view lisp/gs.el @ 103616:af77bf73dfe0

* verilog-mode.el (verilog-beg-of-statement) (verilog-endcomment-reason-re): Support unique case and priority case. (verilog-basic-complete-re): Support localparam lineup. (verilog-beg-of-statement-1): Fix for robustness, unique case. (verilog-set-auto-endcomments): Fix for unique case, always_comb commenting. (verilog-leap-to-case-head): Now support *nested* unique & priority case statements. (verilog-auto-lineup): Make just declarations the default (as it had been). (verilog-leap-to-case-head): Support priority/unique case statements. (verilog-auto-lineup): Rework to give users radio buttons to select the various styles of automatic lineup (verilog-error-regexp-alist): Rework to support the XEmacs style of error regular expressions from compilers, lint tools & simulators. Note that GNU Emacs has made it impossible for a mode to load such things. (electric-verilog-terminate-line, verilog-indent-declaration) (verilog-auto-wiure): Rework for radio button selection of auto-lineup selection of specification of auto lineup. (verilog-beg-of-statement-1): Redesign to support proper operation in additional code, based on testing with auto-lineup. (verilog-calculate-indent, assignments & declarations) (verilog-backward-token): Enhance to support auto-lineup of assignments & declarations. (verilog-in-directive-p, verilog-at-struct-p): New function for easy test of whether we are. (verilog-pretty-declarations, verilog-pretty-expr): Massive rework to support safe execution at almost anyline. (verilog-calc-1): Properly support indenting deep inside generate blocks. (verilog-init-font) Remove definition & use of verilog-init-font, as it is redundant with font-lock-defaults. (verilog-mode): Alter the definition of verilog-font-lock-defualts to avoid circular calls if syntax-ppss is a function (as is the case now in 22.x GNU Emacs) as that function would sometimes call itself, leading to (nearly) infinite recursion (verilog-ovm-begin-re, verilog-ovm-end-re) (verilog-ovm-statement-re, verilog-leap-to-head) (verilog-backward-token): Add support for OVM macros. Some are complete statements, and others open and close scopes like begin and end. (verilog-defun-level-not-generate-re, verilog-defun-level-re) (verilog-defun-level-generate-only-re): Really fix the defun-list compilation issue (verilog-calc-1) (verilog-beg-of-statement): Enhance support for coverpoint, constraint and cross statements (verilog-defun-level-list, verilog-generate-defun-level-list) (verilog-all-defun-level-list): Redo these specifications - it is too hard to support eval-when compile aggregation of lists also built at when-compile time. (verilog-defun-level-list): Place defconsts of variables used in building regular expressions which are built in eval-when-compile bodies in the same eval-when-compile body to facilitate compile without load. (verilog-beg-block-re-ordered): Support indenting virtual/protected tasks and functions. (verilog-defun-level-list,verilog-in-generate-region-p) (verilog-backward-ws&directives, verilog-calc-1): Speed up indentation of some module items (generate items). (verilog-forward-sexp, verilog-leap-to-head): Support stepping across virtual/protected tasks and functions. * verilog-mode.el (verilog-auto-arg, verilog-auto-arg-sort): Allow sorting AUTOARG lists. Suggested by Andrea Fedeli. (verilog-read-sub-decls-line): Fix AUTOWIRE signals getting lost in concatenations. Reported by Yishay Belkind. (verilog-auto-ascii-enum): Support one-hot state machines in AUTOASCIIENUM. Suggested by Lloyd Gomez. (verilog-auto-inst, verilog-auto-inst-port): Include interface modport in AUTOINST and add vl-modport for users. Reported by David Rogoff. (verilog-auto-inout-module, verilog-auto-inst) (verilog-decls-get-interfaces, verilog-insert-definition) (verilog-insert-one-definition, verilog-read-decls) (verilog-read-sub-decls, verilog-read-sub-decls-sig) (verilog-sig-modport, verilog-signals-combine-bus) (verilog-subdecls-get-interfaces): Fix expansion of SystemVerilog interfaces in AUTOINOUTMODULE, AUTOINOUTCOMP, and AUTOINST. Suggested by David Rogoff. (verilog-repair-open-comma): Fix non-insertion of comma when `DEFINE occurs in V2K argument list. Reported by Lane Brooks. (verilog-make-width-expression): Simplify [A-1:0] expression widths to just {A{1'b0}}. (verilog-mode): Cleanup checkdoc warnings. (verilog-auto-inout-module, verilog-signals-matching-dir-re): Add third optional regexp to AUTOINOUTMODULE to allow selecting only inputs/outputs or data type. Suggested by Vasu Kandadi. (next-error-last-buffer): Fix byte-compiler warning. (verilog-auto, verilog-auto-insert-lisp, verilog-auto-inst) (verilog-delete-auto): Add AUTOINSERTLISP to insert arbitrary lisp or shell command text during AUTO expansion. Suggested by Tad Truex. (verilog-read-sub-decls-expr, verilog-read-sub-decls-line) (verilog-read-sub-decls-sig, verilog-symbol-detick-text): Fix dotted nets {a.b,c.d} and excaped identifiers being mis-included in AUTOINOUT. Reported by Matthew Lovell. (verilog-read-always-signals-recurse): Fix AUTORESET "if (a<=b)" causing use of <= assignments. Reported by Alex Reed. (verilog-read-decls): Fix triand, trior, wand, wor to be recognized by AUTOWIRE. Reported by Spencer Isaacson. (verilog-extended-complete-re): Support import "DPI-C" functions. (verilog-read-always-signals-recurse): Fix AUTORESET of "x <= y[a+1:a+1]" to not include a in reset list. Reported by Dan Dever. (verilog-insert-date, verilog-insert-year) (verilog-sk-header-tmpl): Fix verilog-header inserting error on Windows systems. Reported by Michael Potts. (verilog-read-module-name): Fix AUTOINST when the child module declaration's name is a tick define. Reported by Elliot Mednick. (verilog-read-decls): Fix V2K parameter bit subscripts getting passed to next parameter's definition. Reported by Bruce T. (verilog-read-decls): Fix detecting "parameter int" when using AUTOINSTPARAM. Reported by Bruce T. (verilog-goto-defun): Fix goto not finding modules unless first perform a verilog-auto expansion. Suggested by Lawrence Butcher. (verilog-mode): Expand -f flag arguments on entry to mode so verilog-goto-defun will work. Reported by Lawrence Butcher. (verilog-getopt): Expand environment variables in -f file arguments. Suggested by Lawrence Butcher. (verilog-set-define): Fix "Symbol's value as variable is void" when reading enumerations. (verilog-auto-ascii-enum): Fix duplicate labels in AUTOASCIIENUM. Suggested by Stephen Peltan. (verilog-read-defines): Fix reading of enumerations in include files. Reported by Steve Peltan.
author Dan Nicolaescu <dann@ics.uci.edu>
date Sun, 28 Jun 2009 17:52:45 +0000
parents d500b86d2189
children c97e346c34f6
line wrap: on
line source

;;; gs.el --- interface to Ghostscript

;; Copyright (C) 1998, 2001, 2002, 2003, 2004, 2005, 2006, 2007,
;;   2008, 2009 Free Software Foundation, Inc.

;; Maintainer: FSF
;; Keywords: internal

;; This file is part of GNU Emacs.

;; GNU Emacs 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 3 of the License, or
;; (at your option) any later version.

;; GNU Emacs 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 GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.

;;; Commentary:

;; This code is experimental.  Don't use it.

;;; Code:

(defvar gs-program "gs"
  "The name of the Ghostscript interpreter.")


(defvar gs-device "x11"
  "The Ghostscript device to use to produce images.")


(defvar gs-options
  '("-q"
    ;"-dNOPAUSE"
    "-dSAFER"
    "-dBATCH"
    "-sDEVICE=<device>"
    "<file>")
  "List of command line arguments to pass to Ghostscript.
Arguments may contain place-holders `<file>' for the name of the
input file, and `<device>' for the device to use.")
(put 'gs-options 'risky-local-variable t)

(defun gs-options (device file)
  "Return a list of command line options with place-holders replaced.
DEVICE is the value to substitute for the place-holder `<device>',
FILE is the value to substitute for the place-holder `<file>'."
  (mapcar #'(lambda (option)
	      (setq option (replace-regexp-in-string "<device>" device option)
		    option (replace-regexp-in-string "<file>" file option)))
	  gs-options))

;; The GHOSTVIEW property (taken from gv 3.5.8).
;;
;; Type:
;;
;; STRING
;;
;; Parameters:
;;
;; BPIXMAP ORIENT LLX LLY URX URY XDPI YDPI [LEFT BOTTOM TOP RIGHT]
;;
;; Scanf format: "%d %d %d %d %d %d %f %f %d %d %d %d"
;;
;; Explanation of parameters:
;;
;; BPIXMAP: pixmap id of the backing pixmap for the window.  If no
;; pixmap is to be used, this parameter should be zero.  This
;; parameter must be zero when drawing on a pixmap.
;;
;; ORIENT: orientation of the page.  The number represents clockwise
;; rotation of the paper in degrees.  Permitted values are 0, 90, 180,
;; 270.
;;
;; LLX, LLY, URX, URY: Bounding box of the drawable.  The bounding box
;; is specified in PostScript points in default user coordinates.
;;
;; XDPI, YDPI: Resolution of window.  (This can be derived from the
;; other parameters, but not without roundoff error.  These values are
;; included to avoid this error.)
;;
;; LEFT, BOTTOM, TOP, RIGHT: (optional) Margins around the window.
;; The margins extend the imageable area beyond the boundaries of the
;; window.  This is primarily used for popup zoom windows.  I have
;; encountered several instances of PostScript programs that position
;; themselves with respect to the imageable area.  The margins are
;; specified in PostScript points.  If omitted, the margins are
;; assumed to be 0.

(declare-function x-display-mm-width "xfns.c" (&optional terminal))
(declare-function x-display-pixel-width "xfns.c" (&optional terminal))

(defun gs-width-in-pt (frame pixel-width)
  "Return, on FRAME, pixel width PIXEL-WIDTH tranlated to pt."
  (let ((mm (* (float pixel-width)
	       (/ (float (x-display-mm-width frame))
		  (float (x-display-pixel-width frame))))))
    (/ (* 25.4 mm) 72.0)))

(declare-function x-display-mm-height "xfns.c" (&optional terminal))
(declare-function x-display-pixel-height "xfns.c" (&optional terminal))

(defun gs-height-in-pt (frame pixel-height)
  "Return, on FRAME, pixel height PIXEL-HEIGHT tranlated to pt."
  (let ((mm (* (float pixel-height)
	       (/ (float (x-display-mm-height frame))
		  (float (x-display-pixel-height frame))))))
    (/ (* 25.4 mm) 72.0)))

(declare-function x-change-window-property "xfns.c"
		  (prop value &optional frame type format outer-p))

(defun gs-set-ghostview-window-prop (frame spec img-width img-height)
  "Set the `GHOSTVIEW' window property of FRAME.
SPEC is a GS image specification.  IMG-WIDTH is the width of the
requested image, and IMG-HEIGHT is the height of the requested
image in pixels."
  (let* ((box (plist-get (cdr spec) :bounding-box))
	 (llx (elt box 0))
	 (lly (elt box 1))
	 (urx (elt box 2))
	 (ury (elt box 3))
	 (rotation (or (plist-get (cdr spec) :rotate) 0))
	 ;; The pixel width IMG-WIDTH of the pixmap gives the
	 ;; dots, URX - LLX give the inch.
	 (in-width (/ (- urx llx) 72.0))
	 (in-height (/ (- ury lly) 72.0))
	 (xdpi (/ img-width in-width))
	 (ydpi (/ img-height in-height)))
    (x-change-window-property "GHOSTVIEW"
			      (format "0 %d %d %d %d %d %g %g"
				      rotation llx lly urx ury xdpi ydpi)
			      frame)))

(declare-function x-display-grayscale-p "xfns.c" (&optional terminal))

(defun gs-set-ghostview-colors-window-prop (frame pixel-colors)
  "Set the `GHOSTVIEW_COLORS' environment variable depending on FRAME."
  (let ((mode (cond ((x-display-color-p frame) "Color")
		    ((x-display-grayscale-p frame) "Grayscale")
		    (t "Monochrome"))))
    (x-change-window-property "GHOSTVIEW_COLORS"
			      (format "%s %s" mode pixel-colors)
			      frame)))

(declare-function x-window-property "xfns.c"
		  (prop &optional frame type source delete-p vector-ret-p))

;;;###autoload
(defun gs-load-image (frame spec img-width img-height window-and-pixmap-id
			    pixel-colors)
  "Load a PS image for display on FRAME.
SPEC is an image specification, IMG-HEIGHT and IMG-WIDTH are width
and height of the image in pixels.  WINDOW-AND-PIXMAP-ID is a string of
the form \"WINDOW-ID PIXMAP-ID\".  Value is non-nil if successful."
  (unwind-protect
      (let ((file (plist-get (cdr spec) :file))
	    gs
 	    (timeout 40))
 	;; Wait while property gets freed from a previous ghostscript process
 	;; sit-for returns nil as soon as input starts being
	;; available, so if we want to give GhostScript a reasonable
 	;; chance of starting up, we better use sleep-for.  We let
 	;; sleep-for wait only half the time because if input is
 	;; available, it is more likely that we don't care that much
 	;; about garbled redisplay and are in a hurry.
 	(while (and
 		;; Wait while the property is not yet available
 		(not (zerop (length (x-window-property "GHOSTVIEW"
 						       frame))))
 		;; The following was an alternative condition: wait
 		;; while there is still a process running.  The idea
 		;; was to avoid contention between processes.  Turned
 		;; out even more sluggish.
 		;; (get-buffer-process "*GS*")
 		(not (zerop timeout)))
 	  (unless (sit-for 0 100 t)
 	    (sleep-for 0 50))
 	  (setq timeout (1- timeout)))

 	;; No use waiting longer.  We might want to try killing off
 	;; stuck processes, but there is no point in doing so: either
 	;; they are stuck for good, in which case the user would
 	;; probably be responsible for that, and killing them off will
 	;; make debugging harder, or they are not.  In that case, they
 	;; will cause incomplete displays.  But the same will happen
 	;; if they are killed, anyway. The whole is rather
 	;; disconcerting, and fast scrolling through a dozen images
 	;; will make Emacs freeze for a while.  The alternatives are a)
 	;; proper implementation not waiting at all but creating
 	;; appropriate queues, or b) permanently bad display due to
 	;; bad cached images.  So remember that this
 	;; is just a hack and if people don't like the behavior, they
 	;; will most likely like the easy alternatives even less.
 	;; And at least the image cache will make the delay apparent
 	;; just once.
	(gs-set-ghostview-window-prop frame spec img-width img-height)
	(gs-set-ghostview-colors-window-prop frame pixel-colors)
	(setenv "GHOSTVIEW" window-and-pixmap-id)
	(setq gs (apply 'start-process "gs" "*GS*" gs-program
			(gs-options gs-device file)))
	(set-process-query-on-exit-flag gs nil)
	gs)
    nil))


;(defun gs-put-tiger ()
;  (let* ((ps-file "/usr/local/share/ghostscript/5.10/examples/tiger.ps")
;	  (spec `(image :type postscript
;			:pt-width 200 :pt-height 200
;			:bounding-box (22 171 567 738)
;			:file ,ps-file)))
;    (put-text-property 1 2 'display spec)))
;

(provide 'gs)

;; arch-tag: 06ab51b8-4932-4cfe-9f60-b924a8edb3f0
;;; gs.el ends here