2017-05-25 12:56:00

1 files changed, 242 insertions, 220 deletions

diff --git a/tex/context/base/mkiv/font-ots.lua b/tex/context/base/mkiv/font-ots.lua
index 16c2ce735..1230475b3 100644
--- a/tex/context/base/mkiv/font-ots.lua
+++ b/tex/context/base/mkiv/font-ots.lua
@@ -155,11 +155,10 @@ local report_process     = logs.reporter("fonts","otf process")
 local report_warning     = logs.reporter("fonts","otf warning")
 local report_run         = logs.reporter("fonts","otf run")
 
-registertracker("otf.replacements", "otf.singles,otf.multiples,otf.alternatives,otf.ligatures")
-registertracker("otf.positions","otf.marks,otf.kerns,otf.cursive")
-registertracker("otf.actions","otf.replacements,otf.positions")
-registertracker("otf.injections","nodes.injections")
-registertracker("otf.sample","otf.steps,otf.actions,otf.analyzing")
+registertracker("otf.substitutions", "otf.singles","otf.multiples","otf.alternatives","otf.ligatures")
+registertracker("otf.positions",     "otf.marks","otf.kerns","otf.cursive")
+registertracker("otf.actions",       "otf.substitutions","otf.positions")
+registertracker("otf.sample",        "otf.steps","otf.substitutions","otf.positions","otf.analyzing")
 
 local nuts               = nodes.nuts
 local tonode             = nuts.tonode
@@ -192,6 +191,7 @@ local getdir             = nuts.getdir
 local getwidth           = nuts.getwidth
 
 local ischar             = nuts.is_char
+local usesfont           = nuts.uses_font
 
 local insert_node_after  = nuts.insert_after
 local copy_node          = nuts.copy
@@ -932,7 +932,8 @@ end
 
 function handlers.gpos_single(head,start,dataset,sequence,kerns,rlmode,step,i,injection)
     local startchar = getchar(start)
-    if step.format == "pair" then
+--     if step.format == "pair" then
+    if step.format == "pair" or type(kerns) == "table" then
         local dx, dy, w, h = setpair(start,factor,rlmode,sequence.flags[4],kerns,injection)
         if trace_kerns then
             logprocess("%s: shifting single %s by (%p,%p) and correction (%p,%p)",pref(dataset,sequence),gref(startchar),dx,dy,w,h)
@@ -3830,132 +3831,213 @@ otf.helpers.pardirstate = pardirstate
 -- optimizations the principles of processing the features hasn't changed much since
 -- the beginning.
 
-local function featuresprocessor(head,font,attr,direction)
+do
 
-    local sequences = sequencelists[font] -- temp hack
+    -- experimental speedup (only with hyphenated text and multiple fonts per processing)
+    --
+    -- at some point this might become true by default
 
-    if not sequencelists then
-        return head, false
-    end
+    local fastdisc = false  directives.register("otf.fastdisc",function(v) fastdisc = v end)
 
-    nesting = nesting + 1
+    function otf.featuresprocessor(head,font,attr,direction,n)
 
-    if nesting == 1 then
-        currentfont  = font
-        tfmdata      = fontdata[font]
-        descriptions = tfmdata.descriptions -- only needed in gref so we could pass node there instead
-        characters   = tfmdata.characters   -- but this branch is not entered that often anyway
-  local resources    = tfmdata.resources
-        marks        = resources.marks
-        classes      = resources.classes
-        threshold,
-        factor       = getthreshold(font)
-        checkmarks   = tfmdata.properties.checkmarks
+        local sequences = sequencelists[font] -- temp hack
 
-    elseif currentfont ~= font then
+        if not sequencelists then
+            return head, false
+        end
 
-        report_warning("nested call with a different font, level %s, quitting",nesting)
-        nesting = nesting - 1
-        return head, false
+        nesting = nesting + 1
 
-    end
+        if nesting == 1 then
+            currentfont  = font
+            tfmdata      = fontdata[font]
+            descriptions = tfmdata.descriptions -- only needed in gref so we could pass node there instead
+            characters   = tfmdata.characters   -- but this branch is not entered that often anyway
+      local resources    = tfmdata.resources
+            marks        = resources.marks
+            classes      = resources.classes
+            threshold,
+            factor       = getthreshold(font)
+            checkmarks   = tfmdata.properties.checkmarks
 
-    -- some 10% faster when no dynamics but hardly measureable on real runs .. but: it only
-    -- works when we have no other dynamics as otherwise the zero run will be applied to the
-    -- whole stream for which we then need to pass another variable which we won't
+        elseif currentfont ~= font then
 
-    -- if attr == 0 then
-    --     attr = false
-    -- end
+            report_warning("nested call with a different font, level %s, quitting",nesting)
+            nesting = nesting - 1
+            return head, false
 
-    head = tonut(head)
+        end
 
-    if trace_steps then
-        checkstep(head)
-    end
+        -- some 10% faster when no dynamics but hardly measureable on real runs .. but: it only
+        -- works when we have no other dynamics as otherwise the zero run will be applied to the
+        -- whole stream for which we then need to pass another variable which we won't
 
-    local initialrl = direction == "TRT" and -1 or 0
+        -- if attr == 0 then
+        --     attr = false
+        -- end
 
-    local done      = false
-    local datasets  = otf.dataset(tfmdata,font,attr)
-    local dirstack  = { } -- could move outside function but we can have local runs
-    sweephead       = { }
-
-    -- Keeping track of the headnode is needed for devanagari. (I generalized it a bit
-    -- so that multiple cases are also covered.) We could prepend a temp node.
-
-    -- We don't goto the next node when a disc node is created so that we can then treat
-    -- the pre, post and replace. It's a bit of a hack but works out ok for most cases.
-
-    for s=1,#datasets do
-        local dataset      = datasets[s]
-        ----- featurevalue = dataset[1] -- todo: pass to function instead of using a global
-        local attribute    = dataset[2]
-        local sequence     = dataset[3] -- sequences[s] -- also dataset[5]
-        local rlparmode    = initialrl
-        local topstack     = 0
-        local typ          = sequence.type
-        local gpossing     = typ == "gpos_single" or typ == "gpos_pair" -- store in dataset
-        local handler      = handlers[typ]
-        local steps        = sequence.steps
-        local nofsteps     = sequence.nofsteps
-        if not steps then
-            -- this permits injection, watch the different arguments
-            local h, d, ok = handler(head,head,dataset,sequence,nil,nil,nil,0,font,attr)
-            if ok then
-                done = true
-                if h then
-                    head = h
-                end
-            end
-        elseif typ == "gsub_reversecontextchain" then
-            -- this is a limited case, no special treatments like 'init' etc
-            local start  = find_node_tail(head)
-            local rlmode = 0 -- how important is this .. do we need to check for dir?
-            while start do
-                local char = ischar(start,font)
-                if char then
-                    local a -- happens often so no assignment is faster
-                    if attr then
-                        a = getattr(start,0)
+        head = tonut(head)
+
+        if trace_steps then
+            checkstep(head)
+        end
+
+        local initialrl = direction == "TRT" and -1 or 0
+
+        local done      = false
+        local datasets  = otf.dataset(tfmdata,font,attr)
+        local dirstack  = { } -- could move outside function but we can have local runs
+        sweephead       = { }
+
+        -- Keeping track of the headnode is needed for devanagari. (I generalized it a bit
+        -- so that multiple cases are also covered.) We could prepend a temp node.
+
+        -- We don't goto the next node when a disc node is created so that we can then treat
+        -- the pre, post and replace. It's a bit of a hack but works out ok for most cases.
+
+        local discs = fastdisc and n and n > 1 and setmetatableindex(function(t,k)
+            local v = usesfont(k,font)
+            t[k] = v
+            return v
+        end)
+
+        for s=1,#datasets do
+            local dataset      = datasets[s]
+            ----- featurevalue = dataset[1] -- todo: pass to function instead of using a global
+            local attribute    = dataset[2]
+            local sequence     = dataset[3] -- sequences[s] -- also dataset[5]
+            local rlparmode    = initialrl
+            local topstack     = 0
+            local typ          = sequence.type
+            local gpossing     = typ == "gpos_single" or typ == "gpos_pair" -- store in dataset
+            local handler      = handlers[typ]
+            local steps        = sequence.steps
+            local nofsteps     = sequence.nofsteps
+            if not steps then
+                -- this permits injection, watch the different arguments
+                local h, d, ok = handler(head,head,dataset,sequence,nil,nil,nil,0,font,attr)
+                if ok then
+                    done = true
+                    if h then
+                        head = h
                     end
-                    if not a or (a == attr) then
-                        for i=1,nofsteps do
-                            local step = steps[i]
-                            local lookupcache = step.coverage
-                            if lookupcache then
-                                local lookupmatch = lookupcache[char]
-                                if lookupmatch then
-                                    -- todo: disc?
-                                    local ok
-                                    head, start, ok = handler(head,start,dataset,sequence,lookupmatch,rlmode,step,i)
-                                    if ok then
-                                        done = true
-                                        break
+                end
+            elseif typ == "gsub_reversecontextchain" then
+                -- this is a limited case, no special treatments like 'init' etc
+                local start  = find_node_tail(head)
+                local rlmode = 0 -- how important is this .. do we need to check for dir?
+                while start do
+                    local char = ischar(start,font)
+                    if char then
+                        local a -- happens often so no assignment is faster
+                        if attr then
+                            a = getattr(start,0)
+                        end
+                        if not a or (a == attr) then
+                            for i=1,nofsteps do
+                                local step = steps[i]
+                                local lookupcache = step.coverage
+                                if lookupcache then
+                                    local lookupmatch = lookupcache[char]
+                                    if lookupmatch then
+                                        -- todo: disc?
+                                        local ok
+                                        head, start, ok = handler(head,start,dataset,sequence,lookupmatch,rlmode,step,i)
+                                        if ok then
+                                            done = true
+                                            break
+                                        end
                                     end
+                                else
+                                    report_missing_coverage(dataset,sequence)
                                 end
-                            else
-                                report_missing_coverage(dataset,sequence)
                             end
-                        end
-                        if start then
+                            if start then
+                                start = getprev(start)
+                            end
+                        else
                             start = getprev(start)
                         end
                     else
                         start = getprev(start)
                     end
-                else
-                    start = getprev(start)
                 end
-            end
-        else
-            local start  = head
-            local rlmode = initialrl
-            if nofsteps == 1 then -- happens often
-                local step = steps[1]
-                local lookupcache = step.coverage
-                if not lookupcache then
-                    report_missing_coverage(dataset,sequence)
+            else
+                local start  = head
+                local rlmode = initialrl
+                if nofsteps == 1 then -- happens often
+                    local step = steps[1]
+                    local lookupcache = step.coverage
+                    if not lookupcache then
+                        report_missing_coverage(dataset,sequence)
+                    else
+                        while start do
+                            local char, id = ischar(start,font)
+                            if char then
+                             -- local a = attr and getattr(start,0)
+                             -- if a then
+                             --     a = (a == attr) and (not attribute or getprop(start,a_state) == attribute)
+                             -- else
+                             --     a = not attribute or getprop(start,a_state) == attribute
+                             -- end
+                                local a -- happens often so no assignment is faster
+                                if attr then
+                                    if getattr(start,0) == attr and (not attribute or getprop(start,a_state) == attribute) then
+                                        a = true
+                                    end
+                                elseif not attribute or getprop(start,a_state) == attribute then
+                                    a = true
+                                end
+                                if a then
+                                    local lookupmatch = lookupcache[char]
+                                    if lookupmatch then
+                                        local ok
+                                        head, start, ok = handler(head,start,dataset,sequence,lookupmatch,rlmode,step,1)
+                                        if ok then
+                                            done = true
+                                        end
+                                    end
+                                    if start then
+                                        start = getnext(start)
+                                    end
+                                else
+                                   start = getnext(start)
+                                end
+                            elseif char == false then
+                               -- whatever glyph
+                               start = getnext(start)
+                            elseif id == glue_code then
+                                -- happens often
+                               start = getnext(start)
+                            elseif id == disc_code then
+                                if not discs or discs[start] == true then
+                                    local ok
+                                    if gpossing then
+                                        start, ok = kernrun(start,k_run_single,             font,attr,lookupcache,step,dataset,sequence,rlmode,handler)
+                                    elseif typ == "gsub_ligature" then
+                                        start, ok = testrun(start,t_run_single,c_run_single,font,attr,lookupcache,step,dataset,sequence,rlmode,handler)
+                                    else
+                                        start, ok = comprun(start,c_run_single,             font,attr,lookupcache,step,dataset,sequence,rlmode,handler)
+                                    end
+                                    if ok then
+                                        done = true
+                                    end
+                            else
+                                start = getnext(start)
+                            end
+                            elseif id == math_code then
+                                start = getnext(end_of_math(start))
+                            elseif id == dir_code then
+                                start, topstack, rlmode = txtdirstate(start,dirstack,topstack,rlparmode)
+                            elseif id == localpar_code then
+                                start, rlparmode, rlmode = pardirstate(start)
+                            else
+                                start = getnext(start)
+                            end
+                        end
+                    end
+
                 else
                     while start do
                         local char, id = ischar(start,font)
@@ -3975,37 +4057,54 @@ local function featuresprocessor(head,font,attr,direction)
                                 a = true
                             end
                             if a then
-                                local lookupmatch = lookupcache[char]
-                                if lookupmatch then
-                                    local ok
-                                    head, start, ok = handler(head,start,dataset,sequence,lookupmatch,rlmode,step,1)
-                                    if ok then
-                                        done = true
+                                for i=1,nofsteps do
+                                    local step        = steps[i]
+                                    local lookupcache = step.coverage
+                                    if lookupcache then
+                                        local lookupmatch = lookupcache[char]
+                                        if lookupmatch then
+                                            -- we could move all code inline but that makes things even more unreadable
+                                            local ok
+                                            head, start, ok = handler(head,start,dataset,sequence,lookupmatch,rlmode,step,i)
+                                            if ok then
+                                                done = true
+                                                break
+                                            elseif not start then
+                                                -- don't ask why ... shouldn't happen
+                                                break
+                                            end
+                                        end
+                                    else
+                                        report_missing_coverage(dataset,sequence)
                                     end
                                 end
                                 if start then
                                     start = getnext(start)
                                 end
                             else
-                               start = getnext(start)
+                                start = getnext(start)
                             end
                         elseif char == false then
                            -- whatever glyph
-                           start = getnext(start)
+                            start = getnext(start)
                         elseif id == glue_code then
                             -- happens often
-                           start = getnext(start)
+                            start = getnext(start)
                         elseif id == disc_code then
-                            local ok
-                            if gpossing then
-                                start, ok = kernrun(start,k_run_single,             font,attr,lookupcache,step,dataset,sequence,rlmode,handler)
-                            elseif typ == "gsub_ligature" then
-                                start, ok = testrun(start,t_run_single,c_run_single,font,attr,lookupcache,step,dataset,sequence,rlmode,handler)
+                            if not discs or discs[start] == true then
+                                local ok
+                                if gpossing then
+                                    start, ok = kernrun(start,k_run_multiple,               font,attr,steps,nofsteps,dataset,sequence,rlmode,handler)
+                                elseif typ == "gsub_ligature" then
+                                    start, ok = testrun(start,t_run_multiple,c_run_multiple,font,attr,steps,nofsteps,dataset,sequence,rlmode,handler)
+                                else
+                                    start, ok = comprun(start,c_run_multiple,               font,attr,steps,nofsteps,dataset,sequence,rlmode,handler)
+                                end
+                                if ok then
+                                    done = true
+                                end
                             else
-                                start, ok = comprun(start,c_run_single,             font,attr,lookupcache,step,dataset,sequence,rlmode,handler)
-                            end
-                            if ok then
-                                done = true
+                                start = getnext(start)
                             end
                         elseif id == math_code then
                             start = getnext(end_of_math(start))
@@ -4018,94 +4117,20 @@ local function featuresprocessor(head,font,attr,direction)
                         end
                     end
                 end
+            end
 
-            else
-                while start do
-                    local char, id = ischar(start,font)
-                    if char then
-                     -- local a = attr and getattr(start,0)
-                     -- if a then
-                     --     a = (a == attr) and (not attribute or getprop(start,a_state) == attribute)
-                     -- else
-                     --     a = not attribute or getprop(start,a_state) == attribute
-                     -- end
-                        local a -- happens often so no assignment is faster
-                        if attr then
-                            if getattr(start,0) == attr and (not attribute or getprop(start,a_state) == attribute) then
-                                a = true
-                            end
-                        elseif not attribute or getprop(start,a_state) == attribute then
-                            a = true
-                        end
-                        if a then
-                            for i=1,nofsteps do
-                                local step        = steps[i]
-                                local lookupcache = step.coverage
-                                if lookupcache then
-                                    local lookupmatch = lookupcache[char]
-                                    if lookupmatch then
-                                        -- we could move all code inline but that makes things even more unreadable
-                                        local ok
-                                        head, start, ok = handler(head,start,dataset,sequence,lookupmatch,rlmode,step,i)
-                                        if ok then
-                                            done = true
-                                            break
-                                        elseif not start then
-                                            -- don't ask why ... shouldn't happen
-                                            break
-                                        end
-                                    end
-                                else
-                                    report_missing_coverage(dataset,sequence)
-                                end
-                            end
-                            if start then
-                                start = getnext(start)
-                            end
-                        else
-                            start = getnext(start)
-                        end
-                    elseif char == false then
-                       -- whatever glyph
-                        start = getnext(start)
-                    elseif id == glue_code then
-                        -- happens often
-                        start = getnext(start)
-                    elseif id == disc_code then
-                        local ok
-                        if gpossing then
-                            start, ok = kernrun(start,k_run_multiple,               font,attr,steps,nofsteps,dataset,sequence,rlmode,handler)
-                        elseif typ == "gsub_ligature" then
-                            start, ok = testrun(start,t_run_multiple,c_run_multiple,font,attr,steps,nofsteps,dataset,sequence,rlmode,handler)
-                        else
-                            start, ok = comprun(start,c_run_multiple,               font,attr,steps,nofsteps,dataset,sequence,rlmode,handler)
-                        end
-                        if ok then
-                            done = true
-                        end
-                    elseif id == math_code then
-                        start = getnext(end_of_math(start))
-                    elseif id == dir_code then
-                        start, topstack, rlmode = txtdirstate(start,dirstack,topstack,rlparmode)
-                    elseif id == localpar_code then
-                        start, rlparmode, rlmode = pardirstate(start)
-                    else
-                        start = getnext(start)
-                    end
-                end
+            if trace_steps then -- ?
+                registerstep(head)
             end
-        end
 
-        if trace_steps then -- ?
-            registerstep(head)
         end
 
-    end
+        nesting = nesting - 1
+        head    = tonode(head)
 
-    nesting = nesting - 1
-    head    = tonode(head)
+        return head, done
+    end
 
-    return head, done
 end
 
 -- so far
@@ -4119,13 +4144,13 @@ function otf.registerplugin(name,f)
     end
 end
 
-local function plugininitializer(tfmdata,value)
+function otf.plugininitializer(tfmdata,value)
     if type(value) == "string" then
         tfmdata.shared.plugin = plugins[value]
     end
 end
 
-local function pluginprocessor(head,font)
+function otf.pluginprocessor(head,font)
     local s = fontdata[font].shared
     local p = s and s.plugin
     if p then
@@ -4138,7 +4163,7 @@ local function pluginprocessor(head,font)
     end
 end
 
-local function featuresinitializer(tfmdata,value)
+function otf.featuresinitializer(tfmdata,value)
     -- nothing done here any more
 end
 
@@ -4148,18 +4173,15 @@ registerotffeature {
     default      = true,
     initializers = {
         position = 1,
-        node     = featuresinitializer,
-        plug     = plugininitializer,
+        node     = otf.featuresinitializer,
+        plug     = otf.plugininitializer,
     },
     processors   = {
-        node     = featuresprocessor,
-        plug     = pluginprocessor,
+        node     = otf.featuresprocessor,
+        plug     = otf.pluginprocessor,
     }
 }
 
-otf.nodemodeinitializer = featuresinitializer
-otf.featuresprocessor   = featuresprocessor
-
 -- This can be used for extra handlers, but should be used with care!
 
 otf.handlers = handlers -- used in devanagari
@@ -4289,9 +4311,9 @@ local function spaceinitializer(tfmdata,value) -- attr
                                         if type(kern) ~= "table" then
                                             left[k] = kern
                                         elseif single then
-                                            left[k] = v[3]
+                                            left[k] = kern[3]
                                         else
-                                            local one = v[1]
+                                            local one = kern[1]
                                             if one then
                                                 left[k] = one[3]
                                             end