home *** CD-ROM | disk | FTP | other *** search
-
- So many things changed on the aes side that you're best off just reading
- the new gemfast function reference doc. Below is a list of changes I
- started, but I'm not sure I even recorded every new function, let alone
- the descriptions of the changes. Still, I've been sitting on this release
- for six months already, and I'm not going to hold it up any longer just
- because I don't have time to flesh out these notes right now. ::sigh::
-
- Version notes for AESFAST...
-
- v1.8 02/21/91 Miscellanious fixes and additions.
-
- - apl_malloc New stuff. As the utilities get more complex, they
- apl_free need more system-level services. You can provide the
- services by coding these routines, or let default
- routines be picked up from the aesfast library.
-
- - apl_vopen New. These functions open and close a VDI workstation
- apl_vclose and implement the (new) concept of a single shared
- apl_vshared workstation which all gemfast utilties will use. Your
- application can use it too, or can also use this to
- open a separate (private) workstation for its own use.
-
- - appl_init Now appl_init/exit are vectored through apl_xinit/xexit
- appl_exit functions via macros in GEMFAST.H. This revectoring
- apl_xinit is required to use new utilties. The xinit/xexit
- apl_xexit functions just set up some new gl_whatever global vars.
-
- - apl_cleanup
-
- - frm_dsdial Renamed to frm_dsdialog(), and switched around the
- order of the parms. A macro in GEMFAST.H provides
- compatibility with existing code. Also, any string that
- starts with 0x7F is now centered. It also recognizes
- several new values in the graphicflag parameter.
-
- - frm_nldialog New. This builds and conducts a dynamic string dialog
- like frm_dsdialog() does, but the text lines are
- passed as one huge string with NL delimiting each line.
-
- - frm_dialog New. This conducts a complete dialog, or just part
- of one, based on the flags you pass to it.
-
- - frm_printf New. Does about what you'd think: you pass it parms
- frm_vprintf like you would to printf() (or vprintf()), and it makes
- a dialog out of it. (You also pass buttons, and it
- returns the exit button number, so this is really
- what form_alert() always should have been).
-
- - frm_confine
- - frm_desktop
- - frm_enableblit
- - frm_sizes
- - frm_error
- frm_verror
- - frmx_dial
-
- - exterror
- - exterrset
-
- - fsl_dialog New. This conducts a complete file selector dialog.
-
- - grfx_dragbox
-
- - grf_blit New. Blit between screen and memory buffer.
-
- - grf_memblit New. Either allocates a memory buffer and saves a
- piece of the screen into it, or restores the screen
- then frees the buffer.
-
- - mnu_bar
- mnu_enable
- mnu_disable
-
- - obj_clcalc
-
- - obj_mxuserdef
-
- - obj_ppstring
-
- - rc_intersect Now returns FALSE when the width or height of the
- resulting rectangle is zero. Also, now copes with
- negative x/y values properly (but width and height
- better be >= 0).
-
- - rc_gtov The gtov now calcs ending x as grect->x+grect->w-1.
- rc_vtog Likewise for the y. (It didn't used to apply the -1
- adjustment.) The vtog has the inverse adjustment.
- I hope this doesn't break any existing code, but the
- old way was wrong. For example, it would convert
- the mono desktop grect to (0,0,640,400) and if you pass
- that to VDI v_bar() you end up drawing a box 1 pixel
- bigger than the screen.
-
- - rc_intersect Now returns FALSE when the width or height are zero.
-
- - rc_confine
-
- - rsc_rrbuttons
-
- - rsc_cubuttons
-
- - rsc_gtrees New. Gets the address of multiple resource trees in
- one call. Like what rsc_gstrings() does for strings.
-
- - wind_update Now vectored through wnd_update() function. The new
- wnd_update function treats wind_update() calls in a stack-like
- fashion, just like graf_mouse(). If you do BEG_UPDATE
- 3 times, then it takes 3 END_UPDATE calls before the
- real wind_update(END_UPDATE) is passed to the AES.
- Same deal with BEG/END_MCTRL. This is required for the
- new utility functions, which need to change the update
- status without hosing up the application. Since both
- the utils and the ap are forced through wnd_update(),
- the stack-like logic keeps it all straight.
-
-
- - frm_mdo
- - frm_menu
- - frm_dsmenu
- - frm_nlmenu
- - rc_ptinrect
- - frm_move
- - frm_init
- - frm_cleanup
- - frm_start
- - frm_draw
- - frm_do
- - frm_finish
-
- - frm_question
- - frm_qmenu
- - frm_qerror
- - frm_qtext
- - wnd_top
-
-
-