[Emacs-diffs] master b95f53f: * lisp/info.el: Cleanup bytepos/charpos issues

[Stefan Monnier]

branch: master
commit b95f53f0383b5df82e0b711f5027881950893f4f
Author: Stefan Monnier <address@hidden>
Commit: Stefan Monnier <address@hidden>

    * lisp/info.el: Cleanup bytepos/charpos issues
    
    * lisp/international/mule-util.el: Use lexical-binding.
    (filepos-to-bufferpos): New function.
    * lisp/info.el (Info-find-in-tag-table-1): Use 0-based file positions.
    (Info-find-node-2): Use filepos-to-bufferpos (bug#20704).
    (Info-read-subfile, Info-search): Use 0-based file positions.
---
 etc/NEWS                        |    2 +
 lisp/info.el                    |   41 +++++++++++++++++----------------
 lisp/international/mule-util.el |   47 +++++++++++++++++++++++++++++++++-----
 3 files changed, 63 insertions(+), 27 deletions(-)

diff --git a/etc/NEWS b/etc/NEWS
index 571adad..76b4772 100644
--- a/etc/NEWS
+++ b/etc/NEWS
@@ -851,6 +851,8 @@ behavior, set `diff-switches' to `-c'.
 
 * Lisp Changes in Emacs 25.1
 
+** New function `filepos-to-bufferpos'.
+
 ** The default value of `load-read-function' is now `read'.
 
 ** New hook `pre-redisplay-functions', a bit easier to use than 
pre-redisplay-function.
diff --git a/lisp/info.el b/lisp/info.el
index 9602337..413928b 100644
--- a/lisp/info.el
+++ b/lisp/info.el
@@ -1009,7 +1009,7 @@ REGEXP is a regular expression matching nodes or 
references.  Its first
 group should match `Node:' or `Ref:'.
 CASE-FOLD t means search for a case-insensitive match.
 If a match was found, value is a list (FOUND-ANCHOR POS MODE), where
-FOUND-ANCHOR is non-nil if a `Ref:' was matched, POS is the position
+FOUND-ANCHOR is non-nil if a `Ref:' was matched, POS is the file position
 where the match was found, and MODE is `major-mode' of the buffer in
 which the match was found."
   (let ((case-fold-search case-fold))
@@ -1020,7 +1020,7 @@ which the match was found."
       (beginning-of-line)
       (when (re-search-forward regexp nil t)
        (list (string-equal "Ref:" (match-string 1))
-             (+ (point-min) (read (current-buffer)))
+              (read (current-buffer))
              major-mode)))))
 
 (defun Info-find-in-tag-table (marker regexp &optional strict-case)
@@ -1029,7 +1029,7 @@ MARKER specifies the buffer and position to start 
searching at.
 REGEXP is a regular expression matching nodes or references.  Its first
 group should match `Node:' or `Ref:'.
 If a match was found, value is a list (FOUND-ANCHOR POS MODE), where
-FOUND-ANCHOR is non-nil if a `Ref:' was matched, POS is the position
+FOUND-ANCHOR is non-nil if a `Ref:' was matched, POS is the file position
 where the match was found, and MODE is `major-mode' of the buffer in
 which the match was found.
 This function tries to find a case-sensitive match first, then a
@@ -1187,15 +1187,18 @@ is non-nil)."
 
                  (when found
                    ;; FOUND is (ANCHOR POS MODE).
-                   (setq guesspos (nth 1 found))
+                   (let ((filepos (nth 1 found))) ;File position in bytes.
 
-                   ;; If this is an indirect file, determine which
-                   ;; file really holds this node and read it in.
-                   (unless (eq (nth 2 found) 'Info-mode)
-                     ;; Note that the current buffer must be the
-                     ;; *info* buffer on entry to
-                     ;; Info-read-subfile.  Thus the hackery above.
-                     (setq guesspos (Info-read-subfile guesspos)))
+                      ;; If this is an indirect file, determine which
+                      ;; file really holds this node and read it in.
+                      (unless (eq (nth 2 found) 'Info-mode)
+                        ;; Note that the current buffer must be the
+                        ;; *info* buffer on entry to
+                        ;; Info-read-subfile.  Thus the hackery above.
+                        (setq filepos (Info-read-subfile filepos)))
+
+                      (setq guesspos
+                            (filepos-to-bufferpos filepos 'approximate)))
 
                    ;; Handle anchor
                    (when (nth 0 found)
@@ -1203,8 +1206,7 @@ is non-nil)."
                      (throw 'foo t)))))
 
              ;; Else we may have a node, which we search for:
-             (goto-char (max (point-min)
-                             (- (byte-to-position guesspos) 1000)))
+             (goto-char (max (point-min) (- guesspos 1000)))
 
              ;; Now search from our advised position (or from beg of
              ;; buffer) to find the actual node.  First, check
@@ -1506,7 +1508,7 @@ is non-nil)."
 ;; Note that on entry to this function the current-buffer must be the
 ;; *info* buffer; not the info tags buffer.
 (defun Info-read-subfile (nodepos)
-  ;; NODEPOS is either a position (in the Info file as a whole,
+  ;; NODEPOS is either a position in bytes (in the Info file as a whole,
   ;; not relative to a subfile) or the name of a subfile.
   (let (lastfilepos
        lastfilename)
@@ -1523,7 +1525,7 @@ is non-nil)."
                        thisfilepos thisfilename)
                    (search-forward ": ")
                    (setq thisfilename  (buffer-substring beg (- (point) 2)))
-                   (setq thisfilepos (+ (point-min) (read (current-buffer))))
+                   (setq thisfilepos (read (current-buffer)))
                    ;; read in version 19 stops at the end of number.
                    ;; Advance to the next line.
                    (forward-line 1)
@@ -1554,7 +1556,7 @@ is non-nil)."
     ;; Don't add the length of the skipped summary segment to
     ;; the value returned to `Info-find-node-2'.  (Bug#14125)
     (if (numberp nodepos)
-       (+ (- nodepos lastfilepos) (point-min)))))
+       (- nodepos lastfilepos))))
 
 (defun Info-unescape-quotes (value)
   "Unescape double quotes and backslashes in VALUE."
@@ -2013,10 +2015,9 @@ If DIRECTION is `backward', search in the reverse 
direction."
                        (re-search-backward "\\(^.*\\): [0-9]+$")
                      (re-search-forward "\\(^.*\\): [0-9]+$"))
                    (goto-char (+ (match-end 1) 2))
-                   (setq list (cons (cons (+ (point-min)
-                                             (read (current-buffer)))
-                                          (match-string-no-properties 1))
-                                    list))
+                   (push (cons (read (current-buffer))
+                                (match-string-no-properties 1))
+                          list)
                    (goto-char (if backward
                                    (1- (match-beginning 0))
                                  (1+ (match-end 0)))))
diff --git a/lisp/international/mule-util.el b/lisp/international/mule-util.el
index eae787b..92c540c 100644
--- a/lisp/international/mule-util.el
+++ b/lisp/international/mule-util.el
@@ -1,4 +1,4 @@
-;;; mule-util.el --- utility functions for multilingual environment (mule)
+;;; mule-util.el --- utility functions for multilingual environment (mule)  
-*- lexical-binding:t -*-
 
 ;; Copyright (C) 1997-1998, 2000-2015 Free Software Foundation, Inc.
 ;; Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
@@ -30,8 +30,7 @@
 
 ;;; Code:
 
-;;; String manipulations while paying attention to multibyte
-;;; characters.
+;;; String manipulations while paying attention to multibyte characters.
 
 ;;;###autoload
 (defsubst string-to-list (string)
@@ -49,7 +48,6 @@
   (if (integerp obj)
       (aset string idx obj)
     (let ((len1 (length obj))
-         (len2 (length string))
          (i 0))
       (while (< i len1)
        (aset string (+ idx i) (aref obj i))
@@ -90,7 +88,6 @@ defaults to `truncate-string-ellipsis'."
     (setq ellipsis truncate-string-ellipsis))
   (let ((str-len (length str))
        (str-width (string-width str))
-       (ellipsis-len (if ellipsis (length ellipsis) 0))
        (ellipsis-width (if ellipsis (string-width ellipsis) 0))
        (idx 0)
        (column 0)
@@ -129,8 +126,8 @@ defaults to `truncate-string-ellipsis'."
              tail-padding ellipsis))))
 
 
-;;; Nested alist handler.  Nested alist is alist whose elements are
-;;; also nested alist.
+;;; Nested alist handler.
+;; Nested alist is alist whose elements are also nested alist.
 
 ;;;###autoload
 (defsubst nested-alist-p (obj)
@@ -313,6 +310,42 @@ per-character basis, this may not be accurate."
                                  (throw 'tag3 charset)))
                          charset-list)
                    nil)))))))))
+
+;;;###autoload
+(defun filepos-to-bufferpos (byte &optional quality coding-system)
+  "Try to return the buffer position corresponding to a particular file 
position.
+The file position is given as a (0-based) BYTE count.
+The function presumes the file is encoded with CODING-SYSTEM, which defaults
+to `buffer-file-coding-system'.
+QUALITY can be:
+  `approximate', in which case we may cut some corners to avoid
+    excessive work.
+  nil, in which case we may return nil rather than an approximation."
+  ;; `exact', in which case we may end up re-(en|de)coding a large
+  ;;   part of the file.
+  (unless coding-system (setq coding-system buffer-file-coding-system))
+  (let ((eol (coding-system-eol-type coding-system))
+        (type (coding-system-type coding-system))
+        (pm (save-restriction (widen) (point-min))))
+    (pcase type
+      (`utf-8
+       (when (coding-system-get coding-system :bom)
+         (setq byte (max 0 (- byte 3))))
+       (let (pos lines (eol-offset 0))
+         (while
+             (progn
+               (setq pos (byte-to-position (+ pm byte (- eol-offset))))
+               (setq lines (1- (line-number-at-pos pos)))
+               (not (= lines eol-offset)))
+           (setq eol-offset (+ eol-offset lines)))
+         pos))
+      ;; FIXME: What if it's a 2-byte charset?  Are there such beasts?
+      (`charset (+ pm byte))
+      (_
+       (pcase quality
+         (`approximate (+ pm (byte-to-position byte)))
+         ;; (`exact ...)
+         )))))
 
 (provide 'mule-util)