed among the
	available windows for reporting.

	Because the registering of the result search masses should create
	a "wnd_desc" descriptive string for the results window, that
	g_object_data was not available, and when one would try to "find
	masses" inside the search masses results set, the program would
	crash, because as part of displaying the window in which found
	masses are listed, there is the construction of a string making
	use of the "wnd_desc" string created above. Because that string
	was not available, the program crashed.

	All this is fixed now. Details for each function are below.


	* polyxedit-ui-searchmass-options.c
	(polyxedit_searchmass_opt_wnd_mass_entry_activate): fixed problem
	with registering the results' window so that reporting is enabled
	for the results of the search mass operation (it was not possible
	to ask for reporting of the results, because the window containing
	the results did not show up in the window management
	window). During the registration of the window, creation of the
	"wnd_desc" textual description of the purpose of the results
	window that did not exist in the prior version of the software,
	which provoked crashes.
	(polyxedit_searchmass_opt_wnd_search_button): same as above.

	* polyxedit-ui-searchmass-results.c
	(polyxedit_searchmass_res_wnd_setup): fixed bug due to calling
	glade_xml_get_widget (xml, "identity_number_entry"); instead of
	glade_xml_get_widget (xml, "polseq_id_number_entry"); Indeed, the
	glade file has the "polseq_id_number_entry" widget name for the
	entry where the polymer sequence identity number has to be
	displayed.

	* polyxmass-ui-note.c (polyxmass_note_setup_wnd): changed
	"identity_number_entry" widget name with the
	"polseq_id_number_entry" (also the corresponding glade file).

2006-02-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.9.1 (no change regarding the libpolyxmass dependency).
	
	* polyxmass-polchemdefctxt.c
	(polyxmass_polchemdefctxt_init_with_type): fixed a bug due to not
	setting properly the polchemdef->type value after rendering a
	polymer chemistry definition from file (bug occurs only if a
	polymer chemistry definition file is rendered as an alias).

2006-01-06  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.9.0 that depends on libpolyxmass10:0:0 (version 0.9.0)

	* polyxmass-main.c (main): comply with const gchar *.

	* polyxmass-config.c (polyxmass_config_userspec): comply with
	const gchar *.

	* polyxcalc-ui-calculator.c
	(polyxcalc_ui_calculator_wnd_fill_atoms_combobox): memleak fix.

	* polyxmass-polchemdefctxt.c
	(polyxmass_polchemdefctxt_init_with_type): memleak fix due to not
	acknowledging that the 'as' PxmAtomSpec was allocated.

2006-01-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-main.c (polyxedit_main_prepare_new_context_open_file):
	memleak fix.

2005-12-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-ui-findrep.c
	(polyxmass_findrep_replace_monomer_in_seqed_widget_with_monomer_triad):
	fixed bugs due to not handling the graphical rendering of the
	monomer modifications upon Find/Replace operations.

	* polyxmass-ui-seqed-widget-mnm-modif.c
	(polyxmass_seqed_widget_mnm_modif_un_modify_monomer): fixed a bug
	due to assigning a pointer variable a bad value.

	* polyxmass-ui-findrep.c
	(polyxmass_findrep_replace_in_seqed_widget_with_findrepopt): fixed
	a bug due to not setting the current_idx to a proper value
	depending on the fact that the replacement was backward or not
	(which makes a huge difference).

2005-12-19  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Run the following command : gettextize --copy --force --intl to
	force the copying (and not the symlinking) of all the files
	required by intltool and gettext. The instructions after that run
	were :

	8<---------------------------------------------------------------
	Updating EXTRA_DIST in m4/Makefile.am (backup is in m4/Makefile.am~)
	Adding an entry to m4/ChangeLog (backup is in m4/ChangeLog~)

	Please run 'aclocal -I m4' to regenerate the aclocal.m4 file.
	You need aclocal from GNU automake 1.5 (or newer) to do this.
	Then run 'autoconf' to regenerate the configure file.

	You might also want to copy the convenience header file gettext.h
	from the /usr/share/gettext directory into your package.
	It is a wrapper around <libintl.h> that implements the configure --disable-nls
	option.

	Press Return to acknowledge the previous two paragraphs.
	--------------------------------------------------------------->8
	
	All of which I did diligently. 

	After that run the command :

	intltoolize  --automake --copy --force

	which will copy the files intltool-extract.in intltool-merge.in
	and intltool-update.in thus removing their symbolic links.q


	And next I issued '$ make maintainer-clean ; ./removable-files'
	and '$ ./autogen.sh' which ran fine ! 
	
	There is no single symbolic link in the source tree anymore.
	
	Finally, make works flawlessly !

	* polyxmass-ui-seqed-widget.c
	(polyxmass_seqed_widget_integrate_monomer_at_idx): changed if
	(PXM_SEQED_WIDGET (widget)->last_point_1_idx >= length) to idx >=
	length for the monicon insertion.
	(polyxmass_seqed_widget_remove_monomer): added

	g_signal_emit (G_OBJECT (widget),
	seqed_widget_signals[SEQUENCE_MODIFIED], 0);

	as soon as a monomer is removed from the polymer sequence
	monomerGPA!
	
2005-12-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Added pairs : polyxmass-findrepopt.[ch],
	polyxmass-ui-findrep.[ch] and
	polyxmass-ui-findrep-options.[ch]. These pairs of files were
	needed to implement the Find/Replace functionality. The
	Find/Replace functionality is what missed most to the
	polyxmass-bin package. It is now coded and apparently functional.

	* Whole_project : modified code where needed to reflect changes in
	the way libpolyxmass' functions characterize what should be
	compared/duplicated when dealing with objects. The main changes
	are of this kind : change 'PXM_UNDEF_PROP_CMP' to
	'PXM_CMP_NO_DEEP', for example. No changes whatsoever in the
	functionality. Other example, 'PXM_FINDMASSOPT_DUP_DEEP_YES'
	changed to 'PXM_DUP_DEEP'. This simplification in libpolyxmass
	leads to a nice simplification in polyxmass-bin.
	
2005-12-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-ui-seqed-widget.c
	(polyxmass_seqed_widget_integrate_monomer_at_idx): added param to
	function to let the caller specify if the graphical rendering
	should take into account evental modification in the monomer.
	(polyxmass_seqed_widget_integrate_monomer_at_point): added param
	to function to let the caller specify if the graphical rendering
	should take into account evental modification in the monomer.
	(polyxmass_seqed_widget_integrate_sequence_at_point): added param
	to function to let the caller specify if the graphical rendering
	should take into account evental modification in the monomer.
	(polyxmass_seqed_widget_integrate_sequence_at_point): added call
	to pxmchem_monomer_free (monomer_new); in case the monomer integration fails.

2005-11-13  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-pka-ph-pi.c
	(polyxedit_pka_ph_pi_wnd_make_report_text_format): fixed a glitch
	due to passing a end_idx too large to the function making a string
	with a given coordinate (pxmchem_polymer_make_codes_string). This
	value was corrected in pxmchem-polymer.c OK but issued a warning
	message in the console window.

2005-11-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-ui-seqed-widget.c
	(polyxmass_seqed_widget_canvas_event): changed the behaviour of
	the seqed_widget such that when the mouse goes onto the canvas,
	the PXM_SEQED_WIDGET (seqed_widget)->canvas->sw scroll window gets
	the focus through a gtk_widget_grab_focus(). This means that in a
	top level window where two or more seqed_widget objects are
	packed, it is easy to go from one to the other.

	* Whole_project: the window management system and the
	traceability of the sequence windows and result windows now relies
	on the seqed_widget that is considered as the pointer that is used
	to make the ID Number of the sequence. It is no more the editctxt
	object. This is because it does make sense that the widget that
	actually displays the sequence on which the operations are
	performed is used for the ID number definition. This is in view of
	the fact that the editctxt will become some day obsolete, when
	we'll implement crosslinking between two or more sequences
	displayed in a given window or when we'll tackle the problem of
	the polymer ramification. The notion of PxmEditCtxt will slowly
	vanish for a more flexible idea that the central core of
	operations is the PxmPolymer object that is attached to the
	seqed_widget object... We'll see.

2005-11-11  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-ui-seqed-widget-context-menu.c: the contextual menu of
	the seqed_widget now contains the menu items pertaining to the
	polymer/monomer annotations (see below).

	* The files polyxedit-ui-note.[ch] were changed to
	polyxmass-ui-note.[ch] as now the seqed_widget might be the place
	where annotations are performed.

2005-11-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-monicon.c: the file pair polyxedit-monicon.[ch] was
	renamed polyxmass-monicon.[ch] since the monomer icons are no more
	exclusively in use in the polymer sequence editor window, because
	it is now possible to embed a seqed_widget anywhere in the modules
	of the polyxmass program...

	* Lots of commenting of the code where explanations are necessary,
	in particular with respect to resource management, with
	reffing/unreffing around the widgets...

	* polyxmass-pixbuf-rendering.c
	(polyxmass_pixbuf_rendering_render_no_modif): changes all over the
	place to take into account that the pixbufGData g_data_list
	from/to which find/store GdkPixbuf objects is now not a member of
	the polchemdefctxt, but of the seqed_widget on behalf of which the
	rendering is being performed.
	(polyxmass_pixbuf_rendering_monicon_make_canvas_item): because the
	canvas_item instances created in here are stored in PxmMonicon
	objects that HAVE to be FREED later, the canvas_item is reffed
	upon creation. This is so that upon freeing of the monicon, the
	function responsible for freeing the monicon can safely call
	gtk_object_destroy on the canvas_item contained in the monicon
	object!
	(polyxmass_pixbuf_rendering_cursor_monicon_render): exactly the
	same as above!

	AND NOW YES there is no more crashes due to destroying objects
	that were already destroyed!

	* polyxmass-polchemdefctxt.h (struct _PxmPolchemdefCtxt): removed
	the pixbufGData g_data_list object as it now belongs to each
	seqed_widget being created. This was necessary because each
	seqed_widget displays the GtkRange slider where the user might set
	the size of the monomer icons. These monomer icons are actually a
	wrap around GdkPixbuf objects that are stored in the pixbufGData
	data list. Since each seqed_widget can set different sizes for
	these pixbufs, it became necessary to have separated
	pixbufs. Prior to this change all the pixbufs allocated in a given
	polymer chemistry definition context were stored there
	(polchemdefctxt->pixbufGData).  This was rendered obsolete by the
	fact that each seqed_widget might render pixbufs at different
	sizes. Note that the logic making a given pixbuf allocated ONLY
	ONCE for a given seqed_widget is still valid !

	* polyxedit-ui-findrep-options.c
	(polyxedit_findrep_opt_setup_wnd): started coding the find/replace
	window interface. First design using the Glade program. I started
	coding the seqed_widget stuff in the setup_wnd (), and the window
	will embed two seqed_widget objects : the one for the sequence to
	be found and the one for the sequence to be used as a replacement
	(if applicable). Started doing proof of concept with a fake
	polymer object allocated for each seqed_widget object (each
	seqed_widget instance requires to be wired to a polymer and a
	polchemdefctxt through references).

	* polyxmass-ui-seqed-widget.c (polyxmass_seqed_widget_init):
	removed the constructor function and moved its content to this
	init function. Felt it was not essential to bother with a
	constructor... Also corrected a bug that was due to missing the
	g_object_ref () step for each item manually created using the
	widget_new () functions.

2005-11-09  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-ui-seqed-widget.c
	(polyxmass_seqed_widget_constructor): added a
	post_pack_constructor where the seqed_widget continues to be
	constructed after some critical data have been set to it by the
	user function : polchemdefctxt and polymer, for example, that are
	set to members of the seqed_widget as references (and not
	allocated objects). That function is reponsible for triggering the
	first rendering of the sequence located in seqed_widget->polymer,
	for example. These functionalities have thus been removed from the
	polyxedit-main.c polyxedit_main_prepare_new_context_open_file
	editctxt initialization function. The idea is that the most
	important part of displaying and handling a polymer sequence
	should be happening at the seqed_widget level, so that it has the
	most self-contained behaviour.

2005-11-01  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-mnmcompos.c
	(polyxedit_mnmcompos_wnd_execute_button): corrected a bug due to
	decrementing a pointer value instead of an integer value : was
	*(start_idx)--; and should have been (*start_idx)--; with gint
	*start_idx. This led to erroneous behaviour in the monomer
	composition window.

	* polyxmass-ui-seqed-widget.c
	(polyxmass_seqed_widget_constructor): Finished doing the
	extraction of all the seqed_widget functionality out of the
	sequence editor window. Now the PxmSeqedWidget contains all the
	functionality it needs to fulfill its tasks : scroll handling,
	selection handling, editing of the sequence by having a pointer to
	the polchemdefctxt at hand... Going to commit that huge code
	work.

	* polyxedit-ui-note.c (polyxedit_note_setup_wnd): corrected a bug
	due to destroying twice the note setting window. 

2005-10-31  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Whole_project: The way masses are updated is now based on
	signals that are emitted for the PxmSeqedWidget seqed_widget and
	that can be set to be trapped by the widget hosting the
	seqed_widget. Actions can then be taken to recompute masses and
	display them. Note that it is possible that a widget embedding a
	PxmSeqedWidget does not connect to these signals. The polymer
	sequence editor widget of course does, as it wants to know if the
	sequence changes or not, so that it can refresh the mass data in
	the mass display window.

	All the same is valid for the "sequence modified" bit of
	information.

	* polyxmass-pixbuf-rendering.c
	(polyxmass_pixbuf_rendering_render_modif): big big rework of the
	sequence editor routines. As I felt I would need a sequence
	editing widget with a canvas and all the functionality therein, I
	decided to write a brand new PxmSeqedWidget (variable name is
	seqed_widget, usually) that is a composite widget of base class
	GTK_VBOX in which I packed all the widgets necessary to both
	render a polymer sequence, handle the cases where the monomer
	codes is more than 1 letter-long and when keyboard keys are wrong
	(the GTK_ENTRY widgets), to show the position of the monomer that
	is currently moused
	(GTK_LABEL).... All the copy/paste X-Window clipboard selections
	stuff is now totally integrated in the PxmSeqedWidget widget. That
	is by packing such widget somewhere, the widget brings clipboard
	functionality at no cost.  The polymer sequence editor handling is
	thus simplified. Significant code rework, with files removed and
	new files created.

2005-10-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Done the modification that provides the right Free Software
	Foundation address in all the relevant files. Copied the new
	version of file COPYING in the source tree.

	* Finished doing all the modifications related to allowing the
	user to modify the properties of a polymer sequence
	(name/code). Also changed in all the glade files that used to
	display that data in such a manner that these data always appear
	in the exact same manner whatever the window where they are
	displayed. Each time it is possible to hide the polymer sequence
	properties and to redisplay them at will.

2005-09-02  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.9. Dependent on libpolyxmass --version 0.8.9.

	* Makefile.am: Put back the sgml source file for the man
	page. This is to comply with the Debian Free Software Guidelines.

	* polyxedit-ui-cleave-results.c
	(polyxedit_cleave_results_wnd_find_button): fixed a bug due to
	automatically freeing non-allocated string from a
	g_object_set_data_full () function with g_free as (GDestroyNotify)
	function.

	* polyxedit-ui-searchmass-results.c
	(polyxedit_searchmass_results_wnd_find_button): fixed a bug due to
	automatically freeing non-allocated string from a
	g_object_set_data_full () function with g_free as (GDestroyNotify)
	function.

	* polyxedit-ui-fragment-results.c
	(polyxedit_fragment_results_wnd_find_button): fixed a bug due to
	automatically freeing non-allocated string from a
	g_object_set_data_full () function with g_free as (GDestroyNotify)
	function.

2005-08-29  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Thus, function calls in polyxmass-bin were changed also. Mainly
	compute -> calculate transitions.

2005-08-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Removed polyxedit.glade which is not used since a long time
	already.

	* Huge work on internationalization. The i18n framework was
	updated and the translation of the polyxmass-bin.pot file was done
	in fr.po. In the process, a great number of modifications were
	performed in the source. However, no functionality was modified
	whatsoever. Only were performed "cosmetic" and language-related
	modifications generally aimed at clarifying the wording of
	messages output to the user. The GUI was totally translated into
	french. Since french is less compact, as a language, than english,
	there might be GUI items that now are too long. I will change the
	translations on a case-by-case basis. Files were renamed
	occasionally so that their file name is more eloquent about the
	code therein.

2005-08-02  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-ui-window-mngmt.c
	(polyxmass_window_mngmt_wnd_report_clipboard_button): fixed bug
	due to calling the report-making function without having a window
	item being selected in the window management treeview. The crash
	was due to the assertion report!=NULL. Thanks to Lionel Mamane for
	this bug report.
	(polyxmass_window_mngmt_wnd_delete_event): added call to
	polyxmass_timeoutmsg_messages_remove ((GtkWindow *) window); so
	that crashes are prevented when the window is closed before the
	last timed-out message has vanished.

2005-07-31  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* for whole project called intltoolize, which created
	intltool-extract.in intltool-merge.in and intltool-update.in
	symbolic links in srcdir.

	* Makefile.am (localedir): added this definition.
	(INCLUDES): added -DLOCALEDIR=\"$(localedir)\"

	* polyxmass-main.c (main): made a number of setlocale () calls
	with proper categories and locales. This is because Jérôme
	Pansanel found that polyxcalc would fail with LC_ALL=fr_FR
	(decimal separator is ',' in that locale). 

2005-07-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.8. First release compiled using gcc-4.

	* Put the -Werror flag in the INCLUDES directive of the
	src/Makefile.am file. This way, I can remove it from the
	autogen.sh file and from the configure script-related CFLAGS
	settings, and thus can get away with warnings from the configure
	checks (which GNU polyxmass are not responsible for). 

	* configure.in: incremented to 0.8.8 the requirement for
	libpolyxmass version.
	
2005-07-15  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Makefile.am: removed the call to docbook-to-man polyxmass.sgml >
	polyxmass.1 because this utility is not available on other
	distributions with the same name. So the man page is going to be
	shipped as is, without auto-regeneration during a make.

2005-07-11  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-searchmass-options.c
	(polyxedit_searchmass_opt_search_masses): improved the feedback to
	the user such that each time a new mass is being searched it is
	displayed, thus providing a feeling of the work going on.

2005-07-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-searchmass-results.c
	(polyxedit_searchmass_res_wnd_sort_name_column): added this
	function to sort the name column of the displayed oligomers. This
	function is required by the findmass results "name" column of the
	displaying treeview.

	* polyxedit-ui-findmass-results.c
	(polyxedit_findmass_results_wnd_make_report_text_format): improved
	the reporting format.
	(polyxedit_findmass_res_setup_oligs_treeview): improved the
	function so that depending on the kind of oligomers reported
	(cleavage or fragmentation), the proper oligomer name sorting
	function is called.
	

	* polyxedit-ui-cleave-results.c
	(polyxedit_cleave_results_wnd_make_report_text_format): improved
	the reporting format.

	* polyxedit-ui-fragment-results.c
	(polyxedit_fragment_results_wnd_make_report_text_format): improved
	the reporting format.

2005-07-08  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-window-menu.c
	(polyxedit_seqed_window_menu_chemistry_mz_ratio_calc): added the
	window registering call so that the newly created m/z ratio
	calc. window gets listed in the window of window management.

	* polyxcalc-ui-calculator.c
	(polyxcalc_ui_calculator_wnd_mz_ratio_calculator_button): added
	the window registering call so that the newly created m/z ratio
	calc. window gets listed in the window of window management.

	* polyxmass-ui-mz-ratio.c (polyxmass_mz_ratio_wnd_setup): changed
	the setup of the window so that the function now knows which
	module is calling that window. The title window is adapted to the
	situation.
	(polyxmass_mz_ratio_wnd_make_report_text_format): added this
	reporting function that I forgot to code during the last days.

2005-07-01  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-reportopt.[ch]: changed files
	polyxmass-ui-reportopt.[ch] to this new filename, since the
	reporting options will be governed differently dealing with each
	differnent module (polyxdef and polyxcalc). More modularity is
	better.

2005-06-29  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-cleave-results.c
	(polyxedit_cleave_res_wnd_sort_coordinates_column): corrected a
	bug in the sorting algorithm.
	(polyxedit_cleave_res_wnd_sort_name_column): added this function
	to sort correctly the cleavage fragments according to their
	specific name syntax.

	* polyxedit-ui-fragment-results.c
	(polyxedit_fragment_res_wnd_sort_name_column): added this function
	to sort correctly the fragmentation oligomers according to their
	specific name syntax.

2005-06-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-fragment-results.c
	(polyxedit_fragment_res_wnd_really_close): added this function to
	allow automatically closing the window when the corresponding
	polymer sequence editor window is closed. This is also the case
	for the following windows: fragment-options, cleave-results and
	cleave-options.

2005-06-13  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-ui-filechoosers.c
	(polyxmass_filechoosers_polchemdef_select_button_clicked):
	corrected a number of bugs created during the cleaning of the
	interfaces in the polyxmass_filechoosers_show_wnd function and
	related ones (see 2005-05-22).

2005-05-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Modified the edit main menu item from the canvas contextual menu
	as it was a gtk stock menu not properly recognized using
	MacOSX/Fink platform. Thanks to Mark Tracy for finding this bug.

2005-05-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-ui-filechoosers.c (polyxmass_filechoosers_show_wnd):
	cleaned interfaces to calling the different types of file choosing
	windows (atom definition, polymer chemistry definition,
	initialization of a new polymer sequence).

2005-04-25  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in : removed the test of the pkg-config version number.

	* changed the handling of all the var = strlen () calls, so that
	the returned value is of type gsize (equivalent to size_t). Then,
	the returned gsize value is ensured to be less than G_MAXINT and
	finally the value is converted to gint with a simple cast. Bug
	discoverd by Debian users on Sparc workstations.

2005-04-05  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.7 with a dependency to libpolyxmass >=0.8.7 (this
	release should coincide with the first Debian package release. I
	am immensely grateful to Lionel Mamane for working on that with me
	and teaching me a lot of things (often Debian GNU/Linux-related,
	but not only).

2005-04-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-window-menu.[ch] and
	polyxedit-ui-seqed-canvas-menu.[ch] were added to replace the
	polyxedit-ui-seqed-canvasmenu.[ch] pair. The functionality of the
	old pair of files is now distributed in the two new pairs. The
	'window' pair hold now only the functionality that is related to
	the whole polymer sequence editor window. The 'canvas' pair
	(that's the contextual menu) hold now the functionality that is
	only related to a single right-clicked monomer icon. This changed
	was suggested by Lionel Elie Mamane, and I'm grateful to him for
	this.
	

	* polyxedit-ui-seqed-wnd.c (polyxedit_seqed_wnd_setup): changed
	the colors with which the code editing feedback text is displayed:
	dark green for the currently edited (and still valid) monomer code
	feedback message and dark red for the erroneous code feedback
	message.
	(polyxedit_seqed_wnd_setup): changed also (with glade-2) the
	graphical interface of the sequence editor. Should look cleaner
	and also be easier to gain more space for the sequence by shutting
	down frames.

2005-03-29  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-menu.c
	(polyxedit_seqed_canvas_menu_view_completionslist): added this
	menuitem to the canvas contextual menu so that user not knowing
	that TAB provides the list of available monomer code completions
	can find it easily. (Thanks to Lionel Mamane for this suggestion).

2005-03-28  Filippo Rusconi  <filippo rusconi at nowhere dot org>

        * all_glade_files: changed the GtkEntry widgets to have no frame
	each time the GtkEntry has as the sole purpose to produce text for
	user's consumption. Instead, if the GtkEntry is to receive data
	from the user, then the frame is set to visible.

	* polyxedit-ui-seqed-clipboard.c
	(polyxedit_seqed_clipboard_import_wnd_setup): finalized the
	writing of all the functions that will allow the user to purify a
	badly formatted sequence to be pasted in the sequence editor. A
	number of facilities allow for an easy purification, such as
	removing all the digits and spaces in one go... an other goodies.

2005-03-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-clipboard.c
	(polyxedit_seqed_clipboard_selection_received): fixed a
	potentially very big memory leak due to not freeing the allocated
	array of monomers resulting from parsing the sequence string being
	imported into the sequence editor in the form of a clipboard
	paste.

	* polyxedit-ui-seqed-clipboard.[ch]: added this pair of source
	files to handle all the X Window system clipboard functions
	related to the sequence editor.

2005-03-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* autogen.sh (CFLAGS): removed the -z flag to CFLAGS that was a
	sequel of a previous -Wl,-z,defs flag that was to be passed to the
	linker. After this removal, a number of warning: unused variable
	appeared upon compiling polyxmass-bin. Thus code cleanup was
	performed.

2005-03-21  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* autogen.sh : changed the file so that all the autotools-related
	files are updated neatly.

	* TODO : added items as a result of a long discussion with Lionel
	Mamane.

2005-03-15  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-kbd.c (polyxedit_seqed_kbd_*_handler): In all
	the handlers, the code was improved for the case where the
	selection was destroyed by moving the cursor with the keyboard
	keys. This ensure that when a new selection is started, the
	selection's bordering monomers have the proper index value, and
	thus the proper position, which ensures that the selection polygon
	gets drawn correctly.

	* polyxedit-ui-cleave-results.c
	(polyxedit_cleave_res_oligs_treeview_row_activated): Implemented
	this function so that when the user double-clicks onto an oligomer
	item its corresponding sequence gets selected in the editor window
	so that it is easy to see where the oligomer is located.

2005-03-14  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_select_all_sequence): changed this function
	to polyxedit_seqed_wnd_select_sequence () with two more params:
	start_idx and end_idx, so that a single function can either select
	the whole sequence (either start_idx or end_idx == -1) or the
	sequence portion as defined by these two indices. Changed calls to
	the older function so that they still work with the new one.

2005-03-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-calculations-opt.c
	(polyxedit_calculations_opt_ionizerule_activate_entry): changed
	the way the recalculations are performed when the ionization data
	are changed. Previously, only the entry that received the <Enter>
	was updated, even if other text entries had changed. The mass
	results were thus counter-intuitive, as data was displayed in the
	different edit boxes, but the masses were recalculated by only
	taking into account the text entry widget(s) in which <Enter> was
	pressed. Now, logically, all the three actform, charge, level
	GtkEntry widgets are updated whenever <Entry> is pressed in
	whatever of these three widgets. Thus, mass results should always
	be consistent with what's displayed in the three GtkEntry widgets.

2005-03-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxcalc-ui-calculator.c
	(polyxcalc_ui_calculator_wnd_univ_chem_elem_checkbutton): fixed a
	GUI bug due to calling a gtk+ widget with a bad name.

2005-03-09  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-polymer-modif.c
	(polyxedit_ui_polymer_modif_modify_button): removed a debug_print
	statement that caused text to be output on stdout.

2005-03-08  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* autogen.sh (CFLAGS): added -O2 for the optimization.  Added
	--add-missing --force-missing for the automake call, so that the
	last system-wide files are always used.

2005-02-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in: changed version to 0.8.7-no release yet. Just
	making packages for Debian and discovering glitches...

	* polyxmass.sgml: corrected the <!ENTITY dhsection
	"<manvolnum>1</manvolnum>"> from 7 to 1.

	* Makefile.am (EXTRA_DIST): changed the man page name from
	polyxmass-bin.7 to polyxmass.1. Note that the install directory
	was also changed from man7 to man1. Also, the sgml page was
	changed from polyxmass-bin.sgml to polyxmass.sgml. The Makefile.am
	file was changed accordingly.

2005-02-05  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.6

	* whole_package: changed some functions to report errors in the
	form of g_critical messages, instead of g_error, so that no crash
	occurs when something fails. This is true for cases where the user
	tries to open files for which dependencies are not satisfied, like
	when the polyxmass-common package is not installed or when the
	installation is messed up somehow. The message that is returned is
	clear enough to let the user understand that something is not
	correct in her installation, like either polyxmass-common missing
	or some files being absent from locations where they were expected
	to be found.

	* configure.in (LIBPOLYXMASS_REQUIRED): incremented to 0.8.6

	* polyxcalc-calcctxt.c (polyxcalc_calcctxt_free): corrected a bug
	to polyxmass_timeoutmsg_messages_remove for
	calcctxt->calculator_wnd prior to checking that this window
	pointer was actually non-NULL. This yielded a crash when the
	polyxcalc_calcctxt_free() was called before that the window was
	actually correctly set up.

	* Added two new libpopt flags to ask what the current XML file
	format version is for either the polymer chemistry definition or
	the polymer sequence (--polchemdefversion and
	--polymersequenceversion).

	* whole_project: set the log handlers for critical, warning,
	message g_log messages and replaced all the calls to g_log to
	each corresponding g_critical, g_warning, g_message logging
	functions. Each such call is now handled nicely with each
	corresponding handler that sends the message to the console window
	so that the user may see the messages even if polyxmass was not
	started from a terminal.
	
	* configure.in (XXXXXX_REQUIRED): Removed the dependency on the
	polyxmass-common package. This will allow the program to be
	compile on a machine that has no data installed on it. Thanks
	Lionel Mamane for this suggestion.

2005-02-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-ui-console.c (polyxmass_console_setup_wnd): added this
	pair of [ch] files to the project to handle the message-logging
	console window. Added the corresponding polyxmass-console.glade
	file in the srcdir/glade directory.

2005-02-02  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxdef-defctxt.c (polyxdef_defctxt_free): corrected a bug to
	polyxmass_timeoutmsg_messages_remove for defctxt->definition_wnd
	prior to checking that this window pointer was actually
	non-NULL. This yielded a crash when the polyxdef_defctxt_free()
	was called before that the window was actually correctly set up.

2005-01-31  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.5

	* polyxedit-ui-self-speak.c (whole_file: each fputs() return value
	check was corrected as the previous version was not faulty in
	GNU/Linux but was in MacOSX/Fink. This is a known bug that was
	corrected earlier elsewhere in the project (the return value of
	the fputs() function (the fputs () return value is not the same on
	all platforms). Thanks to Marc Dellinger for pointing this out.

2005-01-25  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-searchmass-results.c
	(polyxedit_searchmass_res_setup_oligs_treeview): changed to
	setting-up of the coordinates column of the treeview so that it
	gets sorted correctly by calling automatically the sorting
	function polyxedit_searchmass_res_wnd_sort_coordinates_column.
	(polyxedit_searchmass_res_wnd_sort_coordinates_column): added this
	function to ensure the coordinates [xx-yy] column would sort
	correctly according to xx when its header is clicked.

	* polyxedit-ui-mnmcompos.c
	(polyxedit_mnmcompos_wnd_setup_treeview): the "Name" and "Code"
	columns are now sortable in the treeview that shows the monomeric
	composositio of the polymer sequence.

2005-01-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-cleave-results.c
	(polyxedit_cleave_res_setup_oligs_treeview): changed to setting-up
	of the coordinates column of the treeview so that it gets sorted
	correctly by calling automatically the sorting function
	polyxedit_cleave_res_wnd_sort_coordinates_column.
	(polyxedit_cleave_res_wnd_sort_coordinates_column): added this
	function to ensure the coordinates [xx-yy] column would sort
	correctly according to xx when its header is clicked.

	* polyxedit-ui-seqed-wnd.c (polyxedit_seqed_wnd_delete_event):
	corrected a small bug due to passing the sequence window pointer
	to gtk_dialog_add_buttons (GTK_DIALOG (here_pointer), instead of
	the dialog window pointer.

	* polyxedit-ui-completions.c (polyxedit_completions_wnd_setup):
	the monomer "Code" and "Name" columns of the completions list are
	listed are now sortable.

	* Release 0.8.4

	* polyxedit-main.c (polyxedit_main_prepare_new_context_open_file):
	changed the error reporting from LEVEL_ERROR (that is abortion of
	the program) to LEVEL_CRITICAL when failure to open a polymer
	definition context is encountered (thanks to Mark Tracy for
	suggesting that modification).

	* Whole_program: added the polyxmass-ui-mz-ratio.[ch] pair of
	files to handle the calculation of m/z ratios. That functionality
	might be called either from the polyxcalc module or from the
	polyxedit sequence contextual menu.

2005-01-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-wnd.c (polyxedit_seqed_wnd_delete_event):
	corrected a bug due to calling this function from the handler of
	the Close contextual menu with a NULL window pointer. Added code
	to go searching that pointer value in the seqeditorctxt object
	pointed to by the editctxt pointer that is passed to the function.

	* in file glade/polyxedit-seqeditor.glade changed a number of
	standard items in the contextual sequence editor menu due to
	GTK_ACCEL_* error messages and crashes.

2005-01-11  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* in file glade/polyxedit-acidobasic.glade changed the graphical
	interface to ensure that fields requiring input are clearly
	detectable, in contrast with fields where data are provided as the
	result of calculations (thanks Mark Tracy for pointing this
	problem out).

2004-12-08  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* release 0.8.3

	* Removed fake dependencies on libvorbis header files, since the
	program does not use those libraries. The inclusions date back to
	the developing phase of the self-speaking feature of polymer
	sequences.

2004-12-07  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Added the functionality of electrical charge/isoelectric point
	computations for the polymer sequence. Two sets of files were
	added: polyxedit-ui-pka-ph-pi.[ch] and
	polyxedit-acidobasic.[ch]. This new feature is to be considered
	experimental.

2004-10-23  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* release 0.8.2
	
2004-11-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxdef-ui-polchemdef.c
	(polyxdef_ui_polchemdef_wnd_setup_window): corrected a bug due to
	setting up the monomers' notebook widget twice, which resulted in
	the insertion of two new items each time the "Add" button was
	clicked in the monomers' notebook page. The proper behaviour is
	now ensured.

2004-11-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxdef-main.c (polyxdef_main_prepare_new_context): added a
	call to pxmchem_polchemdef_update_delim_codes () after reading a
	polymer definition from xml file, so that the delim_codes string
	is available when the user starts working on cleavage
	specifications; this string is needed because the cleavage
	specification parsing requires it.

2004-10-07  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.1 

	* polyxcalc-main.c (polyxcalc_main_prepare_new_context): now uses
	the polyxmass_polchemdefctxt_init_with_type () function to load
	the polymer chemistry definition that the user might have
	asked. Otherwise simply loads the atoms for the 'basic' atom
	definition.

2004-10-05  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-self-speak.c (polyxedit_self_speak_wnd_setup): same
	as below. 

	* polyxedit-ui-monomer-modif.c
	(polyxedit_ui_monomer_modif_setup_wnd): improved the user
	experience by setting up a callback mechanism that takes care of
	closing this window if the "parent" sequence editor window is
	closed. As a side effect corrected a bug that was due to not
	emptying the array of timed-out messages before closing the
	window, which had as effect the crash of the pogram when the
	time-out would happen and the window was not there anymore...

	* polyxedit-rendering.c (polyxedit_rendering_pixbuf_render_modif):
	adapted the code to comply with the new fact that the modif
	specifications are now located in another array,
	polchemdefctxt->mdfspecGPA, and no more in
	polchemdefctxt->mnmspecGPA.

2004-10-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-self-speak.c
	(polyxedit_self_speak_wnd_create_polseq_playlist): added the
	capability to let modifications self-speak themselves out.

	* polyxedit-main.c (polyxedit_main_prepare_new_context_open_file):
	this function now calls the
	polyxmass_polchemdefctxt_init_with_type () without duplicating the
	creation of the polchemdefctxt.
	
2004-10-02  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxmass-self-speak.c
	(polyxmass_self_speak_get_silent_sound_file_name): will parse the
	sounds.dic file and return the filename of the silent sound.

2004-09-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-self-speak.c: added this file to handle the
	self-speak feature for polymer sequences that become able to send
	to a playlist file all the sound filenames that are required to
	play a given polymer sequence.

	* polyxmass-self-speak.c: added this file to handle a number of
	tasks related to things like reading the monomer-sound.dic config
	file...

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_export_sound_playlist): added this function
	to serve as relay between the canvas menu and the acutal function
	doing the work.

	* polyxedit-ui-seqed-menu.c
	(polyxedit_canvas_menu_edit_export_sound_playlist): simple menu
	handler for the contextual menu.

2004-09-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

 	* Release 0.8.0

2004-08-30  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-note.c
	(polyxedit_note_wnd_selected_item_propagate): corrected a bug that
	would stop propagating a note in all the monomers of the range is
	one such monomer had a "NOTE"-named property in its propGPA
	array. That is erroneous: a monomer might have as many
	"NOTE"-named properties in its propGPA array! What is not possible
	is that a monomer has two "NOTE"-name property object that have
	the same name themselves, that is that have the same (PxmNote
	*)->name string.

2004-08-25  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in (CONFDIR): will be /etc/polyxmass if no
	--sysconfdir has been passed to the configure script.

	* this ChangeLog entry marks the start of the version 0.8.0 of the
	GNU polyxmass package, as it is born by pooling all my code
	previously written for modules polyxdef, polyxcalc and polyxedit
	into one single binary package: polyxmass. Below are the three
	ChangeLog files from those three modules concatenated together.

2004-07-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in (POLYXMASS_DATA_REQUIRED): changed requirement from
	0.7.0 to 0.7.1 since location of data files changed.

	* polyxedit-config.c: changed user_polyxdef_cfg_file from
	".polyxmass.d/polyxedit.conf" to ".polyxmass/polyxedit.conf".

	* autogen.sh (CFLAGS): added switch --sysconfdir=/etc

	* Makefile.am (polyxeditconf_datadir): now in
	$(sysconfdir)/polyxmass

	* configure.in: changed the CONFDIR specification to point to
	$(sysconfdir)/polyxmass, with autogen and configure saying
	"--sysconfdir=/etc".

2004-07-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.7.2

	* polyxedit-poledtctxt.c (polyxedit_poledtctxt_free): corrected
	bug due to leaving the completions' listview window around after
	the related sequence editor window was closed. Now the closing of
	the completions's listview window is automatic when the sequence
	editor window is closed.

2004-07-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.7.1 compiled using the new libraries Glib, Gtk+... and
	lipbolyxmass version 0.0.0.
	
	* configure.in: AC_INIT(polyxedit, 0.7.0 changed to
	AC_INIT(polyxedit, 0.7.1
	(GLIB_REQUIRED): changed from old version requirements to new
	ones. Same for all the other libraries that were upgraded.

	* polyxedit-ui-findmass-options.c
	(polyxedit_findmass_opt_setup_wnd): changed GtkOptionMenu to
	GtkComboBox because the latter one was dropped from Gtk+ v2.4.0.

	* polyxedit-ui-searchmass-options.c
	(polyxedit_searchmass_opt_setup_wnd): changed GtkOptionMenu to
	GtkComboBox because the latter one was dropped from Gtk+ v2.4.0.

	* polyxedit-globals.c (polyxedit_globals_timeout_message_set):
	changed gtk_timeout_add to g_timeout_add (due to upgrade of
	glib-2.0 and co libraries since last march).
	(polyxedit_globals_timeout_message_stop): changed
	g_object_timeout_remove into g_source_remove (due to upgrade of
	glib-2.0 and co libraries since last march).

2004-03-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* whole_project: remodelled all the function calls that used to be
	performed when comparing/duplicating PxmProp-related or PxmYYYY
	objects, as the framework for comparing and duplicating properties
	(PxmProp objects) has vastly been enhanced in the
	libpxmutils/libpxmchem libraries. This is a major feature
	enhancement, although not directly visible to the user. These
	changes are made with a very powerful find/replace feature in
	mind, which is to be developed soon.

	* whole_project: refactored the manner in which timed-out messages
	can be displayed in the messages_entry GtkEntry wiget. There is
	now a framework ready to be used easily that ensures that pending
	timed-out messages are removed before a window is closed. This way
	a number of crashes are avoided. The crashes due to timing-out of
	messages displayed in a GtkEntry belonging to a no longer-existing
	window are the only crashes that I have encountered in the program
	so far.

2004-03-09  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-note.c
	(polyxedit_note_wnd_selected_item_remove_range): Bogota,
	Colombia. Removed a bug due to freeing the reference monomer'prop
	object instead of removing it from each of the iterated monomers
	in the range. That bug drove to removing twice a noteval object
	out of the reference monomer's "NOTE" prop object.

2004-02-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>
 
        * Release 0.7.0 without major source code changes, only changes in
	the package. The release is aimed at synchronizing all the
	packages on the same value for the milestone Debian GNU/Linux. The
	tar.gz packages are now fully synchronized with the .deb packages
	and the $(package).spec.in file in the source tarball with which
	rpm packages should be able to be prepared.
 
2004-02-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-note.c (polyxedit_note_wnd_note_add_button):
	corrected a number of seemingly erroneous pointer casts from
	GtkTreeModel * to GtkTreeStore * and vice-versa... These problems
	appeared when compiling with -g -O0 -z -Wl -Wall -Werror
	-D_REENTRANT.

2004-01-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.6

	* Full Gnuification of the package. A number of new developments,
	mostly in the sequence annotation field. Going to release this.

2004-01-23  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-note.c
	(polyxedit_note_wnd_reset_note_related_widgets): added this
	function to handle cases when no more selection is present in the
	treeview, and thus nothing should be visible in the note-related
	widgets. This is cosmetic, but improve the user experience with
	this complex window.

2004-01-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in (ALL_LINGUAS): added ALL_LINGUAS="fr" because
	automake would complain in $(srcdir) that this variable is not
	defined while still using gettext...

	* Makefile.in.in (MKINSTALLDIRS): change this file in $(srcdir)/po
	so that the make install target would not failed, not finding the
	mkinstalldirs script... See the file itself for details.

2003-12-14  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_lr_end_modif_button_clicked): added this
	function, in conjunction with a modification of the polymer
	sequence editor gui-defining glade file
	(glade/polyxedit-seqeditor.glade), so that the left and right
	ends' modification data can be shown/hidden using a single click
	onto the frame title (button with no border). This is similar to
	the polymer sequence data frame on top of the polymer sequence
	editor window.
	(polyxedit_seqed_wnd_setup): added a button in the left and right
	en modifs frame so that the user can call from it the polymer
	sequence modification window.

2003-11-16  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_update_monomer_flag): corrected a bug due to
	not unref'ing the allocated GdkGC *gc, which yielded massive
	memory leaks.

2003-11-15  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-rendering.c
	(polyxedit_rendering_monicon_make_canvas_item): changed
	(GtkSignalFunc) to G_CALLBACK ().

	* polyxedit-polseqctxt.c (polyxedit_polseqctxt_free): fixed memory
	leak due to not calling g_free (polseqctxt) (Thanks Valgrind).

2003-11-14  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.5 

	* configure.in updated the
	dependencies. LIB_PXMUTILS_REQUIRED=0.6.2
	LIB_PXMCHEM_REQUIRED=0.6.3 POLYXMASSDATA_REQUIRED=0.6.3

2003-11-13  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Improved the interface for the polymer sequence editor so that
	the status of the monomer under the mouse cursor is displayed
	automatically in the form of a flag with three elements. Written
	the code that updates this flag. See the documentation in the
	polyxmassdata package.
 
2003-11-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-wnd.c (polyxedit_seqed_wnd_canvas_event):
	corrected a nasty bug due to not bypassing a mouse click out of
	sequence drawing area follow by a drag to the sequence area,
	button release: CRASH BOOM ! The mouse event handling function has
	been corrected and its design bettered somewhat.
	(polyxedit_seqed_wnd_render_polseq): when the sequence cursor is
	drawn for the first time, it is drawn left of the very first
	monomer of the sequence. And we did not make sure that the first
	point of the selection rectangle (poledtctxt->selection_rect) had
	meaningful values. Hence crashes.

2003-11-11  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-note.c
	(polyxedit_note_wnd_monomer_note_remove_range): continued
	improving the PxmNote stuff. Now it is possible to add notes to a
	range of monomers in one go. It is also possible to change a note
	throughout a range of monomers. It is also possible to remove a
	note throughout a range of monomers. More flexible than
	that... Thanks to Mark Tracy for suggesting that feature.

	

2003-11-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-globals.c (polyxedit_globals_timeout_stop): corrected
	a bug that would crash the program if a timeout would be triggered
	after the window in which it was created was closed. That was a
	bad format of the call if (TRUE == G_IS_OBJECT (entry)) which was
	actually if (TRUE == G_IS_OBJECT (entry) == TRUE) -:(. Now seems
	to work. Same bug corrected in the polyxdef and polyxcalc modules.

2003-10-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.4

	* polyxedit-ui-note.c
	(polyxedit_note_wnd_edited_note_apply_button): Implementation of a
	set of functionalities aimed at allowing unlimited and flexible
	annotation to polymer sequences and/or monomers. This system is
	virtually applicable to any chemical entity that has the
	propGPA-base property system already implemented (oligomers, for
	example).

2003-10-26  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_get_mnm_idx_with_xy): added code to check if
	the cursor is located in the margin of the sequence display
	area. If so the index that is returned is -1 and it becomes the
	responsibility of the caller to check for that specific
	value. Functions have been changed in that respect. For example,
	the mouse button press/mouse motion handling function was changed
	to check for that -1 returned value. No action is taken if the
	returned value is -1. As soon as the mouse cursor returns in the
	sequence display area, then the mouse button press/mouse motion
	handling function takes actions properly. Also, the position of
	the last moused monomer is updated to " " if the cursor is not
	over a monomer icon.
	(polyxedit_seqed_wnd_get_mnm_idx_with_xy): same as above but when
	the y position is less than zero, which mean that the cursor in in
	the upper margin of the canvas area.
	(polyxedit_seqed_wnd_get_mnm_idx_with_xy): same but for the bottom
	of the sequence display area.

	* polyxedit-poledtctxt.h (struct _PxmPoledtctxt): changed
	last_point_idx to last_point_1_idx (global change throughout all
	the source code). Added two other variants for buttons 2 and 3 of
	the mouse.

	* Makefile.am (SUBDIRS): removed the polyxedit/doc subdirectory as
	this directory was misleading: the documentation is in the
	polyxmassdata package not in each GUI module.

	* removed from the main Makefile.am the pixmaps directory from the
	'SUBDIRS =' line.

	* removed the polyxedit/pixmaps subdirectory because it contained
	two graphics files that actually belong to the glade
	subdirectory. See below.

	* Makefile.am (GRAPHIC_FILES): in the glade subdir. added
	provision to distribute graphics files (icons, for menus, for
	example). Glade/libglade would not load graphic files in
	subdirectories of the glade dir so I removed the pixmaps directory
	alltogether. Distributed graphics go in ghe glade dir now.

2003-10-25  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-kbd.c (polyxedit_seqed_kbd_key_press_event):
	added handler for GDK_F10, that acts like right-clicking onto the
	canvas, thus eliciting the popping up of the menu.

	* polyxedit-ui-seqed-menu.c: added this pair of .c/.h files to
	handle the configuration of the menu that is popped up when the
	mouse right button is clicked in the sequence canvas. The menu
	items that were removed from the program's main window have landed
	in this contextual menu.

2003-10-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-wnd.c (polyxedit_seqed_wnd_canvas_event):
	added a handler that will popup the canvas' menu when the right
	button of the mouse is clicked (that's the X button mouse number
	3).

	* polyxedit-ui-main-wnd.c (polyxedit_setup_app_wnd): removed a
	number of menu handlers as these menus have been removed from the
	main program window. The menus that were removed here are now
	handled in each polymer sequence editing window, as handlers to
	the contextual menu that is popped up when the right mouse button
	is clicked onto the canvas. The model that is retained here, is
	the one that made The Gimp easy to use, that can be adapted to our
	case simpley by stating that "Each polymer sequence has its own
	menu".

2003-10-21  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.3

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_remove_sequence_indices): this function now
	no more checks if a selection is there or not. It is the
	responsibility of the caller that it checks that there is a
	selection before calling this function. All the calling functions
	have been changed to comply with this new policy.
	(polyxedit_seqed_wnd_selection_received): corrected a bug due to
	pasting a sequence in an empty editor window, without checking
	that there was no current selection before calling the function
	that removes the selected oligomer. See above.

	* polyxedit-ui-main-wnd.c
	(polyxedit_main_select_all_menuitem_activate): added handler along
	with menu item for people who do not know that Ctrl+A works the
	same.

	* polyxedit-ui-seqed-kbd.c (polyxedit_seqed_kbd_alpha_handler):
	added code to handle the Ctrl+A keystroke that corresponds to
	selecting the whole sequence.

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_update_sel_polygon): now returns TRUE if the
	selection polygon was changed, and FALSE if nothing was touched.
	(polyxedit_seqed_wnd_select_all_sequence): added this function so
	that it is easy to select all the sequence in one go.

	* polyxedit-ui-main-wnd.c (polyxedit_main_copy_menuitem_activate):
	added handler along with menu item for people who do not know that
	Ctrl+C works the same.

	* polyxedit-ui-main-wnd.c (polyxedit_main_cut_menuitem_activate):
	added handler along with menu item for people who do not know that
	Ctrl+X works the same.

	* polyxedit-ui-main-wnd.c (polyxedit_main_paste_menuitem_activate):
	added handler along with menu item for people who do not know that
	Ctrl+V works the same.

2003-10-20 Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.2

2003-10-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_remove_selection_polygon): removed the call
	to the function that is responsible for updating the selection
	masses, because the call could make a crash if the value of the
	present cursor location had not been previously changed...

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_remove_sequence_indices): cleaned up the code
	that made all the checks pertaining to the value of the start and
	end parameters.

	* polyxedit-ui-seqed-wnd.c (polyxedit_seqed_wnd_clipboard_cut):
	new function that works.

	* polyxedit-ui-seqed-kbd.c (polyxedit_seqed_kbd_evaluate_code):
	not being able to remove the selected oligomer now is an
	abortion-worth error.
	(polyxedit_seqed_kbd_delete_handler): same as above.

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_integrate_monomer_at_idx): removed the two
	function calls that elicited the recomputation of the selected and
	whole sequence masses. That task is left to the caller.

	* polyxedit-ui-seqed-kbd.c
	(polyxedit_seqed_kbd_backspace_handler): cleaned code by pooling
	at the end of the function the two mass updating function.
	(polyxedit_seqed_kbd_delete_handler): same as above.

	* polyxedit-ui-seqed-wnd.c (polyxedit_seqed_wnd_remove_monomer):
	removed the two function calls that elicited the recomputation of
	the selected and whole sequence masses, because this function is
	called by other functions before readjusting the last pointed
	monomer index, which resulted in a crash due to trying to compute
	a mass for a selection greater than the actual size of the polymer
	sequnce. The functions that call this function have to ensure that
	they call the two mass updating functions themselves. This
	function now only performs array removals.

	* polyxedit-ui-seqed-kbd.c (for all the cursor moving keystrokes):
	if selection is changed, added code to call for ownership of the
	primary X selection. That works. For the page up/down keystrokes
	that modify the selection, changed code so that scrolling is much
	more intuitive, with the cursor being always visible either at the
	bottom of the sequence canvas area or at its top depending on the
	direction of the current selection operation (downwards or
	upwards).

2003-10-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-kbd.c (up/down/home/end handlers): exactly
	the same as for the left_ and right_ handlers described below.
	
	* polyxedit-ui-seqed-kbd.c (polyxedit_seqed_kbd_left_handler):
	added code to allow to resume a keyboard selection process that
	was left by releasing all the keyboard keys (with a selected
	portion of the sequence). If shift is pressed and the left/right
	cursor keys are used, the selection is resumed where it was left.

	* polyxedit-ui-seqed-kbd.c (polyxedit_seqed_kbd_right_handler):
	added code to allow to resume a keyboard selection process that
	was left by releasing all the keyboard keys (with a selected
	portion of the sequence). If shift is pressed and the left/right
	cursor keys are used, the selection is resumed where it was left.

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_integrate_sequence_at_point): added that
	function to handle the copy/paste/selection stuff.
	(polyxedit_seqed_wnd_integrate_monomer_at_idx): corrected bug
	-with no effect- due to using poledtctxt->last_point_idx instead
	of using the passed idx parameter value.
	(polyxedit_seqed_wnd_integrate_monomer_at_point): added that
	function that will automatically compute where the insertion has
	to be performed (which is where the point location is).
	(polyxedit_seqed_wnd_get_selection_indices): changed the behaviour
	of the window, where the start and end parameters are no more
	compulsory, so that the function is usable also just to know if a
	sequence portion is selected in the editor window. Changed all the
	function-using code accordingly.

2003-10-17  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in: corrected small glitch that gave a bad config
	directory name in the summary (end of file;
	"${sysconfdir}/polyxmass.d" (which is /etc/polyxmass.d) instead of
	the directory actually computed in this same file:
	${CONFDIR}). Thanks to Austin Acton for pointing that glitch.

2003-08-05  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.1

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_sequence_data_button_clicked): added this
	function to handle button clicks onto the button that was added in
	place of the polymer sequence data frame's label widget. If the
	vbox inside of the frame is visible, then it is made invisible,
	and vice-versa.

	* polyxedit-seqeditor.glade: changed the graphical interface of
	the polymer sequence editor so that the user may hide the polymer
	sequence data (name and ID) by just clicking onto a button that is
	packed in place of the label of the frame widget. This button acts
	as a swith Show/Hide the contents of the frame widget.

	* polyxedit-ui-seqed-wnd.c (polyxedit_seqed_wnd_setup): added code
	to make the editor window widget aware of the new polymer sequence
	data frame widget along with its button that will act as a switch
	Show/Hide the polymer sequence data-related widgets.

2003-07-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.0

	* polyxedit-main.c (polyxedit_configure_userspec): changed code so
	that when the user has no "datadir" data directory configure in
	her own polyxedit.conf file, the userspec->datadir becomes
	automatically the "userspec->configdir/polseqs" directory, which
	means that this directory becomes the polseqs directory in the
	user's own polyxmassdata directory if it not the default installed
	one any more.

2003-07-23  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-globals.h: Changed include
	/usr/include/gtk-2.0/gdk/gdkkeysyms.h to gdk/gdkkeysyms.h (Thanks
	Mark Tracy for pointing this).

2003-07-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Makefile.am (INCLUDES): added -I$(prefix)/include.
	(polyxedit_LDFLAGS): added -L$(prefix)/lib. Both changes to allow
	the package to find the header files of libpxmutils, for example,
	when this lib is installed in exotic places like /opt, for
	example.

2003-07-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-main-wnd.c (polyxedit_main_saveas_menuitem_activate):
	corrected a small bug that prevented the file names to be
	"selected" in the file selection window's file list according to
	the pattern "*.pxm" (completion feature of the GtkFileSelection
	object). 

	* polyxedit-ui-main-wnd.c (polyxedit_main_new_menuitem_activate):
	corrected a small bug that prevented the file names to be
	"selected" in the file selection window's file list according to
	the pattern "*.pxm" (completion feature of the GtkFileSelection
	object).

	* all.package : the polyxmass software suite's configuration
	directory is now in $(prefix)/etc/polyxmass.d. All required
	changes have been performed, in particular in
	polyxedit-config.c. That's a big step that requires that all the
	modules in the polyxmass software suite be updated
	accordingly. Hence the big leap to 0.6.0.

2003-07-17  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* release 0.5.2

2003-07-15  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-main-wnd.c (polyxedit_main_open_menuitem_activate):
	corrected a small bug that prevented the file names to be
	"selected" in the file selection window's file list according to
	the pattern "*.pxm" (completion feature of the GtkFileSelection
	object).
	(polyxedit_main_open_menuitem_activate): made the sequence file
	selection window to be able to cope with multiple files selected.

2003-06-21  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-rw-polseq.c (polyxedit_rw_polseq_read_from_filesel):
	change the G_LOG_LEVEL from ERROR (abortion of the program) to
	WARNING so that when a polymer sequence file is not found the
	program does not exit abruptly. Now the program send a message to
	the console.
	(polyxedit_rw_polseq_read): exactly as above for
	polyxedit_rw_polseq_read_from_filesel.

	* polyxedit-main.c (main): added a loop to iterate in the command
	line parameters remaining after that the registered options were
	parsed by libpopt. These "leftover" parameters are considered to
	be polymer sequence files.

2003-06-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* release 0.5.1

2003-06-09  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxedit-ui-seqed-wnd.c
	(polyxedit_seqed_wnd_sel_polygon_event): corrected a bug due to
	not returning FALSE, so that the canvas did think that the
	redrawing was handled fully (which is not true). This modification
	was required after the project passed to using the 2.2 version of
	the Gtk+ library and dependencies.


/* GNU polyxmass - the massist's program.
   -------------------------------------- 
   Copyright (C) 2000,2001,2002 Filippo Rusconi

   http://www.polyxmass.org

   This file is part of the "GNU polyxmass" project.

   The "GNU polyxmass" project is an official GNU project (see
   www.gnu.org) released ---in its entirety--- under the GNU General
   Public License and was started at the Centre National de la
   Recherche Scientifique (FRANCE), that granted me the formal
   authorization to publish it under this Free Software License.

   This software is free software; you can redistribute it and/or
   modify it under the terms of the GNU  General Public
   License as published by the Free Software Foundation; either
   version 2 of the License, or (at your option) any later version.
   
   This software is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
   General Public License for more details.
   
   You should have received a copy of the GNU  General Public
   License along with this software; if not, write to the
   Free Software Foundation, Inc., 59 Temple Place - Suite 330,
   Boston, MA 02111-1307, USA.
*/
2004-07-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in (POLYXMASS_DATA_REQUIRED): changed requirement from
	0.7.0 to 0.7.1 since location of data files changed.

	* polyxcalc-config.c: changed user_polyxdef_cfg_file from
	".polyxmass.d/polyxcalc.conf" to ".polyxmass/polyxcalc.conf".

	* autogen.sh (CFLAGS): added switch --sysconfdir=/etc

	* Makefile.am (polyxcalcconf_datadir): now in
	$(sysconfdir)/polyxmass

	* configure.in (CONFDIR): changed the CONFDIR specification to
	point to $(sysconfdir)/polyxmass, with autogen and configure
	saying "--sysconfdir=/etc". 

2004-07-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.7.1

	* configure.in (LIBPOLYXMASS_REQUIRED): replaced the requirements
	for libpxmutils and libpxmchem with libpolyxmass, as now the two
	former libraries have been merged into one library: libpolyxmass.

	* polyxcalc-globals.h: changed the inclusion of header files for
	libraries libpxmutils and libpxmchem as now one library is used:
	#include <libpolyxmass/libpolyxmass.h>

	* whole_project: note that the library requirements for
	libpxmutils and libpxmchem did not changed from 0.7.0 to 0.7.1.
	
	* configure.in: switch AC_INIT(polyxcalc, 0.7.0, to
	AC_INIT(polyxcalc, 0.7.1,

	* polyxcalc-ui-main.c (polyxcalc_setup_app_wnd): big set of
	changes to comply with the ever-changing Gtk+
	libraries. GtkOptionMenu is no longer working as before, so I took
	profit of the changed to improve the way polyxcalc works (mainly
	replacing unintuitive option menus with buttons laid out correctly
	in the window). Added functionality to hide widgets when these are
	of no importance to the user, so that much space can be gained
	thus reducing the footprint of the program window.

2004-06-30  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxcalc-globals.c (polyxcalc_globals_timeout_message_set):
	changed gtk_timeout_add to g_timeout_add (due to upgrade of
	glib-2.0 and co libraries since last march).
	(polyxcalc_globals_timeout_message_stop): changed
	g_object_timeout_remove into g_source_remove (due to upgrade of
	glib-2.0 and co libraries since last march).
	

2004-03-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* whole_project: refactored the manner in which timed-out messages
	can be displayed in the messages_entry GtkEntry wiget. There is
	now a framework ready to be used easily that ensures that pending
	timed-out messages are removed before a window is closed. This way
	a number of crashes will be avoided when timed-out messages will
	be used later (for the moment this framework is not yet used).
	
2004-02-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>
 
        * Release 0.7.0 without source code changes, only changes in the
	package. The release is aimed at synchronizing all the packages on
	the same value for the milestone Debian GNU/Linux. The tar.gz
	packages are now fully synchronized with the .deb packages and the
	$(package).spec.in file in the source tarball with which rpm
	packages should be able to be prepared.
 
2004-01-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.5

	* Full GNUification of the package. No source code change.

2004-01-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in (ALL_LINGUAS): added ALL_LINGUAS="fr" because
	automake would complain in $(srcdir) that this variable is not
	defined while still using gettext...

	* Makefile.in.in (MKINSTALLDIRS): changed that file in
	$(srcdir)/po so that the installation would not fail at make
	install target due to not finding the mkinstalldirs target. See
	the file proper for details.

2003-11-14  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.4 (just recompile with new libs).

	* configure.in Updated dependencies: LIB_PXMUTILS_REQUIRED=0.6.2
	LIB_PXMCHEM_REQUIRED=0.6.3 POLYXMASSDATA_REQUIRED=0.6.3


2003-11-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxcalc-globals.c (polyxcalc_globals_timeout_stop): corrected
	a bug that would crash the program if a timeout would be triggered
	after the window in which it was created was closed. That was a
	bad format of the call if (TRUE == G_IS_OBJECT (entry)) which was
	actually if (TRUE == G_IS_OBJECT (entry) == TRUE) -:(. Now seems
	to work.

2003-10-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.3 (just recompile with new libs).

2003-10-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxcalc-ui-main.c (polyxcalc_main_new_menuitem_activate):
	memory leak fixes using memprof. Small leaks, but leaks...

2003-10-19  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.2: just a recompilation with requirement of
	libpxmchem lib version 0.6.1 due to a bug in that library. No
	source code changes.

2003-10-17  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in: corrected small glitch that gave a bad config
	directory name in the summary (end of file;
	"${sysconfdir}/polyxmass.d" (which is /etc/polyxmass.d) instead of
	the directory actually computed in this same file:
	${CONFDIR}). Thanks to Austin Acton for pointing that glitch.

2003-08-05  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.1

2003-08-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxcalc-ui-main.c (polyxcalc_main_wnd_delete_event): removed a
	call to destroy the recorder window if it was still present, since
	this window is set as a full datum to the data of the window, with
	a destroying function to be called whenever the program is
	closed. This call made a core dump at each closing of the program
	which was harmless but inelegant.
	
2003-08-02  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxcalc-ui-chempad.c
	(polyxcalc_chempad_fill_chempadkey_array): removed this function
	and replace its functionality (with improvements) within a new
	function: polyxcalc_chempad_parse_file().
	(polyxcalc_chempad_wnd_fill_layout): now handles the case where
	only chempad_columns or chempad_rows is defined in the
	chempad.conf file. This way the layout is computed at best. It is
	no more required that both value be defined in the conf file, as
	this was heretical from a geometrical standpoint. The
	corresponding polyxmassdata/chempad.conf and
	polyxmassdata/protein/chempad.conf files were modified by removing
	the chempad_row$3 line, retaining the chempad_columns$3 line.

	* polyxcalc-ui-chempad.h (struct _PxmChempadgeom): added this
	structure to ease the handling of the chempad geometry.

2003-07-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.0

2003-07-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Makefile.am (INCLUDES): added -I$(prefix)/include.
	(polyxcalc_LDFLAGS): added -L$(prefix)/lib. Both changes to allow
	the package to find the header files of libpxmutils, for example,
	when this lib is installed in exotic places like /opt, for
	example.

2003-07-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxcalc-ui-main.c (polyxcalc_main_new_menuitem_activate):
	corrected a small bug that prevented the file names to be
	"selected" in the file selection window's file list according to
	the pattern "*.xml" (completion feature of the GtkFileSelection
	object). 

2003-07-21  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxcalc-calcmass.c (polxyxcalc_calcmass_account_sequence):
	corrected bug due to freeing two times the monomer GPA in case the
	polymer sequence could not be successfully parsed and made into an
	array of monomers (thanks to Mark Tracy for reporting this).

	* polyxcalc-ui-main.c (polyxcalc_setup_app_wnd): added code to set
	to sensitive=FALSE the polymer chemistry-specific frames when no
	poldef is passed to the window setup function. This is foolproof
	code.
	(polyxcalc_main_open_poldef_from_fileselection): If a new polymer
	chemistry definition file is successfully opened, and its data
	parsed, then ensure that the polymer chemistry-specific frames
	[monomers, modifications, sequence] are effectively made
	sensitive, since the user will be able to enter
	monomer/modif/sequence requirements.

2003-07-17  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* release 0.5.2

	* corrected the dependencies with respect to the libpxmutils,
	libpxmchem and polyxmassdata packages.

2003-06-21  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxcalc-ui-main.c (polyxcalc_clear_all_chem_data_activate):
	added the code to ensure that the count entry related to the
	polymer sequence would also be reset to empty. This one did not
	get erased as all the other ones.

2003-06-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.5.1

	* polyxcalc-ui-main.c
	(polyxcalc_remove_initial_from_result_masses_activate): added this
	function that is symmetrical to the one that adds the initial to
	result.

2003-02-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxcalc-globals.h (polyxcalc_globals_copy_files): removed
	declaration of this function for which no definition was given,
	since it is actually pxmutils_globals_copy_files () in the
	libpxmutils library !



/* GNU polyxmass - the massist's program.
   -------------------------------------- 
   Copyright (C) 2000,2001,2002 Filippo Rusconi

   http://www.polyxmass.org

   This file is part of the "GNU polyxmass" project.

   The "GNU polyxmass" project is an official GNU project (see
   www.gnu.org) released ---in its entirety--- under the GNU General
   Public License and was started at the Centre National de la
   Recherche Scientifique (FRANCE), that granted me the formal
   authorization to publish it under this Free Software License.

   This software is free software; you can redistribute it and/or
   modify it under the terms of the GNU  General Public
   License as published by the Free Software Foundation; either
   version 2 of the License, or (at your option) any later version.
   
   This software is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
   General Public License for more details.
   
   You should have received a copy of the GNU  General Public
   License along with this software; if not, write to the
   Free Software Foundation, Inc., 59 Temple Place - Suite 330,
   Boston, MA 02111-1307, USA.
*/
2004-07-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in (POLYXMASS_DATA_REQUIRED): changed requirement from
	0.7.0 to 0.7.1 since location of data files changed.

	* polyxdef-config.c: changed user_polyxdef_cfg_file from
	".polyxmass.d/polyxdef.conf" to ".polyxmass/polyxdef.conf".

	* autogen.sh (CFLAGS): added switch --sysconfdir=/etc

	* Makefile.am (polyxdefconf_datadir): now in
	$(sysconfdir)/polyxmass

	* configure.in (CONFDIR): changed the CONFDIR specification to
	point to $(sysconfdir)/polyxmass, with autogen and configure
	saying "--sysconfdir=/etc".

2004-07-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.7.1

	* polyxdef-globals.h: changed the inclusion of header files for
	libraries libpxmutils/libpxmchem to the location of libpolyxmass'
	header files: #include <libpolyxmass/libpolyxmass.h>

	* configure.in: switch AC_INIT(polyxdef, 0.7.0, to
	AC_INIT(polyxdef, 0.7.1,

	* configure.in (LIBPOLYXMASS_REQUIRED): replaced the requirements
	for libpxmutils and libpxmchem with libpolyxmass, as now the two
	former libraries have been merged into one library: libpolyxmass.

2004-06-30  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxdef-globals.c (polyxdef_globals_timeout_message_set):
	changed gtk_object_timeout_add to g_timeout_add (due to upgrade of
	glib-2.0 and co libraries since last march).
	(polyxdef_globals_timeout_message_stop): changed
	g_object_timeout_remove into g_source_remove (due to upgrade of
	glib-2.0 and co libraries since last march).

2004-03-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>
	
	* whole_project: refactored the manner in which timed-out messages
	can be displayed in the messages_entry GtkEntry wiget. There is
	now a framework ready to be used easily that ensures that pending
	timed-out messages are removed before a window is closed. This way
	a number of crashes are avoided. The crashes due to timing-out of
	messages displayed in a GtkEntry belonging to a no longer-existing
	window are the only crashes that I have encountered in the program
	so far.

2004-02-16  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxdef-ui-main.c (polyxdef_main_save_poldef_to_file):
	corrected the test that checked the return value out of the
	fputs() call to be compliant with the MacOSX platform. This bug
	already had been corrected in the polyxedit module, but for some
	reason now is creeping up in polyxdef. Now should be OK. Once
	again thanks to Mark Tracy for pointing that bug to me.

2004-02-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>
 
        * Release 0.7.0 without source code changes, only changes in the
	package. The release is aimed at synchronizing all the packages on
	the same value for the milestone Debian GNU/Linux. The tar.gz
	packages are now fully synchronized with the .deb packages and the
	$(package).spec.in file in the source tarball with which rpm
	packages should be able to be prepared.
 
2004-01-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.5
	
	* GNUification of the package. No source code changes... Now the
	development is performed under Debian GNU/Linux unstable.

2004-01-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in (ALL_LINGUAS): added ALL_LINGUAS="fr" because
	automake would complain in $(srcdir) that this variable is not
	defined while still using gettext...

	* Makefile.in.in (MKINSTALLDIRS): change this file in $(srcdir)/po
	so that the make install target would not failed, not finding the
	mkinstalldirs script... See the file itself for details.

2003-11-14  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.4 (just a recompile with new libs.)

	* configure.in: dependencies updated LIB_PXMUTILS_REQUIRED=0.6.2
	LIB_PXMCHEM_REQUIRED=0.6.3 POLYXMASSDATA_REQUIRED=0.6.3

2003-11-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxdef-globals.c (polyxdef_globals_timeout_stop): corrected a
	bug that would crash the program if a timeout would be triggered
	after the window in which it was created was closed. That was a
	bad format of the call if (TRUE == G_IS_OBJECT (entry)) which was
	actually if (TRUE == G_IS_OBJECT (entry) == TRUE) -:(. Now seems
	to work.

2003-10-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.3 (just a recompile with new libs.)

	* Makefile.am (glade_files_DATA): Changed the file so that
	graphics files in the glade directory are automatically
	distributed.

2003-10-19  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.2: just a recompilation with requirement of
	libpxmchem lib version 0.6.1 due to a bug in that library. No
	source code changes.

2003-09-11  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.1

2003-10-17  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in: corrected small glitch that gave a bad config
	directory name in the summary (end of file;
	"${sysconfdir}/polyxmass.d" (which is /etc/polyxmass.d) instead of
	the directory actually computed in this same file:
	${CONFDIR}). Thanks to Austin Acton for pointing that glitch.

2003-08-31  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxdef-ui-poldef.c (polyxdef_setup_poldef_wnd): fixed a bug
	that was due to trying to access an inexistent ionizerule pointer
	in the PxmPoldef that was allocated with the New menu. 

2003-07-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.0

2003-07-23  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* polyxdef-globals.h: Changed include
	/usr/include/gtk-2.0/gdk/gdkkeysyms.h to gdk/gdkkeysyms.h (Thanks
	Mark Tracy for pointing this).

2003-07-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Makefile.am (INCLUDES): added -I$(prefix)/include.
	(polyxdef_LDFLAGS): added -L$(prefix)/lib. Both changes to allow
	the package to find the header files of libpxmutils, for example,
	when this lib is installed in exotic places like /opt, for
	example.

	* polyxdef-ui-main.c (polyxdef_main_open_menuitem_activate):
	corrected a small ui bug by which the *.xml pattern was not
	correctly being completed in the file selection window.

	* all.package : the polyxmass software suite's configuration
	directory is now in $(prefix)/etc/polyxmass.d. All required
	changes have been performed, in particular in
	polyxdef-config.c. That's a big step that requires that all the
	modules in the polyxmass software suite be updated
	accordingly. Hence the big leap to 0.6.0.

2003-07-17  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.5.1

	* Added this ChangeLog file. 



/* GNU polyxmass - the massist's program.
   -------------------------------------- 
   Copyright (C) 2000,2001,2002 Filippo Rusconi

   http://www.polyxmass.org

   This file is part of the "GNU polyxmass" project.

   The "GNU polyxmass" project is an official GNU project (see
   www.gnu.org) released ---in its entirety--- under the GNU General
   Public License and was started at the Centre National de la
   Recherche Scientifique (FRANCE), that granted me the formal
   authorization to publish it under this Free Software License.

   This software is free software; you can redistribute it and/or
   modify it under the terms of the GNU  General Public
   License as published by the Free Software Foundation; either
   version 2 of the License, or (at your option) any later version.
   
   This software is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
   General Public License for more details.
   
   You should have received a copy of the GNU  General Public
   License along with this software; if not, write to the
   Free Software Foundation, Inc., 59 Temple Place - Suite 330,
   Boston, MA 02111-1307, USA.
*/
2006-04-09  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.9.1 with -version-info 11:0:0

	* configure.in: added AM_MAINTAINER_MODE to configure.in.

	* pxmchem-masscalc.c (pxmchem_masscalc_polymer): fixed a bug that
	was injected during the removal of the log message in this
	function (the if statement was there while the action to be taken
	was removed, thus setting the next function call as the new action
	to be taken...). Removed the if statement itself.

2006-03-26  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-elemcompcalc.c (various_places): removed a log message
	warning that sequence is empty. This is no more required at this
	stage of development.

2006-03-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-masscalc.c (various_places): removed a log message
	warning that sequence is empty. This is no more required at this
	stage of development.

2006-01-06  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.9.0 with -version-info 10:0:0

	* pxmchem-polymer.c (pxmchem_polymer_render_xml_file): memleak
	fix.

	* libpolyxmass-atomspec.c
	(libpolyxmass_atomspec_init_from_dic_line): memleak fix due to not
	freeing the contents of a GString but only the container.

	* libpolyxmass-monomerspec.c
	(libpolyxmass_monomerspec_init_from_sounds_dic_line): changed
	g_string_truncate with free and new replacements.
	(libpolyxmass_monomerspec_init_from_sounds_dic_line): twice.

	* libpolyxmass-modifspec.c
	(libpolyxmass_modifspec_init_from_sounds_dic_line): changed
	g_string_truncate with free and new replacements.

	* libpolyxmass-note.c (libpolyxmass_note_prop_render_xml_note):
	memleak fix.

2006-01-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-atomspec.c
	(libpolyxmass_atomspec_init_from_dic_line): memleak fix.

	* libpolyxmass-config.c
	(libpolyxmass_config_get_user_polyxmass_config_dir): code cleanup.
	(libpolyxmass_config_get_user_polchem_defs_atom_defs_dic_dir):
	memleak fix.

	* pxmchem-polchemdef.c (pxmchem_polchemdef_render_xml_file):
	memleak fix.

2005-12-28  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-note.c (libpolyxmass_noteval_cmp): fixed a bug in
	the comparisons of gint and gdouble.

	* libpolyxmass-prop.c (libpolyxmass_prop_cmp): fixed a bug due to
	not checking that the compared prop object had at least an
	identical name prior to asserting that their custom comparison
	function pointers were identical.

2005-12-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Whole_project: Merged code with the branch that was set up to
	correct the Debian bug# . Also, in that branch was a modification
	in the way the intl-tools were setup : using no more symbolic
	links but pure file copies. Here is the explanation out of the
	ChangeLog of that branch :

	2005-11-15  Filippo Rusconi  <filippo rusconi at nowhere dot org>

        * Run the following command : gettextize --copy --force --intl to
        force the copying (and not the symlinking) of all the files
        required by intltool and gettext. The instructions after that run
        were :

        8<~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
        Updating EXTRA_DIST in m4/Makefile.am (backup is in
        m4/Makefile.am~) Adding an entry to m4/ChangeLog (backup is in
        m4/ChangeLog~) Updating Makefile.am (backup is in Makefile.am~)
        Adding an entry to ChangeLog (backup is in ChangeLog~)

        Please run 'aclocal -I m4' to regenerate the aclocal.m4 file.  You
        need aclocal from GNU automake 1.5 (or newer) to do this.  Then
        run 'autoconf' to regenerate the configure file.

        You might also want to copy the convenience header file gettext.h
        from the /usr/share/gettext directory into your package.  It is a
        wrapper around <libintl.h> that implements the configure
        --disable-nls option.

        Press Return to acknowledge the previous two paragraphs.
        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~>8

        All of which I did diligently. And next I issued '$ make
        maintainer-clean ; ./removable-files' and '$ ./autogen.sh' which
        ran almost fine, as it complained that

        'cannot stat `po/POTFILES': No such file or directory
        ./configure: line 27885: po/POTFILES.tmp: No such file or
        directory'... Funny because po/POTFILES is there... and then it
        creates it :

        config.status: creating po/POTFILES

        So, apparently this is because I call make maintainer-clean, which
        removes critical files... No problem, then.

        There is no single symbolic link in the source tree anymore.

        Finally, make works flawlessly !




2005-12-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Whole_library_code: same as below but for all the functions
	concerned with this and in any file. Basically, each time it was
	required, changed 

	result += strcmp (string1, string2);

	with 

	if (0 != strcmp (string1, string2))
	result++;

	* pxmchem-monomer.c (pxmchem_monomer_cmp): corrected a bug due to
	summing the returned value from strcmp () calls and expecting the
	sum to be 0 only if the different string were all identical. But
	this is not the case : 'K' and 'Q' amino-acids are returned
	identical if their code, name and formula are compared !

2005-12-15  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-globals.h: added PxmHowDup and PxmHowCmp
	enumerations so that we can get rid of all the PxmTypeEnum
	enumerations which were too complicated to deal with. These
	enumeration simply specify if duplication/comparison of
	PxmType/PxmProp objects should be DEEP or NO_DEEP. Performed all
	the changes in all the source files. Compiles OK and is tested by
	using the new library version with polyxmass-bin. 

2005-10-29  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-globals.h: added typedef enum _PxmMasscalcTarget
	PxmMasscalcTarget; to indicate if mass update should be for whole
	sequence or selected sequence or both. In use in the polyxmass-bin
	code.

2005-10-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-cleave.c (pxmchem_cleave_parse_cleavage_site): corrected
	a misfit during the changes about comparing the empty string below.

2005-10-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-monomer.c (pxmchem_monomer_fill_array_from_string): code
	cleanup with comparing an empty string with character '\x0'
	instead of comparing it with "\x0". Thanks to Lionel Elie Mamane
	for pointing this.
	(pxmchem_monomer_fill_array_from_string_check_errors): same as above.
	(pxmchem_monomer_count_from_string): same as above.

	* pxmchem-cleave.c (pxmchem_cleave_parse_cleavage_site): code
	cleanup with comparing an empty string with character '\x0'
	instead of comparing it with "\x0". Thanks to Lionel Elie Mamane
	for pointing this.

2005-10-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Done the modification that provides the right Free Software
	Foundation address in all the relevant files. Copied the new
	version of file COPYING in the source tree.

2005-09-01  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release with -version-info 9:0:0, because function names have
	been changed.

	* Makefile.am: Put back the sgml source file for the man
	page. This is to comply with the Debian Free Software Guidelines.

2005-08-31  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* debian/control (Description): removed 
	Replaces: libpolyxmass8
	from section Package: libpolyxmass9.

2005-08-29  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-polymer.c
	(pxmchem_polymer_calculate_length_from_string): changed
	pxmchem_polymer_compute_length_from_string() to
	pxmchem_polymer_calculate_length_from_string().

	* libpolyxmass-acidobasic.h: changed
	libpolyxmass_acidobasic_compute_charge_ratio () to
	libpolyxmass_acidobasic_calculate_charge_ratio (). This means that
	the version of the lib should be changed.

2005-08-13  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Whole_project: tuned the internationalization of the software
	and started (and almost finished) doing the fr.po translation. No
	functionality was modified. However function names could be
	changed.

2005-08-01  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-globals.c (libpolyxmass_globals_strtoi): corrected
	a portability bug with the s390 platform. strlen (endptr) is %zu
	and not %zd. On an Intel platform this works OK. Thanks to Lionel
	Mamane for this fix.

2005-07-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.8 with -version-info 8:0:0 (compiled with gcc-4.0).

2005-07-17  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* whole-library-source : removed the libxml2 specific calls
	xmlStrcmp and replaced using the standard strcmp call. Due to
	switching to gcc-4.0, a number of corrections were made to
	accomodate the difference between guchar *(that is xmlChar *) and
	gchar * that is our classical signed char.

	* Makefile.am: removed the call to docbook-to-man
	libpolyxmass.sgml > libpolyxmass.7 because this utility is not
	available on other distribution with the same name. So the man
	page is going to be shipped as is, without auto-regeneration
	during a make.

2005-07-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-searchmass.c (pxmchem_searchmass_polymer): improved the
	oligomer naming scheme. Now the oligomer name is in the form
	"x#y", with x being the number of the search mass operation
	(particularly necessary if the search is for multiple masses) and
	y being the number of the oligomer being found by that mass. For
	example if we had a number of masses to search, we may have two
	oligomers complying with mass search number 3. We would thus have
	2 oligomers of names "3#1" and "3#2".

	* pxmchem-fragment.c (pxmchem_fragment_polymer_end_left): modified
	code so that the construction of the fragmentation oligomer always
	tells if it contains a left/right-end monomer modification. This
	is because the reporting functions in the polyxedit module require
	this. Note that this change does not involve any other change that
	this one, meaning that the mass calculations are not modified in
	any way.

2005-07-09  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-oligomer.c (pxmchem_oligomer_with_options_new): changed
	the way the oligomer gets a number of properties to indicate if it
	corresponds to either or both the LEFT or RIGHT end of the parent
	polymer sequence. This set of prop objects set to the
	oligomer->propGPA array do not change the way masses were
	computed. The changes are necessary so that when the user clicks
	onto an oligomer item in a set of cleavage results the oligomer
	data display correctly if it is LEFT/RIGHT-end modified,
	irrespective of the fact that the mass of the potential
	modification is calculated or not.

2005-06-29  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-fragment.c: changed the naming convention for the
	fragmentation oligomers to ease automatic parsing for sorting in
	the treeviews...
	(pxmchem_fragment_parse_name): added this function in order to
	parse the fragmentation oligomer's name according to its specific
	syntax.

	* pxmchem-cleave.c: changed the naming convention for the cleavage
	oligomers to ease automatic parsing for sorting in the
	treeviews...

2005-06-28  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-polymer.c
	(pxmchem_polymer_format_txt_string_polseqinfo): added this
	function for use in the reporting function in the GUI binary.
	(pxmchem_polymer_format_txt_string_polseq_with_mnm_GPA): added this
	function for use in the reporting function in the GUI binary.
	(pxmchem_polymer_format_txt_string_polymer_prop): added this
	function for use in the reporting function in the GUI binary.

	* pxmchem-monomer.c (pxmchem_monomer_format_txt_string_monomer):
	added function to produce a character string representing the
	monomer prop data.

2005-06-17  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-plugin.[ch]: changed the pxmchem-plugin.[ch] pair
	of files to this new name because we are dealing with data which
	are not only chemical...

	* libpolyxmass-reportopt.[ch] : added this new pair of files to
	deal with the way the reporting is configured when the user asks
	that the polymer sequence data, or results data are to be exported
	(as text or latex strings, for example).

2005-06-13  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-polchemdefspec.c (libpolyxmass_polchemdefspec_dup):
	added this new function to allow easy duplication of
	PxmPolchemdefSpec objects.

2005-06-06  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-atom.c (pxmchem_atom_get_ptr_top_by_symbol): added this
	function to use with the isotopic pattern calculation functions.
	(pxmchem_atom_get_ptr_by_symbol): same as above.
	(pxmchem_atom_get_ptr_bottom_by_symbol): added this function to
	use with the isotopic pattern calculation functions.

2005-06-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-atom.c (pxmchem_atom_get_highest_abund_isotope_idx):
	added this function to return the index of the isotope that has
	the highest abundance in the array of abundances in an atom
	object.
	(pxmchem_atom_get_lowest_mass_isotope_idx): added this function to
	return the index of the isotope that has the lowest mass in the
	array of masses in an atom object.
	(pxmchem_atom_set_mono_in_atom_by_highest_abund): changed the
	pxmchem_atom_set_mono_in_atom() function into this function, as it
	was considering the mono mass to be the mass of the most abundant
	isotope. However, depending on the readings, the mono mass can
	also be the mass of the lightest isotope (which is not necessarily
	the most abundant). This function, logically, makes use of the
	pxmchem_atom_get_highest_abund_isotope_idx() function.
	(pxmchem_atom_set_mono_in_atom_by_lowest_mass): This function was
	added to find the monoisotopic mass in an atom, according to the
	definition that the monoisotopic mass is the mass of the lightest
	isotope (and NOT the most abundant). Boron (or Bore) is an example
	for which the lightest isotope is not the most abundant. By this
	monoisotopic definition, the monoisotopic mass of B should be the
	lightest (<mass>10.0129380000</mass>
	<abund>19.9000000000</abund>). The most abundant isotope has the
	heaviest mass (<mass>11.0093053000</mass>
	<abund>80.1000000000</abund>). This function, logically, makes use
	of the pxmchem_atom_get_lowest_mass_isotope_idx() function.
	(pxmchem_atom_render_xml_node_atom): changed this function so that
	it calls the pxmchem_atom_set_mono_in_atom_by_lowest_mass()
	function to set the monoisotopic mass of the currently rendered
	atom. This is because the definition of the monoisotopic mass is
	the mass of the lightest isotope, not the mass of the most
	abundant isotope.

2005-04-25  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-globals.c (libpolyxmass_globals_strtoi): corrected
	a string format such that %zd is the format for the value returned
	by strlen (). Thanks to Kurt Roeckx <kurt@roeckx.be>.

	* configure.in : removed the test of the pkg-config version number.

	* changed the handling of all the var = strlen () calls, so that
	the returned value is of type gsize (equivalent to size_t). Then,
	the returned gsize value is ensured to be less than G_MAXINT and
	finally the value is converted to gint with a simple cast. Bug
	discoverd by Debian users on Sparc workstations.

2005-04-05  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.7 with -version-info 7:0:0 (this release should
	coincide with the first Debian package release. Thanks to Lionel
	Mamane for working on that with me).

2005-04-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-polymer.c (pxmchem_polymer_format_xml_string_DTD): added
	a small license blurb about the DTD.

2005-03-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-monomer.c
	(pxmchem_monomer_fill_array_from_string_check_errors): added this
	function so that it is possible to get all the positions in the
	sequence string where errors are found. This is important to let
	the user know that an imported sequence (for example from a
	copy/paste clipboard operation) contains invalid characters/codes.

2005-03-08  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* autogen.sh (CFLAGS): added -O2 for the optimization.  Added
	--add-missing --force-missing for the automake call; libtoolize
	--copy --force; autoheader --force; autoconf --force; autopoint
	--force so that the last system-wide files are always used. Also
	added code to copy to $srcdir the config.sub and config.guess
	files from /usr/share/misc
	(automake package files).

2005-02-28  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in : removed the libpolyxmass.spec.in template file.

2005-02-09  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-polymer-plugins.h: PxmPolymer *
	(*pxmchem_polseqdata_xml_node_render_plugin) declared as extern
	for MacOSX compatibility (Thanks Mark Tracy).

	* pxmchem-polchemdef-plugins.h (PxmPolchemdef *
	pxmchem_polchemdefdata_xml_node_render_plugin): declared as extern
	for MacOSX compatibility (Thanks Mark Tracy).

2005-02-05  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.6 with -version-info 6:0:2

	* whole_library: improved the handling of polymer chemistry
	definition or polymer sequence XML files loading errors. There
	should no more be abortion of the program when an error occurs
	during the loading of any file (Thanks Mark Tracy for pointing
	that un-friendly behaviour).
	
	* added pxmchem-polymer-plugins.[ch] to handle the versioning of
	the functions that will render the <polseqdata version="xx"> xml
	node and modified acccordingly the function that reads the polymer
	sequence from the xml file itself.

	* added pxmchem-polchemdef-plugins.[ch] to handle the versioning
	of the functions that will render the <polchemdefdata
	version="xx"> xml node and modified acccordingly the function that
	reads the polymer definition chemistry from the xml file itself.

2005-02-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Makefile.am (INCLUDES): added the
	-DG_LOG_DOMAIN=\"libpolyxmass\"\ flag so that libpolyxmass reports
	errors with proper domain name.

	* pxmchem-polymer.c (pxmchem_polymer_format_xml_string_DTD): added
	a version attribute to the <polseqdata> element of the DTD.

2005-02-02  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-polchemdef.c (pxmchem_polchemdef_set_version): added
	function to set value of 'version' member variable.
	(pxmchem_polchemdef_validate_all): added verification that the
	version member is a string OK.
	(pxmchem_polchemdef_format_xml_string_DTD): corrected a bug due to
	writing a bad DTD description for the cleavage rule element
	contained in the cleavage rule (was writing lr-rule instead of
	clr).
	(pxmchem_polchemdef_get_current_xml_version): added this function
	to return the string corresponding to what the library considers
	the current xml file version for storing PxmPolchemdef data. This
	version string is #define POLCHEMDEF_XML_VERSION "0.1" in
	pxmchem-polchemdef.h

	* pxmchem-polchemdef.h (struct _PxmPolchemdef): added the
	'version' gchar* member to hold a string like "0.8.5" that
	corresponds to the version of the polymer chemistry definition XML
	file.

2005-01-31  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.5 with -version-info 5:0:1

2005-01-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-globals.c
	(libpolyxmass_globals_parse_range_coordinates): added this
	function to parse [xx-yy] strings and put the integer xx value
	into an integer variable, and the same for yy.

	* Release 0.8.4 with -version-info 4:0:0

	* libpolyxmass-config.c
	(libpolyxmass_config_get_user_polyxmass_config_dir): added code to
	create the .polyxmass/atom-defs and .polyxmass/polchem-defs user
	configuration directories during the configuration directory
	creation.

2005-01-21  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-masspair.h (struct _PxmMasspair): added the
	PxmIonizerule *ionizerule member to the structure as it is of
	evident necessity that one will want to know how a given mass pair
	was computed. This is a very small overhead as the PxmIonizerule
	struct is very small and PxmMasspair is not very repeatedly used
	throughout the program.

	* autogen.sh: removed the -z switch in the CFLAGS because that
	switch had the effect that gcc did not warn me when an error (typo
	in a function, for example) did occur.

2005-01-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-masscalc.c (pxmchem_masscalc_account_ionizerule):
	changed the ionizerule-accounting function name from
	pxmchem_masscalc_polymer_account_ionizerule to
	pxmchem_masscalc_account_ionizerule as this function does not
	involve a polymer whatsoever, and thus can be used for other
	calculations. 
	(pxmchem_masscalc_decount_ionizerule): added this function which
	computes the molecular mass of an analyte given the m/z ratio that
	was computed for it and given the ionization rules that were used
	for that m/z computation. This function just reverts the action of
	the pxmchem_masscalc_account_ionizerule function.

2004-12-07  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.3, with -version-info 3:0:0

2004-11-26  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-polymer.c (pxmchem_polymer_write_xml_file): corrected a
	bug that would corrupt the name of a sequence file after the first
	writing of the sequence to that file. This is due to freeing the
	file character string, and using the same freed pointer to
	allocate through duplicatioon another (corrupt) string. This bug
	would be produced by opening a sequence, modifying it, saving
	(File->Save)... that would be ok, but then another File->Save
	would fail because the filename is corrupt by the end of the first
	saving command.

2004-11-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-pka pair of files changed to
	libpolyxmass-chemgroup. Made all the required changes to the code.

2004-11-07  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-polymer.c (pxmchem_polymer_get_right_end_modif_name):
	corrected a bug due to casting the gchar * prop->data to (PxmModif
	*) totally forgetting that a modification is a PxmProp with name
	"MODIF" and with data "Acetylation", for example. Same for
	pxmchem_polymer_get_left_end_modif_name.

2004-11-05  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-modif.c (pxmchem_modif_GPA_empty): added this function
	to ease the process of emptying arrays of modif objects.

	* libpolyxmass-pka.[ch]: added this pair of files to bring new
	pKa-related types to the library: PxmPka and PxmPkaRule. These
	types are used in the calculations performed in
	libpolyxmass-acidobasic.c and in the polyxmass binary file.

	* libpolyxmass-acidobasic.[ch]: added this pair of files where
	calculations about pH/pKa/charges are performed.

2004-10-23  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.2

	* libpolyxmass-polchemdefspec.c
	(libpolyxmass_polchemdefspec_parse_polchem_defs_cat_files): same
	as below, but for the catalogue files that list all the polymer
	chemistry definitions and the location of their respective data
	files.

	* libpolyxmass-atomspec.c
	(libpolyxmass_atomspec_parse_atom_defs_cat_files): changed the
	function so that it parses all the files with suffix
	"atom-defs-cat" found in the directory ".../atom-defs" either in
	the user or in the system configuration directory.
	(libpolyxmass_atomspec_parse_polchem_defs_atom_defs_dic_files):
	same as above but for the dictionary files that make the link
	between the polymer chemistry definition names and the names of
	the atom definitions to be used by them.

	* libpolyxmass-config.c
	(libpolyxmass_config_get_user_polchem_defs_cat_dir and similar):
	changed the functions (and their name) to return the directory
	where catalogue or dictionary files should be found for further
	parsing. Before, these functions used to return the main
	catalogue/dictionary file.

2004-10-07  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in: release 0.8.1

2004-10-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* whole_library: separated the modif-related stuff away from the
	monomer-related stuff in libpolyxmass-monomerspec files. Now the
	modif-related material is in libpolyxmass-modifspec.[ch] files.

2004-10-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-modifspec.h (struct _PxmModifSpec): added the
	libpolyxmass-modifspec.[ch] file pair to separate the
	modifications' handling from the monomers'. 

	* libpolyxmass-monomerspec.h (struct _PxmMonomerSpec): added
	mnm_code_sound and mdf_name_sound gchar * members to hold the
	absolute file name of the monomer's name sound

2004-09-28  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-monomerspec.c
	(libpolyxmass_monomerspec_get_ptr_top_by_modif): added an
	equivalent _get_ptr_ to each _get_index_ because these
	ptr-returning functions are more useful in the end...

2004-09-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-monomerspec.c (libpolyxmass_monomerspec_set_name):
	added this function to set the name of the monomer for which the
	PxmMonomerSpec's data are being filled up.
	(libpolyxmass_monomerspec_get_ptr_by_code): added this series of
	three function so that they return a pointer to the found item,
	instead of its index. These function variants are useful to be
	able to modify the found item through the pointer returned.

	* libpolyxmass-monomerspec.h (struct _PxmMonomerSpec): added
	monomer's 'name' member for use in the sound self-speak new
	functions in the polyxmass-bin module.

2004-09-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.8.0 of the package, with library -version-info 1:0:0.

	Note that this release corresponds to the big overhaul of the
	whole GNU polyxmass project, with the three distinct modules
	polyxdef, polyxcalc and polyxedit being grouped into one single
	binary package.

	Also note that a slew of modifications were performed to enhance
	greatly the configurability of the package. For example, the
	functions in libpolyxmass-config.c are totally new in this
	release.
	
2004-08-31  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-globals.c
	(libpolyxmass_globals_parse_masses_in_text): corrected a bug in
	the masses' text parsing that would bring no masses if a string
	with two spaces was encountered.

2004-08-29  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-cleavespec.c (whole_project): changed the PxmLRrule type
	into the PxmCleaveRule type and PxmCleavespec into PxmCleaveSpec,
	so that the naming of the types is more consistent with the pair
	PxmFragSpec and PxmFragRule.

	Also, the change was performed in the polymer chemistry definition
	DTD that sits in all the polymer chemistry definition files.

2004-08-25  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in (CONFDIR): will be /etc/polyxmass if no
	--sysconfdir has been passed to the configure script.

2004-08-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-atom.c (pxmchem_atom_check_symbol_syntax): corrected a
	nasty bug that would found that a single-character atom symbol is
	invalid!
	(pxmchem_atom_array_validate_all): improved the quality of the
	atom validation code. Now checks each atom symbol for syntax
	correctness. Also, now the massGPA and abundGPA arrays of
	monoisotopic data are required to at least have one element in
	them.

2004-08-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-atom.c (pxmchem_atom_array_validate_all): added this
	function so that the polyxdef module can validate the atom array
	with informative feedback on errors encountered.
	(pxmchem_atom_check_data_integrity): added this function.

2004-08-08  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* libpolyxmass-globals.c (g_ptr_array_insert_val): modified
	g_return_val_if_fail (index < array->len, NULL); into
	g_return_val_if_fail (index <= array->len, NULL); so that it
	becomes possible to "insert" data at index 0 if the array is
	empty. First testing proves this is possible.

	* pxmchem-atom.c (pxmchem_atom_format_xml_string_atomdefdata):
	added this function so that storing atom definitions to a file is
	easy from the other GNU polyxmass modules.
	(pxmchem_atomdef_format_xml_string_DTD): added this function in
	the process of creating an infrastructure to easily store to disk
	atom definitions in xml format.
	(pxmchem_atom_set_name): new function
	(pxmchem_atom_set_symbol): new function


2004-08-05  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-atom.c (pxmchem_atom_GPA_empty): added this function to
	empty the array of atom without freeing it. Useful in the polyxdef
	module.

2004-08-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-polchemdef.c (pxmchem_polchemdef_new): polchemdef->ionizerule is now
	created upon creation of the polymer definition instance. The fact
	that it was not allocated upon allocation of the PxmPolchemdef was
	leading to messy code.

2004-07-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-globals.c (pxmutils_globals_copy_files): improved error
	reporting for source/destination file open error. 

	* pxmutils-config.c: changed the general configuration file from
	polyxmassdata.conf to polyxmass-data.conf. This is a big change
	that will bump current version to 1:0:0 right before next release.

2004-07-23  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Makefile.am (man): changed manpage from .3 to .7 so that
	it goes to the library calls manpage section 7.

2004-07-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in (dnl): commented out (dnl) the gtk-doc paragraph
	since gtk-doc is not used today.

2004-07-21  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Makefile.am (EXTRA_DIST): changed manpage from .1 to .3 so that
	it goes to the library calls manpage section 3.

2004-07-03  gettextize  <bug-gnu-gettext@gnu.org>

	* right_before_release: -version-info 0:0:0
	
	* Makefile.am (SUBDIRS): Remove intl.
	(EXTRA_DIST): Add config.rpath.

2004-07-03 Filippo Rusconi  <filippo rusconi at nowhere dot org>
	
	* Merge_of_the_libpxmutils_and_libxpmchem_libraries: into one
	single library name libpolyxmass. Below are ---sequentially--
	listed the ChangeLog file for libpxmutils (bottom part of the file
	starting from here) and the ChangeLog file for libpxmchem (top of
	bottom part from here). 

2004-03-30  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-polymer.c (pxmchem_polymer_find_prop_in_sequence): added
	this function to allow easy searching of monomers in a polymer
	sequence that contain a prop object matching the prop object
	passed as parameter.

2004-03-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* whole_libray: improved very much the handling of the PxmYYYY
	object's prop and non-prop duplication and comparison
	functions. These functions are now fully configurable depending on
	the data that are pointed to by the data member of the PxmProp
	object (that incorporates a PxmYYYY object, like PxmMonomer or
	PxmOligomer, for example). This big feature enhancement is made in
	view of a find/replace feature in the polyxedit module of the GNU
	polyxmass software suite. See libpxchem for comparable changes
	also.

2004-03-11  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-polymer.c (pxmchem_polymer_find_sequence_motif): started
	working on this new function. This function should become a very
	powerful function allowing the search operation to be performed
	using sequence motifs, along with conditions like chemical
	modifications, or notes, or whatever prop objects.

2004-03-08  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* whole_library: applied required changes so that the new prop
	schema could be enforced everywhere: the duplication and
	comparison custom functions for PxmProp objects are now always
	used with a parameter that helps telling how these operations are
	to be carried over. See libpxmutils for details.

2004-03-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-modif.c (pxmchem_modif_cmp): added this function and
	related material.
	(pxmchem_modif_dup_no_prop): removed this function since the other
	one (pxmchem_modif_dup) now has a new parameter indicating the
	depth of the copy.

	* pxmchem-modif.h: added enum _PxmModifCmp in order to better code
	the PxmModif-comparing function.

2004-02-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

        * Release 0.7.0 (-version-info 4:2:0) without source code changes,
	only changes in the package. The release is aimed at synchronizing
	all the packages on the same value for the milestone Debian
	GNU/Linux. The tar.gz packages are now fully synchronized with the
	.deb packages and the $(package).spec.in file in the source
	tarball with which rpm packages should be able to be prepared.
 
2004-01-31  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* THANKS: updated file.

2004-01-30  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* added  -Wl -z -D_REENTRANT to the autogen.sh CFLAGS line so that the
	library gets compiled according to the Debian GNU/Linux standards.

2004-01-24 Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.4 (version-info 4:1:0)
	
2004-01-19  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in: release 0.6.4 (version-info 4:1:0)

	* pxmchem-plugin.c (pxmchem_prop_choose_render_xml_prop_plugin):
	removed the "COMMENT" prop name from the plugin function since now
	the "NOTE" prop name will be used instead. Advantage is that a
	NOTE of name "COMMENT" can have as many values as necessary.
	(pxmchem_prop_choose_format_xml_prop_plugin): removed the
	"COMMENT" prop name from the plugin function since now the "NOTE"
	prop name will be used instead. Advantage is that a NOTE of name
	"COMMENT" can have as many values as necessary.

	* Full GNUification of the package.

2004-01-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Makefile.in.in (MKINSTALLDIRS): modified this file so that make
	install does not fail using mkinstalldirs. See this file itself
	for detailed explanations.

2003-11-15  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-atom.c (pxmchem_atom_render_xml_file): added call to
	xmlFreeDoc (xml_doc); if an error occured prior to returning -1.
	(pxmchem_atom_free): corrected a bug that was due to not testing
	the NULLiness of the abundGPA array when freeing the atom. That
	bug did lead to memory leaks. Thanks Valgrind.

2003-11-14  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.3 (version-info 4:0:0)

	* configure.in (LIB_PXMUTILS_REQUIRED): set to
	LIB_PXMUTILS_REQUIRED=0.6.2.

2003-11-11  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-polymer.c (pxmchem_polymer_write_xml_file): added code
	to update the file name in the file member of the polymer->plminfo
	structure so that the calling function can use that string to
	update the window title of the sequence's window. Useful also
	because the sequence remains self-knowledgeable of what file it
	was last written to.

2003-10-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.2 (version-info 3:0:0)

	* pxmchem-cleave.c (pxmchem_cleave_parse_cleavage_site):
	rearranged code so that the logic that was set up in the
	pxmchem_monomer_extract_code_from_string() is correctly exploited.

	* pxmchem-monomer.c (pxmchem_monomer_extract_code_from_string):
	further tweaking of that function, so that if a character is not
	ascii alpha, it does return the index+1 where that character was
	found, and does not print out a nasty error message, so that
	functions that need to trap non-ascii character (and for which
	such chars occurring are not a fatal error) can get the info
	without blocking the parsing of the string.

	* pxmchem-polymer.c (pxmchem_polymer_write_xml_file): corrected
	the code that tested the returned value from the fputs () call. It
	used to test if (result <= 0 || result == EOF) --> error. It now
	tests if (result < 0 || result == EOF) --> error. That is because
	the GNU glibc fputs() return a non-negative number on success, or
	EOF on error. The Mac libSystem.B libc library returns 0 on
	success and EOF on error. Thanks to Mark Tracy for pointing that
	bug.


2003-10-26  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-monomer.c (pxmchem_monomer_extract_code_from_string):
	corrected a bug for which an error was not correctly reported if
	currently parsed char was not ascii_alpha. Thanks Mark Tracy for
	reporting it.

2003-10-23  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-init.c (pxmchem_init_library): fixed memleak ('file'
	string).

2003-10-19  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.1
	
2003-10-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmchem-monomer.c (pxmchem_monomer_fill_array_from_string):
	corrected a bug which could yield either a crash or a memory leak:
	allocating an array (fillGPA param) if this was NULL, but without
	never freeing it once the function would return. No single call
	was passing a NULL fillGPA so the bug was without effect in the
	library. Also changed the [pxmchem_monomer_GPA_free () +
	g_ptr_array_new ()] combination with a single call to
	pxmchem_monomer_GPA_empty (). That is done when the 'empty_first'
	parameter is TRUE.

2003-07-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.0

2003-07-23  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in (LIB_PXMUTILS_REQUIRED): set to 0.6.0.

	* Makefile.am (libpxmchem_la_LDFLAGS): set -version-info 1:0:0
	[interface changes in the library].

	* pxmchem-plugin.h (gchar *): pxmchem_atom_GPA made extern, and
	redeclared in pxmchem-plugin.c (Thanks to Mark Tracy for this port
	to Mac OS X).

	* pxmchem-plugin.h (gchar *): pxmchem_globals_atom_num_format and
	all siblings made extern, and redeclared in pxmchem-plugin.c
	(Thanks to Mark Tracy for this port to Mac OS X).

	* pxmchem-plugin.h (gchar *):
	pxmchem_prop_formatter_xml_prop_plugin made extern, and redeclared
	in pxmchem-plugin.c (Thanks to Mark Tracy for this port to Mac OS
	X).

	* pxmchem-plugin.h (gchar *):
	pxmchem_prop_renderer_xml_prop_plugin made extern, and redeclared
	in pxmchem-plugin.c (Thanks to Mark Tracy for this port to Mac OS
	X).

2003-07-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Makefile.am (INCLUDES): added -I$(prefix)/include.
	(libpxmchem_la_LDFLAGS): added -L$(prefix)/lib. Both changes to
	allow the package to find the header files of libpxmutils, for
	example, when this lib is installed in exotic places like /opt,
	for example.

	* all.package : switched to using $(prefix)/etc/polyxmass.d as the
	main polyxmass software suite's configuration directory. All
	required changes were done. That's a big step that requires that
	all the modules in the polyxmass software suite be updated
	accordingly. Hence the big leap to 0.6.0.


2003-06-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.5.1

2003-06-09  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* README: Added a description of the reason of existence of this
	library.

	* pxmchem-globals.c: Started versioning at 0:0:0

2002-09-20  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxm-chem.c (whole_file): started this project of library which
	should implement all the chemistry that will be used for the
	description of the polymer and its elements and for later mass
	calculations. For example, this library will hold all the struct
	definition that describe atoms, monomers, oligomers, polymers and
	all needed accessory structures and functions. The code was taken
	from the first version of polyxmass (version 0.11 actually) and
	modified.

2004-03-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* whole_libray: improved very much the handling of the PxmProp
	object's custom duplication and comparison functions. These
	functions are now fully configurable depending on the data that
	are pointed to by the data member of the PxmProp object. This big
	feature enhancement is made in view of a find/replace feature in
	the polyxedit module of the GNU polyxmass software suite. See
	libpxchem for comparable changes also.

2004-02-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.7.0 (-version-info 3:1:0) without source code changes,
	only changes in the package. The release is aimed at synchronizing
	all the packages on the same value for the milestone Debian
	GNU/Linux. The tar.gz packages are now fully synchronized with the
	.deb packages and the $(package).spec.in file in the source
	tarball with which rpm packages should be able to be prepared.

2004-01-31  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* updated the THANKS file that was desperately empty !

2004-01-30  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* added -Wl -z -D_REENTRANT to the autogen.sh CFLAGS line so that the
	library gets compiled according to the Debian GNU/Linux standards.

2004-01-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.3 (-version-info 3:0:0)

	* GNUification of the whole package.
	
2004-01-19  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* configure.in: Release 0.6.3 (-version-info 3:0:0)
	
2004-01-14  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-prop.c (pxmutils_prop_cmp): added this function that
	will call the prop->custom_cmp function to compare the two prop
	objects after having made sure that everything is ok.

2004-01-12  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-globals.c (pxmutils_globals_strtod): changed code so
	that if something goes wrong in the conversion the original value
	pointed to by the parameter val is not changed. This was already
	the safe implementation in the sister function strtoi.

2004-01-04  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-note.h (struct _PxmNote): changed the structure so that
	it now contains a GPtrArray* in place of the value. This allows
	for the note to contain more than one value. Further, since this
	note is now conceived to receive any number of PxmNoteVal objects
	in this array, it becomes really interesting.
	(struct _PxmNoteVal): this structure is the value structure for a
	PxmNote object. Any number of instances of this type go in the
	array in the PxmNote object. Two members allow full qualification
	of the value contained in the PxmNoteVal instance: a value
	-allocated gpointer- and a type -gint- of that value.

	All relevant stuff has been rewritten/changed.

2004-01-03  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Makefile.in.in (MKINSTALLDIRS): modified this file in the
	$(srcdir)/po directory so that make install does not fail. See
	file itself for explanations of the modification. This happens
	with the switch to Debian GNU/Linux, as I did not have this
	problem with the RedHat9 distro.

2003-12-02  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-xxxxspec.c (pxmutils_polchemdefspec_fill_array_from_dict):
	started refining the parsing code so that it becomes more robust
	when a line in the file begins with an arbitrary number of space
	characters... and may either contain valid strings after these
	space chars, or end with a new line. Basically, the new code will
	skip any space char found at the beginning of the line. Once these
	chars have been skipped, analysis of the remainder characters can
	start...

	* pxmutils-xxxxspec.c (pxmutils_monomerspec_fill_array_from_dict):
	started refining the parsing code so that it becomes more robust
	when a line in the file begins with an arbitrary number of space
	characters... and may either contain valid strings after these
	space chars, or end with a new line. Basically, the new code will
	skip any space char found at the beginning of the line. Once these
	chars have been skipped, analysis of the remainder characters can
	start...

2003-11-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-globals.c (pxmutils_globals_strtod): reverted the
	previous changes about duplication of the string passed as
	argument, this because I removed the g_strstrip () function
	call...

2003-11-26  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-globals.c (pxmutils_globals_strtoi): changed the code
	so that this function no more calls the pxmutils_globals_strtol ()
	function. The code is more self-contained and cleaner now.
	(pxmutils_globals_strtol): removed this function which was not
	used in any way.

2003-11-25  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-globals.c (pxmutils_globals_strtoi): corrected a bug
	due to not correctly testing the long_val value obtained after
	strtol conversion. 

2003-11-16  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-globals.c (pxmutils_globals_strtod): now makes a copy
	of the string passed as parameter so that the strip functions
	works on a copy and not on the original. Same for functions below.
	(pxmutils_globals_strtol): see above.
	(pxmutils_globals_strtoi): see above.

2003-11-14  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.2

2003-11-10  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-note.c (pxmutils_note_prop_dup): corrected a bug that
	led to allocation of two PxmNote objects instead of one.
	(pxmutils_note_prop_render_xml_node_note): corrected a bug that
	led the property that was constructed to be returned without first
	setting the function pointers to the appropriate note_prop
	duplication, freeing and comparing functions. That led to
	unexpected behaviour when calling pxmutils_prop_dup (prop) on that
	prop, because the duplication function was one that considered
	that the two members are strings ! which is not the case in a
	note_prop property!

2003-11-09  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-prop.c (pxmutils_prop_find_prop): corrected a bug which
	made the comparison of the props' data asymmetrical by calling the
	data comparing function passed as parameter using a prop->data
	first parameter and a prop second parameter. That yielded a
	crash. Now the pxmutils_prop_data_cmp_func () is called with two
	parameters prop->data of type gpointer.

	* pxmutils-note.c (pxmutils_note_cmp): corrected a bug which made
	a false positive (return 0) if the NOTE_BOTH param was
	passed. This bug made two prop objects look identical even if they
	were not.

	* pxmutils-prop.c (pxmutils_prop_gpointer_data_prop_cmp):
	corrected a bug which made a false positive (return 0) if the
	PROP_BOTH param was passed. This bug made two prop objects look
	identical even if they were not.

	* pxmutils-note.c (pxmutils_note_prop_cmp): corrected a bug which
	made a false positive (return 0) if the PROP_BOTH param was
	passed. This bug made two prop objects look identical even if they
	were not.

	* pxmutils-prop.c (pxmutils_prop_dflt_prop_cmp): corrected a bug
	which made a false positive (return 0) if the PROP_BOTH param was
	passed. This bug made two prop objects look identical even if they
	were not.

2003-10-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.1

	* pxmutils-globals.c (pxmutils_globals_copy_files): changed the
	testing of the fputs () return value so that other libc's are
	handled softly. (Thanks to Mark Tracy for this one.)

2003-10-24  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-note.c (pxmutils_note_prop_render_xml_node_note): Added
	this new pair of .c/.h files to deal with PxmNote objects that are
	going to be used in the polyxedit module in order to let the users
	annotate the polymer sequence both at the level of the polymer
	object or at the level of the monomer objects. These are additions
	of interfaces to the library and should elicit a change in the
	libtool version numbering scheme of the library.

2003-10-23  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxmutils-config.c (pxmutils_config_get_user_polyxmassdata_dir):
	fixed memleak ('line' string).

2003-07-27  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.6.0

	* -version-info 0:0:0 (no change) because there was no change in
	any code. Only packaging changes.

2003-07-22  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Makefile.am (libpxmutils_la_LDFLAGS): added -L$(prefix)/lib.
	(INCLUDES): added -I$(prefix)/include. Both changes to allow the
	package to find the header files of libpxmutils, for example, when
	this lib is installed in exotic places like /opt, for example.

2003-07-21  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* all.package : switched to using $(prefix)/etc/polyxmass.d as the
	main polyxmass software suite's configuration directory. All
	required changes were done. That's a big step that requires that
	all the modules in the polyxmass software suite be updated
	accordingly. Hence the big leap to 0.6.0.

2003-06-18  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* Release 0.5.1

2003-06-09  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* README: Added a description of the library's reason of existence.

	* pxmutils-whole-project: Started versioning at 0:0:0

2002-09-19  Filippo Rusconi  <filippo rusconi at nowhere dot org>

	* pxm-prop.c (whole_file): started this project of library which
	should implement the pxm_prop object. The code was taken from the
	first version of polyxmass (version 0.11 actually) and modified.




/* GNU polyxmass - the massist's program.
   -------------------------------------- 
   Copyright (C) 2000,2001,2002,2003,2004 Filippo Rusconi

   http://www.polyxmass.org

   This file is part of the "GNU polyxmass" project.

   The "GNU polyxmass" project is an official GNU project (see
   www.gnu.org) released ---in its entirety--- under the GNU General
   Public License and was started at the Centre National de la
   Recherche Scientifique (FRANCE), that granted me the formal
   authorization to publish it under this Free Software License.

   This software is free software; you can redistribute it and/or
   modify it under the terms of the GNU  General Public
   License as published by the Free Software Foundation; either
   version 2 of the License, or (at your option) any later version.
   
   This software is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
   General Public License for more details.
   
   You should have received a copy of the GNU  General Public
   License along with this software; if not, write to the
   Free Software Foundation, Inc., 59 Temple Place - Suite 330,
   Boston, MA 02111-1307, USA.
*/