(defmacro xml-node-name (node)
"Return the tag associated with NODE.
The tag is a lower-case symbol."
(list 'car node))
(defmacro xml-node-attributes (node)
"Return the list of attributes of NODE.
The list can be nil."
(list 'nth 1 node))
(defmacro xml-node-children (node)
"Return the list of children of NODE.
This is a list of nodes, and it can be nil."
(list 'cddr node))
(defun xml-get-children (node child-name)
"Return the children of NODE whose tag is CHILD-NAME.
CHILD-NAME should be a lower case symbol."
(let ((children (xml-node-children node))
match)
(while children
(if (car children)
(if (equal (xml-node-name (car children)) child-name)
(set 'match (append match (list (car children))))))
(set 'children (cdr children)))
match))
(defun xml-get-attribute (node attribute)
"Get from NODE the value of ATTRIBUTE.
An empty string is returned if the attribute was not found."
(if (xml-node-attributes node)
(let ((value (assoc attribute (xml-node-attributes node))))
(if value
(cdr value)
""))
""))
(defun xml-parse-file (file &optional parse-dtd)
"Parse the well-formed XML FILE.
If FILE is already edited, this will keep the buffer alive.
Returns the top node with all its children.
If PARSE-DTD is non-nil, the DTD is parsed rather than skipped."
(let ((keep))
(if (get-file-buffer file)
(progn
(set-buffer (get-file-buffer file))
(setq keep (point)))
(find-file file))
(let ((xml (xml-parse-region (point-min)
(point-max)
(current-buffer)
parse-dtd)))
(if keep
(goto-char keep)
(kill-buffer (current-buffer)))
xml)))
(defun xml-parse-region (beg end &optional buffer parse-dtd)
"Parse the region from BEG to END in BUFFER.
If BUFFER is nil, it defaults to the current buffer.
Returns the XML list for the region, or raises an error if the region
is not a well-formed XML file.
If PARSE-DTD is non-nil, the DTD is parsed rather than skipped,
and returned as the first element of the list"
(let (xml result dtd)
(save-excursion
(if buffer
(set-buffer buffer))
(goto-char beg)
(while (< (point) end)
(if (search-forward "<" end t)
(progn
(forward-char -1)
(if (null xml)
(progn
(set 'result (xml-parse-tag end parse-dtd))
(cond
((listp (car result))
(set 'dtd (car result))
(add-to-list 'xml (cdr result)))
(t
(add-to-list 'xml result))))
(error "XML files can have only one toplevel tag")))
(goto-char end)))
(if parse-dtd
(cons dtd (reverse xml))
(reverse xml)))))
(defun xml-parse-tag (end &optional parse-dtd)
"Parse the tag that is just in front of point.
The end tag must be found before the position END in the current buffer.
If PARSE-DTD is non-nil, the DTD of the document, if any, is parsed and
returned as the first element in the list.
Returns one of:
- a list : the matching node
- nil : the point is not looking at a tag.
- a cons cell: the first element is the DTD, the second is the node"
(cond
((looking-at "<\\?")
(search-forward "?>" end)
(skip-chars-forward " \t\n")
(xml-parse-tag end))
((looking-at "<!\\[CDATA\\[")
(let ((pos (match-end 0)))
(unless (search-forward "]]>" end t)
(error "CDATA section does not end anywhere in the document"))
(buffer-substring-no-properties pos (match-beginning 0))))
((looking-at "<!DOCTYPE")
(let (dtd)
(if parse-dtd
(set 'dtd (xml-parse-dtd end))
(xml-skip-dtd end))
(skip-chars-forward " \t\n")
(if dtd
(cons dtd (xml-parse-tag end))
(xml-parse-tag end))))
((looking-at "<!--")
(search-forward "-->" end)
(skip-chars-forward " \t\n")
(xml-parse-tag end))
((looking-at "</")
'())
((looking-at "<\\([^/> \t\n]+\\)")
(let* ((node-name (match-string 1))
(children (list (intern node-name)))
(case-fold-search nil) pos)
(goto-char (match-end 1))
(set 'children (append children (list (xml-parse-attlist end))))
(if (looking-at "/>")
(progn
(forward-char 2)
(skip-chars-forward " \t\n")
(append children '("")))
(if (eq (char-after) ?>)
(progn
(forward-char 1)
(skip-chars-forward " \t\n")
(while (not (looking-at (concat "</" node-name "[ \t\n]*>")))
(cond
((looking-at "</")
(error (concat
"XML: invalid syntax -- invalid end tag (expecting "
node-name
") at pos " (number-to-string (point)))))
((= (char-after) ?<)
(set 'children (append children (list (xml-parse-tag end)))))
(t
(set 'pos (point))
(search-forward "<" end)
(forward-char -1)
(let ((string (buffer-substring-no-properties pos (point)))
(pos 0))
(set 'children (append children
(list (xml-substitute-special string))))))))
(goto-char (match-end 0))
(skip-chars-forward " \t\n")
(if (> (point) end)
(error "XML: End tag for %s not found before end of region"
node-name))
children
)
(error "XML: Invalid attribute list")
))))
(t (error "XML: Invalid character"))
))
(defun xml-parse-attlist (end)
"Return the attribute-list that point is looking at.
The search for attributes end at the position END in the current buffer.
Leaves the point on the first non-blank character after the tag."
(let ((attlist '())
name)
(skip-chars-forward " \t\n")
(while (looking-at "\\([a-zA-Z_:][-a-zA-Z0-9._:]*\\)[ \t\n]*=[ \t\n]*")
(set 'name (intern (match-string 1)))
(goto-char (match-end 0))
(unless (looking-at "\"\\([^\"]+\\)\"")
(unless (looking-at "'\\([^']+\\)'")
(error "XML: Attribute values must be given between quotes")))
(if (assoc name attlist)
(error "XML: each attribute must be unique within an element"))
(set 'attlist (append attlist
(list (cons name (match-string-no-properties 1)))))
(goto-char (match-end 0))
(skip-chars-forward " \t\n")
(if (> (point) end)
(error "XML: end of attribute list not found before end of region"))
)
attlist
))
(defun xml-skip-dtd (end)
"Skip the DTD that point is looking at.
The DTD must end before the position END in the current buffer.
The point must be just before the starting tag of the DTD.
This follows the rule [28] in the XML specifications."
(forward-char (length "<!DOCTYPE"))
(if (looking-at "[ \t\n]*>")
(error "XML: invalid DTD (excepting name of the document)"))
(condition-case nil
(progn
(forward-word 1) (skip-chars-forward " \t\n")
(if (looking-at "\\[")
(re-search-forward "\\][ \t\n]*>" end)
(search-forward ">" end)))
(error (error "XML: No end to the DTD"))))
(defun xml-parse-dtd (end)
"Parse the DTD that point is looking at.
The DTD must end before the position END in the current buffer."
(let (dtd type element end-pos)
(forward-char (length "<!DOCTYPE"))
(skip-chars-forward " \t\n")
(if (looking-at ">")
(error "XML: invalid DTD (excepting name of the document)"))
(looking-at "\\sw+")
(set 'dtd (list 'dtd (match-string-no-properties 0)))
(goto-char (match-end 0))
(skip-chars-forward " \t\n")
(if (looking-at "SYSTEM")
(error "XML: Don't know how to handle external DTDs"))
(if (not (= (char-after) ?\[))
(error "XML: Unknown declaration in the DTD"))
(forward-char 1)
(while (and (not (looking-at "[ \t\n]*\\]"))
(<= (point) end))
(cond
((looking-at
"[\t \n]*<!ELEMENT[ \t\n]+\\([a-zA-Z0-9.%;]+\\)[ \t\n]+\\([^>]+\\)>")
(setq element (intern (match-string-no-properties 1))
type (match-string-no-properties 2))
(set 'end-pos (match-end 0))
(cond
((string-match "^EMPTY[ \t\n]*$" type) (set 'type 'empty))
((string-match "^ANY[ \t\n]*$" type) (set 'type 'any))
((string-match "^(\\(.*\\))[ \t\n]*$" type) (set 'type (xml-parse-elem-type (match-string-no-properties 1 type))))
((string-match "^%[^;]+;[ \t\n]*$" type) nil)
(t
(error "XML: Invalid element type in the DTD")))
(if (assoc element dtd)
(error "XML: elements declaration must be unique in a DTD (<%s>)"
(symbol-name element)))
(set 'dtd (append dtd (list (list element type))))
(goto-char end-pos)
)
(t
(error "XML: Invalid DTD item"))
)
)
(search-forward ">" end)
dtd
))
(defun xml-parse-elem-type (string)
"Convert a STRING for an element type into an elisp structure."
(let (elem modifier)
(if (string-match "(\\([^)]+\\))\\([+*?]?\\)" string)
(progn
(setq elem (match-string 1 string)
modifier (match-string 2 string))
(if (string-match "|" elem)
(set 'elem (append '(choice)
(mapcar 'xml-parse-elem-type
(split-string elem "|"))))
(if (string-match "," elem)
(set 'elem (append '(seq)
(mapcar 'xml-parse-elem-type
(split-string elem ","))))
)))
(if (string-match "[ \t\n]*\\([^+*?]+\\)\\([+*?]?\\)" string)
(setq elem (match-string 1 string)
modifier (match-string 2 string))))
(if (and (stringp elem)
(string= elem "#PCDATA"))
(set 'elem 'pcdata))
(cond
((string= modifier "+")
(list '+ elem))
((string= modifier "*")
(list '* elem))
((string= modifier "?")
(list '? elem))
(t
elem))))
(defun xml-substitute-special (string)
"Return STRING, after subsituting special XML sequences."
(while (string-match "&" string)
(set 'string (replace-match "&" t nil string)))
(while (string-match "<" string)
(set 'string (replace-match "<" t nil string)))
(while (string-match ">" string)
(set 'string (replace-match ">" t nil string)))
(while (string-match "'" string)
(set 'string (replace-match "'" t nil string)))
(while (string-match """ string)
(set 'string (replace-match "\"" t nil string)))
string)
(defun xml-debug-print (xml)
(while xml
(xml-debug-print-internal (car xml) "")
(set 'xml (cdr xml)))
)
(defun xml-debug-print-internal (xml &optional indent-string)
"Outputs the XML tree in the current buffer.
The first line indented with INDENT-STRING."
(let ((tree xml)
attlist)
(unless indent-string
(set 'indent-string ""))
(insert indent-string "<" (symbol-name (xml-node-name tree)))
(set 'attlist (xml-node-attributes tree))
(while attlist
(insert " ")
(insert (symbol-name (caar attlist)) "=\"" (cdar attlist) "\"")
(set 'attlist (cdr attlist)))
(insert ">")
(set 'tree (xml-node-children tree))
(while tree
(cond
((listp (car tree))
(insert "\n")
(xml-debug-print-internal (car tree) (concat indent-string " "))
)
((stringp (car tree))
(insert (car tree))
)
(t
(error "Invalid XML tree")))
(set 'tree (cdr tree))
)
(insert "\n" indent-string
"</" (symbol-name (xml-node-name xml)) ">")
))
(provide 'xml)