beta 2011.09.27 20:05

55 files changed, 783 insertions, 283 deletions

diff --git a/context/data/scite/lexers/scite-context-lexer-lua.lua b/context/data/scite/lexers/scite-context-lexer-lua.lua
index 62577d4a9..1b55be55f 100644
--- a/context/data/scite/lexers/scite-context-lexer-lua.lua
+++ b/context/data/scite/lexers/scite-context-lexer-lua.lua
@@ -100,10 +100,10 @@ local shortcomment  = token("comment", dashes * lexer.nonnewline^0)
 local longcomment   = token("comment", dashes * longcomment)
 
 local shortstring   = token("quote",  dquote)
-                    * token("string", (escaped + (1-dquote))^0 )
+                    * token("string", (escaped + (1-dquote))^0)
                     * token("quote",  dquote)
                     + token("quote",  squote)
-                    * token("string", (escaped + (1-squote))^0 )
+                    * token("string", (escaped + (1-squote))^0)
                     * token("quote",  squote)
 
 local longstring    = token("quote",  longonestart)
diff --git a/context/data/scite/lexers/scite-context-lexer-mps.lua b/context/data/scite/lexers/scite-context-lexer-mps.lua
index afde63bcc..d374f7f82 100644
--- a/context/data/scite/lexers/scite-context-lexer-mps.lua
+++ b/context/data/scite/lexers/scite-context-lexer-mps.lua
@@ -8,13 +8,15 @@ local info = {
 
 local lexer = lexer
 local global, string, table, lpeg = _G, string, table, lpeg
-local token, style, colors, exact_match, no_style = lexer.token, lexer.style, lexer.colors, lexer.exact_match, lexer.style_nothing
+local token, exact_match = lexer.token, lexer.exact_match
 local P, R, S, V, C, Cmt = lpeg.P, lpeg.R, lpeg.S, lpeg.V, lpeg.C, lpeg.Cmt
-local type, next, pcall, loadfile = type, next, pcall, loadfile
+local type = type
 
 module(...)
 
-local metafunlexer = _M
+local metafunlexer      = _M
+
+local context           = lexer.context
 
 local metafunhelpers    = { }
 local metafunconstants  = { }
@@ -23,14 +25,14 @@ local primitivecommands = { }
 
 do
 
-    local definitions = lexer.context.loaddefinitions("scite-context-data-metapost")
+    local definitions = context.loaddefinitions("scite-context-data-metapost")
 
     if definitions then
         plaincommands     = definitions.plain      or { }
         primitivecommands = definitions.primitives or { }
     end
 
-    local definitions = lexer.context.loaddefinitions("scite-context-data-metafun")
+    local definitions = context.loaddefinitions("scite-context-data-metafun")
 
     if definitions then
         metafunhelpers   = definitions.helpers   or { }
@@ -44,16 +46,9 @@ local whitespace = metafunlexer.WHITESPACE -- triggers states
 local space      = lexer.space -- S(" \n\r\t\f\v")
 local any        = lexer.any
 
-local digit      = R("09")
-local sign       = S("+-")
-local period     = P(".")
 local dquote     = P('"')
 local cstoken    = R("az","AZ") + P("_")
-local number     = sign^-1 * (                     -- at most one
-                        digit^1 * period * digit^0 -- 10.0 10.
-                      + digit^0 * period * digit^1 -- 0.10 .10
-                      + digit^1                    -- 10
-                   )
+local number     = context.patterns.real
 
 local cstokentex = R("az","AZ","\127\255") + S("@!?_")
 
@@ -66,10 +61,10 @@ local constant   = token('data',      exact_match(metafunconstants))
 local helper     = token('command',   exact_match(metafunhelpers))
 local plain      = token('plain',     exact_match(plaincommands))
 local quoted     = token('quote',     dquote)
-                 * token('string',    P(1-dquote)^1)
+                 * token('string',    P(1-dquote)^0)
                  * token('quote',     dquote)
 local primitive  = token('primitive', exact_match(primitivecommands))
-local identifier = token('default',   cstoken)
+local identifier = token('default',   cstoken^1)
 local number     = token('number',    number)
 local special    = token('special',   S("#()[]<>=:\"")) -- or else := <> etc split
 local texlike    = token('string',    P("\\") * cstokentex^1)
@@ -86,12 +81,12 @@ _rules = {
     { 'number',     number     },
     { 'quoted',     quoted     },
     { 'special',    special    },
---     { 'texlike',    texlike    },
+ -- { 'texlike',    texlike    },
     { 'extra',      extra      },
     { 'rest',       rest       },
 }
 
-_tokenstyles = lexer.context.styleset
+_tokenstyles = context.styleset
 
 _foldsymbols = {
     _patterns = {
diff --git a/context/data/scite/lexers/scite-context-lexer-pdf-object.lua b/context/data/scite/lexers/scite-context-lexer-pdf-object.lua
new file mode 100644
index 000000000..1de006813
--- /dev/null
+++ b/context/data/scite/lexers/scite-context-lexer-pdf-object.lua
@@ -0,0 +1,115 @@
+local info = {
+    version   = 1.002,
+    comment   = "scintilla lpeg lexer for pdf",
+    author    = "Hans Hagen, PRAGMA-ADE, Hasselt NL",
+    copyright = "PRAGMA ADE / ConTeXt Development Team",
+    license   = "see context related readme files",
+}
+
+local lexer = lexer
+local token = lexer.token
+local P, R, S, C, V = lpeg.P, lpeg.R, lpeg.S, lpeg.C, lpeg.V
+local global = _G
+
+module(...)
+
+local objectlexer       = _M
+
+local context           = lexer.context
+local patterns          = context.patterns
+
+local whitespace        = objectlexer.WHITESPACE -- triggers states
+
+local space             = lexer.space
+local somespace         = space^1
+
+local newline           = S("\n\r")
+local real              = patterns.real
+local cardinal          = patterns.cardinal
+
+local lparent           = P("(")
+local rparent           = P(")")
+local langle            = P("<")
+local rangle            = P(">")
+local escape            = P("\\")
+local anything          = P(1)
+local unicodetrigger    = P("feff")
+
+local nametoken         = 1 - space - S("<>/[]()")
+local name              = P("/") * nametoken^1
+
+local p_string          = P { ( escape * anything + lparent * V(1) * rparent + (1 - rparent) )^0 }
+
+local t_spacing         = token(whitespace, space^1)
+local t_spaces          = token(whitespace, space^1)^0
+
+local p_stream          = P("stream")
+local p_endstream       = P("endstream")
+----- p_obj             = P("obj")
+local p_endobj          = P("endobj")
+local p_reference       = P("R")
+
+local p_objectnumber    = patterns.cardinal
+local p_comment         = P('%') * (1-S("\n\r"))^0
+
+local string            = token("quote",    lparent)
+                        * token("string",   p_string)
+                        * token("quote",    rparent)
+local unicode           = token("quote",    langle)
+                        * token("plain",    unicodetrigger)
+                        * token("string",   (1-rangle)^1)
+                        * token("quote",    rangle)
+local whatsit           = token("quote",    langle)
+                        * token("string",   (1-rangle)^1)
+                        * token("quote",    rangle)
+local keyword           = token("command",  name)
+local constant          = token("constant", name)
+local number            = token('number',   real)
+local reference         = token("number",   cardinal)
+                        * t_spacing
+                        * token("number",   cardinal)
+                        * t_spacing
+                        * token("keyword",  p_reference)
+local t_comment         = token("comment",  p_comment)
+
+--    t_openobject      = token("number",  p_objectnumber)
+--                      * t_spacing
+--                      * token("number",  p_objectnumber)
+--                      * t_spacing
+--                      * token("keyword", p_obj)
+local t_closeobject     = token("keyword", p_endobj)
+
+local t_opendictionary  = token("grouping", P("<<"))
+local t_closedictionary = token("grouping", P(">>"))
+
+local t_openarray       = token("grouping", P("["))
+local t_closearray      = token("grouping", P("]"))
+
+local t_stream          = token("keyword", p_stream)
+--                         * token("default", newline * (1-newline*p_endstream*newline)^1 * newline)
+                        * token("default", (1 - p_endstream)^1)
+                        * token("keyword", p_endstream)
+
+local t_dictionary      = { "dictionary",
+                            dictionary = t_opendictionary * (t_spaces * keyword * t_spaces * V("whatever"))^0 * t_spaces * t_closedictionary,
+                            array      = t_openarray * (t_spaces * V("whatever"))^0 * t_spaces * t_closearray,
+                            whatever   = V("dictionary") + V("array") + constant + reference + string + unicode + number + whatsit,
+                        }
+
+local t_object          = { "object", -- weird that we need to catch the end here (probably otherwise an invalid lpeg)
+                            object     = t_spaces * (V("dictionary") * t_spaces * t_stream^-1 + V("array") + t_spaces) * t_spaces * t_closeobject,
+                            dictionary = t_opendictionary * (t_spaces * keyword * t_spaces * V("whatever"))^0 * t_spaces * t_closedictionary,
+                            array      = t_openarray * (t_spaces * V("whatever"))^0 * t_spaces * t_closearray,
+                            whatever   = V("dictionary") + V("array") + constant + reference + string + unicode + number + whatsit,
+                        }
+
+_shared = {
+    dictionary = t_dictionary,
+}
+
+_rules = {
+    { 'whitespace', t_spacing },
+    { 'object',     t_object  },
+}
+
+_tokenstyles = context.styleset
diff --git a/context/data/scite/lexers/scite-context-lexer-pdf-xref.lua b/context/data/scite/lexers/scite-context-lexer-pdf-xref.lua
new file mode 100644
index 000000000..8988fbbb4
--- /dev/null
+++ b/context/data/scite/lexers/scite-context-lexer-pdf-xref.lua
@@ -0,0 +1,41 @@
+local info = {
+    version   = 1.002,
+    comment   = "scintilla lpeg lexer for pdf",
+    author    = "Hans Hagen, PRAGMA-ADE, Hasselt NL",
+    copyright = "PRAGMA ADE / ConTeXt Development Team",
+    license   = "see context related readme files",
+}
+
+local lexer = lexer
+local token = lexer.token
+local P = lpeg.P
+local global = _G
+
+module(...)
+
+local pdflexer          = _M
+local objectlexer       = lexer.load("scite-context-lexer-pdf-object")
+
+local context           = lexer.context
+local patterns          = context.patterns
+
+local whitespace        = pdflexer.WHITESPACE -- triggers states
+
+local space             = patterns.space
+local spacing           = patterns.spacing
+
+local t_spacing         = token(whitespace, spacing)
+
+local p_trailer         = P("trailer")
+
+local t_xref            = token("default", (1-p_trailer)^1)
+                        * token("keyword", p_trailer)
+                        * t_spacing
+                        * objectlexer._shared.dictionary
+
+_rules = {
+    { 'whitespace', t_spacing },
+    { 'xref',       t_xref    },
+}
+
+_tokenstyles = context.styleset
diff --git a/context/data/scite/lexers/scite-context-lexer-pdf.lua b/context/data/scite/lexers/scite-context-lexer-pdf.lua
new file mode 100644
index 000000000..ebc3fba4f
--- /dev/null
+++ b/context/data/scite/lexers/scite-context-lexer-pdf.lua
@@ -0,0 +1,63 @@
+local info = {
+    version   = 1.002,
+    comment   = "scintilla lpeg lexer for pdf",
+    author    = "Hans Hagen, PRAGMA-ADE, Hasselt NL",
+    copyright = "PRAGMA ADE / ConTeXt Development Team",
+    license   = "see context related readme files",
+}
+
+local lexer = lexer
+local token = lexer.token
+local P, R, S = lpeg.P, lpeg.R, lpeg.S
+local global = _G
+
+module(...)
+
+local pdflexer          = _M
+local objectlexer       = lexer.load("scite-context-lexer-pdf-object")
+local xreflexer         = lexer.load("scite-context-lexer-pdf-xref")
+
+local context           = lexer.context
+local patterns          = context.patterns
+
+local whitespace        = pdflexer.WHITESPACE -- triggers states
+
+local space             = patterns.space
+local spacing           = patterns.spacing
+local nospacing         = patterns.nospacing
+local anything          = patterns.anything
+local restofline        = patterns.restofline
+
+local t_spacing         = token(whitespace, spacing)
+local t_rest            = token("default",  nospacing) -- anything
+
+local p_obj             = P("obj")
+local p_endobj          = P("endobj")
+local p_xref            = P("xref")
+local p_startxref       = P("startxref")
+
+local p_objectnumber    = patterns.cardinal
+local p_comment         = P('%') * restofline
+
+local t_comment         = token("comment", p_comment)
+local t_openobject      = token("number",  p_objectnumber)
+                        * t_spacing
+                        * token("number",  p_objectnumber)
+                        * t_spacing
+                        * token("keyword", p_obj)
+                        * t_spacing^0
+local t_closeobject     = token("keyword", p_endobj)
+
+local t_openxref        = token("keyword", p_xref)
+local t_closexref       = token("keyword", p_startxref)
+
+lexer.embed_lexer(pdflexer, objectlexer, t_openobject, t_closeobject)
+lexer.embed_lexer(pdflexer, xreflexer,   t_openxref,   t_closexref)
+
+_rules = {
+    { 'whitespace', t_spacing },
+    { 'comment',    t_comment },
+    { 'rest',       t_rest    },
+}
+
+_tokenstyles = context.styleset
diff --git a/context/data/scite/lexers/scite-context-lexer-tex.lua b/context/data/scite/lexers/scite-context-lexer-tex.lua
index cea0aa544..8204ae3a0 100644
--- a/context/data/scite/lexers/scite-context-lexer-tex.lua
+++ b/context/data/scite/lexers/scite-context-lexer-tex.lua
@@ -34,9 +34,9 @@ local info = {
 
 local lexer = lexer
 local global, string, table, lpeg = _G, string, table, lpeg
-local token, style, colors, exact_match, no_style = lexer.token, lexer.style, lexer.colors, lexer.exact_match, lexer.style_nothing
+local token, exact_match = lexer.token, lexer.exact_match
 local P, R, S, V, C, Cmt, Cp, Cc, Ct = lpeg.P, lpeg.R, lpeg.S, lpeg.V, lpeg.C, lpeg.Cmt, lpeg.Cp, lpeg.Cc, lpeg.Ct
-local type, next, pcall, loadfile, setmetatable = type, next, pcall, loadfile, setmetatable
+local type, next = type, next
 local find, match, lower = string.find, string.match, string.lower
 
 module(...)
@@ -45,14 +45,16 @@ local contextlexer = _M
 local cldlexer     = lexer.load('scite-context-lexer-cld')
 local mpslexer     = lexer.load('scite-context-lexer-mps')
 
-local commands   = { en = { } }
-local primitives = { }
-local helpers    = { }
-local constants  = { }
+local commands     = { en = { } }
+local primitives   = { }
+local helpers      = { }
+local constants    = { }
+
+local context      = lexer.context
 
 do -- todo: only once, store in global
 
-    local definitions = lexer.context.loaddefinitions("scite-context-data-interfaces")
+    local definitions = context.loaddefinitions("scite-context-data-interfaces")
 
     if definitions then
         for interface, list in next, definitions do
@@ -72,14 +74,14 @@ do -- todo: only once, store in global
         end
     end
 
-    local definitions = lexer.context.loaddefinitions("scite-context-data-context")
+    local definitions = context.loaddefinitions("scite-context-data-context")
 
     if definitions then
         helpers   = definitions.helpers   or { }
         constants = definitions.constants or { }
     end
 
-    local definitions = lexer.context.loaddefinitions("scite-context-data-tex")
+    local definitions = context.loaddefinitions("scite-context-data-tex")
 
     if definitions then
         local function add(data)
@@ -109,9 +111,10 @@ local knowncommand = Cmt(cstoken^1, function(_,i,s)
     return currentcommands[s] and i
 end)
 
-local wordpattern = lexer.context.wordpattern
-local checkedword = lexer.context.checkedword
-local setwordlist = lexer.context.setwordlist
+local wordtoken   = context.patterns.wordtoken
+local wordpattern = context.patterns.wordpattern
+local checkedword = context.checkedword
+local setwordlist = context.setwordlist
 local validwords  = false
 
 -- % language=uk
@@ -121,11 +124,11 @@ local knownpreamble = Cmt(#P("% "), function(input,i,_) -- todo : utfbomb
         validwords = false
         local s, e, word = find(input,'^(.+)[\n\r]',i) -- combine with match
         if word then
-            local interface = match(word,"interface=(..)")
+            local interface = match(word,"interface=([a-z]+)")
             if interface then
                 currentcommands  = commands[interface] or commands.en or { }
             end
-            local language = match(word,"language=(..)")
+            local language = match(word,"language=([a-z]+)")
             validwords = language and setwordlist(language)
         end
     end
@@ -170,6 +173,8 @@ end)
 
 -- experiment: keep space with whatever ... less tables
 
+-- 10pt
+
 local commentline            = P('%') * (1-S("\n\r"))^0
 local endline                = S("\n\r")^1
 
@@ -194,7 +199,10 @@ local p_csname               = backslash * (cstoken^1 + P(1))
 local p_grouping             = S("{$}")
 local p_special              = S("#()[]<>=\"")
 local p_extra                = S("`~%^&_-+/\'|")
-local p_text                 = cstoken^1 --maybe add punctuation and space
+local p_text                 = wordtoken^1 --maybe add punctuation and space
+
+local p_number               = context.patterns.real
+local p_unit                 = P("pt") + P("bp") + P("sp") + P("mm") + P("cm") + P("cc") + P("dd")
 
 -- no looking back           = #(1-S("[=")) * cstoken^3 * #(1-S("=]"))
 
@@ -283,10 +291,11 @@ local primitive              = token('primitive', p_primitive  )
 local ifprimitive            = token('primitive', p_ifprimitive)
 local csname                 = token('user',      p_csname     )
 local grouping               = token('grouping',  p_grouping   )
+local number                 = token('number',    p_number     )
+                             * token('constant',  p_unit       )
 local special                = token('special',   p_special    )
 local extra                  = token('extra',     p_extra      )
------ text                   = token('default',   p_text       )
------ word                   = token("okay",      p_word       )
+local text                   = token('default',   p_text       )
 local word                   = p_word
 
 ----- startluacode           = token("grouping", P("\\startluacode"))
@@ -306,7 +315,7 @@ end
 local function stopdisplaylua(_,i,s)
     local ok = luatag == s
     if ok then
-cldlexer._directives.cld_inline = false
+        cldlexer._directives.cld_inline = false
         luastatus = false
     end
     return ok
@@ -344,7 +353,7 @@ local function stopinlinelua_e(_,i,s) -- }
         lualevel = lualevel - 1
         local ok = lualevel <= 0
         if ok then
-cldlexer._directives.cld_inline = false
+            cldlexer._directives.cld_inline = false
             luastatus = false
         end
         return ok
@@ -395,7 +404,7 @@ _rules = {
     { "whitespace",  spacing     },
     { "preamble",    preamble    },
     { "word",        word        },
- -- { "text",        text        },
+    { "text",        text        }, -- non words
     { "comment",     comment     },
     { "constant",    constant    },
     { "helper",      helper      },
@@ -405,12 +414,13 @@ _rules = {
     { "csname",      csname      },
  -- { "whatever",    specialword }, -- not yet, crashes
     { "grouping",    grouping    },
+ -- { "number",      number      },
     { "special",     special     },
     { "extra",       extra       },
     { "rest",        rest        },
 }
 
-_tokenstyles = lexer.context.styleset
+_tokenstyles = context.styleset
 
 local folds = {
     ["\\start"] = 1, ["\\stop" ] = -1,
diff --git a/context/data/scite/lexers/scite-context-lexer-txt.lua b/context/data/scite/lexers/scite-context-lexer-txt.lua
new file mode 100644
index 000000000..07dff2970
--- /dev/null
+++ b/context/data/scite/lexers/scite-context-lexer-txt.lua
@@ -0,0 +1,69 @@
+local lexer = lexer
+local token = lexer.token
+local P, S, Cmt = lpeg.P, lpeg.S, lpeg.Cmt
+local find, match = string.find, string.match
+
+module(...)
+
+local textlexer   = _M
+
+local context     = lexer.context
+
+local whitespace  = textlexer.WHITESPACE -- triggers states
+
+local space       = lexer.space
+local any         = lexer.any
+
+local wordtoken   = context.patterns.wordtoken
+local wordpattern = context.patterns.wordpattern
+local checkedword = context.checkedword
+local setwordlist = context.setwordlist
+local validwords  = false
+
+-- [#!-%] language=uk
+
+local p_preamble = Cmt(#(S("#!-%") * P(" ")), function(input,i,_) -- todo: utf bomb
+    if i == 1 then -- < 10 then
+        validwords = false
+        local s, e, line = find(input,'^[#!%-%%](.+)[\n\r]',i)
+        if line then
+            local language = match(line,"language=([a-z]+)")
+            if language then
+                validwords = setwordlist(language)
+            end
+        end
+    end
+    return false
+end)
+
+local t_preamble =
+    token('preamble', p_preamble)
+
+local t_word =
+    Cmt(wordpattern, function(_,i,s)
+        if validwords then
+            return checkedword(validwords,s,i)
+        else
+            return true, { "text", i }
+        end
+    end)
+
+local t_text =
+    token("default", wordtoken^1)
+
+local t_rest =
+    token("default", (1-wordtoken-space)^1)
+
+local t_spacing =
+    token(whitespace, space^1)
+
+_rules = {
+    { "whitespace", t_spacing  },
+    { "preamble",   t_preamble },
+    { "word",       t_word     }, -- words >= 3
+    { "text",       t_text     }, -- non words
+    { "rest",       t_rest     },
+}
+
+_tokenstyles = lexer.context.styleset
+
diff --git a/context/data/scite/lexers/scite-context-lexer-xml.lua b/context/data/scite/lexers/scite-context-lexer-xml.lua
index 0441585c1..f8892bbce 100644
--- a/context/data/scite/lexers/scite-context-lexer-xml.lua
+++ b/context/data/scite/lexers/scite-context-lexer-xml.lua
@@ -14,15 +14,17 @@ local info = {
 
 local lexer = lexer
 local global, string, table, lpeg = _G, string, table, lpeg
-local token, style, colors, exact_match, no_style = lexer.token, lexer.style, lexer.colors, lexer.exact_match, lexer.style_nothing
+local token, exact_match = lexer.token, lexer.exact_match
 local P, R, S, V, C, Cmt = lpeg.P, lpeg.R, lpeg.S, lpeg.V, lpeg.C, lpeg.Cmt
-local type, setmetatable = type, setmetatable
+local type = type
 local match, find = string.match, string.find
 
 module(...)
 
 local examplelexer     = _M
 
+local context          = lexer.context
+
 local whitespace       = examplelexer.WHITESPACE -- triggers states
 
 local space            = lexer.space -- S(" \t\n\r\v\f")
@@ -49,9 +51,9 @@ local closecdata       = P("]]>")
 
 local entity           = ampersand * (1-semicolon)^1 * semicolon
 
-local wordpattern = lexer.context.wordpattern
-local checkedword = lexer.context.checkedword
-local setwordlist = lexer.context.setwordlist
+local wordpattern = context.patterns.wordpattern
+local checkedword = context.patterns.checkedword
+local setwordlist = context.setwordlist
 local validwords  = false
 
 -- <?xml version="1.0" encoding="UTF-8" language="uk" ?>
@@ -72,7 +74,6 @@ local p_preamble = Cmt(#P("<?xml "), function(input,i,_) -- todo: utf bomb
     return false
 end)
 
-
 local p_word =
     Cmt(wordpattern, function(_,i,s)
         if validwords then
@@ -190,7 +191,7 @@ _rules = {
     { "rest",        p_rest        },
 }
 
-_tokenstyles = lexer.context.styleset
+_tokenstyles = context.styleset
 
 _foldsymbols = { -- somehow doesn't work yet
     _patterns = {
diff --git a/context/data/scite/lexers/scite-context-lexer.lua b/context/data/scite/lexers/scite-context-lexer.lua
index a2bb35a57..13b8d77a8 100644
--- a/context/data/scite/lexers/scite-context-lexer.lua
+++ b/context/data/scite/lexers/scite-context-lexer.lua
@@ -41,7 +41,11 @@ local type, next, setmetatable, rawset = type, next, setmetatable, rawset
 
 dofile(_LEXERHOME .. '/lexer.lua')
 
-lexer.context = lexer.context or { }
+lexer.context    = lexer.context or { }
+local context    = lexer.context
+
+context.patterns = context.patterns or { }
+local patterns   = context.patterns
 
 local locations = {
  -- lexer.context.path,
@@ -87,7 +91,7 @@ end
 --     end
 -- end
 
-function lexer.context.loaddefinitions(name)
+function context.loaddefinitions(name)
     for i=1,#locations do
         local data = collect(locations[i] .. "/" .. name)
         if data then
@@ -98,7 +102,7 @@ end
 
 -- maybe more efficient:
 
-function lexer.context.word_match(words,word_chars,case_insensitive)
+function context.word_match(words,word_chars,case_insensitive)
     local chars = '%w_' -- maybe just "" when word_chars
     if word_chars then
         chars = '^([' .. chars .. gsub(word_chars,'([%^%]%-])', '%%%1') ..']+)'
@@ -128,16 +132,42 @@ end
 
 -- nicer (todo: utf):
 
-local defaults = R("az","AZ","\127\255","__")
+local idtoken = R("az","AZ","\127\255","__")
+local digit   = R("09")
+local sign    = S("+-")
+local period  = P(".")
+local space   = S(" \n\r\t\f\v")
+
+patterns.idtoken  = idtoken
+
+patterns.digit    = digit
+patterns.sign     = sign
+patterns.period   = period
+
+patterns.cardinal = digit^1
+patterns.integer  = sign^-1 * digit^1
 
-function lexer.context.exact_match(words,word_chars,case_insensitive)
+patterns.real     =
+    sign^-1 * (                    -- at most one
+        digit^1 * period * digit^0 -- 10.0 10.
+      + digit^0 * period * digit^1 -- 0.10 .10
+      + digit^1                    -- 10
+   )
+
+patterns.restofline = (1-S("\n\r"))^1
+patterns.space      = space
+patterns.spacing    = space^1
+patterns.nospacing  = (1-space)^1
+patterns.anything   = P(1)
+
+function context.exact_match(words,word_chars,case_insensitive)
     local characters = concat(words)
     local pattern -- the concat catches _ etc
     if word_chars == true or word_chars == false or word_chars == nil then
         word_chars = ""
     end
     if type(word_chars) == "string" then
-        pattern = S(characters) + defaults
+        pattern = S(characters) + idtoken
         if case_insensitive then
             pattern = pattern + S(upper(characters)) + S(lower(characters))
         end
@@ -166,7 +196,6 @@ function lexer.context.exact_match(words,word_chars,case_insensitive)
     end
 end
 
-
 -- spell checking (we can only load lua files)
 
 -- return {
@@ -185,13 +214,13 @@ local function splitwords(words)
     return lpegmatch(splitter,words)
 end
 
-function lexer.context.setwordlist(tag,limit) -- returns hash (lowercase keys and original values)
+function context.setwordlist(tag,limit) -- returns hash (lowercase keys and original values)
     if not tag or tag == "" then
         return false
     elseif lists[tag] ~= nil then
         return lists[tag]
     else
-        local list = lexer.context.loaddefinitions("spell-" .. tag)
+        local list = context.loaddefinitions("spell-" .. tag)
         if not list or type(list) ~= "table" then
             lists[tag] = false
             return nil
@@ -207,9 +236,10 @@ function lexer.context.setwordlist(tag,limit) -- returns hash (lowercase keys an
     end
 end
 
-lexer.context.wordpattern = R("az","AZ","\127\255")^3 -- todo: if limit and #s < limit then
+patterns.wordtoken   = R("az","AZ","\127\255")
+patterns.wordpattern = patterns.wordtoken^3 -- todo: if limit and #s < limit then
 
-function lexer.context.checkedword(validwords,s,i) -- ,limit
+function context.checkedword(validwords,s,i) -- ,limit
     if not validwords then
         return true, { "text", i }
     else
@@ -302,7 +332,7 @@ local splitlines = ( (
   + (                           newline   ) / function()    action_n()    end
 ) )^0
 
-function lexer.context.fold(text, start_pos, start_line, start_level)
+function context.fold(text, start_pos, start_line, start_level)
     if text == '' then
         return { }
     end
@@ -413,7 +443,7 @@ function lexer.context.fold(text, start_pos, start_line, start_level)
     return folds
 end
 
-function lexer.context.lex(text,init_style)
+function context.lex(text,init_style)
     local lexer = global._LEXER
     local grammar = lexer._GRAMMAR
     if not grammar then
@@ -499,11 +529,11 @@ end
 
 -- todo: keywords: one lookup and multiple matches
 
-function lexer.context.token(name, patt)
+function context.token(name, patt)
     return Ct(patt * Cc(name) * Cp())
 end
 
-lexer.fold        = lexer.context.fold
-lexer.lex         = lexer.context.lex
-lexer.token       = lexer.context.token
-lexer.exact_match = lexer.context.exact_match
+lexer.fold        = context.fold
+lexer.lex         = context.lex
+lexer.token       = context.token
+lexer.exact_match = context.exact_match
diff --git a/context/data/scite/scite-context-external.properties b/context/data/scite/scite-context-external.properties
index 14bdd222f..021c72133 100644
--- a/context/data/scite/scite-context-external.properties
+++ b/context/data/scite/scite-context-external.properties
@@ -24,6 +24,8 @@ lexer.$(file.patterns.context)=lpeg_scite-context-lexer-tex
 lexer.$(file.patterns.tex)=lpeg_scite-context-lexer-tex
 lexer.$(file.patterns.lua)=lpeg_scite-context-lexer-lua
 lexer.$(file.patterns.example)=lpeg_scite-context-lexer-xml
+lexer.$(file.patterns.text)=lpeg_scite-context-lexer-txt
+lexer.$(file.patterns.pdf)=lpeg_scite-context-lexer-pdf
 
 comment.block.lpeg_scite-context-lexer-tex=%
 comment.block.at.line.start.lpeg_scite-context-lexer-tex=1
@@ -37,9 +39,6 @@ comment.block.at.line.start.lpeg_scite-context-lexer-lua=1
 comment.block.lpeg_scite-context-lexer-cld=--
 comment.block.at.line.start.lpeg_scite-context-lexer-cld=1
 
-#~ comment.block.lpeg_scite-context-lexer-xml=<!--
-#~ comment.block.at.line.start.lpeg_scite-context-lexer-xml=1
-
 comment.block.lpeg_props=#
 comment.block.at.line.start.lpeg_props=1
 
diff --git a/context/data/scite/scite-context.properties b/context/data/scite/scite-context.properties
index 03df9fb75..5e57d8fcd 100644
--- a/context/data/scite/scite-context.properties
+++ b/context/data/scite/scite-context.properties
@@ -60,6 +60,18 @@ lexer.$(file.patterns.lua)=lua
 command.compile.$(file.patterns.lua)=mtxrun --script "$(FileNameExt)"
 command.go.$(file.patterns.lua)=mtxrun --script "$(FileNameExt)"
 
+# Test: patterns
+
+file.patterns.text=*.txt
+
+filter.text=Text |$(file.patterns.text)|
+
+file.patterns.pdf=*.pdf
+
+filter.pdf=PDF |$(file.patterns.pdf)|
+
+# lexer.$(file.patterns.txt)=txt
+
 # Commands: some scripts
 
 if PLAT_WIN
diff --git a/metapost/context/base/mp-char.mp b/metapost/context/base/mp-char.mp
index 786c1f904..fd9e33848 100644
--- a/metapost/context/base/mp-char.mp
+++ b/metapost/context/base/mp-char.mp
@@ -37,7 +37,6 @@ grid_width   := 60pt ; grid_height  := 40pt ;
 shape_width  := 45pt ; shape_height := 30pt ;
 
 chart_offset :=  2pt ;
-
 color chart_background_color ; chart_background_color := white ;
 
 %D test mode
diff --git a/scripts/context/lua/mtx-check.lua b/scripts/context/lua/mtx-check.lua
index e0ab3b882..38ab2254a 100644
--- a/scripts/context/lua/mtx-check.lua
+++ b/scripts/context/lua/mtx-check.lua
@@ -6,6 +6,9 @@ if not modules then modules = { } end modules ['mtx-check'] = {
     license   = "see context related readme files"
 }
 
+local P, R, S, V, C, CP, CC, lpegmatch = lpeg.P, lpeg.R, lpeg.S, lpeg.V, lpeg.C, lpeg.Cp, lpeg.Cc, lpeg.match
+local gsub, sub, format = string.gsub, string.sub, string.format
+
 local helpinfo = [[
 --convert             check tex file for errors
 ]]
@@ -31,21 +34,23 @@ validator.direct = false
 validator.printer = print
 validator.tracer  = print
 
-local message = function(position, kind)
+local message = function(position, kind, extra)
     local ve = validator.errors
-    ve[#ve+1] = { kind, position, validator.n }
+    ve[#ve+1] = { kind, position, validator.n, extra }
     if validator.direct then
-        validator.printer(string.format("%s error at position %s (line %s)", kind, position, validator.n))
+        if extra then
+            validator.printer(format("%s error at position %s (line %s) (%s)",kind,position,validator.n,extra))
+        else
+            validator.printer(format("%s error at position %s (line %s)",kind,position,validator.n))
+        end
     end
 end
 local progress = function(position, data, kind)
     if validator.trace then
-        validator.tracer(string.format("%s at position %s: %s", kind, position, data or ""))
+        validator.tracer(format("%s at position %s: %s", kind, position, data or ""))
     end
 end
 
-local P, R, S, V, C, CP, CC = lpeg.P, lpeg.R, lpeg.S, lpeg.V, lpeg.C, lpeg.Cp, lpeg.Cc
-
 local i_m, d_m = P("$"), P("$$")
 local l_s, r_s = P("["), P("]")
 local l_g, r_g = P("{"), P("}")
@@ -61,44 +66,56 @@ local newline = crlf + cr + lf
 
 local line = newline / function() validator.n = validator.n + 1 end
 
---  local grammar = P { "tokens",
---      ["tokens"]   = (V("whatever") + V("grouped") +  V("setup") + V("display") + V("inline") + V("errors") + 1)^0,
---      ["whatever"] = line + esc * 1 + C(P("%") * (1-line)^0),
---      ["grouped"]  = CP() * C(l_g * (V("whatever") + V("grouped") + V("setup") + V("display") + V("inline") + (1 - l_g - r_g))^0 * r_g) * CC("group") / progress,
---      ["setup"]    = CP() * C(l_s * (V("whatever") + V("grouped") + V("setup") + V("display") + V("inline") + (1 - l_s - r_s))^0 * r_s) * CC("setup") / progress,
---      ["display"]  = CP() * C(d_m * (V("whatever") + V("grouped") + (1 - d_m))^0 * d_m) * CC("display") / progress,
---      ["inline"]   = CP() * C(i_m * (V("whatever") + V("grouped") + (1 - i_m))^0 * i_m) * CC("inline") / progress,
---      ["errors"]   = (V("gerror") + V("serror") + V("derror") + V("ierror")) * true,
---      ["gerror"]   = CP() * (l_g + r_g) * CC("grouping") / message,
---      ["serror"]   = CP() * (l_s + r_g) * CC("setup error") / message,
---      ["derror"]   = CP() * d_m * CC("display math error") / message,
---      ["ierror"]   = CP() * i_m * CC("inline math error") / message,
---  }
-
 local startluacode = P("\\startluacode")
 local stopluacode  = P("\\stopluacode")
 
 local somecode  = startluacode * (1-stopluacode)^1 * stopluacode
 
+local stack = { }
+
+local function push(p,s)
+-- print("start",p,s)
+    table.insert(stack,{ p, s })
+end
+
+local function pop(p,s)
+-- print("stop",p,s)
+    local top = table.remove(stack)
+    if not top then
+        message(p,"missing start")
+    elseif top[2] ~= s then
+        message(p,"missing stop",format("see line %s",top[1]))
+    else
+        -- okay
+    end
+end
+
+local cstoken = R("az","AZ","\127\255")
+
+local start   = CP() * P("\\start") * C(cstoken^0) / push
+local stop    = CP() * P("\\stop")  * C(cstoken^0) / pop
+
 local grammar = P { "tokens",
-    ["tokens"]   = (V("ignore") + V("whatever") + V("grouped") +  V("setup") + V("display") + V("inline") + V("errors") + 1)^0,
-    ["whatever"] = line + esc * 1 + C(P("%") * (1-line)^0),
-    ["grouped"]  = l_g * (V("whatever") + V("grouped") + V("setup") + V("display") + V("inline") + (1 - l_g - r_g))^0 * r_g,
-    ["setup"]    = l_s * (okay + V("whatever") + V("grouped") + V("setup") + V("display") + V("inline") + (1 - l_s - r_s))^0 * r_s,
-    ["display"]  = d_m * (V("whatever") + V("grouped") + (1 - d_m))^0 * d_m,
-    ["inline"]   = i_m * (V("whatever") + V("grouped") + (1 - i_m))^0 * i_m,
-    ["errors"]   = (V("gerror")+ V("serror") + V("derror") + V("ierror")),
-    ["gerror"]   = CP() * (l_g + r_g) * CC("grouping") / message,
-    ["serror"]   = CP() * (l_s + r_g) * CC("setup error") / message,
-    ["derror"]   = CP() * d_m * CC("display math error") / message,
-    ["ierror"]   = CP() * i_m * CC("inline math error") / message,
-    ["ignore"]   = somecode,
+    ["tokens"]      = (V("ignore") + V("start") + V("stop") + V("whatever") + V("grouped") + V("setup") + V("display") + V("inline") + V("errors") + 1)^0,
+    ["start"]       = start,
+    ["stop"]        = stop,
+    ["whatever"]    = line + esc * 1 + C(P("%") * (1-line)^0),
+    ["grouped"]     = l_g * (V("whatever") + V("grouped") + V("setup") + V("display") + V("inline") + (1 - l_g - r_g))^0 * r_g,
+    ["setup"]       = l_s * (okay + V("whatever") + V("grouped") + V("setup") + V("display") + V("inline") + (1 - l_s - r_s))^0 * r_s,
+    ["display"]     = d_m * (V("whatever") + V("grouped") + (1 - d_m))^0 * d_m,
+    ["inline"]      = i_m * (V("whatever") + V("grouped") + (1 - i_m))^0 * i_m,
+    ["errors"]      = (V("gerror")+ V("serror") + V("derror") + V("ierror")),
+    ["gerror"]      = CP() * (l_g + r_g) * CC("grouping") / message,
+    ["serror"]      = CP() * (l_s + r_g) * CC("setup error") / message,
+    ["derror"]      = CP() * d_m * CC("display math error") / message,
+    ["ierror"]      = CP() * i_m * CC("inline math error") / message,
+    ["ignore"]      = somecode,
 }
 
 function validator.check(str)
     validator.n = 1
     validator.errors = { }
-    grammar:match(str)
+    lpegmatch(grammar,str)
 end
 
 --~ str = [[
@@ -109,6 +126,12 @@ end
 --~ ]]
 --~ str = string.rep(str,10)
 
+local remapper = {
+    ["\n"] = " <lf> ",
+    ["\r"] = " <cr> ",
+    ["\t"] = " <tab> ",
+}
+
 function scripts.checker.check(filename)
     local str = io.loaddata(filename)
     if str then
@@ -117,15 +140,15 @@ function scripts.checker.check(filename)
         if #errors > 0 then
             for k=1,#errors do
                 local v = errors[k]
-                local kind, position, line = v[1], v[2], v[3]
-                local data = str:sub(position-30,position+30)
-                data = data:gsub("(.)", {
-                    ["\n"] = " <lf> ",
-                    ["\r"] = " <cr> ",
-                    ["\t"] = " <tab> ",
-                })
-                data = data:gsub("^ *","")
-                print(string.format("% 5i  %-10s  %s", line, kind, data))
+                local kind, position, line, extra = v[1], v[2], v[3], v[4]
+                local data = sub(str,position-30,position+30)
+                data = gsub(data,".", remapper)
+                data = gsub(data,"^ *","")
+                if extra then
+                    print(format("% 5i  %-10s  %s (%s)", line, kind, data, extra))
+                else
+                    print(format("% 5i  %-10s  %s", line, kind, data))
+                end
             end
         else
             print("no error")
diff --git a/scripts/context/lua/mtx-context.lua b/scripts/context/lua/mtx-context.lua
index 43dcd3455..82346f394 100644
--- a/scripts/context/lua/mtx-context.lua
+++ b/scripts/context/lua/mtx-context.lua
@@ -626,7 +626,7 @@ scripts.context.interfaces = {
     de = "cont-de",
     fr = "cont-fr",
     nl = "cont-nl",
-    cz = "cont-cz",
+    cs = "cont-cs",
     it = "cont-it",
     ro = "cont-ro",
     pe = "cont-pe",
diff --git a/scripts/context/lua/mtx-epub.lua b/scripts/context/lua/mtx-epub.lua
index a4b96d3be..0dc533c4a 100644
--- a/scripts/context/lua/mtx-epub.lua
+++ b/scripts/context/lua/mtx-epub.lua
@@ -62,7 +62,7 @@ local package = [[
         <dc:language>en</dc:language>
         <dc:identifier id="%s" >urn:uuid:%s</dc:identifier>
         <dc:creator opf:file-as="Self, My" opf:role="aut">MySelf</dc:creator>
-        <dc:date>%s</dc:date> 
+        <dc:date>%s</dc:date>
     </metadata>
 
     <manifest>
@@ -168,13 +168,13 @@ function scripts.epub.make()
 
     local filename = environment.files[1]
 
-    if filename and filename ~= "" then
+    if filename and filename ~= "" and type(filename) == "string" then
 
         filename = file.basename(filename)
         local specfile = file.replacesuffix(filename,"specification")
         local specification = lfs.isfile(specfile) and dofile(specfile) or { }
 
-     -- inspect(specification)
+--         inspect(specification)
 
         local name       = specification.name       or file.removesuffix(filename)
         local identifier = specification.identifier or os.uuid(true)
@@ -215,7 +215,10 @@ function scripts.epub.make()
 
         local function copythem(files)
             for i=1,#files do
-                copyone(files[i])
+                local filename = files[i]
+                if type(filename) == "string" then
+                    copyone(filename)
+                end
             end
         end
 
diff --git a/scripts/context/lua/mtxrun.lua b/scripts/context/lua/mtxrun.lua
index 7061c54d5..3e6d9e303 100644
--- a/scripts/context/lua/mtxrun.lua
+++ b/scripts/context/lua/mtxrun.lua
@@ -3719,7 +3719,7 @@ local function utf16_to_utf8_be(t)
             if right then
                 local now = 256*left + right
                 if more > 0 then
-                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000
+                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000 -- the 0x10000 smells wrong
                     more = 0
                     r = r + 1
                     result[r] = utfchar(now)
@@ -3747,7 +3747,7 @@ local function utf16_to_utf8_le(t)
             if right then
                 local now = 256*right + left
                 if more > 0 then
-                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000
+                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000 -- the 0x10000 smells wrong
                     more = 0
                     r = r + 1
                     result[r] = utfchar(now)
diff --git a/scripts/context/stubs/mswin/mtxrun.lua b/scripts/context/stubs/mswin/mtxrun.lua
index 7061c54d5..3e6d9e303 100644
--- a/scripts/context/stubs/mswin/mtxrun.lua
+++ b/scripts/context/stubs/mswin/mtxrun.lua
@@ -3719,7 +3719,7 @@ local function utf16_to_utf8_be(t)
             if right then
                 local now = 256*left + right
                 if more > 0 then
-                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000
+                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000 -- the 0x10000 smells wrong
                     more = 0
                     r = r + 1
                     result[r] = utfchar(now)
@@ -3747,7 +3747,7 @@ local function utf16_to_utf8_le(t)
             if right then
                 local now = 256*right + left
                 if more > 0 then
-                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000
+                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000 -- the 0x10000 smells wrong
                     more = 0
                     r = r + 1
                     result[r] = utfchar(now)
diff --git a/scripts/context/stubs/unix/mtxrun b/scripts/context/stubs/unix/mtxrun
index 7061c54d5..3e6d9e303 100755
--- a/scripts/context/stubs/unix/mtxrun
+++ b/scripts/context/stubs/unix/mtxrun
@@ -3719,7 +3719,7 @@ local function utf16_to_utf8_be(t)
             if right then
                 local now = 256*left + right
                 if more > 0 then
-                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000
+                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000 -- the 0x10000 smells wrong
                     more = 0
                     r = r + 1
                     result[r] = utfchar(now)
@@ -3747,7 +3747,7 @@ local function utf16_to_utf8_le(t)
             if right then
                 local now = 256*right + left
                 if more > 0 then
-                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000
+                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000 -- the 0x10000 smells wrong
                     more = 0
                     r = r + 1
                     result[r] = utfchar(now)
diff --git a/tex/context/base/back-exp.lua b/tex/context/base/back-exp.lua
index 8a588ce6a..11ed26366 100644
--- a/tex/context/base/back-exp.lua
+++ b/tex/context/base/back-exp.lua
@@ -25,6 +25,7 @@ local utfchar, utfbyte, utfsub, utfgsub = utf.char, utf.byte, utf.sub, utf.gsub
 local insert, remove = table.insert, table.remove
 local topoints = number.topoints
 local utfvalues = string.utfvalues
+local fromunicode16 = fonts.mappings.fromunicode16
 
 local trace_export      = false  trackers.register  ("export.trace",     function(v) trace_export = v end)
 local less_state        = false  directives.register("export.lessstate", function(v) less_state   = v end)
@@ -1825,10 +1826,12 @@ local function collectresults(head,list) -- is last used (we also have currentat
                                 local u = fc.tounicode
                                 if u and u ~= "" then
                                     -- tracing
-                                    for s in gmatch(u,"....") do -- is this ok?
-                                        nofcurrentcontent = nofcurrentcontent + 1
-                                        currentcontent[nofcurrentcontent] = utfchar(tonumber(s,16))
-                                    end
+--                                     for s in gmatch(u,"....") do -- is this ok?
+--                                         nofcurrentcontent = nofcurrentcontent + 1
+--                                         currentcontent[nofcurrentcontent] = utfchar(tonumber(s,16))
+--                                     end
+                                    nofcurrentcontent = nofcurrentcontent + 1
+                                    currentcontent[nofcurrentcontent] = utfchar(fromunicode16(u))
                                 else
                                     nofcurrentcontent = nofcurrentcontent + 1
                                     currentcontent[nofcurrentcontent] = utfchar(c)
@@ -2188,9 +2191,6 @@ local function stopexport(v)
     local result = allcontent(tree) -- also does some housekeeping and data collecting
     --
     local files = {
-        xhtmlfile,
-     -- stylefilename,
-     -- imagefilename,
     }
     local results = concat {
         wholepreamble(),
@@ -2218,6 +2218,7 @@ local function stopexport(v)
         else
             xhtmlfile = file.addsuffix(xhtmlfile,"xhtml")
         end
+        files[#files+1] = xhtmlfile
         report_export("saving xhtml variant in '%s",xhtmlfile)
         local xmltree = cleanxhtmltree(xml.convert(results))
         xml.save(xmltree,xhtmlfile)
diff --git a/tex/context/base/bibl-bib.lua b/tex/context/base/bibl-bib.lua
index d7c195576..444f7e9bc 100644
--- a/tex/context/base/bibl-bib.lua
+++ b/tex/context/base/bibl-bib.lua
@@ -204,12 +204,12 @@ function bibtex.toxml(session,options)
     local strip = options.strip -- todo: interface
     local entries = session.entries
     r = r + 1 ; result[r] = format("<?xml version='1.0' standalone='yes'?>")
-    result[#result+1] = format("<bibtex>")
+    r = r + 1 ; result[r] = format("<bibtex>")
     for id, categories in next, session.data do
         id = lower(gsub(id,"^@",""))
         for name, entry in next, categories do
             if not entries or entries[name] then
-                result[#result+1] = format("<entry tag='%s' category='%s'>",lower(name),id)
+                r = r + 1 ; result[r] = format("<entry tag='%s' category='%s'>",lower(name),id)
                 for key, value in next, entry do
                     value = gsub(value,"\\(.)",ihatethis)
                     value = lpegmatch(escaped_pattern,value)
diff --git a/tex/context/base/cont-new.mkii b/tex/context/base/cont-new.mkii
index 78cc2795c..20a06246e 100644
--- a/tex/context/base/cont-new.mkii
+++ b/tex/context/base/cont-new.mkii
@@ -11,7 +11,7 @@
 %C therefore copyrighted by \PRAGMA. See mreadme.pdf for
 %C details.
 
-\newcontextversion{2011.09.19 23:43}
+\newcontextversion{2011.09.27 20:05}
 
 %D This file is loaded at runtime, thereby providing an
 %D excellent place for hacks, patches, extensions and new
diff --git a/tex/context/base/cont-new.mkiv b/tex/context/base/cont-new.mkiv
index 9c2ac5258..41106bfc6 100644
--- a/tex/context/base/cont-new.mkiv
+++ b/tex/context/base/cont-new.mkiv
@@ -11,7 +11,7 @@
 %C therefore copyrighted by \PRAGMA. See mreadme.pdf for
 %C details.
 
-\newcontextversion{2011.09.19 23:43}
+\newcontextversion{2011.09.27 20:05}
 
 %D This file is loaded at runtime, thereby providing an
 %D excellent place for hacks, patches, extensions and new
diff --git a/tex/context/base/context-version.pdf b/tex/context/base/context-version.pdf
index 0a3d4b524..1347294c2 100644
--- a/tex/context/base/context-version.pdf
+++ b/tex/context/base/context-version.pdf
diff --git a/tex/context/base/context-version.png b/tex/context/base/context-version.png
index 70f13ed78..4fe6a22e5 100644
--- a/tex/context/base/context-version.png
+++ b/tex/context/base/context-version.png
diff --git a/tex/context/base/context.mkii b/tex/context/base/context.mkii
index 9355e5357..454c8fcae 100644
--- a/tex/context/base/context.mkii
+++ b/tex/context/base/context.mkii
@@ -20,7 +20,7 @@
 %D your styles an modules.
 
 \edef\contextformat {\jobname}
-\edef\contextversion{2011.09.19 23:43}
+\edef\contextversion{2011.09.27 20:05}
 
 %D For those who want to use this:
 
diff --git a/tex/context/base/context.mkiv b/tex/context/base/context.mkiv
index 5aaf50cea..0a44b2efb 100644
--- a/tex/context/base/context.mkiv
+++ b/tex/context/base/context.mkiv
@@ -20,7 +20,7 @@
 %D your styles an modules.
 
 \edef\contextformat {\jobname}
-\edef\contextversion{2011.09.19 23:43}
+\edef\contextversion{2011.09.27 20:05}
 
 %D For those who want to use this:
 
diff --git a/tex/context/base/font-con.lua b/tex/context/base/font-con.lua
index 9028443f0..f6b415bc7 100644
--- a/tex/context/base/font-con.lua
+++ b/tex/context/base/font-con.lua
@@ -403,21 +403,35 @@ function constructors.scale(tfmdata,specification)
     local sharedkerns   = { }
     --
     for unicode, character in next, characters do
-        local chr, description, index
+        local chr, description, index, touni
         if changed then
-            -- basemode hack
+            -- basemode hack (we try to catch missing tounicodes, e.g. needed for ssty in math cambria)
             local c = changed[unicode]
             if c then
-                description = descriptions[c] or character
+                description = descriptions[c] or descriptions[unicode] or character
                 character = characters[c] or character
                 index = description.index or c
+                if tounicode then
+                    touni = tounicode[index] -- nb: index!
+                    if not touni then -- goodie
+                        local d = descriptions[unicode] or characters[unicode]
+                        local i = d.index or unicode
+                        touni = tounicode[i] -- nb: index!
+                    end
+                end
             else
                 description = descriptions[unicode] or character
                 index = description.index or unicode
+                if tounicode then
+                    touni = tounicode[index] -- nb: index!
+                end
             end
         else
             description = descriptions[unicode] or character
             index = description.index or unicode
+            if tounicode then
+                touni = tounicode[index] -- nb: index!
+            end
         end
         local width  = description.width
         local height = description.height
@@ -460,15 +474,12 @@ function constructors.scale(tfmdata,specification)
                 }
             end
         end
+        if touni then
+            chr.tounicode = touni
+        end
     --  if trace_scaling then
     --    report_defining("t=%s, u=%s, i=%s, n=%s c=%s",k,chr.tounicode or "",index or 0,description.name or '-',description.class or '-')
     --  end
-        if tounicode then
-            local tu = tounicode[index] -- nb: index!
-            if tu then
-                chr.tounicode = tu
-            end
-        end
         if hasquality then
             -- we could move these calculations elsewhere (saves calculations)
             local ve = character.expansion_factor
diff --git a/tex/context/base/font-map.lua b/tex/context/base/font-map.lua
index b30462f23..7f5305f9a 100644
--- a/tex/context/base/font-map.lua
+++ b/tex/context/base/font-map.lua
@@ -98,6 +98,15 @@ local function tounicode16sequence(unicodes)
     return concat(t)
 end
 
+local function fromunicode16(str)
+    if #str == 4 then
+        return tonumber(str,16)
+    else
+        local l, r = match(str,"(....)(....)")
+        return (tonumber(l,16)- 0xD800)*0x400  + tonumber(r,16) - 0xDC00
+    end
+end
+
 --~ This is quite a bit faster but at the cost of some memory but if we
 --~ do this we will also use it elsewhere so let's not follow this route
 --~ now. I might use this method in the plain variant (no caching there)
@@ -122,6 +131,7 @@ mappings.loadlumtable        = loadlumtable
 mappings.makenameparser      = makenameparser
 mappings.tounicode16         = tounicode16
 mappings.tounicode16sequence = tounicode16sequence
+mappings.fromunicode16       = fromunicode16
 
 local separator   = S("_.")
 local other       = C((1 - separator)^1)
@@ -177,7 +187,9 @@ function mappings.addtounicode(data,filename)
         if unic == -1 or unic >= private or (unic >= 0xE000 and unic <= 0xF8FF) or unic == 0xFFFE or unic == 0xFFFF then
             local unicode = lumunic and lumunic[name] or unicodevector[name]
             if unicode then
-                originals[index], tounicode[index], ns = unicode, tounicode16(unicode), ns + 1
+                originals[index] = unicode
+                tounicode[index] = tounicode16(unicode)
+                ns               = ns + 1
             end
             -- cidmap heuristics, beware, there is no guarantee for a match unless
             -- the chain resolves
@@ -186,7 +198,9 @@ function mappings.addtounicode(data,filename)
                 if foundindex then
                     unicode = cidcodes[foundindex] -- name to number
                     if unicode then
-                        originals[index], tounicode[index], ns = unicode, tounicode16(unicode), ns + 1
+                        originals[index] = unicode
+                        tounicode[index] = tounicode16(unicode)
+                        ns               = ns + 1
                     else
                         local reference = cidnames[foundindex] -- number to name
                         if reference then
@@ -194,16 +208,23 @@ function mappings.addtounicode(data,filename)
                             if foundindex then
                                 unicode = cidcodes[foundindex]
                                 if unicode then
-                                    originals[index], tounicode[index], ns = unicode, tounicode16(unicode), ns + 1
+                                    originals[index] = unicode
+                                    tounicode[index] = tounicode16(unicode)
+                                    ns               = ns + 1
                                 end
                             end
                             if not unicode then
                                 local foundcodes, multiple = lpegmatch(uparser,reference)
                                 if foundcodes then
+                                    originals[index] = foundcodes
                                     if multiple then
-                                        originals[index], tounicode[index], nl, unicode = foundcodes, tounicode16sequence(foundcodes), nl + 1, true
+                                        tounicode[index] = tounicode16sequence(foundcodes)
+                                        nl               = nl + 1
+                                        unicode          = true
                                     else
-                                        originals[index], tounicode[index], ns, unicode = foundcodes, tounicode16(foundcodes), ns + 1, foundcodes
+                                        tounicode[index] = tounicode16(foundcodes)
+                                        ns               = ns + 1
+                                        unicode          = foundcodes
                                     end
                                 end
                             end
@@ -214,19 +235,8 @@ function mappings.addtounicode(data,filename)
             -- a.whatever or a_b_c.whatever or a_b_c (no numbers)
             if not unicode then
                 local split = lpegmatch(ligsplitter,name)
-                local nplit = (split and #split) or 0
-                if nplit == 0 then
-                    -- skip
-                elseif nplit == 1 then
-                    local base = split[1]
-                    unicode = unicodes[base] or unicodevector[base]
-                    if unicode then
-                        if type(unicode) == "table" then
-                            unicode = unicode[1]
-                        end
-                        originals[index], tounicode[index], ns = unicode, tounicode16(unicode), ns + 1
-                    end
-                else
+                local nplit = split and #split or 0
+                if nplit >= 2 then
                     local t, n = { }, 0
                     for l=1,nplit do
                         local base = split[l]
@@ -244,25 +254,38 @@ function mappings.addtounicode(data,filename)
                     if n == 0 then -- done then
                         -- nothing
                     elseif n == 1 then
-                        originals[index], tounicode[index], nl, unicode = t[1], tounicode16(t[1]), nl + 1, true
+                        originals[index] = t[1]
+                        tounicode[index] = tounicode16(t[1])
                     else
-                        originals[index], tounicode[index], nl, unicode = t, tounicode16sequence(t), nl + 1, true
+                        originals[index] = t
+                        tounicode[index] = tounicode16sequence(t)
                     end
+                    nl = nl + 1
+                    unicode = true
+                else
+                    -- skip: already checked and we don't want privates here
                 end
             end
-            -- last resort
+            -- last resort (we might need to catch private here as well)
             if not unicode then
                 local foundcodes, multiple = lpegmatch(uparser,name)
                 if foundcodes then
                     if multiple then
-                        originals[index], tounicode[index], nl, unicode = foundcodes, tounicode16sequence(foundcodes), nl + 1, true
+                        originals[index] = foundcodes
+                        tounicode[index] = tounicode16sequence(foundcodes)
+                        nl               = nl + 1
+                        unicode          = true
                     else
-                        originals[index], tounicode[index], ns, unicode = foundcodes, tounicode16(foundcodes), ns + 1, foundcodes
+                        originals[index] = foundcodes
+                        tounicode[index] = tounicode16(foundcodes)
+                        ns               = ns + 1
+                        unicode          = foundcodes
                     end
                 end
             end
          -- if not unicode then
-         --     originals[index], tounicode[index] = 0xFFFD, "FFFD"
+         --     originals[index] = 0xFFFD
+         --     tounicode[index] = "FFFD"
          -- end
         end
     end
diff --git a/tex/context/base/l-unicode.lua b/tex/context/base/l-unicode.lua
index 6295ecb7d..246171aec 100644
--- a/tex/context/base/l-unicode.lua
+++ b/tex/context/base/l-unicode.lua
@@ -218,7 +218,7 @@ local function utf16_to_utf8_be(t)
             if right then
                 local now = 256*left + right
                 if more > 0 then
-                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000
+                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000 -- the 0x10000 smells wrong
                     more = 0
                     r = r + 1
                     result[r] = utfchar(now)
@@ -246,7 +246,7 @@ local function utf16_to_utf8_le(t)
             if right then
                 local now = 256*right + left
                 if more > 0 then
-                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000
+                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000 -- the 0x10000 smells wrong
                     more = 0
                     r = r + 1
                     result[r] = utfchar(now)
diff --git a/tex/context/base/math-tag.lua b/tex/context/base/math-tag.lua
index 49138ac3b..0ac5b0897 100644
--- a/tex/context/base/math-tag.lua
+++ b/tex/context/base/math-tag.lua
@@ -87,6 +87,8 @@ end
 
 -- todo: check function here and keep attribute the same
 
+-- todo: variants -> original
+
 local actionstack = { }
 
 process = function(start) -- we cannot use the processor as we have no finalizers (yet)
@@ -117,6 +119,7 @@ process = function(start) -- we cannot use the processor as we have no finalizer
             else
                 tag = "mo"
             end
+-- print(start,a,tag)
             set_attribute(start,a_tagged,start_tagged(tag,a))
             stop_tagged()
             break -- okay?
diff --git a/tex/context/base/math-vfu.lua b/tex/context/base/math-vfu.lua
index 9aa2f400a..31c13cf8c 100644
--- a/tex/context/base/math-vfu.lua
+++ b/tex/context/base/math-vfu.lua
@@ -653,11 +653,12 @@ function vfmath.define(specification,set,goodies)
                                     ref = { { 'slot', s, index } }
                                     si[index] = ref
                                 end
+                                local italic = fci.italic
                                 local t = {
-                                    width    = fci.width,
+                                    width    = fci.width + italic, -- watch this !
                                     height   = fci.height,
                                     depth    = fci.depth,
-                                    italic   = fci.italic,
+                                    italic   = italic,
                                     commands = ref,
                                 }
                                 local n = fci.next
diff --git a/tex/context/base/page-lay.mkiv b/tex/context/base/page-lay.mkiv
index f76a03727..7ff3420d9 100644
--- a/tex/context/base/page-lay.mkiv
+++ b/tex/context/base/page-lay.mkiv
@@ -586,13 +586,13 @@
    \recalculatelayout}
 
 \def\checkcurrentoddlayout
-  {\ifcsname\??ly\v!odd\c!state\endcsname
-     \doifvalue{\??ly\v!odd\c!state}\v!start{\changetolayout\v!odd}%
+  {\ifcsname\namedlayouthash\v!odd\c!state\endcsname
+     \doif{\namedlayoutparameter\v!odd\c!state}\v!start{\changetolayout\v!odd}%
    \fi}
 
 \def\checkcurrentevenlayout
-  {\ifcsname\??ly\v!even\c!state\endcsname
-     \doifvalue{\??ly\v!even\c!state}\v!start{\changetolayout\v!even}%
+  {\ifcsname\namedlayouthash\v!even\c!state\endcsname
+     \doif{\namedlayoutparameter\v!even\c!state}\v!start{\changetolayout\v!even}%
    \fi}
 
 \ifx\lastpage\undefined
@@ -613,11 +613,11 @@
    \fi}
 
 \def\checkcurrentlayout % public and used in naw, so keep this name
-  {\ifcsname\??ly\realfolio\c!state\endcsname
-     \doifvalue{\??ly\realfolio\c!state}\v!start{\changetolayout\realfolio}%
-   \else\ifcsname\??ly\reverserealfolio\c!state\endcsname
-     \doifvalue{\??ly\reverserealfolio\c!state}\v!start{\changetolayout\reverserealfolio}%
-   \else\ifcsname\??ly\v!current\c!state\endcsname
+  {\ifcsname\namedlayouthash\realfolio\c!state\endcsname
+     \doif{\namedlayoutparameter\realfolio\c!state}\v!start{\changetolayout\realfolio}%
+   \else\ifcsname\namedlayouthash\reverserealfolio\c!state\endcsname
+     \doif{\namedlayoutparameter\reverserealfolio\c!state}\v!start{\changetolayout\reverserealfolio}%
+   \else\ifcsname\namedlayouthash\v!current\c!state\endcsname
      \changetolayout\v!current % no start test ?
    \else
      \doifoddpageelse\checkcurrentoddlayout\checkcurrentevenlayout
diff --git a/tex/context/base/page-set.mkiv b/tex/context/base/page-set.mkiv
index 7ef196d53..0c3dfff30 100644
--- a/tex/context/base/page-set.mkiv
+++ b/tex/context/base/page-set.mkiv
@@ -2480,7 +2480,7 @@
    \setcolumnsetspanhsize\mofcolumns\!!countc % a/b used
    \hsize\columnsetspanhsize
    \setbox\scratchbox\vbox\bgroup
-     \dostartframedtext[cs:#1][\v!none]% geen nils placement
+     \dostartframedtext{cs:#1}[\v!none]% geen nils placement
      % spoils spacing : \vskip-\struttotal\par\verticalstrut\par
      \ifnum\columnsetlevel>\zerocount
        \namedframedtextparameter{cs:#1}\c!before
diff --git a/tex/context/base/s-pre-61.tex b/tex/context/base/s-pre-61.tex
index d4e06d0a7..21e461320 100644
--- a/tex/context/base/s-pre-61.tex
+++ b/tex/context/base/s-pre-61.tex
@@ -234,6 +234,7 @@
 \StartSteps
 
 \startitemize
+\item Fiona Apple \FlushStep
 \item Tori Amos \FlushStep
 \item Kate Bush \FlushStep
 \item Heather Nova \FlushStep
diff --git a/tex/context/base/s-pre-70.mkiv b/tex/context/base/s-pre-70.mkiv
index 9f76fc84a..2c8bb2d26 100644
--- a/tex/context/base/s-pre-70.mkiv
+++ b/tex/context/base/s-pre-70.mkiv
@@ -119,16 +119,16 @@
   [state=start]
 
 \startinteractionmenu[bottom]
-    \txt
+    \starttxt
         \interactionbar
           [alternative=d,
            symbol=yes,
            color=white,
            contrastcolor=textcolor]
-    \\
+    \stoptxt
     \hfilll
-    \but [previouspage] < < < \\
-    \but [nextpage] > > > \\
+    \startbut [previouspage] < < < \stopbut
+    \startbut [nextpage]     > > > \stopbut
 \stopinteractionmenu
 
 %D Instead of the normal symbols we use more punky ones.
@@ -150,7 +150,7 @@
 %D Run this file with the command: \type {context --mode=demo s-pre-70}
 %D in order to get an example.
 
-\doifnotmode{demo} {\endinput}
+\continueifinputfile{s-pre-70.mkiv}
 
 \usemodule[pre-60] % use the stepper
 
@@ -159,14 +159,18 @@
 \title {Punk for dummies}
 
 \dorecurse{10} {
+
     \title{Just a few dummy pages}
-    \StartSteps \startitemize[packed]
-    \startitemize
-        \startitem bla \FlushStep \stopitem
-        \startitem bla bla \FlushStep \stopitem
-        \startitem bla bla bla \FlushStep \stopitem
-        \startitem bla bla bla bla \FlushStep \stopitem
-    \stopitemize \StopSteps
+
+    \StartSteps
+        \startitemize[packed]
+            \startitem bla \FlushStep \stopitem
+            \startitem bla bla \FlushStep \stopitem
+            \startitem bla bla bla \FlushStep \stopitem
+            \startitem bla bla bla bla \FlushStep \stopitem
+        \stopitemize
+    \StopSteps
+
 }
 
 \stoptext
diff --git a/tex/context/base/s-pre-71.mkiv b/tex/context/base/s-pre-71.mkiv
index a52718576..38dae61df 100644
--- a/tex/context/base/s-pre-71.mkiv
+++ b/tex/context/base/s-pre-71.mkiv
@@ -145,7 +145,7 @@
 \let\StartText\starttext
 \let\StopText \stoptext
 
-\doifnotmode{demo}{\endinput}
+\continueifinputfile{s-pre-71.mkiv}
 
 \starttext
 
diff --git a/tex/context/base/scrn-but.mkvi b/tex/context/base/scrn-but.mkvi
index 7370d944f..693996d8a 100644
--- a/tex/context/base/scrn-but.mkvi
+++ b/tex/context/base/scrn-but.mkvi
@@ -739,7 +739,7 @@
 \unexpanded\def\scrn_menu_but#content\\{\scrn_menu_but_start#content\stopbut} \let\stopbut\relax
 \unexpanded\def\scrn_menu_got#content\\{\scrn_menu_got_start#content\stopgot} \let\stopgot\relax
 \unexpanded\def\scrn_menu_nop#content\\{\scrn_menu_nop_start#content\stopnop} \let\stopnop\relax
-\unexpanded\def\scrn_menu_txt#content\\{\scrn_menu_nop_start#content\stoptxt} \let\stoptxt\relax
+\unexpanded\def\scrn_menu_txt#content\\{\scrn_menu_txt_start#content\stoptxt} \let\stoptxt\relax
 \unexpanded\def\scrn_menu_rul#content\\{\scrn_menu_rul_start#content\stoprul} \let\stoprul\relax
 \unexpanded\def\scrn_menu_com#content\\{\scrn_menu_com_start#content\stopcom} \let\stopcom\relax
 
diff --git a/tex/context/base/sort-ini.lua b/tex/context/base/sort-ini.lua
index d9a6bb64d..8640ba825 100644
--- a/tex/context/base/sort-ini.lua
+++ b/tex/context/base/sort-ini.lua
@@ -96,10 +96,11 @@ local validmethods      = table.tohash {
 }
 
 local predefinedmethods = {
-    [v_before] = "mm,mc,uc",
-    [v_after]  = "pm,mc,uc",
-    [v_first]  = "pc,mm,uc",
-    [v_last]   = "mc,mm,uc",
+    [v_default] = "zc,pc,zm,pm,uc",
+    [v_before]  = "mm,mc,uc",
+    [v_after]   = "pm,mc,uc",
+    [v_first]   = "pc,mm,uc",
+    [v_last]    = "mc,mm,uc",
 }
 
 sorters = {
@@ -113,7 +114,7 @@ sorters = {
         digitsoffset      = digitsoffset,
         digitsmaximum     = digitsmaximum,
         defaultlanguage   = v_default,
-        defaultmethod     = v_before,
+        defaultmethod     = v_default,
         defaultdigits     = v_numbers,
     }
 }
diff --git a/tex/context/base/spac-ali.mkiv b/tex/context/base/spac-ali.mkiv
index 686bcfb83..ffc1f2434 100644
--- a/tex/context/base/spac-ali.mkiv
+++ b/tex/context/base/spac-ali.mkiv
@@ -244,6 +244,10 @@
      \let\raggedtopcommand   \vfilll % used with \framed for
    \fi}                              % instance in tables
 
+\setvalue{\@@ragged@@command\v!last}%
+  {\appendtoks\centeredlastline\to\everyraggedcommand
+   \!!donecfalse}
+
 \setvalue{\@@ragged@@command\v!flushleft }{\getvalue{\@@ragged@@command\v!right }}
 \setvalue{\@@ragged@@command\v!flushright}{\getvalue{\@@ragged@@command\v!left  }}
 \setvalue{\@@ragged@@command\v!center    }{\getvalue{\@@ragged@@command\v!middle}}
@@ -358,8 +362,8 @@
 
 \let\updateraggedskips\relax
 
-\def\setraggedskips#1#2#3#4#5#6#7% never change this name
-  {\def\updateraggedskips{\dosetraggedskips{#1}{#2}{#3}{#4}{#5}{#6}{#7}}%
+\unexpanded\def\setraggedskips#1#2#3#4#5#6#7% never change this name
+  {\unexpanded\def\updateraggedskips{\dosetraggedskips{#1}{#2}{#3}{#4}{#5}{#6}{#7}}%
    \updateraggedskips}
 
 \def\dosetraggedskips#1#2#3#4#5#6#7%
@@ -419,10 +423,12 @@
 
 % defaults
 
-\def\raggedfillamount    {1fil}
-\def\raggedhalffillamount{.5fil}
-\def\raggedspaceamount   {\interwordspace} % {.3333em}
-\def\raggedxspaceamount  {.5em}
+\def\raggedfillamount        {1fil}
+\def\raggednegativefillamount{-1fil}
+\def\raggeddoublefillamount  {2fil}
+\def\raggedhalffillamount    {.5fil}
+\def\raggedspaceamount       {\interwordspace} % {.3333em}
+\def\raggedxspaceamount      {.5em}
 
 \unexpanded\def\notragged
   {\raggedstatus\zerocount
@@ -445,6 +451,16 @@
    \setraggedskips2\middleraggedness\middleraggedness\raggedspaceamount
      \raggedxspaceamount\zeropoint\zeropoint}
 
+\unexpanded\def\centeredlastline
+  {\setraggedskips
+     \zerocount
+     \raggedfillamount
+     \raggednegativefillamount
+     \zeropoint
+     \zeropoint
+     \raggeddoublefillamount
+     \zeropoint}
+
 %D We used to have:
 %D
 %D \starttyping
@@ -561,6 +577,8 @@
 \installalign {l2r}            {\lefttoright}
 \installalign {r2l}            {\righttoleft}
 
+\installalign \v!last          {\centeredlastline}
+
 \newcount\hyphenminoffset
 
 \ifx\sethyphenationvariables\undefined \let\sethyphenationvariables\relax \fi
diff --git a/tex/context/base/status-files.pdf b/tex/context/base/status-files.pdf
index a72d471d4..c8c6b27a5 100644
--- a/tex/context/base/status-files.pdf
+++ b/tex/context/base/status-files.pdf
diff --git a/tex/context/base/status-lua.pdf b/tex/context/base/status-lua.pdf
index 27829fd27..e4e11892c 100644
--- a/tex/context/base/status-lua.pdf
+++ b/tex/context/base/status-lua.pdf
diff --git a/tex/context/base/strc-doc.lua b/tex/context/base/strc-doc.lua
index fff357574..7532ac63f 100644
--- a/tex/context/base/strc-doc.lua
+++ b/tex/context/base/strc-doc.lua
@@ -595,6 +595,9 @@ function sections.typesetnumber(entry,kind,...) -- kind='section','number','pref
         local firstprefix, lastprefix = 0, 16
         if segments then
             local f, l = match(tostring(segments),"^(.-):(.+)$")
+            if l == "*" then
+                l = 100 -- new
+            end
             if f and l then
                 -- 0:100, chapter:subsubsection
                 firstprefix = tonumber(f) or sections.getlevel(f) or 0
diff --git a/tex/context/base/strc-flt.mkiv b/tex/context/base/strc-flt.mkiv
index 3a6aea0ab..219ff812a 100644
--- a/tex/context/base/strc-flt.mkiv
+++ b/tex/context/base/strc-flt.mkiv
@@ -912,7 +912,7 @@
    \flushsidefloats % hoort eigenlijk niet hier
    \docomplexplacefloat[#1][\v!text,#2,\v!left][#3]}
 
-\def\docomplexreserveblock#1#2%
+\def\dodocomplexreserveblock#1#2%
   {\begingroup
    \edef\currentfloat{#1}%
    \let\currentfloatcaption\currentfloat
@@ -921,7 +921,7 @@
    \endgroup}
 
 \long\def\docomplexreserveblock[#1][#2][#3][#4]#5%
-  {\getvalue{\e!place#1}[#3][#4]{#5}{\docomplexreserveblock{#1}{#2}{#1}}}
+  {\getvalue{\e!place#1}[#3][#4]{#5}{\dodocomplexreserveblock{#1}{#2}}}
 
 \def\docomplexstartreservetextblock[#1][#2][#3][#4]%
   {\flushsidefloats % hoort eigenlijk niet hier
diff --git a/tex/context/base/strc-reg.mkiv b/tex/context/base/strc-reg.mkiv
index 0467eac96..4871b40ce 100644
--- a/tex/context/base/strc-reg.mkiv
+++ b/tex/context/base/strc-reg.mkiv
@@ -706,9 +706,9 @@
 \def\withregisterpagecommand#1#2#3#4%
   {\def\currentregisterpageindex{#2}%
    \iflocation
-     \goto{\registerparameter\c!pagecommand{#4}}[internal(#2)]%
+     \goto{\applyprocessor{#1}{\registerparameter\c!pagecommand{#4}}}[internal(#2)]%
    \else
-     \registerparameter\c!pagecommand{#4}%
+     \applyprocessor{#1}{\registerparameter\c!pagecommand{#4}}%
    \fi}
 
 \def\registeronepage#1#2#3#4% #1:processor content
diff --git a/tex/context/base/strc-ren.mkiv b/tex/context/base/strc-ren.mkiv
index 0d82b7f6f..11a2ee2b3 100644
--- a/tex/context/base/strc-ren.mkiv
+++ b/tex/context/base/strc-ren.mkiv
@@ -25,12 +25,6 @@
 
 \newevery \everyheadstart \relax
 
-% \unexpanded\def\placeheadmargintexts
-%   {\the\everyheadstart
-%    \doif{\headparameter\c!margintext}\v!yes\placemargincontent}
-
-\let\placeheadmargintexts\relax
-
 \unexpanded\def\setupheadcomponentfont#1#2%
   {\dontconvertfont
    \ifconditional\headisdisplay
@@ -77,9 +71,8 @@
   {\begingroup
      \setupheadcomponentfont\c!textstyle\c!textcolor
      \headparameter\c!commandbefore
-     \placeheadmargintexts
-     \ifcsname\??nh\currenthead\c!deeptextcommand\endcsname
-       \expandafter\let\expandafter\deepstructuretitlecommand\csname\??nh\currenthead\c!deeptextcommand\endcsname
+     \ifcsname\currentheadhash\c!deeptextcommand\endcsname
+       \expandafter\let\expandafter\deepstructuretitlecommand\csname\currentheadhash\c!deeptextcommand\endcsname
      \fi
      \ifconditional\headisdisplay % \ifdisplaysectionhead
        % struts can be nilled with \setnostrut
@@ -100,9 +93,8 @@
 \def\doplaceheadnumbercomponent#1%
   {\begingroup
      \setupheadcomponentfont\c!numberstyle\c!numbercolor
-     \placeheadmargintexts
-     \ifcsname\??nh\currenthead\c!deepnumbercommand\endcsname
-       \expandafter\let\expandafter\deepstructurenumbercommand\csname\??nh\currenthead\c!deepnumbercommand\endcsname
+     \ifcsname\currentheadhash\c!deepnumbercommand\endcsname
+       \expandafter\let\expandafter\deepstructurenumbercommand\csname\currentheadhash\c!deepnumbercommand\endcsname
      \fi
      \ifconditional\headisdisplay % \ifdisplaysectionhead
        % can be nilled with \setnostrut
@@ -149,7 +141,7 @@
       % maybe auto: backreference when given, else list
    \fi\fi}
 
-\unexpanded\def\placeheadtext
+\unexpanded\def\placecurrentheadtext
   {\beginheadplacement
    \setheadmarking
    \doresettructureheadnumbercontent
@@ -161,13 +153,13 @@
      \setbox\sectionheadbox\ifvertical\vbox\else\hbox\fi \headreferenceattributes
        {\dosetlocalgridsnapping{\headparameter\c!internalgrid}%
         \doresettructureheadnumbercontent
-        \dosetfontattribute{\??nh\currenthead}\c!style % but we don't want color to influence user command, todo: get the if-else out of it
+        \dosetheadfontattribute\c!style
         \setinlineheadreferenceattributes
         \headparameter\c!command{}{\doplaceheadtextcomponent\getheadtitle}}%
    \fi
    \endheadplacement{\getheadsyncs}}
 
-\unexpanded\def\placeheadnumbertext
+\unexpanded\def\placecurrentheadnumbertext
   {\beginheadplacement
    \setheadmarking
    \doiftextelse{\getheadnumber}\dosettructureheadnumbercontent\doresettructureheadnumbercontent
@@ -178,13 +170,13 @@
      \docheckheadreference
      \setbox\sectionheadbox\ifvertical\vbox\else\hbox\fi \headreferenceattributes
        {\dosetlocalgridsnapping{\headparameter\c!internalgrid}%
-        \dosetfontattribute{\??nh\currenthead}\c!style
+        \dosetheadfontattribute\c!style
         \setinlineheadreferenceattributes
         \headparameter\c!command{\doplaceheadnumbercomponent\getheadnumber}{\doplaceheadtextcomponent\getheadtitle}}%
    \fi
    \endheadplacement{\getheadsyncs}}
 
-\unexpanded\def\placeheadempty
+\unexpanded\def\placecurrentheadempty
   {\hbox \headreferenceattributes {\getheadsyncs}}
 
 %D \starttyping
@@ -461,7 +453,7 @@
       \fi
       #2}}
 
-\unexpanded\def\placeheadmargin#1#2%
+\unexpanded\def\placeheadmarginalternative#1#2%
   {\vbox
      {\localheadsetup
       \begstrut % use one \strut here!
@@ -473,8 +465,8 @@
       \fi
       {#2}}}
 
-\defineheadplacement[\v!inmargin][\v!vertical]#1#2{\placeheadmargin{#1}{#2}}
-\defineheadplacement[\v!margin]  [\v!vertical]#1#2{\placeheadmargin{#1}{#2}}
+\defineheadplacement[\v!inmargin][\v!vertical]#1#2{\placeheadmarginalternative{#1}{#2}}
+\defineheadplacement[\v!margin]  [\v!vertical]#1#2{\placeheadmarginalternative{#1}{#2}}
 
 \defineheadplacement[\v!middle][\v!vertical]#1#2%
   {\vbox
@@ -496,7 +488,7 @@
    {\begstrut#2}%
    \egroup}
 
-\unexpanded\def\placeheadlohi#1#2#3%
+\unexpanded\def\placeheadlohialternative#1#2#3%
   {\ifconditional\headshownumber
      \setbox0\hbox{#2}
      \setbox2=#1{\localheadsetup\advance\hsize-\wd0\relax#3}%
@@ -508,8 +500,8 @@
 % onder/boven lijnt het nummer op de onderste/bovenste regel
 % uit van een meerregelige kop
 
-\defineheadplacement[\v!bottom][\v!vertical]#1#2{\placeheadlohi\vbox{#1}{#2}}
-\defineheadplacement[\v!top]   [\v!vertical]#1#2{\placeheadlohi\vtop{#1}{#2}}
+\defineheadplacement[\v!bottom][\v!vertical]#1#2{\placeheadlohialternative\vbox{#1}{#2}}
+\defineheadplacement[\v!top]   [\v!vertical]#1#2{\placeheadlohialternative\vtop{#1}{#2}}
 
 % helpers
 
diff --git a/tex/context/base/strc-sec.mkiv b/tex/context/base/strc-sec.mkiv
index a2f011941..d2836d6eb 100644
--- a/tex/context/base/strc-sec.mkiv
+++ b/tex/context/base/strc-sec.mkiv
@@ -426,8 +426,8 @@
      [#3]%
    \reportcurrentstructure}
 
-\unexpanded\def\placeheadtext  {\dosingleempty\doplaceheadtext  } % use with care
-\unexpanded\def\placeheadnumber{\dosingleempty\doplaceheadnumber} % use with care
+\unexpanded\def\placeheadtext  {\dosingleempty\place_head_text  } % use with care
+\unexpanded\def\placeheadnumber{\dosingleempty\place_head_number} % use with care
 
 \ifdefined\setupheadcomponentfont \else
 
@@ -439,7 +439,7 @@
 
 \fi
 
-\def\doplaceheadtext[#1]%
+\def\place_head_text[#1]%
   {\dontleavehmode
    \begingroup
    \edef\currenthead{#1}%
@@ -449,7 +449,7 @@
    \endgraf
    \endgroup}
 
-\def\doplaceheadnumber[#1]%
+\def\place_head_number[#1]%
   {\dontleavehmode
    \begingroup
    \edef\currenthead{#1}%
@@ -504,20 +504,20 @@
        \let\getheadtitle\fullheadtitle
        \ifconditional\headshownumber
          \let\getheadnumber\fullheadnumber
-         \placeheadnumbertext
+         \placecurrentheadnumbertext
        \else
-         \placeheadtext
+         \placecurrentheadtext
        \fi
        \doheadspacingafteryes
      \else\ifconditional\headhidden
        \doregisterhead\currenthead{#2}{#3}% after optional \page
        \let\getheadsyncs\theheadsynchonization
-       \placeheadhidden % only something when tracing
+       \placecurrentheadhidden % only something when tracing
      \else
        \doheadspacingbeforenop % toegevoegd ivm subpaginanr / tug sheets
        \doregisterhead\currenthead{#2}{#3}% after optional \page
        \let\getheadsyncs\theheadsynchonization
-       \placeheadempty % just flush 'm
+       \placecurrentheadempty % just flush 'm
        \doheadspacingafternop
      \fi\fi
    \else
@@ -526,18 +526,18 @@
        \doregisterhead\currenthead{#2}{#3}% after optional \page
        \let\getheadsyncs\theheadsynchonization
        \let\getheadtitle\fullheadtitle
-       \placeheadtext
+       \placecurrentheadtext
        \doheadspacingafteryes
      \else\ifconditional\headhidden
        \doregisterhead\currenthead{#2}{#3}% after optional \page
        \let\getheadsyncs\theheadsynchonization
-       \placeheadhidden % only something when tracing
+       \placecurrentheadhidden % only something when tracing
      \else
        % do nothing / should be vbox to 0pt
        \doheadspacingbeforenop
        \doregisterhead\currenthead{#2}{#3}% after optional \page
        \let\getheadsyncs\theheadsynchonization
-       \placeheadempty % just flush 'm
+       \placecurrentheadempty % just flush 'm
        \doheadspacingafternop
      \fi\fi
    \fi
@@ -553,21 +553,21 @@
 
 % typesetting
 
-\unexpanded\def\placeheadnumbertext % dummy, will be overloaded
+\unexpanded\def\placecurrentheadnumbertext
   {\setheadmarking
    \getheadnumber/\getheadtitle
    \getheadsyncs}
 
-\unexpanded\def\placeheadtext % dummy, will be overloaded
+\unexpanded\def\placecurrentheadtext
   {\setheadmarking
    \getheadtitle
    \getheadsyncs}
 
-\unexpanded\def\placeheadempty % dummy, will be overloaded
+\unexpanded\def\placecurrentheadempty
   {\setheadmarking
    \getheadsyncs}
 
-\unexpanded\def\placeheadhidden
+\unexpanded\def\placecurrentheadhidden
   {\setxvalue{\currenthead:sync}%
      {\noexpand\setgvalue{\currenthead:sync}{}%
       \noexpand\pagetype[\currentheadcoupling]% hm also number
diff --git a/tex/context/interface/cont-cs.xml b/tex/context/interface/cont-cs.xml
index 03e68bbc5..7af35cd94 100644
--- a/tex/context/interface/cont-cs.xml
+++ b/tex/context/interface/cont-cs.xml
@@ -13,6 +13,8 @@
 
 <cd:interface xmlns:cd="http://www.pragma-ade.com/commands" name="context test" language="cs" version="2006.08.02">
 
+  <!-- there are some more -->
+
   <cd:define name="align">
     <cd:constant type="uvnitr"/>
     <cd:constant type="vnejsi"/>
@@ -25,6 +27,10 @@
     <cd:constant type="normalni"/>
     <cd:constant type="ne"/>
     <cd:constant type="ano"/>
+    <cd:constant type="siroky"/>
+    <cd:constant type="posledni"/>
+    <cd:constant type="r2l"/>
+    <cd:constant type="l2r"/>
   </cd:define>
 
   <cd:define name="symalign">
diff --git a/tex/context/interface/cont-de.xml b/tex/context/interface/cont-de.xml
index 659fdf3f0..7d164e861 100644
--- a/tex/context/interface/cont-de.xml
+++ b/tex/context/interface/cont-de.xml
@@ -13,6 +13,8 @@
 
 <cd:interface xmlns:cd="http://www.pragma-ade.com/commands" name="context test" language="de" version="2006.08.02">
 
+  <!-- there are some more -->
+
   <cd:define name="align">
     <cd:constant type="innen"/>
     <cd:constant type="aussen"/>
@@ -25,6 +27,10 @@
     <cd:constant type="normal"/>
     <cd:constant type="nein"/>
     <cd:constant type="ja"/>
+    <cd:constant type="breit"/>
+    <cd:constant type="letzte"/>
+    <cd:constant type="r2l"/>
+    <cd:constant type="l2r"/>
   </cd:define>
 
   <cd:define name="symalign">
diff --git a/tex/context/interface/cont-en.xml b/tex/context/interface/cont-en.xml
index 46568857a..8cc1bac6c 100644
--- a/tex/context/interface/cont-en.xml
+++ b/tex/context/interface/cont-en.xml
@@ -13,6 +13,8 @@
 
 <cd:interface xmlns:cd="http://www.pragma-ade.com/commands" name="context test" language="en" version="2006.08.02">
 
+  <!-- there are some more -->
+
   <cd:define name="align">
     <cd:constant type="inner"/>
     <cd:constant type="outer"/>
@@ -25,6 +27,10 @@
     <cd:constant type="normal"/>
     <cd:constant type="no"/>
     <cd:constant type="yes"/>
+    <cd:constant type="broad"/>
+    <cd:constant type="last"/>
+    <cd:constant type="r2l"/>
+    <cd:constant type="l2r"/>
   </cd:define>
 
   <cd:define name="symalign">
diff --git a/tex/context/interface/cont-fr.xml b/tex/context/interface/cont-fr.xml
index 081ea1802..7b109fd68 100644
--- a/tex/context/interface/cont-fr.xml
+++ b/tex/context/interface/cont-fr.xml
@@ -13,6 +13,8 @@
 
 <cd:interface xmlns:cd="http://www.pragma-ade.com/commands" name="context test" language="fr" version="2006.08.02">
 
+  <!-- there are some more -->
+
   <cd:define name="align">
     <cd:constant type="interieur"/>
     <cd:constant type="exterieur"/>
@@ -25,6 +27,10 @@
     <cd:constant type="normal"/>
     <cd:constant type="non"/>
     <cd:constant type="oui"/>
+    <cd:constant type="large"/>
+    <cd:constant type="dernier"/>
+    <cd:constant type="r2l"/>
+    <cd:constant type="l2r"/>
   </cd:define>
 
   <cd:define name="symalign">
diff --git a/tex/context/interface/cont-it.xml b/tex/context/interface/cont-it.xml
index b739b60cb..e8574d2f6 100644
--- a/tex/context/interface/cont-it.xml
+++ b/tex/context/interface/cont-it.xml
@@ -13,6 +13,8 @@
 
 <cd:interface xmlns:cd="http://www.pragma-ade.com/commands" name="context test" language="it" version="2006.08.02">
 
+  <!-- there are some more -->
+
   <cd:define name="align">
     <cd:constant type="interno"/>
     <cd:constant type="esterno"/>
@@ -25,6 +27,10 @@
     <cd:constant type="normale"/>
     <cd:constant type="no"/>
     <cd:constant type="si"/>
+    <cd:constant type="ampio"/>
+    <cd:constant type="ultimo"/>
+    <cd:constant type="r2l"/>
+    <cd:constant type="l2r"/>
   </cd:define>
 
   <cd:define name="symalign">
diff --git a/tex/context/interface/cont-nl.xml b/tex/context/interface/cont-nl.xml
index b3036a5a9..cb585d771 100644
--- a/tex/context/interface/cont-nl.xml
+++ b/tex/context/interface/cont-nl.xml
@@ -13,6 +13,8 @@
 
 <cd:interface xmlns:cd="http://www.pragma-ade.com/commands" name="context test" language="nl" version="2006.08.02">
 
+  <!-- there are some more -->
+
   <cd:define name="align">
     <cd:constant type="binnen"/>
     <cd:constant type="buiten"/>
@@ -25,6 +27,10 @@
     <cd:constant type="normaal"/>
     <cd:constant type="nee"/>
     <cd:constant type="ja"/>
+    <cd:constant type="ruim"/>
+    <cd:constant type="laatste"/>
+    <cd:constant type="r2l"/>
+    <cd:constant type="l2r"/>
   </cd:define>
 
   <cd:define name="symalign">
diff --git a/tex/context/interface/cont-pe.xml b/tex/context/interface/cont-pe.xml
index 956087029..e271cb846 100644
--- a/tex/context/interface/cont-pe.xml
+++ b/tex/context/interface/cont-pe.xml
@@ -13,6 +13,8 @@
 
 <cd:interface xmlns:cd="http://www.pragma-ade.com/commands" name="context test" language="pe" version="2006.08.02">
 
+  <!-- there are some more -->
+
   <cd:define name="align">
     <cd:constant type="داخلی"/>
     <cd:constant type="خارجی"/>
@@ -25,6 +27,10 @@
     <cd:constant type="نرمال"/>
     <cd:constant type="نه"/>
     <cd:constant type="بله"/>
+    <cd:constant type="گسترده"/>
+    <cd:constant type="آخرین"/>
+    <cd:constant type="r2l"/>
+    <cd:constant type="l2r"/>
   </cd:define>
 
   <cd:define name="symalign">
diff --git a/tex/context/interface/cont-ro.xml b/tex/context/interface/cont-ro.xml
index d91aa40f1..dc198d252 100644
--- a/tex/context/interface/cont-ro.xml
+++ b/tex/context/interface/cont-ro.xml
@@ -13,6 +13,8 @@
 
 <cd:interface xmlns:cd="http://www.pragma-ade.com/commands" name="context test" language="ro" version="2006.08.02">
 
+  <!-- there are some more -->
+
   <cd:define name="align">
     <cd:constant type="intern"/>
     <cd:constant type="extern"/>
@@ -25,6 +27,10 @@
     <cd:constant type="normal"/>
     <cd:constant type="nu"/>
     <cd:constant type="da"/>
+    <cd:constant type="broad"/>
+    <cd:constant type="ultim"/>
+    <cd:constant type="r2l"/>
+    <cd:constant type="l2r"/>
   </cd:define>
 
   <cd:define name="symalign">
diff --git a/tex/generic/context/luatex/luatex-fonts-merged.lua b/tex/generic/context/luatex/luatex-fonts-merged.lua
index 28ac0347a..b60a0b79b 100644
--- a/tex/generic/context/luatex/luatex-fonts-merged.lua
+++ b/tex/generic/context/luatex/luatex-fonts-merged.lua
@@ -1,6 +1,6 @@
 -- merged file : luatex-fonts-merged.lua
 -- parent file : luatex-fonts.lua
--- merge date  : 09/19/11 23:43:44
+-- merge date  : 09/27/11 20:05:02
 
 do -- begin closure to overcome local limits and interference
 
@@ -3520,21 +3520,35 @@ function constructors.scale(tfmdata,specification)
     local sharedkerns   = { }
     --
     for unicode, character in next, characters do
-        local chr, description, index
+        local chr, description, index, touni
         if changed then
-            -- basemode hack
+            -- basemode hack (we try to catch missing tounicodes, e.g. needed for ssty in math cambria)
             local c = changed[unicode]
             if c then
-                description = descriptions[c] or character
+                description = descriptions[c] or descriptions[unicode] or character
                 character = characters[c] or character
                 index = description.index or c
+                if tounicode then
+                    touni = tounicode[index] -- nb: index!
+                    if not touni then -- goodie
+                        local d = descriptions[unicode] or characters[unicode]
+                        local i = d.index or unicode
+                        touni = tounicode[i] -- nb: index!
+                    end
+                end
             else
                 description = descriptions[unicode] or character
                 index = description.index or unicode
+                if tounicode then
+                    touni = tounicode[index] -- nb: index!
+                end
             end
         else
             description = descriptions[unicode] or character
             index = description.index or unicode
+            if tounicode then
+                touni = tounicode[index] -- nb: index!
+            end
         end
         local width  = description.width
         local height = description.height
@@ -3577,15 +3591,12 @@ function constructors.scale(tfmdata,specification)
                 }
             end
         end
+        if touni then
+            chr.tounicode = touni
+        end
     --  if trace_scaling then
     --    report_defining("t=%s, u=%s, i=%s, n=%s c=%s",k,chr.tounicode or "",index or 0,description.name or '-',description.class or '-')
     --  end
-        if tounicode then
-            local tu = tounicode[index] -- nb: index!
-            if tu then
-                chr.tounicode = tu
-            end
-        end
         if hasquality then
             -- we could move these calculations elsewhere (saves calculations)
             local ve = character.expansion_factor
@@ -4569,6 +4580,15 @@ local function tounicode16sequence(unicodes)
     return concat(t)
 end
 
+local function fromunicode16(str)
+    if #str == 4 then
+        return tonumber(str,16)
+    else
+        local l, r = match(str,"(....)(....)")
+        return (tonumber(l,16)- 0xD800)*0x400  + tonumber(r,16) - 0xDC00
+    end
+end
+
 --~ This is quite a bit faster but at the cost of some memory but if we
 --~ do this we will also use it elsewhere so let's not follow this route
 --~ now. I might use this method in the plain variant (no caching there)
@@ -4593,6 +4613,7 @@ mappings.loadlumtable        = loadlumtable
 mappings.makenameparser      = makenameparser
 mappings.tounicode16         = tounicode16
 mappings.tounicode16sequence = tounicode16sequence
+mappings.fromunicode16       = fromunicode16
 
 local separator   = S("_.")
 local other       = C((1 - separator)^1)
@@ -4648,7 +4669,9 @@ function mappings.addtounicode(data,filename)
         if unic == -1 or unic >= private or (unic >= 0xE000 and unic <= 0xF8FF) or unic == 0xFFFE or unic == 0xFFFF then
             local unicode = lumunic and lumunic[name] or unicodevector[name]
             if unicode then
-                originals[index], tounicode[index], ns = unicode, tounicode16(unicode), ns + 1
+                originals[index] = unicode
+                tounicode[index] = tounicode16(unicode)
+                ns               = ns + 1
             end
             -- cidmap heuristics, beware, there is no guarantee for a match unless
             -- the chain resolves
@@ -4657,7 +4680,9 @@ function mappings.addtounicode(data,filename)
                 if foundindex then
                     unicode = cidcodes[foundindex] -- name to number
                     if unicode then
-                        originals[index], tounicode[index], ns = unicode, tounicode16(unicode), ns + 1
+                        originals[index] = unicode
+                        tounicode[index] = tounicode16(unicode)
+                        ns               = ns + 1
                     else
                         local reference = cidnames[foundindex] -- number to name
                         if reference then
@@ -4665,16 +4690,23 @@ function mappings.addtounicode(data,filename)
                             if foundindex then
                                 unicode = cidcodes[foundindex]
                                 if unicode then
-                                    originals[index], tounicode[index], ns = unicode, tounicode16(unicode), ns + 1
+                                    originals[index] = unicode
+                                    tounicode[index] = tounicode16(unicode)
+                                    ns               = ns + 1
                                 end
                             end
                             if not unicode then
                                 local foundcodes, multiple = lpegmatch(uparser,reference)
                                 if foundcodes then
+                                    originals[index] = foundcodes
                                     if multiple then
-                                        originals[index], tounicode[index], nl, unicode = foundcodes, tounicode16sequence(foundcodes), nl + 1, true
+                                        tounicode[index] = tounicode16sequence(foundcodes)
+                                        nl               = nl + 1
+                                        unicode          = true
                                     else
-                                        originals[index], tounicode[index], ns, unicode = foundcodes, tounicode16(foundcodes), ns + 1, foundcodes
+                                        tounicode[index] = tounicode16(foundcodes)
+                                        ns               = ns + 1
+                                        unicode          = foundcodes
                                     end
                                 end
                             end
@@ -4685,19 +4717,8 @@ function mappings.addtounicode(data,filename)
             -- a.whatever or a_b_c.whatever or a_b_c (no numbers)
             if not unicode then
                 local split = lpegmatch(ligsplitter,name)
-                local nplit = (split and #split) or 0
-                if nplit == 0 then
-                    -- skip
-                elseif nplit == 1 then
-                    local base = split[1]
-                    unicode = unicodes[base] or unicodevector[base]
-                    if unicode then
-                        if type(unicode) == "table" then
-                            unicode = unicode[1]
-                        end
-                        originals[index], tounicode[index], ns = unicode, tounicode16(unicode), ns + 1
-                    end
-                else
+                local nplit = split and #split or 0
+                if nplit >= 2 then
                     local t, n = { }, 0
                     for l=1,nplit do
                         local base = split[l]
@@ -4715,25 +4736,38 @@ function mappings.addtounicode(data,filename)
                     if n == 0 then -- done then
                         -- nothing
                     elseif n == 1 then
-                        originals[index], tounicode[index], nl, unicode = t[1], tounicode16(t[1]), nl + 1, true
+                        originals[index] = t[1]
+                        tounicode[index] = tounicode16(t[1])
                     else
-                        originals[index], tounicode[index], nl, unicode = t, tounicode16sequence(t), nl + 1, true
+                        originals[index] = t
+                        tounicode[index] = tounicode16sequence(t)
                     end
+                    nl = nl + 1
+                    unicode = true
+                else
+                    -- skip: already checked and we don't want privates here
                 end
             end
-            -- last resort
+            -- last resort (we might need to catch private here as well)
             if not unicode then
                 local foundcodes, multiple = lpegmatch(uparser,name)
                 if foundcodes then
                     if multiple then
-                        originals[index], tounicode[index], nl, unicode = foundcodes, tounicode16sequence(foundcodes), nl + 1, true
+                        originals[index] = foundcodes
+                        tounicode[index] = tounicode16sequence(foundcodes)
+                        nl               = nl + 1
+                        unicode          = true
                     else
-                        originals[index], tounicode[index], ns, unicode = foundcodes, tounicode16(foundcodes), ns + 1, foundcodes
+                        originals[index] = foundcodes
+                        tounicode[index] = tounicode16(foundcodes)
+                        ns               = ns + 1
+                        unicode          = foundcodes
                     end
                 end
             end
          -- if not unicode then
-         --     originals[index], tounicode[index] = 0xFFFD, "FFFD"
+         --     originals[index] = 0xFFFD
+         --     tounicode[index] = "FFFD"
          -- end
         end
     end