# Improved XML highlighter by: Brian Candler <B.Candler@pobox.com>

# Define no. sync lines
# You can say:
# -200     means 200 lines
# -        means always start parsing from beginning of file when we lose sync
#          if nothing is specified, the default is -50

-

# Define colors
#
# bold inverse blink dim underline italic
# white cyan magenta blue yellow green red black
# bg_white bg_cyan bg_magenta bg_blue bg_yellow bg_green bg_red bg_black

# The underlines are here right now just because I want to distinguish which
# bits have been coloured (say) CdataStart, CdataBody, CdataEnd. And that's
# because I think it may be useful to make that distinction for some people.

=Idle
=Error		red bold
=Tag		blue
=Attr		cyan
=Constant	blue bold
=EntityRef	magenta
=Decl		cyan
=CommentStart	green
=CommentBody	green
=CommentEnd	green
=PIStart	magenta bold
=PIBody		magenta
=PIEnd		magenta bold
=CdataStart	blue bold
=CdataBody	bold
=CdataEnd	blue bold

# http://www.w3.org/TR/2004/REC-xml-20040204/
#
# NOTE: For UNICODE compatibility, the ranges
#	"A-Za-z_:"		-- first character of Name
#	"A-Za-z0-9._:-"		-- subsequent characters of Name
# ought to be replaced with some appropriate Unicode character classes

:content Idle
	*		content
	"<"		tag		recolor=-1
	"&"		entityref	recolor=-1

# > is allowed
#	">"		error		noeat recolor=-1


# In several contexts a space is an error, and since a coloured space is
# not visible, we colour the next non-space character as well.

:error Error
	*		error_visible	noeat recolor=-1
	" \t\r\n"	error

:error_visible Error
	*		content

# Matched: &

:entityref EntityRef
	*		error		noeat recolor=-1
	"A-Za-z0-9._:-"	entityref
	"#"		entityref
	";"		content

# Matched: <

:tag Tag
	*		error		noeat recolor=-1
	"/"		end_tag
	"A-Za-z_:"	start_or_empty_tag
	"?"		pi_start	recolor=-2
	"!"		decl		recolor=-2 buffer

# Matched: </

:end_tag Tag
	*		error		noeat recolor=-1
	"A-Za-z_:"	end_tag2

:end_tag2 Tag
	*		error		noeat recolor=-1
	"A-Za-z0-9._:-"	end_tag2
	" \t\r\n"	end_tag2_sp	recolor=-1
	">"		content

:end_tag2_sp Idle
	*		end_tag3	noeat

:end_tag3 Tag
	*		error		noeat recolor=-1
	" \t\r\n"	end_tag3_sp	recolor=-1
	">"		content

:end_tag3_sp Idle
	*		end_tag_3	noeat

# Matched: <tag

:start_or_empty_tag Tag
	*		error		noeat recolor=-1
	"A-Za-z0-9._:-"	start_or_empty_tag
	"/"		empty_tag
	" \t\r\n"	start_or_empty_tag_sp	recolor=-1
	">"		content

:start_or_empty_tag_sp Idle
	*		tag_space	noeat

# Matched: <tag/

:empty_tag Tag
	*		error		noeat recolor=-1
	">"		content

# Matched: <tag SPACE

:tag_space Attr
	*		error		noeat recolor=-1
	" \t\r\n"	tag_space_sp	recolor=-1
	"A-Za-z_:"	attr
	"/"		empty_tag	recolor=-1
	">"		close_tag	recolor=-1

:tag_space_sp Idle
	*		tag_space	noeat

# Matched: <tag attr

:attr Attr
	*		error		noeat recolor=-1
	"A-Za-z0-9._:-" attr
	" \t\r\n"	attr_sp		recolor=-1
	"="		string

:attr_sp Idle
	*		attr_before	noeat

:attr_before Attr
	*		error		noeat recolor=-1
	" \t\r\n"	attr_sp		recolor=-1
	"="		string

# Matched: <tag attr=

:string Attr
	*		error		noeat recolor=-1
	" \t\r\n"	string_sp	recolor=-1
	"\""		astring		recolor=-1
	"'"		bstring		recolor=-1

:string_sp Idle
	*		string		noeat

# Matched: <tag attr="

:astring Constant
	*		astring
	"<"		error		noeat recolor=-1
	"&"		achar		recolor=-1
	"\""		endstring

#	">"		error		noeat recolor=-1


:bstring Constant
	*		bstring
	"<"		error		noeat recolor=-1
	"&"		bchar		recolor=-1
	"'"		endstring

#	">"		error		noeat recolor=-1


:achar EntityRef
	*		error		noeat recolor=-1
	"A-Za-z0-9._:-"	achar
	"#"		achar
	";"		astring

:bchar EntityRef
	*		error		noeat recolor=-1
	"A-Za-z0-9._:-"	bchar
	"#"		bchar
	";"		bstring

# Matched: <tag attr="foo"

:endstring Attr
	*		error		noeat recolor=-1
	" \t\r\n"	tag_space_sp
	"/"		empty_tag	recolor=-1
	">"		close_tag	recolor=-1

# This state is just to recolor the final ">" at the end of <tag attr="val">

:close_tag Tag
	*		content		noeat

# Matched: <?

:pi_start PIStart
	*		pi		noeat recolor=-1

:pi PIBody
	*		pi
	"?"		pi2

:pi2 PIBody
	*		pi
	">"		pi_end		recolor=-2

:pi_end PIEnd
	*		content		noeat recolor=-1

# Matched: <!

:decl Decl
	*		decl		strings
	"!--"		comment_start	recolor=-5
	"![CDATA["	cdata_start	recolor=-10
done
#	"<"		decl_nest
	">"		content

# We allow one level of <...> nesting within declarations

:decl_nest Decl
	*		decl_nest
	">"		decl

# Matched: <!--

:comment_start CommentStart
	*		comment		noeat

:comment CommentBody
	*		comment
	"-"		comment2

:comment2 CommentBody
	*		comment
	"-"		comment3

:comment3 CommentBody
	*		comment_error	noeat recolor=-3
	">"		comment_end	recolor=-3

:comment_end CommentEnd
	*		content		noeat recolor=-1

# For compatibility, the string "--" (double-hyphen) MUST NOT occur within
# comments. [http://www.w3.org/TR/REC-xml/ section 2.5]

:comment_error Error
	*		comment
	"-"		comment_error
	">"		comment_end	recolor=-3

# Matched: <![CDATA[

:cdata_start CdataStart
	*		cdata		noeat

:cdata CdataBody
	*		cdata
	"]"		cdata2

:cdata2 CdataBody
	*		cdata
	"]"		cdata3

:cdata3 CdataBody
	*		cdata
	">"		cdata_end	recolor=-3

:cdata_end CdataEnd
	*		content		noeat recolor=-1