(defgroup sort nil
"Commands to sort text in an Emacs buffer."
:group 'data)
(defcustom sort-fold-case nil
"*Non-nil if the buffer sort functions should ignore case."
:group 'sort
:type 'boolean)
(defun sort-subr (reverse nextrecfun endrecfun &optional startkeyfun endkeyfun)
"General text sorting routine to divide buffer into records and sort them.
Arguments are REVERSE NEXTRECFUN ENDRECFUN &optional STARTKEYFUN ENDKEYFUN.
We divide the accessible portion of the buffer into disjoint pieces
called sort records. A portion of each sort record (perhaps all of
it) is designated as the sort key. The records are rearranged in the
buffer in order by their sort keys. The records may or may not be
contiguous.
Usually the records are rearranged in order of ascending sort key.
If REVERSE is non-nil, they are rearranged in order of descending sort key.
The variable `sort-fold-case' determines whether alphabetic case affects
the sort order.
The next four arguments are functions to be called to move point
across a sort record. They will be called many times from within sort-subr.
NEXTRECFUN is called with point at the end of the previous record.
It moves point to the start of the next record.
It should move point to the end of the buffer if there are no more records.
The first record is assumed to start at the position of point when sort-subr
is called.
ENDRECFUN is called with point within the record.
It should move point to the end of the record.
STARTKEYFUN moves from the start of the record to the start of the key.
It may return either a non-nil value to be used as the key, or
else the key is the substring between the values of point after
STARTKEYFUN and ENDKEYFUN are called. If STARTKEYFUN is nil, the key
starts at the beginning of the record.
ENDKEYFUN moves from the start of the sort key to the end of the sort key.
ENDKEYFUN may be nil if STARTKEYFUN returns a value or if it would be the
same as ENDRECFUN."
(let ((messages (> (- (point-max) (point-min)) 50000)))
(save-excursion
(if messages (message "Finding sort keys..."))
(let* ((sort-lists (sort-build-lists nextrecfun endrecfun
startkeyfun endkeyfun))
(old (reverse sort-lists))
(case-fold-search sort-fold-case))
(if (null sort-lists)
()
(or reverse (setq sort-lists (nreverse sort-lists)))
(if messages (message "Sorting records..."))
(setq sort-lists
(if (fboundp 'sortcar)
(sortcar sort-lists
(cond ((numberp (car (car sort-lists)))
'<)
((consp (car (car sort-lists)))
(function
(lambda (a b)
(> 0 (compare-buffer-substrings
nil (car a) (cdr a)
nil (car b) (cdr b))))))
(t
'string<)))
(sort sort-lists
(cond ((numberp (car (car sort-lists)))
'car-less-than-car)
((consp (car (car sort-lists)))
(function
(lambda (a b)
(> 0 (compare-buffer-substrings
nil (car (car a)) (cdr (car a))
nil (car (car b)) (cdr (car b)))))))
(t
(function
(lambda (a b)
(string< (car a) (car b)))))))))
(if reverse (setq sort-lists (nreverse sort-lists)))
(if messages (message "Reordering buffer..."))
(sort-reorder-buffer sort-lists old)))
(if messages (message "Reordering buffer... Done"))))
nil)
(defun sort-build-lists (nextrecfun endrecfun startkeyfun endkeyfun)
(let ((sort-lists ())
(start-rec nil)
done key)
(while (not (eobp))
(setq start-rec (point)) (setq done nil)
(setq key (catch 'key
(or (and startkeyfun (funcall startkeyfun))
(let ((start (point)))
(funcall (or endkeyfun
(prog1 endrecfun (setq done t))))
(cons start (point))))))
(cond ((prog1 done (setq done nil)))
(endrecfun (funcall endrecfun))
(nextrecfun (funcall nextrecfun) (setq done t)))
(if key (setq sort-lists (cons
(if (and (consp key)
(equal (car key) start-rec)
(equal (cdr key) (point)))
(cons key key)
(cons key (cons start-rec (point))))
sort-lists)))
(and (not done) nextrecfun (funcall nextrecfun)))
sort-lists))
(defun sort-reorder-buffer (sort-lists old)
(let ((inhibit-quit t)
(last (point-min))
(min (point-min)) (max (point-max)))
(goto-char (point-max))
(insert-before-markers " ")
(narrow-to-region min (1- (point-max)))
(while sort-lists
(goto-char (point-max))
(insert-buffer-substring (current-buffer)
last
(nth 1 (car old)))
(goto-char (point-max))
(insert-buffer-substring (current-buffer)
(nth 1 (car sort-lists))
(cdr (cdr (car sort-lists))))
(setq last (cdr (cdr (car old)))
sort-lists (cdr sort-lists)
old (cdr old)))
(goto-char (point-max))
(insert-buffer-substring (current-buffer)
last
max)
(delete-region min max)
(goto-char (point-max))
(narrow-to-region min (1+ (point)))
(delete-region (point) (1+ (point)))))
(defun sort-lines (reverse beg end)
"Sort lines in region alphabetically; argument means descending order.
Called from a program, there are three arguments:
REVERSE (non-nil means reverse order), BEG and END (region to sort).
The variable `sort-fold-case' determines whether alphabetic case affects
the sort order."
(interactive "P\nr")
(save-excursion
(save-restriction
(narrow-to-region beg end)
(goto-char (point-min))
(sort-subr reverse 'forward-line 'end-of-line))))
(defun sort-paragraphs (reverse beg end)
"Sort paragraphs in region alphabetically; argument means descending order.
Called from a program, there are three arguments:
REVERSE (non-nil means reverse order), BEG and END (region to sort).
The variable `sort-fold-case' determines whether alphabetic case affects
the sort order."
(interactive "P\nr")
(save-excursion
(save-restriction
(narrow-to-region beg end)
(goto-char (point-min))
(sort-subr reverse
(function
(lambda ()
(while (and (not (eobp)) (looking-at paragraph-separate))
(forward-line 1))))
'forward-paragraph))))
(defun sort-pages (reverse beg end)
"Sort pages in region alphabetically; argument means descending order.
Called from a program, there are three arguments:
REVERSE (non-nil means reverse order), BEG and END (region to sort).
The variable `sort-fold-case' determines whether alphabetic case affects
the sort order."
(interactive "P\nr")
(save-excursion
(save-restriction
(narrow-to-region beg end)
(goto-char (point-min))
(sort-subr reverse
(function (lambda () (skip-chars-forward "\n")))
'forward-page))))
(defvar sort-fields-syntax-table nil)
(if sort-fields-syntax-table nil
(let ((table (make-syntax-table))
(i 0))
(while (< i 256)
(modify-syntax-entry i "w" table)
(setq i (1+ i)))
(modify-syntax-entry ?\ " " table)
(modify-syntax-entry ?\t " " table)
(modify-syntax-entry ?\n " " table)
(modify-syntax-entry ?\. "_" table) (setq sort-fields-syntax-table table)))
(defcustom sort-numeric-base 10
"*The default base used by `sort-numeric-fields'."
:group 'sort
:type 'integer)
(defun sort-numeric-fields (field beg end)
"Sort lines in region numerically by the ARGth field of each line.
Fields are separated by whitespace and numbered from 1 up.
Specified field must contain a number in each line of the region,
which may begin with \"0x\" or \"0\" for hexadecimal and octal values.
Otherwise, the number is interpreted according to sort-numeric-base.
With a negative arg, sorts by the ARGth field counted from the right.
Called from a program, there are three arguments:
FIELD, BEG and END. BEG and END specify region to sort."
(interactive "p\nr")
(sort-fields-1 field beg end
(lambda ()
(sort-skip-fields field)
(let* ((case-fold-search t)
(base
(if (looking-at "\\(0x\\)[0-9a-f]\\|\\(0\\)[0-7]")
(cond ((match-beginning 1)
(goto-char (match-end 1))
16)
((match-beginning 2)
(goto-char (match-end 2))
8)
(t nil)))))
(string-to-number (buffer-substring (point)
(save-excursion
(forward-sexp 1)
(point)))
(or base sort-numeric-base))))
nil))
(defun sort-fields (field beg end)
"Sort lines in region lexicographically by the ARGth field of each line.
Fields are separated by whitespace and numbered from 1 up.
With a negative arg, sorts by the ARGth field counted from the right.
Called from a program, there are three arguments:
FIELD, BEG and END. BEG and END specify region to sort.
The variable `sort-fold-case' determines whether alphabetic case affects
the sort order."
(interactive "p\nr")
(sort-fields-1 field beg end
(function (lambda ()
(sort-skip-fields field)
nil))
(function (lambda () (skip-chars-forward "^ \t\n")))))
(defun sort-fields-1 (field beg end startkeyfun endkeyfun)
(let ((tbl (syntax-table)))
(if (zerop field) (setq field 1))
(unwind-protect
(save-excursion
(save-restriction
(narrow-to-region beg end)
(goto-char (point-min))
(set-syntax-table sort-fields-syntax-table)
(sort-subr nil
'forward-line 'end-of-line
startkeyfun endkeyfun)))
(set-syntax-table tbl))))
(defun sort-skip-fields (n)
(if (> n 0)
(let ((i (1- n)))
(while (> i 0)
(skip-chars-forward " \t")
(skip-chars-forward "^ \t\n")
(setq i (1- i)))
(skip-chars-forward " \t")
(if (eolp)
(error "Line has too few fields: %s"
(buffer-substring
(save-excursion (beginning-of-line) (point))
(save-excursion (end-of-line) (point))))))
(end-of-line)
(let ((i (1- (- n))))
(while (> i 0)
(skip-chars-backward " \t")
(skip-chars-backward "^ \t\n")
(setq i (1- i)))
(skip-chars-backward " \t"))
(if (bolp)
(error "Line has too few fields: %s"
(buffer-substring
(save-excursion (beginning-of-line) (point))
(save-excursion (end-of-line) (point)))))
(skip-chars-backward "^ \t\n")))
(defvar sort-regexp-fields-regexp)
(defvar sort-regexp-record-end)
(defun sort-regexp-fields-next-record ()
(let ((oldpos (point)))
(and (re-search-forward sort-regexp-fields-regexp nil 'move)
(setq sort-regexp-record-end (match-end 0))
(if (= sort-regexp-record-end oldpos)
(progn
(forward-char 1)
(re-search-forward sort-regexp-fields-regexp nil 'move)
(setq sort-regexp-record-end (match-end 0)))
t)
(goto-char (match-beginning 0)))))
(defun sort-regexp-fields (reverse record-regexp key-regexp beg end)
"Sort the region lexicographically as specified by RECORD-REGEXP and KEY.
RECORD-REGEXP specifies the textual units which should be sorted.
For example, to sort lines RECORD-REGEXP would be \"^.*$\"
KEY specifies the part of each record (ie each match for RECORD-REGEXP)
is to be used for sorting.
If it is \"\\\\digit\" then the digit'th \"\\\\(...\\\\)\" match field from
RECORD-REGEXP is used.
If it is \"\\\\&\" then the whole record is used.
Otherwise, it is a regular-expression for which to search within the record.
If a match for KEY is not found within a record then that record is ignored.
With a negative prefix arg sorts in reverse order.
The variable `sort-fold-case' determines whether alphabetic case affects
the sort order.
For example: to sort lines in the region by the first word on each line
starting with the letter \"f\",
RECORD-REGEXP would be \"^.*$\" and KEY would be \"\\\\=\\<f\\\\w*\\\\>\""
(interactive "P\nsRegexp specifying records to sort:
sRegexp specifying key within record: \nr")
(cond ((or (equal key-regexp "") (equal key-regexp "\\&"))
(setq key-regexp 0))
((string-match "\\`\\\\[1-9]\\'" key-regexp)
(setq key-regexp (- (aref key-regexp 1) ?0))))
(save-excursion
(save-restriction
(narrow-to-region beg end)
(goto-char (point-min))
(let (sort-regexp-record-end
(sort-regexp-fields-regexp record-regexp))
(re-search-forward sort-regexp-fields-regexp)
(setq sort-regexp-record-end (point))
(goto-char (match-beginning 0))
(sort-subr reverse
'sort-regexp-fields-next-record
(function (lambda ()
(goto-char sort-regexp-record-end)))
(function (lambda ()
(let ((n 0))
(cond ((numberp key-regexp)
(setq n key-regexp))
((re-search-forward
key-regexp sort-regexp-record-end t)
(setq n 0))
(t (throw 'key nil)))
(condition-case ()
(cons (match-beginning n)
(match-end n))
(error (throw 'key nil)))))))))))
(defvar sort-columns-subprocess t)
(defun sort-columns (reverse &optional beg end)
"Sort lines in region alphabetically by a certain range of columns.
For the purpose of this command, the region BEG...END includes
the entire line that point is in and the entire line the mark is in.
The column positions of point and mark bound the range of columns to sort on.
A prefix argument means sort into REVERSE order.
The variable `sort-fold-case' determines whether alphabetic case affects
the sort order.
Note that `sort-columns' rejects text that contains tabs,
because tabs could be split across the specified columns
and it doesn't know how to handle that. Also, when possible,
it uses the `sort' utility program, which doesn't understand tabs.
Use \\[untabify] to convert tabs to spaces before sorting."
(interactive "P\nr")
(save-excursion
(let (beg1 end1 col-beg1 col-end1 col-start col-end)
(goto-char (min beg end))
(setq col-beg1 (current-column))
(beginning-of-line)
(setq beg1 (point))
(goto-char (max beg end))
(setq col-end1 (current-column))
(forward-line)
(setq end1 (point))
(setq col-start (min col-beg1 col-end1))
(setq col-end (max col-beg1 col-end1))
(if (search-backward "\t" beg1 t)
(error "sort-columns does not work with tabs -- use M-x untabify"))
(if (not (or (eq system-type 'vax-vms)
(text-properties-at beg1)
(< (next-property-change beg1 nil end1) end1)))
(let ((sort-args (list (if reverse "-rt\n" "-t\n")
(concat "+0." (int-to-string col-start))
(concat "-0." (int-to-string col-end)))))
(when sort-fold-case
(push "-f" sort-args))
(apply #'call-process-region beg1 end1 "sort" t t nil sort-args))
(save-excursion
(save-restriction
(narrow-to-region beg1 end1)
(goto-char beg1)
(sort-subr reverse 'forward-line 'end-of-line
#'(lambda () (move-to-column col-start) nil)
#'(lambda () (move-to-column col-end) nil))))))))
(defun reverse-region (beg end)
"Reverse the order of lines in a region.
From a program takes two point or marker arguments, BEG and END."
(interactive "r")
(if (> beg end)
(let (mid) (setq mid end end beg beg mid)))
(save-excursion
(goto-char beg)
(or (bolp) (forward-line 1))
(setq beg (point))
(goto-char end)
(or (and (eolp) (not (bolp))) (progn (forward-line -1) (end-of-line)))
(setq end (point-marker))
(let (ll (do t))
(while do
(goto-char beg)
(setq ll (cons (buffer-substring (point) (progn (end-of-line) (point)))
ll))
(setq do (/= (point) end))
(delete-region beg (if do (1+ (point)) (point))))
(while (cdr ll)
(insert (car ll) "\n")
(setq ll (cdr ll)))
(insert (car ll)))))
(provide 'sort)