Merge from emacs-24; up to 2014-07-02T00:57:53Z!juri@jurta.org

admin/authors.el

@@ -92,6 +92,7 @@ files.")
     ("Joseph Arceneaux" "Joe Arceneaux")
     ("Joseph M. Kelsey" "Joe Kelsey")	; FIXME ?
     ("Juan León Lahoz García" "Juan-Leon Lahoz Garcia")
+    ("Jürgen Hötzel" "Juergen Hoetzel")
     ("K. Shane Hartman" "Shane Hartman")
     ("Kai Großjohann" "Kai Grossjohann")
     ("Karl Berry" "K. Berry")

doc/lispref/ChangeLog

@@ -1,3 +1,8 @@
+2014-08-28  Eli Zaretskii  <eliz@gnu.org>
+
+	* display.texi (Bidirectional Display): Update the Emacs's class
+	of bidirectional conformance.
+
 2014-08-27  Dmitry Antipov  <dmantipov@yandex.ru>
 
 	* eval.texi (Eval): Mention possible recovery from stack overflow.

doc/lispref/display.texi

@@ -6551,8 +6551,10 @@ positions do not increase monotonically with string or buffer
 position.  In performing this @dfn{bidirectional reordering}, Emacs
 follows the Unicode Bidirectional Algorithm (a.k.a.@: @acronym{UBA}),
 which is described in Annex #9 of the Unicode standard
-(@url{http://www.unicode.org/reports/tr9/}).  Emacs provides a ``Full
-Bidirectionality'' class implementation of the @acronym{UBA}.
+(@url{http://www.unicode.org/reports/tr9/}).  Emacs currently provides
+a ``Non-isolate Bidirectionality'' class implementation of the
+@acronym{UBA}: it does not yet support the isolate directional
+formatting characters introduced with Unicode Standard v6.3.0.
 
 @defvar bidi-display-reordering
 If the value of this buffer-local variable is non-@code{nil} (the

etc/ChangeLog

@@ -1,3 +1,7 @@
+2014-08-28  Glenn Morris  <rgm@gnu.org>
+
+	* emacs.appdata.xml: New file; description from Emacs's homepage.
+
 2014-08-25  Eli Zaretskii  <eliz@gnu.org>
 
 	* NEWS: Mention that string-collate-* functions are supported on

etc/emacs.appdata.xml

@@ -0,0 +1,33 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!-- Copyright (C) 2014 Free Software Foundation, Inc. -->
+<application>
+ <id type="desktop">emacs.desktop</id>
+ <metadata_license>GFDL-1.3</metadata_license>
+ <!-- It is GFDL-1.3+, but appdata-validate complains... -->
+ <project_license>GPL-3.0+ and GFDL-1.3</project_license>
+ <name>GNU Emacs</name>
+ <summary>An extensible text editor</summary>
+ <description>
+  <p>
+   GNU Emacs is an extensible, customizable text editor - and more.
+   At its core is an interpreter for Emacs Lisp, a dialect of the Lisp
+   programming language with extensions to support text editing.
+  </p>
+  <p>The features of GNU Emacs include:</p>
+  <ul>
+   <li>Content-sensitive editing modes, including syntax coloring, for
+   a wide-range of file types</li>
+   <li>Complete built-in documentation, including a tutorial for new users</li>
+   <li>Full Unicode support for nearly all human languages and their scripts</li>
+   <li>Highly customizable, using Emacs Lisp code or a graphical interface</li>
+   <li>Includes a project planner, mail and news reader, debugger
+   interface, calendar, and more</li>
+  </ul>
+ </description>
+ <screenshots>
+  <screenshot type="default" width="632" height="354">http://www.gnu.org/software/emacs/images/appdata.png</screenshot>
+ </screenshots>
+ <url type="homepage">http://www.gnu.org/software/emacs</url>
+ <updatecontact>emacs-devel_at_gnu.org</updatecontact>
+ <project_group>GNU</project_group>
+</application>

lisp/ChangeLog

@@ -1,3 +1,26 @@
+2014-08-28  Michael Albinus  <michael.albinus@gmx.de>
+
+	* emacs-lisp/authors.el (authors-aliases): Addition.
+
+	* net/tramp-adb.el: Spell author name correctly.
+
+2014-08-28  João Távora  <joaotavora@gmail.com>
+
+	* net/shr.el (shr-expand-url): Plain expand-file-name is not enough;
+	use url-expand-file-name.  (Bug#18310)
+
+2014-08-28  Glenn Morris  <rgm@gnu.org>
+
+	* emulation/cua-rect.el (cua--highlight-rectangle):
+	Avoid error at point-min.  (Bug#18309)
+
+2014-08-28  Stefan Monnier  <monnier@iro.umontreal.ca>
+
+	* progmodes/python.el (python-shell-prompt-detect): Remove redundant
+	executable-find (bug#18244).
+
+	* simple.el (self-insert-uses-region-functions): Defvar.
+
 2014-08-28  Glenn Morris  <rgm@gnu.org>
 
 	* subr.el (remq): Revert 2014-08-25 doc change (not always true).

lisp/emulation/cua-rect.el

@@ -794,7 +794,7 @@ If command is repeated at same position, delete the rectangle."
 			       (make-string
 				(- l cl0 (if (and (= le pl) (/= le lb)) 1 0))
 				(if cua--virtual-edges-debug ?. ?\s))
-			       'face (or (get-text-property (1- s) 'face) 'default)))
+			       'face (or (get-text-property (max (1- s) (point-min)) 'face) 'default)))
 		     (if (/= pl le)
 			 (setq s (1- s))))
 		   (cond

lisp/net/shr.el

@@ -589,6 +589,10 @@ size, and full-buffer size."
 	  (url-type parsed)
 	  url)))
 
+(autoload 'url-expand-file-name "url-expand")
+
+;; FIXME This needs some tests writing.
+;; Does it even need to exist, given that url-expand-file-name does?
 (defun shr-expand-url (url &optional base)
   (setq base
 	(if base
@@ -614,7 +618,7 @@ size, and full-buffer size."
 	 (concat (nth 3 base) url))
 	(t
 	 ;; Totally relative.
-	 (concat (car base) (expand-file-name url (cadr base))))))
+	 (url-expand-file-name url (concat (car base) (cadr base))))))
 
 (defun shr-ensure-newline ()
   (unless (zerop (current-column))

lisp/net/tramp-adb.el

@@ -2,7 +2,7 @@
 
 ;; Copyright (C) 2011-2014 Free Software Foundation, Inc.
 
-;; Author: Juergen Hoetzel <juergen@archlinux.org>
+;; Author: Jürgen Hötzel <juergen@archlinux.org>
 ;; Keywords: comm, processes
 ;; Package: tramp
 

lisp/progmodes/python.el

@@ -1905,7 +1905,7 @@ detection and just returns nil."
                 (let ((code-file (python-shell--save-temp-file code)))
                   ;; Use `process-file' as it is remote-host friendly.
                   (process-file
-                   (executable-find python-shell-interpreter)
+                   python-shell-interpreter
                    code-file
                    '(t nil)
                    nil
@@ -2061,11 +2061,14 @@ uniqueness for different types of configurations."
             (or python-shell-virtualenv-path "")
             (mapconcat #'identity python-shell-exec-path "")))))
 
-(defun python-shell-parse-command ()
+(defun python-shell-parse-command ()    ;FIXME: why name it "parse"?
   "Calculate the string used to execute the inferior Python process."
+  ;; FIXME: process-environment doesn't seem to be used anywhere within
+  ;; this let.
   (let ((process-environment (python-shell-calculate-process-environment))
         (exec-path (python-shell-calculate-exec-path)))
     (format "%s %s"
+            ;; FIXME: Why executable-find?
             (executable-find python-shell-interpreter)
             python-shell-interpreter-args)))
 
@@ -2101,11 +2104,10 @@ uniqueness for different types of configurations."
 (defun python-shell-calculate-exec-path ()
   "Calculate exec path given `python-shell-virtualenv-path'."
   (let ((path (append python-shell-exec-path
-                      exec-path nil)))
+                      exec-path nil)))  ;FIXME: Why nil?
     (if (not python-shell-virtualenv-path)
         path
-      (cons (format "%s/bin"
-                    (directory-file-name python-shell-virtualenv-path))
+      (cons (expand-file-name "bin" python-shell-virtualenv-path)
             path))))
 
 (defvar python-shell--package-depth 10)

lisp/simple.el

@@ -374,6 +374,13 @@ Other major modes are defined by comparison with this one."
 
 ;; Making and deleting lines.
 
+(defvar self-insert-uses-region-functions nil
+  "Special hook to tell if `self-insert-command' will use the region.
+It must be called via `run-hook-with-args-until-success' with no arguments.
+Any `post-self-insert-command' which consumes the region should
+register a function on this hook so that things like `delete-selection-mode'
+can refrain from consuming the region.")
+
 (defvar hard-newline (propertize "\n" 'hard t 'rear-nonsticky '(hard))
   "Propertized string representing a hard newline character.")
 

msdos/ChangeLog

@@ -1,3 +1,13 @@
+2014-08-28  Eli Zaretskii  <eliz@gnu.org>
+
+	* sedlibmk.inp (gl_LIBOBJS): Add execinfo.o.  Reported by Juan
+	Manuel Guerrero <juan.guerrero@gmx.de>.
+
+	* sed2v2.inp [DJGPP <= 2.03]: Add a prototype for snprintf, to
+	avoid compilation warning from newer GCC versions that have
+	snprintf as a built-in.  Reported by Juan Manuel Guerrero
+	<juan.guerrero@gmx.de>.
+
 2014-08-09  Eli Zaretskii  <eliz@gnu.org>
 
 	* INSTALL: Fix last change.

msdos/sed2v2.inp

@@ -127,6 +127,8 @@ s/^#undef HAVE_STRTOULL *$/#define HAVE_STRTOULL 1/
 #define HAVE_SNPRINTF 1\
 #else\
 #undef HAVE_SNPRINTF\
+#include <sys/types.h>\
+extern int snprintf (char *__restrict, size_t, const char *__restrict, ...);\
 #endif
 s/^#undef PENDING_OUTPUT_N_BYTES *$/#define PENDING_OUTPUT_N_BYTES fp->_ptr - fp->_base/
 

msdos/sedlibmk.inp

@@ -315,6 +315,7 @@ am__cd = cd
 /^BYTESWAP_H *=/s/@[^@\n]*@/byteswap.h/
 /^DIRENT_H *=/s/@[^@\n]*@//
 /^ERRNO_H *=/s/@[^@\n]*@//
+/^EXECINFO_H *=/s/@[^@\n]*@/execinfo.h/
 /^STDBOOL_H *=/s/@[^@\n]*@//
 /^STDALIGN_H *=/s/@[^@\n]*@/stdalign.h/
 /^STDARG_H *=/s/@[^@\n]*@//
@@ -333,7 +334,7 @@ am__cd = cd
 /am__append_[1-9][0-9]* *=.*gettext\.h/s/@[^@\n]*@/\#/
 /am__append_2 *=.*verify\.h/s/@[^@\n]*@//
 /^@gl_GNULIB_ENABLED_tempname_TRUE@/s/@[^@\n]*@//
-/^gl_LIBOBJS *=/s/@[^@\n]*@/getopt.o getopt1.o memrchr.o sig2str.o time_r.o getloadavg.o pthread_sigmask.o mkostemp.o fpending.o fdatasync.o/
+/^gl_LIBOBJS *=/s/@[^@\n]*@/getopt.o getopt1.o memrchr.o sig2str.o time_r.o getloadavg.o pthread_sigmask.o mkostemp.o fpending.o fdatasync.o execinfo.o/
 /^am__append_[1-9][0-9]* *=/,/^[^ 	]/{
   s/ *inttypes\.h//
   s| *sys/select\.h||

src/ChangeLog

@@ -1,3 +1,20 @@
+2014-08-28  Eli Zaretskii  <eliz@gnu.org>
+
+	* conf_post.h (_GL_EXECINFO_INLINE) [MSDOS]: Don't define.
+
+	* indent.c (Fvertical_motion): Fix vertical motion up through a
+	display property after a newline.  (Bug#18276)
+
+	* xdisp.c (display_line): Don't assume that the call to
+	reseat_at_next_visible_line_start ends up at a character
+	immediately following the newline on the previous line.  Avoids
+	setting the ends_at_zv_p flag on screen lines that are not at or
+	beyond ZV, which causes infloop in redisplay.  For the details, see
+	http://lists.gnu.org/archive/html/emacs-devel/2014-08/msg00368.html.
+
+	* dispnew.c (buffer_posn_from_coords): Fix mirroring of X
+	coordinate for hscrolled R2L screen lines.  (Bug#18277)
+
 2014-08-28  Paul Eggert  <eggert@cs.ucla.edu>
 
 	* sysdep.c (LC_COLLATE, LC_COLLATE_MASK): Give individual defaults

src/conf_post.h

@@ -123,13 +123,6 @@ You lose; /* Emacs for DOS must be compiled with DJGPP */
    so we could reuse it in readlinkat; see msdos.c.  */
 #define opendir sys_opendir
 
-/* The "portable" definition of _GL_INLINE on config.h does not work
-   with DJGPP GCC 3.4.4: it causes unresolved externals in sysdep.c,
-   although lib/execinfo.h is included and the inline functions there
-   are visible.  */
-#if __GNUC__ < 4
-# define _GL_EXECINFO_INLINE inline
-#endif
 /* End of gnulib-related stuff.  */
 
 #define emacs_raise(sig) msdos_fatal_signal (sig)

src/dispnew.c

@@ -5143,9 +5143,8 @@ buffer_posn_from_coords (struct window *w, int *x, int *y, struct display_pos *p
   move_it_to (&it, -1, 0, *y, -1, MOVE_TO_X | MOVE_TO_Y);
 
   /* TO_X is the pixel position that the iterator will compute for the
-     glyph at *X.  We add it.first_visible_x because iterator
-     positions include the hscroll.  */
-  to_x = x0 + it.first_visible_x;
+     glyph at *X.  */
+  to_x = x0;
   if (it.bidi_it.paragraph_dir == R2L)
     /* For lines in an R2L paragraph, we need to mirror TO_X wrt the
        text area.  This is because the iterator, even in R2L
@@ -5159,6 +5158,10 @@ buffer_posn_from_coords (struct window *w, int *x, int *y, struct display_pos *p
        it should be mirrored into zero pixel position.)  */
     to_x = window_box_width (w, TEXT_AREA) - to_x - 1;
 
+  /* We need to add it.first_visible_x because iterator positions
+     include the hscroll. */
+  to_x += it.first_visible_x;
+
   /* Now move horizontally in the row to the glyph under *X.  Second
      argument is ZV to prevent move_it_in_display_line from matching
      based on buffer positions.  */

src/indent.c

@@ -2004,6 +2004,8 @@ whether or not it is currently displayed in some window.  */)
       int first_x;
       bool overshoot_handled = 0;
       bool disp_string_at_start_p = 0;
+      ptrdiff_t nlines = XINT (lines);
+      int vpos_init = 0;
 
       itdata = bidi_shelve_cache ();
       SET_TEXT_POS (pt, PT, PT_BYTE);
@@ -2093,18 +2095,31 @@ whether or not it is currently displayed in some window.  */)
 
 	  overshoot_handled = 1;
 	}
-      if (XINT (lines) <= 0)
+      else if (IT_CHARPOS (it) == PT - 1
+	       && FETCH_BYTE (PT - 1) == '\n'
+	       && nlines < 0)
 	{
-	  it.vpos = 0;
+	  /* The position we started from was covered by a display
+	     property, so we moved to position before the string, and
+	     backed up one line, because the character at PT - 1 is a
+	     newline.  So we need one less line to go up.  */
+	  nlines++;
+	  /* But we still need to record that one line, in order to
+	     return the correct value to the caller.  */
+	  vpos_init = -1;
+	}
+      if (nlines <= 0)
+	{
+	  it.vpos = vpos_init;
 	  /* Do this even if LINES is 0, so that we move back to the
 	     beginning of the current line as we ought.  */
-	  if (XINT (lines) == 0 || IT_CHARPOS (it) > 0)
-	    move_it_by_lines (&it, max (PTRDIFF_MIN, XINT (lines)));
+	  if (nlines == 0 || IT_CHARPOS (it) > 0)
+	    move_it_by_lines (&it, max (PTRDIFF_MIN, nlines));
 	}
       else if (overshoot_handled)
 	{
 	  it.vpos = 0;
-	  move_it_by_lines (&it, min (PTRDIFF_MAX, XINT (lines)));
+	  move_it_by_lines (&it, min (PTRDIFF_MAX, nlines));
 	}
       else
 	{
@@ -2119,13 +2134,13 @@ whether or not it is currently displayed in some window.  */)
 		  it.vpos = 0;
 		  move_it_by_lines (&it, 1);
 		}
-	      if (XINT (lines) > 1)
-		move_it_by_lines (&it, min (PTRDIFF_MAX, XINT (lines) - 1));
+	      if (nlines > 1)
+		move_it_by_lines (&it, min (PTRDIFF_MAX, nlines - 1));
 	    }
 	  else
 	    {
 	      it.vpos = 0;
-	      move_it_by_lines (&it, min (PTRDIFF_MAX, XINT (lines)));
+	      move_it_by_lines (&it, min (PTRDIFF_MAX, nlines));
 	    }
 	}
 

src/xdisp.c

@@ -3413,6 +3413,48 @@ handle_stop (struct it *it)
   if (it->selective_display_ellipsis_p)
     it->saved_face_id = it->face_id;
 
+  /* Here's the description of the semantics of, and the logic behind,
+     the various HANDLED_* statuses:
+
+     HANDLED_NORMALLY means the handler did its job, and the loop
+     should proceed to calling the next handler in order.
+
+     HANDLED_RECOMPUTE_PROPS means the handler caused a significant
+     change in the properties and overlays at current position, so the
+     loop should be restarted, to re-invoke the handlers that were
+     already called.  This happens when fontification-functions were
+     called by handle_fontified_prop, and actually fontified
+     something.  Another case where HANDLED_RECOMPUTE_PROPS is
+     returned is when we discover overlay strings that need to be
+     displayed right away.  The loop below will continue for as long
+     as the status is HANDLED_RECOMPUTE_PROPS.
+
+     HANDLED_RETURN means return immediately to the caller, to
+     continue iteration without calling any further handlers.  This is
+     used when we need to act on some property right away, for example
+     when we need to display the ellipsis or a replacing display
+     property, such as display string or image.
+
+     HANDLED_OVERLAY_STRING_CONSUMED means an overlay string was just
+     consumed, and the handler switched to the next overlay string.
+     This signals the loop below to refrain from looking for more
+     overlays before all the overlay strings of the current overlay
+     are processed.
+
+     Some of the handlers called by the loop push the iterator state
+     onto the stack (see 'push_it'), and arrange for the iteration to
+     continue with another object, such as an image, a display string,
+     or an overlay string.  In most such cases, it->stop_charpos is
+     set to the first character of the string, so that when the
+     iteration resumes, this function will immediately be called
+     again, to examine the properties at the beginning of the string.
+
+     When a display or overlay string is exhausted, the iterator state
+     is popped (see 'pop_it'), and iteration continues with the
+     previous object.  Again, in many such cases this function is
+     called again to find the next position where properties might
+     change.  */
+
   do
     {
       handled = HANDLED_NORMALLY;
@@ -20621,10 +20663,15 @@ display_line (struct it *it)
 	  row->truncated_on_right_p = 1;
 	  it->continuation_lines_width = 0;
 	  reseat_at_next_visible_line_start (it, 0);
-	  if (IT_BYTEPOS (*it) <= BEG_BYTE)
-	    row->ends_at_zv_p = true;
+	  /* We insist below that IT's position be at ZV because in
+	     bidi-reordered lines the character at visible line start
+	     might not be the character that follows the newline in
+	     the logical order.  */
+	  if (IT_BYTEPOS (*it) > BEG_BYTE)
+	    row->ends_at_zv_p =
+	      IT_BYTEPOS (*it) >= ZV_BYTE && FETCH_BYTE (ZV_BYTE - 1) != '\n';
 	  else
-	    row->ends_at_zv_p = FETCH_BYTE (IT_BYTEPOS (*it) - 1) != '\n';
+	    row->ends_at_zv_p = false;
 	  break;
 	}
     }

test/ChangeLog

@@ -1,3 +1,8 @@
+2014-08-28  Glenn Morris  <rgm@gnu.org>
+
+	* automated/python-tests.el (python-shell-calculate-exec-path-2):
+	Update test for today's python.el changes.
+
 2014-08-13  Jan Nieuwenhuizen  <janneke@gnu.org>
 
 	* automated/compile-tests.el (compile--test-error-line): Grok FILE

test/automated/python-tests.el

@@ -1753,7 +1753,7 @@ Using `python-shell-interpreter' and
   "Test `python-shell-exec-path' modification."
   (let* ((original-exec-path exec-path)
          (python-shell-virtualenv-path
-          (directory-file-name user-emacs-directory))
+          (directory-file-name (expand-file-name user-emacs-directory)))
          (exec-path (python-shell-calculate-exec-path)))
     (should (equal
              exec-path