For the release

For the release

[Richard M. Stallman]

I have the impression that many feature changes made were not recorded
in etc/NEWS.  So someone should check the change logs looking for things
that need to be added there.

Re: For the release

[A Soare]

> If you feel for it then please try to look at the indentation functions 
> in mumamo.el, part of nXhtml.
> 

I looked a little. If you refer to html, xml, or similar, yes, the automaton can indent easy these kind of codes. It must be written for this kind of codes just the function xml-indent-rules for example.

It can indent more general every arborescent structure, particularly that for a programming language, because every code express a logic, and a logic is a tree.



____________________________________________________

Avant de prendre le volant, repérez votre itinéraire et visualisez le trafic ! http://itineraire.voila.fr/itineraire.html

Re: For the release

[A Soare]

> I have the impression that many feature changes made were not recorded
> in etc/NEWS.  So someone should check the change logs looking for things
> that need to be added there.
> 
> 

Do not forget to specify that the indentation of lisp is founded on a different system;

I will write soon an identical indentation of C code, using the same automaton, because it takes me just little time.

However the implantation of the automaton in C/parse-partial-sexp, as a new field to teh lisp_parse_state structure, in order to get the indentation of every kind of code (in _every_ major mode) after running parse-partial sexp, as I explained you, will be made by somebody else if they want to, because for me it is clear how it can be done, and maybe somebody wants to do it. I will not spend time on. It's identical to that for lisp, but need to be implanted in C.

struct lisp_parse_state
  {
    int depth;	   /* Depth at end of parsing.  */
    int instring;  /* -1 if not within string, else desired terminator.  */
    int incomment; /* -1 if in unnestable comment else comment nesting */
    int comstyle;  /* comment style a=0, or b=1, or ST_COMMENT_STYLE.  */
    int quoted;	   /* Nonzero if just after an escape char at end of parsing */
    int mindepth;  /* Minimum depth seen while scanning.  */
    /* Char number of most recent start-of-expression at current level */
    EMACS_INT thislevelstart;
    /* Char number of start of containing expression */
    EMACS_INT prevlevelstart;
    EMACS_INT location;	     /* Char number at which parsing stopped.  */
    EMACS_INT comstr_start;  /* Position of last comment/string starter.  */
    Lisp_Object levelstarts; /* Char numbers of starts-of-expression
				of levels (starting from outermost).  */

    Lisp_Object INDENTATION;  <= NEW FIELD

  };





____________________________________________________

Avant de prendre le volant, repérez votre itinéraire et visualisez le trafic ! http://itineraire.voila.fr/itineraire.html

For the release

[Richard M. Stallman]

Could someone please check the fix that Kevin Gal sent in June?


** Fix recognition of shell's `dirs' command.

Is his change right?

Date: Wed, 29 Jun 2005 18:21:28 -0500 (CDT)
From: kevin.gal@verizon.net

When I use tcsh (which echoes command input) in a shell buffer,
entering "M-x dirs" fails because it mistakens the echoed "dirs"
string as the directory string to use as input when changing the
buffers default directory.  The attached gziped patch file contains a
fix.  The patch also binds the "dirs" command to "\e\C-m" in
shell-mode-map (previously unbound) to make it easy to invoke.
Finally, a test has been added to see if "shell-dirstack-query" is
non-nil.  If so, it is not overwritten.

To enable the fix, the user must set comint-process-echoes to t.