annotate etc/ETAGS.EBNF @ 72917:17942cb3949e

(allout-regexp, allout-line-boundary-regexp) (allout-bob-regexp): Correct grouping and boundaries to fix backwards traversal. (allout-depth-specific-regexp, allout-depth-one-regexp): New versions that exploit \\{M\\} regexp syntax, to avoid geometric or worse time in allout-ascend. (allout-doublecheck-at-and-shallower): Identify depth threshold below which topics are checked for and disqualified by containment discontinuities. (allout-hotspot-key-handler): Correctly handle multiple-key strokes. Remove some unused variables. (allout-mode-leaders): Clarify that mode-specific comment-start will be used (set-allout-regexp): Correctly regexp-quote allout regexps to properly accept alternative header-leads and primary bullets with regexp-specific characters (eg, C "/*", mathematica "(*"). Include new regular expressions among those configured. (allout-infer-header-lead-and-primary-bullet): Rename allout-infer-header-lead. (allout-recent-depth): Manifest as a variable as well as a function. (allout-prefix-data): Simplify into an inline instead of a macro, assuming current match data rather than being explicitly passed it. Establish allout-recent-depth value as well as allout-recent-prefix-beginning and allout-recent-prefix-end. (allout-aberrant-container-p): True when an item's immediate offspring discontinuously contained. Useful for disqualifying unintended topic prefixes, likely at low depths. (allout-goto-prefix-doublechecked): Elaborated version of allout-goto-prefix which disqualifies aberrant pseudo-items. (allout-pre-next-prefix): Layer on top of lower-level routines, to get disqualification of aberrant containers. (allout-end-of-prefix, allout-end-of-subtree): Disqualify aberrant containers. (allout-beginning-of-current-entry): Position at start of buffer when in container (depth 0) entry. (nullify-allout-prefix-data): Invalidate allout-recent-* prefix data. (allout-current-bullet): Strip text properties. (allout-get-prefix-bullet): Use right match groups. (allout-beginning-of-line, allout-next-heading): Disqualify aberrant containers. (allout-previous-heading): Disqualify aberrant containers, and change to regular (rather than inline) function, to allow self-recursion. (allout-get-invisibility-overlay): Increment so progress is made when the first overlay is not the sought one. (allout-end-of-prefix): Disqualify aberrant containers. (allout-end-of-line): Cycle something like allout-beginning-of-line. (allout-mode): Make allout-old-style-prefixes (ie, enabling use with outline.el outlines) functional again. Change the primary bullet along with the header-lead - level 1 new-style bullets now work. Engage allout-before-change-handler in mainline emacs, not just xemacs, to do undo handling. (allout-before-change-handler): Expose undo changes occurring in hidden regions. Use allout-get-invisibility-overlay instead of reimplementing it inline. (allout-chart-subtree): Use start rather than end of prefix in charts. Use allout-recent-depth variable. (allout-chart-siblings): Disqualify aberrant topics. (allout-beginning-of-current-entry): Position correctly. (allout-ascend): Use new allout-depth-specific-regexp and allout-depth-one-regexp for linear instead of O(N^2) or worse behavior. (allout-ascend-to-depth): Depend on allout-ascend, rather than reimplementing an algorithm. (allout-up-current-level): Depend on allout-ascend, rather than reimplementing an algorithm. Return to start-point if we fail. (allout-descend-to-depth): Use allout-recent-depth variable instead of function. (allout-next-sibling): On traversal of numerous intervening topics, resort to economical allout-next-sibling-leap. (allout-next-sibling-leap): Specialized version of allout-next-sibling that uses allout-ascend cleverly, to depend on a regexp search to leap large numbers of contained topics, rather than arbitrarily many one-by-one traversals. (allout-next-visible-heading): Disqualify aberrant topics. (allout-previous-visible-heading): Position consistently when interactive. (allout-forward-current-level): Base on allout-previous-sibling rather than (differently) reimplmenting the algorithm. Remove some unused variables. (allout-solicit-alternate-bullet): Present default choice stripped of text properties. (allout-rebullet-heading): Use bullet stripped of text properties. Register changes using allout-exposure-change-hook. Disregard aberrant topics. (allout-shift-in): With universal-argument, make topic a peer of it's former offspring. Simplify the code by separating out allout-shift-out functionality. (allout-shift-out): With universal-argument, make offspring peers of their former container, and its siblings. Implement the functionality here, rather than inappropriately muddling the implementation of allout-shift-in. (allout-rebullet-topic): Respect additional argument for new parent-child separation function. (allout-yank-processing): Use allout-ascend directly. (allout-show-entry): Disqualify aberrant topics. (allout-show-children): Handle discontinuous children gracefully, extending the depth being revealed to expose them and posting a message indicating the situation. (allout-show-to-offshoot): Remove obsolete and incorrect comment. Leave cursor in correct position. (allout-hide-current-subtree): Use allout-ascend directly. Disqualify aberrant topics. (allout-kill-line, allout-kill-topic): Preserve exposure layout in a way that the yanks can restore it, as used to happen. (allout-yank-processing): Restore exposure layout as recorded by allout-kill-*, as used to happen. (allout-annotate-hidden, allout-hide-by-annotation): New routines for preseving and restoring exposure layout across kills. (allout-toggle-subtree-encryption): Run allout-exposure-change-hook. (allout-encrypt-string): Strip text properties. Rearranged order and outline-headings for some of the miscellaneous functions. (allout-resolve-xref): No need to quote the error name in the condition-case handler section. (allout-flatten): Classic recursive (and recursively intensive, without tail-recursion) list-flattener, needed by allout-shift-out when confronted with discontinuous children.

-*- indented-text -*-

See the end of this file for copyright information.

This file contains two sections:

1) An EBNF (Extended Backus-Naur Form) description of the format of
    the tags file created by etags.c and interpreted by etags.el;
2) A discussion of tag names and implicit tag names.

====================== 1) EBNF tag file description =====================

Productions created from current behaviour to aid extensions
Francesco Potorti` <[e-mail]> 2002
----------------

FF ::= #x0c				   /* tag section starter */

LF ::= #x0a				   /* line terminator */

DEL ::= #x7f				   /* pattern terminator */

SOH ::= #x01				   /* name terminator */

regchar ::= [^#x0a#x0c#x7f]		   /* regular character */

regstring ::= { regchar }		   /* regular string */

unsint ::= [0-9] { [0-9] }		   /* non-negative integer */



tagfile ::= { tagsection }		   /* a tags file */

tagsection ::= FF LF ( includesec | regularsec ) LF

includesec ::= filename ",include" [ LF fileprop ]

regularsec ::= filename "," [ unsint ] [ LF fileprop ] { LF tag }

filename ::= regchar regstring		   /* a file name */

fileprop ::= "(" regstring ")"		   /* an elisp alist */

tag ::= directtag | patterntag

directtag ::= DEL realposition		   /* no pattern */

patterntag ::= pattern DEL [ tagname SOH ] position

pattern ::= regstring			   /* a tag pattern */

tagname ::= regchar regstring		   /* a tag name */

position ::= realposition | "," 	   /* charpos,linepos */

realposition ::= "," unsint | unsint "," | unsint "," unsint

==================== end of EBNF tag file description ====================



======================= 2) discussion of tag names =======================

- WHAT ARE TAG NAMES
Tag lines in a tags file are usually made from the above defined pattern
and by an optional tag name.  The pattern is a string that is searched
in the source file to find the tagged line.

- WHY TAG NAMES ARE GOOD
When a user looks for a tag, Emacs first compares the tag with the tag
names contained in the tags file.  If no match is found, Emacs compares
the tag with the patterns.  The tag name is then the preferred way to
look for tags in the tags file, because when the tag name is present
Emacs can find a tag faster and more accurately.  These tag names are
part of tag lines in the tags file, so we call them "explicit".

- WHY IMPLICIT TAG NAMES ARE EVEN BETTER
When a tag line has no name, but a name can be deduced from the pattern,
we say that the tag line has an implicit tag name.  Often tag names are
redundant; this happens when the name of a tag is an easily guessable
substring of the tag pattern.  We define a set of rules to decide
whether it is possible to deduce the tag name from the pattern, and make
an unnamed tag in those cases.  The name deduced from the pattern of an
unnamed tag is the implicit name of that tag.
  When the user looks for a tag, and Emacs finds no explicit tag names
that match it, Emacs then looks for an tag whose implicit tag name
matches the request.  etags.c uses implicit tag names when possible, in
order to reduce the size of the tags file.
  An implicit tag name is deduced from the pattern by discarding the
last character if it is one of ` \f\t\n\r()=,;', then taking all the
rightmost consecutive characters in the pattern which are not one of
those.

===================== end of discussion of tag names =====================

Copyright (c) 2002, 2004, 2006
Free software Foundation, Inc.

COPYING PERMISSIONS:

    This document 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 program 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 program; if not, write to the Free Software
    Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA