diff options
| author | Philipp Gesang <phg42.2a@gmail.com> | 2014-01-07 20:41:32 +0100 |
|---|---|---|
| committer | Philipp Gesang <phg42.2a@gmail.com> | 2014-01-07 20:41:32 +0100 |
| commit | ec9087800c135b68d08bfb54da170d675fdc4f7b (patch) | |
| tree | fdfa0f31566ecd13635f3d058012fe7c23616b59 | |
| parent | 319be72c5d5beb8d05e5a2e572d5860beb5f09f5 (diff) | |
| download | luaotfload-ec9087800c135b68d08bfb54da170d675fdc4f7b.tar.gz | |
[fontloader] import luatex-fonts-inj.lua, luatex-fonts-otn.lua
| -rw-r--r-- | luaotfload-fonts-inj.lua | 526 | ||||
| -rw-r--r-- | luaotfload-fonts-otn.lua | 2848 |
2 files changed, 3374 insertions, 0 deletions
diff --git a/luaotfload-fonts-inj.lua b/luaotfload-fonts-inj.lua new file mode 100644 index 0000000..ae48150 --- /dev/null +++ b/luaotfload-fonts-inj.lua @@ -0,0 +1,526 @@ +if not modules then modules = { } end modules ['node-inj'] = { + version = 1.001, + comment = "companion to node-ini.mkiv", + author = "Hans Hagen, PRAGMA-ADE, Hasselt NL", + copyright = "PRAGMA ADE / ConTeXt Development Team", + license = "see context related readme files", +} + +-- This is very experimental (this will change when we have luatex > .50 and +-- a few pending thingies are available. Also, Idris needs to make a few more +-- test fonts. Btw, future versions of luatex will have extended glyph properties +-- that can be of help. Some optimizations can go away when we have faster machines. + +-- todo: make a special one for context + +local next = next +local utfchar = utf.char + +local trace_injections = false trackers.register("nodes.injections", function(v) trace_injections = v end) + +local report_injections = logs.reporter("nodes","injections") + +local attributes, nodes, node = attributes, nodes, node + +fonts = fonts +local fontdata = fonts.hashes.identifiers + +nodes.injections = nodes.injections or { } +local injections = nodes.injections + +local nodecodes = nodes.nodecodes +local glyph_code = nodecodes.glyph +local kern_code = nodecodes.kern +local nodepool = nodes.pool +local newkern = nodepool.kern + +local traverse_id = node.traverse_id +local insert_node_before = node.insert_before +local insert_node_after = node.insert_after + +local a_kernpair = attributes.private('kernpair') +local a_ligacomp = attributes.private('ligacomp') +local a_markbase = attributes.private('markbase') +local a_markmark = attributes.private('markmark') +local a_markdone = attributes.private('markdone') +local a_cursbase = attributes.private('cursbase') +local a_curscurs = attributes.private('curscurs') +local a_cursdone = attributes.private('cursdone') + +-- This injector has been tested by Idris Samawi Hamid (several arabic fonts as well as +-- the rather demanding Husayni font), Khaled Hosny (latin and arabic) and Kaj Eigner +-- (arabic, hebrew and thai) and myself (whatever font I come across). I'm pretty sure +-- that this code is not 100% okay but examples are needed to figure things out. + +function injections.installnewkern(nk) + newkern = nk or newkern +end + +local cursives = { } +local marks = { } +local kerns = { } + +-- Currently we do gpos/kern in a bit inofficial way but when we have the extra fields in +-- glyphnodes to manipulate ht/dp/wd explicitly I will provide an alternative; also, we +-- can share tables. + +-- For the moment we pass the r2l key ... volt/arabtype tests .. idris: this needs +-- checking with husayni (volt and fontforge). + +function injections.setcursive(start,nxt,factor,rlmode,exit,entry,tfmstart,tfmnext) + local dx, dy = factor*(exit[1]-entry[1]), factor*(exit[2]-entry[2]) + local ws, wn = tfmstart.width, tfmnext.width + local bound = #cursives + 1 + start[a_cursbase] = bound + nxt[a_curscurs] = bound + cursives[bound] = { rlmode, dx, dy, ws, wn } + return dx, dy, bound +end + +function injections.setpair(current,factor,rlmode,r2lflag,spec,tfmchr) + local x, y, w, h = factor*spec[1], factor*spec[2], factor*spec[3], factor*spec[4] + -- dy = y - h + if x ~= 0 or w ~= 0 or y ~= 0 or h ~= 0 then + local bound = current[a_kernpair] + if bound then + local kb = kerns[bound] + -- inefficient but singles have less, but weird anyway, needs checking + kb[2], kb[3], kb[4], kb[5] = (kb[2] or 0) + x, (kb[3] or 0) + y, (kb[4] or 0)+ w, (kb[5] or 0) + h + else + bound = #kerns + 1 + current[a_kernpair] = bound + kerns[bound] = { rlmode, x, y, w, h, r2lflag, tfmchr.width } + end + return x, y, w, h, bound + end + return x, y, w, h -- no bound +end + +function injections.setkern(current,factor,rlmode,x,tfmchr) + local dx = factor*x + if dx ~= 0 then + local bound = #kerns + 1 + current[a_kernpair] = bound + kerns[bound] = { rlmode, dx } + return dx, bound + else + return 0, 0 + end +end + +function injections.setmark(start,base,factor,rlmode,ba,ma,index,baseismark) -- ba=baseanchor, ma=markanchor + local dx, dy = factor*(ba[1]-ma[1]), factor*(ba[2]-ma[2]) -- the index argument is no longer used but when this + local bound = base[a_markbase] -- fails again we should pass it + local index = 1 + if bound then + local mb = marks[bound] + if mb then + -- if not index then index = #mb + 1 end + index = #mb + 1 + mb[index] = { dx, dy, rlmode } + start[a_markmark] = bound + start[a_markdone] = index + return dx, dy, bound + else + report_injections("possible problem, %U is base mark without data (id %a)",base.char,bound) + end + end +-- index = index or 1 + index = index or 1 + bound = #marks + 1 + base[a_markbase] = bound + start[a_markmark] = bound + start[a_markdone] = index + marks[bound] = { [index] = { dx, dy, rlmode, baseismark } } + return dx, dy, bound +end + +local function dir(n) + return (n and n<0 and "r-to-l") or (n and n>0 and "l-to-r") or "unset" +end + +local function trace(head) + report_injections("begin run") + for n in traverse_id(glyph_code,head) do + if n.subtype < 256 then + local kp = n[a_kernpair] + local mb = n[a_markbase] + local mm = n[a_markmark] + local md = n[a_markdone] + local cb = n[a_cursbase] + local cc = n[a_curscurs] + local char = n.char + report_injections("font %s, char %U, glyph %c",n.font,char,char) + if kp then + local k = kerns[kp] + if k[3] then + report_injections(" pairkern: dir %a, x %p, y %p, w %p, h %p",dir(k[1]),k[2],k[3],k[4],k[5]) + else + report_injections(" kern: dir %a, dx %p",dir(k[1]),k[2]) + end + end + if mb then + report_injections(" markbase: bound %a",mb) + end + if mm then + local m = marks[mm] + if mb then + local m = m[mb] + if m then + report_injections(" markmark: bound %a, index %a, dx %p, dy %p",mm,md,m[1],m[2]) + else + report_injections(" markmark: bound %a, missing index",mm) + end + else + m = m[1] + report_injections(" markmark: bound %a, dx %p, dy %p",mm,m and m[1],m and m[2]) + end + end + if cb then + report_injections(" cursbase: bound %a",cb) + end + if cc then + local c = cursives[cc] + report_injections(" curscurs: bound %a, dir %a, dx %p, dy %p",cc,dir(c[1]),c[2],c[3]) + end + end + end + report_injections("end run") +end + +-- todo: reuse tables (i.e. no collection), but will be extra fields anyway +-- todo: check for attribute + +-- We can have a fast test on a font being processed, so we can check faster for marks etc +-- but I'll make a context variant anyway. + +local function show_result(head) + local current = head + local skipping = false + while current do + local id = current.id + if id == glyph_code then + report_injections("char: %C, width %p, xoffset %p, yoffset %p",current.char,current.width,current.xoffset,current.yoffset) + skipping = false + elseif id == kern_code then + report_injections("kern: %p",current.kern) + skipping = false + elseif not skipping then + report_injections() + skipping = true + end + current = current.next + end +end + +function injections.handler(head,where,keep) + local has_marks, has_cursives, has_kerns = next(marks), next(cursives), next(kerns) + if has_marks or has_cursives then + if trace_injections then + trace(head) + end + -- in the future variant we will not copy items but refs to tables + local done, ky, rl, valid, cx, wx, mk, nofvalid = false, { }, { }, { }, { }, { }, { }, 0 + if has_kerns then -- move outside loop + local nf, tm = nil, nil + for n in traverse_id(glyph_code,head) do -- only needed for relevant fonts + if n.subtype < 256 then + nofvalid = nofvalid + 1 + valid[nofvalid] = n + if n.font ~= nf then + nf = n.font + tm = fontdata[nf].resources.marks + end + if tm then + mk[n] = tm[n.char] + end + local k = n[a_kernpair] + if k then + local kk = kerns[k] + if kk then + local x, y, w, h = kk[2] or 0, kk[3] or 0, kk[4] or 0, kk[5] or 0 + local dy = y - h + if dy ~= 0 then + ky[n] = dy + end + if w ~= 0 or x ~= 0 then + wx[n] = kk + end + rl[n] = kk[1] -- could move in test + end + end + end + end + else + local nf, tm = nil, nil + for n in traverse_id(glyph_code,head) do + if n.subtype < 256 then + nofvalid = nofvalid + 1 + valid[nofvalid] = n + if n.font ~= nf then + nf = n.font + tm = fontdata[nf].resources.marks + end + if tm then + mk[n] = tm[n.char] + end + end + end + end + if nofvalid > 0 then + -- we can assume done == true because we have cursives and marks + local cx = { } + if has_kerns and next(ky) then + for n, k in next, ky do + n.yoffset = k + end + end + -- todo: reuse t and use maxt + if has_cursives then + local p_cursbase, p = nil, nil + -- since we need valid[n+1] we can also use a "while true do" + local t, d, maxt = { }, { }, 0 + for i=1,nofvalid do -- valid == glyphs + local n = valid[i] + if not mk[n] then + local n_cursbase = n[a_cursbase] + if p_cursbase then + local n_curscurs = n[a_curscurs] + if p_cursbase == n_curscurs then + local c = cursives[n_curscurs] + if c then + local rlmode, dx, dy, ws, wn = c[1], c[2], c[3], c[4], c[5] + if rlmode >= 0 then + dx = dx - ws + else + dx = dx + wn + end + if dx ~= 0 then + cx[n] = dx + rl[n] = rlmode + end + -- if rlmode and rlmode < 0 then + dy = -dy + -- end + maxt = maxt + 1 + t[maxt] = p + d[maxt] = dy + else + maxt = 0 + end + end + elseif maxt > 0 then + local ny = n.yoffset + for i=maxt,1,-1 do + ny = ny + d[i] + local ti = t[i] + ti.yoffset = ti.yoffset + ny + end + maxt = 0 + end + if not n_cursbase and maxt > 0 then + local ny = n.yoffset + for i=maxt,1,-1 do + ny = ny + d[i] + local ti = t[i] + ti.yoffset = ny + end + maxt = 0 + end + p_cursbase, p = n_cursbase, n + end + end + if maxt > 0 then + local ny = n.yoffset + for i=maxt,1,-1 do + ny = ny + d[i] + local ti = t[i] + ti.yoffset = ny + end + maxt = 0 + end + if not keep then + cursives = { } + end + end + if has_marks then + for i=1,nofvalid do + local p = valid[i] + local p_markbase = p[a_markbase] + if p_markbase then + local mrks = marks[p_markbase] + local nofmarks = #mrks + for n in traverse_id(glyph_code,p.next) do + local n_markmark = n[a_markmark] + if p_markbase == n_markmark then + local index = n[a_markdone] or 1 + local d = mrks[index] + if d then + local rlmode = d[3] + -- + local k = wx[p] + if k then + local x = k[2] + local w = k[4] + if w then + if rlmode and rlmode >= 0 then + -- kern(x) glyph(p) kern(w-x) mark(n) + n.xoffset = p.xoffset - p.width + d[1] - (w-x) + else + -- kern(w-x) glyph(p) kern(x) mark(n) + n.xoffset = p.xoffset - d[1] - x + end + else + if rlmode and rlmode >= 0 then + -- okay for husayni + n.xoffset = p.xoffset - p.width + d[1] + else + -- needs checking: is x ok here? + n.xoffset = p.xoffset - d[1] - x + end + end + else + if rlmode and rlmode >= 0 then + n.xoffset = p.xoffset - p.width + d[1] + else + n.xoffset = p.xoffset - d[1] + end + local w = n.width + if w ~= 0 then + insert_node_before(head,n,newkern(-w/2)) + insert_node_after(head,n,newkern(-w/2)) + end + end + -- -- + if mk[p] then + n.yoffset = p.yoffset + d[2] + else + n.yoffset = n.yoffset + p.yoffset + d[2] + end + -- + if nofmarks == 1 then + break + else + nofmarks = nofmarks - 1 + end + end + else + -- KE: there can be <mark> <mkmk> <mark> sequences in ligatures + end + end + end + end + if not keep then + marks = { } + end + end + -- todo : combine + if next(wx) then + for n, k in next, wx do + -- only w can be nil (kernclasses), can be sped up when w == nil + local x = k[2] + local w = k[4] + if w then + local rl = k[1] -- r2l = k[6] + local wx = w - x + if rl < 0 then -- KE: don't use r2l here + if wx ~= 0 then + insert_node_before(head,n,newkern(wx)) -- type 0/2 + end + if x ~= 0 then + insert_node_after (head,n,newkern(x)) -- type 0/2 + end + else + if x ~= 0 then + insert_node_before(head,n,newkern(x)) -- type 0/2 + end + if wx ~= 0 then + insert_node_after (head,n,newkern(wx)) -- type 0/2 + end + end + elseif x ~= 0 then + -- this needs checking for rl < 0 but it is unlikely that a r2l script + -- uses kernclasses between glyphs so we're probably safe (KE has a + -- problematic font where marks interfere with rl < 0 in the previous + -- case) + insert_node_before(head,n,newkern(x)) -- a real font kern, type 0 + end + end + end + if next(cx) then + for n, k in next, cx do + if k ~= 0 then + local rln = rl[n] + if rln and rln < 0 then + insert_node_before(head,n,newkern(-k)) -- type 0/2 + else + insert_node_before(head,n,newkern(k)) -- type 0/2 + end + end + end + end + if not keep then + kerns = { } + end + -- if trace_injections then + -- show_result(head) + -- end + return head, true + elseif not keep then + kerns, cursives, marks = { }, { }, { } + end + elseif has_kerns then + if trace_injections then + trace(head) + end + for n in traverse_id(glyph_code,head) do + if n.subtype < 256 then + local k = n[a_kernpair] + if k then + local kk = kerns[k] + if kk then + local rl, x, y, w = kk[1], kk[2] or 0, kk[3], kk[4] + if y and y ~= 0 then + n.yoffset = y -- todo: h ? + end + if w then + -- copied from above + -- local r2l = kk[6] + local wx = w - x + if rl < 0 then -- KE: don't use r2l here + if wx ~= 0 then + insert_node_before(head,n,newkern(wx)) + end + if x ~= 0 then + insert_node_after (head,n,newkern(x)) + end + else + if x ~= 0 then + insert_node_before(head,n,newkern(x)) + end + if wx ~= 0 then + insert_node_after(head,n,newkern(wx)) + end + end + else + -- simple (e.g. kernclass kerns) + if x ~= 0 then + insert_node_before(head,n,newkern(x)) + end + end + end + end + end + end + if not keep then + kerns = { } + end + -- if trace_injections then + -- show_result(head) + -- end + return head, true + else + -- no tracing needed + end + return head, false +end diff --git a/luaotfload-fonts-otn.lua b/luaotfload-fonts-otn.lua new file mode 100644 index 0000000..c57be5f --- /dev/null +++ b/luaotfload-fonts-otn.lua @@ -0,0 +1,2848 @@ +if not modules then modules = { } end modules ['font-otn'] = { + version = 1.001, + comment = "companion to font-ini.mkiv", + author = "Hans Hagen, PRAGMA-ADE, Hasselt NL", + copyright = "PRAGMA ADE / ConTeXt Development Team", + license = "see context related readme files", +} + +-- preprocessors = { "nodes" } + +-- this is still somewhat preliminary and it will get better in due time; +-- much functionality could only be implemented thanks to the husayni font +-- of Idris Samawi Hamid to who we dedicate this module. + +-- in retrospect it always looks easy but believe it or not, it took a lot +-- of work to get proper open type support done: buggy fonts, fuzzy specs, +-- special made testfonts, many skype sessions between taco, idris and me, +-- torture tests etc etc ... unfortunately the code does not show how much +-- time it took ... + +-- todo: +-- +-- kerning is probably not yet ok for latin around dics nodes (interesting challenge) +-- extension infrastructure (for usage out of context) +-- sorting features according to vendors/renderers +-- alternative loop quitters +-- check cursive and r2l +-- find out where ignore-mark-classes went +-- default features (per language, script) +-- handle positions (we need example fonts) +-- handle gpos_single (we might want an extra width field in glyph nodes because adding kerns might interfere) +-- mark (to mark) code is still not what it should be (too messy but we need some more extreem husayni tests) +-- remove some optimizations (when I have a faster machine) +-- +-- maybe redo the lot some way (more context specific) + +--[[ldx-- +<p>This module is a bit more split up that I'd like but since we also want to test +with plain <l n='tex'/> it has to be so. This module is part of <l n='context'/> +and discussion about improvements and functionality mostly happens on the +<l n='context'/> mailing list.</p> + +<p>The specification of OpenType is kind of vague. Apart from a lack of a proper +free specifications there's also the problem that Microsoft and Adobe +may have their own interpretation of how and in what order to apply features. +In general the Microsoft website has more detailed specifications and is a +better reference. There is also some information in the FontForge help files.</p> + +<p>Because there is so much possible, fonts might contain bugs and/or be made to +work with certain rederers. These may evolve over time which may have the side +effect that suddenly fonts behave differently.</p> + +<p>After a lot of experiments (mostly by Taco, me and Idris) we're now at yet another +implementation. Of course all errors are mine and of course the code can be +improved. There are quite some optimizations going on here and processing speed +is currently acceptable. Not all functions are implemented yet, often because I +lack the fonts for testing. Many scripts are not yet supported either, but I will +look into them as soon as <l n='context'/> users ask for it.</p> + +<p>Because there are different interpretations possible, I will extend the code +with more (configureable) variants. I can also add hooks for users so that they can +write their own extensions.</p> + +<p>Glyphs are indexed not by unicode but in their own way. This is because there is no +relationship with unicode at all, apart from the fact that a font might cover certain +ranges of characters. One character can have multiple shapes. However, at the +<l n='tex'/> end we use unicode so and all extra glyphs are mapped into a private +space. This is needed because we need to access them and <l n='tex'/> has to include +then in the output eventually.</p> + +<p>The raw table as it coms from <l n='fontforge'/> gets reorganized in to fit out needs. +In <l n='context'/> that table is packed (similar tables are shared) and cached on disk +so that successive runs can use the optimized table (after loading the table is +unpacked). The flattening code used later is a prelude to an even more compact table +format (and as such it keeps evolving).</p> + +<p>This module is sparsely documented because it is a moving target. The table format +of the reader changes and we experiment a lot with different methods for supporting +features.</p> + +<p>As with the <l n='afm'/> code, we may decide to store more information in the +<l n='otf'/> table.</p> + +<p>Incrementing the version number will force a re-cache. We jump the number by one +when there's a fix in the <l n='fontforge'/> library or <l n='lua'/> code that +results in different tables.</p> +--ldx]]-- + +-- action handler chainproc chainmore comment +-- +-- gsub_single ok ok ok +-- gsub_multiple ok ok not implemented yet +-- gsub_alternate ok ok not implemented yet +-- gsub_ligature ok ok ok +-- gsub_context ok -- +-- gsub_contextchain ok -- +-- gsub_reversecontextchain ok -- +-- chainsub -- ok +-- reversesub -- ok +-- gpos_mark2base ok ok +-- gpos_mark2ligature ok ok +-- gpos_mark2mark ok ok +-- gpos_cursive ok untested +-- gpos_single ok ok +-- gpos_pair ok ok +-- gpos_context ok -- +-- gpos_contextchain ok -- +-- +-- todo: contextpos and contextsub and class stuff +-- +-- actions: +-- +-- handler : actions triggered by lookup +-- chainproc : actions triggered by contextual lookup +-- chainmore : multiple substitutions triggered by contextual lookup (e.g. fij -> f + ij) +-- +-- remark: the 'not implemented yet' variants will be done when we have fonts that use them +-- remark: we need to check what to do with discretionaries + +-- We used to have independent hashes for lookups but as the tags are unique +-- we now use only one hash. If needed we can have multiple again but in that +-- case I will probably prefix (i.e. rename) the lookups in the cached font file. + +-- Todo: make plugin feature that operates on char/glyphnode arrays + +local concat, insert, remove = table.concat, table.insert, table.remove +local gmatch, gsub, find, match, lower, strip = string.gmatch, string.gsub, string.find, string.match, string.lower, string.strip +local type, next, tonumber, tostring = type, next, tonumber, tostring +local lpegmatch = lpeg.match +local random = math.random +local formatters = string.formatters + +local logs, trackers, nodes, attributes = logs, trackers, nodes, attributes + +local registertracker = trackers.register + +local fonts = fonts +local otf = fonts.handlers.otf + +local trace_lookups = false registertracker("otf.lookups", function(v) trace_lookups = v end) +local trace_singles = false registertracker("otf.singles", function(v) trace_singles = v end) +local trace_multiples = false registertracker("otf.multiples", function(v) trace_multiples = v end) +local trace_alternatives = false registertracker("otf.alternatives", function(v) trace_alternatives = v end) +local trace_ligatures = false registertracker("otf.ligatures", function(v) trace_ligatures = v end) +local trace_contexts = false registertracker("otf.contexts", function(v) trace_contexts = v end) +local trace_marks = false registertracker("otf.marks", function(v) trace_marks = v end) +local trace_kerns = false registertracker("otf.kerns", function(v) trace_kerns = v end) +local trace_cursive = false registertracker("otf.cursive", function(v) trace_cursive = v end) +local trace_preparing = false registertracker("otf.preparing", function(v) trace_preparing = v end) +local trace_bugs = false registertracker("otf.bugs", function(v) trace_bugs = v end) +local trace_details = false registertracker("otf.details", function(v) trace_details = v end) +local trace_applied = false registertracker("otf.applied", function(v) trace_applied = v end) +local trace_steps = false registertracker("otf.steps", function(v) trace_steps = v end) +local trace_skips = false registertracker("otf.skips", function(v) trace_skips = v end) +local trace_directions = false registertracker("otf.directions", function(v) trace_directions = v end) + +local report_direct = logs.reporter("fonts","otf direct") +local report_subchain = logs.reporter("fonts","otf subchain") +local report_chain = logs.reporter("fonts","otf chain") +local report_process = logs.reporter("fonts","otf process") +local report_prepare = logs.reporter("fonts","otf prepare") +local report_warning = logs.reporter("fonts","otf warning") + +registertracker("otf.verbose_chain", function(v) otf.setcontextchain(v and "verbose") end) +registertracker("otf.normal_chain", function(v) otf.setcontextchain(v and "normal") end) + +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") + +local insert_node_after = node.insert_after +local delete_node = nodes.delete +local copy_node = node.copy +local find_node_tail = node.tail or node.slide +local flush_node_list = node.flush_list +local end_of_math = node.end_of_math + +local setmetatableindex = table.setmetatableindex + +local zwnj = 0x200C +local zwj = 0x200D +local wildcard = "*" +local default = "dflt" + +local nodecodes = nodes.nodecodes +local whatcodes = nodes.whatcodes +local glyphcodes = nodes.glyphcodes +local disccodes = nodes.disccodes + +local glyph_code = nodecodes.glyph +local glue_code = nodecodes.glue +local disc_code = nodecodes.disc +local whatsit_code = nodecodes.whatsit +local math_code = nodecodes.math + +local dir_code = whatcodes.dir +local localpar_code = whatcodes.localpar + +local discretionary_code = disccodes.discretionary + +local ligature_code = glyphcodes.ligature + +local privateattribute = attributes.private + +-- Something is messed up: we have two mark / ligature indices, one at the injection +-- end and one here ... this is bases in KE's patches but there is something fishy +-- there as I'm pretty sure that for husayni we need some connection (as it's much +-- more complex than an average font) but I need proper examples of all cases, not +-- of only some. + +local a_state = privateattribute('state') +local a_markbase = privateattribute('markbase') +local a_markmark = privateattribute('markmark') +local a_markdone = privateattribute('markdone') -- assigned at the injection end +local a_cursbase = privateattribute('cursbase') +local a_curscurs = privateattribute('curscurs') +local a_cursdone = privateattribute('cursdone') +local a_kernpair = privateattribute('kernpair') +local a_ligacomp = privateattribute('ligacomp') -- assigned here (ideally it should be combined) + +local injections = nodes.injections +local setmark = injections.setmark +local setcursive = injections.setcursive +local setkern = injections.setkern +local setpair = injections.setpair + +local markonce = true +local cursonce = true +local kernonce = true + +local fonthashes = fonts.hashes +local fontdata = fonthashes.identifiers + +local otffeatures = fonts.constructors.newfeatures("otf") +local registerotffeature = otffeatures.register + +local onetimemessage = fonts.loggers.onetimemessage or function() end + +otf.defaultnodealternate = "none" -- first last + +-- we share some vars here, after all, we have no nested lookups and less code + +local tfmdata = false +local characters = false +local descriptions = false +local resources = false +local marks = false +local currentfont = false +local lookuptable = false +local anchorlookups = false +local lookuptypes = false +local handlers = { } +local rlmode = 0 +local featurevalue = false + +-- head is always a whatsit so we can safely assume that head is not changed + +-- we use this for special testing and documentation + +local checkstep = (nodes and nodes.tracers and nodes.tracers.steppers.check) or function() end +local registerstep = (nodes and nodes.tracers and nodes.tracers.steppers.register) or function() end +local registermessage = (nodes and nodes.tracers and nodes.tracers.steppers.message) or function() end + +local function logprocess(...) + if trace_steps then + registermessage(...) + end + report_direct(...) +end + +local function logwarning(...) + report_direct(...) +end + +local f_unicode = formatters["%U"] +local f_uniname = formatters["%U (%s)"] +local f_unilist = formatters["% t (% t)"] + +local function gref(n) -- currently the same as in font-otb + if type(n) == "number" then + local description = descriptions[n] + local name = description and description.name + if name then + return f_uniname(n,name) + else + return f_unicode(n) + end + elseif n then + local num, nam = { }, { } + for i=1,#n do + local ni = n[i] + if tonumber(ni) then -- later we will start at 2 + local di = descriptions[ni] + num[i] = f_unicode(ni) + nam[i] = di and di.name or "-" + end + end + return f_unilist(num,nam) + else + return "<error in node mode tracing>" + end +end + +local function cref(kind,chainname,chainlookupname,lookupname,index) -- not in the mood to alias f_ + if index then + return formatters["feature %a, chain %a, sub %a, lookup %a, index %a"](kind,chainname,chainlookupname,lookupname,index) + elseif lookupname then + return formatters["feature %a, chain %a, sub %a, lookup %a"](kind,chainname,chainlookupname,lookupname) + elseif chainlookupname then + return formatters["feature %a, chain %a, sub %a"](kind,chainname,chainlookupname) + elseif chainname then + return formatters["feature %a, chain %a"](kind,chainname) + else + return formatters["feature %a"](kind) + end +end + +local function pref(kind,lookupname) + return formatters["feature %a, lookup %a"](kind,lookupname) +end + +-- We can assume that languages that use marks are not hyphenated. We can also assume +-- that at most one discretionary is present. + +-- We do need components in funny kerning mode but maybe I can better reconstruct then +-- as we do have the font components info available; removing components makes the +-- previous code much simpler. Also, later on copying and freeing becomes easier. +-- However, for arabic we need to keep them around for the sake of mark placement +-- and indices. + +local function copy_glyph(g) -- next and prev are untouched ! + local components = g.components + if components then + g.components = nil + local n = copy_node(g) + g.components = components + return n + else + return copy_node(g) + end +end + +-- start is a mark and we need to keep that one + +local function markstoligature(kind,lookupname,head,start,stop,char) + if start == stop and start.char == char then + return head, start + else + local prev = start.prev + local next = stop.next + start.prev = nil + stop.next = nil + local base = copy_glyph(start) + if head == start then + head = base + end + base.char = char + base.subtype = ligature_code + base.components = start + if prev then + prev.next = base + end + if next then + next.prev = base + end + base.next = next + base.prev = prev + return head, base + end +end + +-- The next code is somewhat complicated by the fact that some fonts can have ligatures made +-- from ligatures that themselves have marks. This was identified by Kai in for instance +-- arabtype: KAF LAM SHADDA ALEF FATHA (0x0643 0x0644 0x0651 0x0627 0x064E). This becomes +-- KAF LAM-ALEF with a SHADDA on the first and a FATHA op de second component. In a next +-- iteration this becomes a KAF-LAM-ALEF with a SHADDA on the second and a FATHA on the +-- third component. + +local function getcomponentindex(start) + if start.id ~= glyph_code then + return 0 + elseif start.subtype == ligature_code then + local i = 0 + local components = start.components + while components do + i = i + getcomponentindex(components) + components = components.next + end + return i + elseif not marks[start.char] then + return 1 + else + return 0 + end +end + +-- eventually we will do positioning in an other way (needs addional w/h/d fields) + +local function toligature(kind,lookupname,head,start,stop,char,markflag,discfound) -- brr head + if start == stop and start.char == char then + start.char = char + return head, start + end + local prev = start.prev + local next = stop.next + start.prev = nil + stop.next = nil + local base = copy_glyph(start) + if start == head then + head = base + end + base.char = char + base.subtype = ligature_code + base.components = start -- start can have components + if prev then + prev.next = base + end + if next then + next.prev = base + end + base.next = next + base.prev = prev + if not discfound then + local deletemarks = markflag ~= "mark" + local components = start + local baseindex = 0 + local componentindex = 0 + local head = base + local current = base + -- first we loop over the glyphs in start .. stop + while start do + local char = start.char + if not marks[char] then + baseindex = baseindex + componentindex + componentindex = getcomponentindex(start) + elseif not deletemarks then -- quite fishy + start[a_ligacomp] = baseindex + (start[a_ligacomp] or componentindex) + if trace_marks then + logwarning("%s: keep mark %s, gets index %s",pref(kind,lookupname),gref(char),start[a_ligacomp]) + end + head, current = insert_node_after(head,current,copy_node(start)) -- unlikely that mark has components + elseif trace_marks then + logwarning("%s: delete mark %s",pref(kind,lookupname),gref(char)) + end + start = start.next + end + -- we can have one accent as part of a lookup and another following + -- local start = components -- was wrong (component scanning was introduced when more complex ligs in devanagari was added) + local start = current.next + while start and start.id == glyph_code do + local char = start.char + if marks[char] then + start[a_ligacomp] = baseindex + (start[a_ligacomp] or componentindex) + if trace_marks then + logwarning("%s: set mark %s, gets index %s",pref(kind,lookupname),gref(char),start[a_ligacomp]) + end + else + break + end + start = start.next + end + end + return head, base +end + +function handlers.gsub_single(head,start,kind,lookupname,replacement) + if trace_singles then + logprocess("%s: replacing %s by single %s",pref(kind,lookupname),gref(start.char),gref(replacement)) + end + start.char = replacement + return head, start, true +end + +local function get_alternative_glyph(start,alternatives,value,trace_alternatives) + local n = #alternatives + if value == "random" then + local r = random(1,n) + return alternatives[r], trace_alternatives and formatters["value %a, taking %a"](value,r) + elseif value == "first" then + return alternatives[1], trace_alternatives and formatters["value %a, taking %a"](value,1) + elseif value == "last" then + return alternatives[n], trace_alternatives and formatters["value %a, taking %a"](value,n) + else + value = tonumber(value) + if type(value) ~= "number" then + return alternatives[1], trace_alternatives and formatters["invalid value %s, taking %a"](value,1) + elseif value > n then + local defaultalt = otf.defaultnodealternate + if defaultalt == "first" then + return alternatives[n], trace_alternatives and formatters["invalid value %s, taking %a"](value,1) + elseif defaultalt == "last" then + return alternatives[1], trace_alternatives and formatters["invalid value %s, taking %a"](value,n) + else + return false, trace_alternatives and formatters["invalid value %a, %s"](value,"out of range") + end + elseif value == 0 then + return start.char, trace_alternatives and formatters["invalid value %a, %s"](value,"no change") + elseif value < 1 then + return alternatives[1], trace_alternatives and formatters["invalid value %a, taking %a"](value,1) + else + return alternatives[value], trace_alternatives and formatters["value %a, taking %a"](value,value) + end + end +end + +local function multiple_glyphs(head,start,multiple,ignoremarks) + local nofmultiples = #multiple + if nofmultiples > 0 then + start.char = multiple[1] + if nofmultiples > 1 then + local sn = start.next + for k=2,nofmultiples do -- todo: use insert_node +-- untested: +-- +-- while ignoremarks and marks[sn.char] then +-- local sn = sn.next +-- end + local n = copy_node(start) -- ignore components + n.char = multiple[k] + n.next = sn + n.prev = start + if sn then + sn.prev = n + end + start.next = n + start = n + end + end + return head, start, true + else + if trace_multiples then + logprocess("no multiple for %s",gref(start.char)) + end + return head, start, false + end +end + +function handlers.gsub_alternate(head,start,kind,lookupname,alternative,sequence) + local value = featurevalue == true and tfmdata.shared.features[kind] or featurevalue + local choice, comment = get_alternative_glyph(start,alternative,value,trace_alternatives) + if choice then + if trace_alternatives then + logprocess("%s: replacing %s by alternative %a to %s, %s",pref(kind,lookupname),gref(start.char),choice,gref(choice),comment) + end + start.char = choice + else + if trace_alternatives then + logwarning("%s: no variant %a for %s, %s",pref(kind,lookupname),value,gref(start.char),comment) + end + end + return head, start, true +end + +function handlers.gsub_multiple(head,start,kind,lookupname,multiple,sequence) + if trace_multiples then + logprocess("%s: replacing %s by multiple %s",pref(kind,lookupname),gref(start.char),gref(multiple)) + end + return multiple_glyphs(head,start,multiple,sequence.flags[1]) +end + +function handlers.gsub_ligature(head,start,kind,lookupname,ligature,sequence) + local s, stop, discfound = start.next, nil, false + local startchar = start.char + if marks[startchar] then + while s do + local id = s.id + if id == glyph_code and s.font == currentfont and s.subtype<256 then + local lg = ligature[s.char] + if lg then + stop = s + ligature = lg + s = s.next + else + break + end + else + break + end + end + if stop then + local lig = ligature.ligature + if lig then + if trace_ligatures then + local stopchar = stop.char + head, start = markstoligature(kind,lookupname,head,start,stop,lig) + logprocess("%s: replacing %s upto %s by ligature %s case 1",pref(kind,lookupname),gref(startchar),gref(stopchar),gref(start.char)) + else + head, start = markstoligature(kind,lookupname,head,start,stop,lig) + end + return head, start, true + else + -- ok, goto next lookup + end + end + else + local skipmark = sequence.flags[1] + while s do + local id = s.id + if id == glyph_code and s.subtype<256 then + if s.font == currentfont then + local char = s.char + if skipmark and marks[char] then + s = s.next + else + local lg = ligature[char] + if lg then + stop = s + ligature = lg + s = s.next + else + break + end + end + else + break + end + elseif id == disc_code then + discfound = true + s = s.next + else + break + end + end + local lig = ligature.ligature + if lig then + if stop then + if trace_ligatures then + local stopchar = stop.char + head, start = toligature(kind,lookupname,head,start,stop,lig,skipmark,discfound) + logprocess("%s: replacing %s upto %s by ligature %s case 2",pref(kind,lookupname),gref(startchar),gref(stopchar),gref(start.char)) + else + head, start = toligature(kind,lookupname,head,start,stop,lig,skipmark,discfound) + end + return head, start, true + else + -- weird but happens (in some arabic font) + start.char = lig + if trace_ligatures then + logprocess("%s: replacing %s by (no real) ligature %s case 3",pref(kind,lookupname),gref(startchar),gref(lig)) + end + return head, start, true + end + else + -- weird but happens + end + end + return head, start, false +end + +--[[ldx-- +<p>We get hits on a mark, but we're not sure if the it has to be applied so +we need to explicitly test for basechar, baselig and basemark entries.</p> +--ldx]]-- + +function handlers.gpos_mark2base(head,start,kind,lookupname,markanchors,sequence) + local markchar = start.char + if marks[markchar] then + local base = start.prev -- [glyph] [start=mark] + if base and base.id == glyph_code and base.font == currentfont and base.subtype<256 then + local basechar = base.char + if marks[basechar] then + while true do + base = base.prev + if base and base.id == glyph_code and base.font == currentfont and base.subtype<256 then + basechar = base.char + if not marks[basechar] then + break + end + else + if trace_bugs then + logwarning("%s: no base for mark %s",pref(kind,lookupname),gref(markchar)) + end + return head, start, false + end + end + end + local baseanchors = descriptions[basechar] + if baseanchors then + baseanchors = baseanchors.anchors + end + if baseanchors then + local baseanchors = baseanchors['basechar'] + if baseanchors then + local al = anchorlookups[lookupname] + for anchor,ba in next, baseanchors do + if al[anchor] then + local ma = markanchors[anchor] + if ma then + local dx, dy, bound = setmark(start,base,tfmdata.parameters.factor,rlmode,ba,ma) + if trace_marks then + logprocess("%s, anchor %s, bound %s: anchoring mark %s to basechar %s => (%p,%p)", + pref(kind,lookupname),anchor,bound,gref(markchar),gref(basechar),dx,dy) + end + return head, start, true + end + end + end + if trace_bugs then + logwarning("%s, no matching anchors for mark %s and base %s",pref(kind,lookupname),gref(markchar),gref(basechar)) + end + end + elseif trace_bugs then + -- logwarning("%s: char %s is missing in font",pref(kind,lookupname),gref(basechar)) + onetimemessage(currentfont,basechar,"no base anchors",report_fonts) + end + elseif trace_bugs then + logwarning("%s: prev node is no char",pref(kind,lookupname)) + end + elseif trace_bugs then + logwarning("%s: mark %s is no mark",pref(kind,lookupname),gref(markchar)) + end + return head, start, false +end + +function handlers.gpos_mark2ligature(head,start,kind,lookupname,markanchors,sequence) + -- check chainpos variant + local markchar = start.char + if marks[markchar] then + local base = start.prev -- [glyph] [optional marks] [start=mark] + if base and base.id == glyph_code and base.font == currentfont and base.subtype<256 then + local basechar = base.char + if marks[basechar] then + while true do + base = base.prev + if base and base.id == glyph_code and base.font == currentfont and base.subtype<256 then + basechar = base.char + if not marks[basechar] then + break + end + else + if trace_bugs then + logwarning("%s: no base for mark %s",pref(kind,lookupname),gref(markchar)) + end + return head, start, false + end + end + end + local index = start[a_ligacomp] + local baseanchors = descriptions[basechar] + if baseanchors then + baseanchors = baseanchors.anchors + if baseanchors then + local baseanchors = baseanchors['baselig'] + if baseanchors then + local al = anchorlookups[lookupname] + for anchor, ba in next, baseanchors do + if al[anchor] then + local ma = markanchors[anchor] + if ma then + ba = ba[index] + if ba then + local dx, dy, bound = setmark(start,base,tfmdata.parameters.factor,rlmode,ba,ma) -- index + if trace_marks then + logprocess("%s, anchor %s, index %s, bound %s: anchoring mark %s to baselig %s at index %s => (%p,%p)", + pref(kind,lookupname),anchor,index,bound,gref(markchar),gref(basechar),index,dx,dy) + end + return head, start, true + else + if trace_bugs then + logwarning("%s: no matching anchors for mark %s and baselig %s with index %a",pref(kind,lookupname),gref(markchar),gref(basechar),index) + end + end + end + end + end + if trace_bugs then + logwarning("%s: no matching anchors for mark %s and baselig %s",pref(kind,lookupname),gref(markchar),gref(basechar)) + end + end + end + elseif trace_bugs then + -- logwarning("%s: char %s is missing in font",pref(kind,lookupname),gref(basechar)) + onetimemessage(currentfont,basechar,"no base anchors",report_fonts) + end + elseif trace_bugs then + logwarning("%s: prev node is no char",pref(kind,lookupname)) + end + elseif trace_bugs then + logwarning("%s: mark %s is no mark",pref(kind,lookupname),gref(markchar)) + end + return head, start, false +end + +function handlers.gpos_mark2mark(head,start,kind,lookupname,markanchors,sequence) + local markchar = start.char + if marks[markchar] then + local base = start.prev -- [glyph] [basemark] [start=mark] + local slc = start[a_ligacomp] + if slc then -- a rather messy loop ... needs checking with husayni + while base do + local blc = base[a_ligacomp] + if blc and blc ~= slc then + base = base.prev + else + break + end + end + end + if base and base.id == glyph_code and base.font == currentfont and base.subtype<256 then -- subtype test can go + local basechar = base.char + local baseanchors = descriptions[basechar] + if baseanchors then + baseanchors = baseanchors.anchors + if baseanchors then + baseanchors = baseanchors['basemark'] + if baseanchors then + local al = anchorlookups[lookupname] + for anchor,ba in next, baseanchors do + if al[anchor] then + local ma = markanchors[anchor] + if ma then + local dx, dy, bound = setmark(start,base,tfmdata.parameters.factor,rlmode,ba,ma,true) + if trace_marks then + logprocess("%s, anchor %s, bound %s: anchoring mark %s to basemark %s => (%p,%p)", + pref(kind,lookupname),anchor,bound,gref(markchar),gref(basechar),dx,dy) + end + return head, start, true + end + end + end + if trace_bugs then + logwarning("%s: no matching anchors for mark %s and basemark %s",pref(kind,lookupname),gref(markchar),gref(basechar)) + end + end + end + elseif trace_bugs then + -- logwarning("%s: char %s is missing in font",pref(kind,lookupname),gref(basechar)) + onetimemessage(currentfont,basechar,"no base anchors",report_fonts) + end + elseif trace_bugs then + logwarning("%s: prev node is no mark",pref(kind,lookupname)) + end + elseif trace_bugs then + logwarning("%s: mark %s is no mark",pref(kind,lookupname),gref(markchar)) + end + return head, start, false +end + +function handlers.gpos_cursive(head,start,kind,lookupname,exitanchors,sequence) -- to be checked + local alreadydone = cursonce and start[a_cursbase] + if not alreadydone then + local done = false + local startchar = start.char + if marks[startchar] then + if trace_cursive then + logprocess("%s: ignoring cursive for mark %s",pref(kind,lookupname),gref(startchar)) + end + else + local nxt = start.next + while not done and nxt and nxt.id == glyph_code and nxt.font == currentfont and nxt.subtype<256 do + local nextchar = nxt.char + if marks[nextchar] then + -- should not happen (maybe warning) + nxt = nxt.next + else + local entryanchors = descriptions[nextchar] + if entryanchors then + entryanchors = entryanchors.anchors + if entryanchors then + entryanchors = entryanchors['centry'] + if entryanchors then + local al = anchorlookups[lookupname] + for anchor, entry in next, entryanchors do + if al[anchor] then + local exit = exitanchors[anchor] + if exit then + local dx, dy, bound = setcursive(start,nxt,tfmdata.parameters.factor,rlmode,exit,entry,characters[startchar],characters[nextchar]) + if trace_cursive then + logprocess("%s: moving %s to %s cursive (%p,%p) using anchor %s and bound %s in rlmode %s",pref(kind,lookupname),gref(startchar),gref(nextchar),dx,dy,anchor,bound,rlmode) + end + done = true + break + end + end + end + end + end + elseif trace_bugs then + -- logwarning("%s: char %s is missing in font",pref(kind,lookupname),gref(startchar)) + onetimemessage(currentfont,startchar,"no entry anchors",report_fonts) + end + break + end + end + end + return head, start, done + else + if trace_cursive and trace_details then + logprocess("%s, cursive %s is already done",pref(kind,lookupname),gref(start.char),alreadydone) + end + return head, start, false + end +end + +function handlers.gpos_single(head,start,kind,lookupname,kerns,sequence) + local startchar = start.char + local dx, dy, w, h = setpair(start,tfmdata.parameters.factor,rlmode,sequence.flags[4],kerns,characters[startchar]) + if trace_kerns then + logprocess("%s: shifting single %s by (%p,%p) and correction (%p,%p)",pref(kind,lookupname),gref(startchar),dx,dy,w,h) + end + return head, start, false +end + +function handlers.gpos_pair(head,start,kind,lookupname,kerns,sequence) + -- todo: kerns in disc nodes: pre, post, replace -> loop over disc too + -- todo: kerns in components of ligatures + local snext = start.next + if not snext then + return head, start, false + else + local prev, done = start, false + local factor = tfmdata.parameters.factor + local lookuptype = lookuptypes[lookupname] + while snext and snext.id == glyph_code and snext.font == currentfont and snext.subtype<256 do + local nextchar = snext.char + local krn = kerns[nextchar] + if not krn and marks[nextchar] then + prev = snext + snext = snext.next + else + if not krn then + -- skip + elseif type(krn) == "table" then + if lookuptype == "pair" then -- probably not needed + local a, b = krn[2], krn[3] + if a and #a > 0 then + local startchar = start.char + local x, y, w, h = setpair(start,factor,rlmode,sequence.flags[4],a,characters[startchar]) + if trace_kerns then + logprocess("%s: shifting first of pair %s and %s by (%p,%p) and correction (%p,%p)",pref(kind,lookupname),gref(startchar),gref(nextchar),x,y,w,h) + end + end + if b and #b > 0 then + local startchar = start.char + local x, y, w, h = setpair(snext,factor,rlmode,sequence.flags[4],b,characters[nextchar]) + if trace_kerns then + logprocess("%s: shifting second of pair %s and %s by (%p,%p) and correction (%p,%p)",pref(kind,lookupname),gref(startchar),gref(nextchar),x,y,w,h) + end + end + else -- wrong ... position has different entries + report_process("%s: check this out (old kern stuff)",pref(kind,lookupname)) + -- local a, b = krn[2], krn[6] + -- if a and a ~= 0 then + -- local k = setkern(snext,factor,rlmode,a) + -- if trace_kerns then + -- logprocess("%s: inserting first kern %s between %s and %s",pref(kind,lookupname),k,gref(prev.char),gref(nextchar)) + -- end + -- end + -- if b and b ~= 0 then + -- logwarning("%s: ignoring second kern xoff %s",pref(kind,lookupname),b*factor) + -- end + end + done = true + elseif krn ~= 0 then + local k = setkern(snext,factor,rlmode,krn) + if trace_kerns then + logprocess("%s: inserting kern %s between %s and %s",pref(kind,lookupname),k,gref(prev.char),gref(nextchar)) + end + done = true + end + break + end + end + return head, start, done + end +end + +--[[ldx-- +<p>I will implement multiple chain replacements once I run into a font that uses +it. It's not that complex to handle.</p> +--ldx]]-- + +local chainmores = { } +local chainprocs = { } + +local function logprocess(...) + if trace_steps then + registermessage(...) + end + report_subchain(...) +end + +local logwarning = report_subchain + +local function logprocess(...) + if trace_steps then + registermessage(...) + end + report_chain(...) +end + +local logwarning = report_chain + +-- We could share functions but that would lead to extra function calls with many +-- arguments, redundant tests and confusing messages. + +function chainprocs.chainsub(head,start,stop,kind,chainname,currentcontext,lookuphash,lookuplist,chainlookupname) + logwarning("%s: a direct call to chainsub cannot happen",cref(kind,chainname,chainlookupname)) + return head, start, false +end + +function chainmores.chainsub(head,start,stop,kind,chainname,currentcontext,lookuphash,lookuplist,chainlookupname,n) + logprocess("%s: a direct call to chainsub cannot happen",cref(kind,chainname,chainlookupname)) + return head, start, false +end + +-- The reversesub is a special case, which is why we need to store the replacements +-- in a bit weird way. There is no lookup and the replacement comes from the lookup +-- itself. It is meant mostly for dealing with Urdu. + +function chainprocs.reversesub(head,start,stop,kind,chainname,currentcontext,lookuphash,replacements) + local char = start.char + local replacement = replacements[char] + if replacement then + if trace_singles then + logprocess("%s: single reverse replacement of %s by %s",cref(kind,chainname),gref(char),gref(replacement)) + end + start.char = replacement + return head, start, true + else + return head, start, false + end +end + +--[[ldx-- +<p>This chain stuff is somewhat tricky since we can have a sequence of actions to be +applied: single, alternate, multiple or ligature where ligature can be an invalid +one in the sense that it will replace multiple by one but not neccessary one that +looks like the combination (i.e. it is the counterpart of multiple then). For +example, the following is valid:</p> + +<typing> +<line>xxxabcdexxx [single a->A][multiple b->BCD][ligature cde->E] xxxABCDExxx</line> +</typing> + +<p>Therefore we we don't really do the replacement here already unless we have the +single lookup case. The efficiency of the replacements can be improved by deleting +as less as needed but that would also make the code even more messy.</p> +--ldx]]-- + +-- local function delete_till_stop(head,start,stop,ignoremarks) -- keeps start +-- local n = 1 +-- if start == stop then +-- -- done +-- elseif ignoremarks then +-- repeat -- start x x m x x stop => start m +-- local next = start.next +-- if not marks[next.char] then +-- local components = next.components +-- if components then -- probably not needed +-- flush_node_list(components) +-- end +-- head = delete_node(head,next) +-- end +-- n = n + 1 +-- until next == stop +-- else -- start x x x stop => start +-- repeat +-- local next = start.next +-- local components = next.components +-- if components then -- probably not needed +-- flush_node_list(components) +-- end +-- head = delete_node(head,next) +-- n = n + 1 +-- until next == stop +-- end +-- return head, n +-- end + +--[[ldx-- +<p>Here we replace start by a single variant, First we delete the rest of the +match.</p> +--ldx]]-- + +function chainprocs.gsub_single(head,start,stop,kind,chainname,currentcontext,lookuphash,currentlookup,chainlookupname,chainindex) + -- todo: marks ? + local current = start + local subtables = currentlookup.subtables + if #subtables > 1 then + logwarning("todo: check if we need to loop over the replacements: %s",concat(subtables," ")) + end + while current do + if current.id == glyph_code then + local currentchar = current.char + local lookupname = subtables[1] -- only 1 + local replacement = lookuphash[lookupname] + if not replacement then + if trace_bugs then + logwarning("%s: no single hits",cref(kind,chainname,chainlookupname,lookupname,chainindex)) + end + else + replacement = replacement[currentchar] + if not replacement or replacement == "" then + if trace_bugs then + logwarning("%s: no single for %s",cref(kind,chainname,chainlookupname,lookupname,chainindex),gref(currentchar)) + end + else + if trace_singles then + logprocess("%s: replacing single %s by %s",cref(kind,chainname,chainlookupname,lookupname,chainindex),gref(currentchar),gref(replacement)) + end + current.char = replacement + end + end + return head, start, true + elseif current == stop then + break + else + current = current.next + end + end + return head, start, false +end + +chainmores.gsub_single = chainprocs.gsub_single + +--[[ldx-- +<p>Here we replace start by a sequence of new glyphs. First we delete the rest of +the match.</p> +--ldx]]-- + +function chainprocs.gsub_multiple(head,start,stop,kind,chainname,currentcontext,lookuphash,currentlookup,chainlookupname) + -- local head, n = delete_till_stop(head,start,stop) + local startchar = start.char + local subtables = currentlookup.subtables + local lookupname = subtables[1] + local replacements = lookuphash[lookupname] + if not replacements then + if trace_bugs then + logwarning("%s: no multiple hits",cref(kind,chainname,chainlookupname,lookupname)) + end + else + replacements = replacements[startchar] + if not replacements or replacement == "" then + if trace_bugs then + logwarning("%s: no multiple for %s",cref(kind,chainname,chainlookupname,lookupname),gref(startchar)) + end + else + if trace_multiples then + logprocess("%s: replacing %s by multiple characters %s",cref(kind,chainname,chainlookupname,lookupname),gref(startchar),gref(replacements)) + end + return multiple_glyphs(head,start,replacements,currentlookup.flags[1]) + end + end + return head, start, false +end + +chainmores.gsub_multiple = chainprocs.gsub_multiple + +--[[ldx-- +<p>Here we replace start by new glyph. First we delete the rest of the match.</p> +--ldx]]-- + +-- char_1 mark_1 -> char_x mark_1 (ignore marks) +-- char_1 mark_1 -> char_x + +-- to be checked: do we always have just one glyph? +-- we can also have alternates for marks +-- marks come last anyway +-- are there cases where we need to delete the mark + +function chainprocs.gsub_alternate(head,start,stop,kind,chainname,currentcontext,lookuphash,currentlookup,chainlookupname) + local current = start + local subtables = currentlookup.subtables + local value = featurevalue == true and tfmdata.shared.features[kind] or featurevalue + while current do + if current.id == glyph_code then -- is this check needed? + local currentchar = current.char + local lookupname = subtables[1] + local alternatives = lookuphash[lookupname] + if not alternatives then + if trace_bugs then + logwarning("%s: no alternative hit",cref(kind,chainname,chainlookupname,lookupname)) + end + else + alternatives = alternatives[currentchar] + if alternatives then + local choice, comment = get_alternative_glyph(current,alternatives,value,trace_alternatives) + if choice then + if trace_alternatives then + logprocess("%s: replacing %s by alternative %a to %s, %s",cref(kind,chainname,chainlookupname,lookupname),gref(char),choice,gref(choice),comment) + end + start.char = choice + else + if trace_alternatives then + logwarning("%s: no variant %a for %s, %s",cref(kind,chainname,chainlookupname,lookupname),value,gref(char),comment) + end + end + elseif trace_bugs then + logwarning("%s: no alternative for %s, %s",cref(kind,chainname,chainlookupname,lookupname),gref(currentchar),comment) + end + end + return head, start, true + elseif current == stop then + break + else + current = current.next + end + end + return head, start, false +end + +chainmores.gsub_alternate = chainprocs.gsub_alternate + +--[[ldx-- +<p>When we replace ligatures we use a helper that handles the marks. I might change +this function (move code inline and handle the marks by a separate function). We +assume rather stupid ligatures (no complex disc nodes).</p> +--ldx]]-- + +function chainprocs.gsub_ligature(head,start,stop,kind,chainname,currentcontext,lookuphash,currentlookup,chainlookupname,chainindex) + local startchar = start.char + local subtables = currentlookup.subtables + local lookupname = subtables[1] + local ligatures = lookuphash[lookupname] + if not ligatures then + if trace_bugs then + logwarning("%s: no ligature hits",cref(kind,chainname,chainlookupname,lookupname,chainindex)) + end + else + ligatures = ligatures[startchar] + if not ligatures then + if trace_bugs then + logwarning("%s: no ligatures starting with %s",cref(kind,chainname,chainlookupname,lookupname,chainindex),gref(startchar)) + end + else + local s = start.next + local discfound = false + local last = stop + local nofreplacements = 0 + local skipmark = currentlookup.flags[1] + while s do + local id = s.id + if id == disc_code then + s = s.next + discfound = true + else + local schar = s.char + if skipmark and marks[schar] then -- marks + s = s.next + else + local lg = ligatures[schar] + if lg then + ligatures, last, nofreplacements = lg, s, nofreplacements + 1 + if s == stop then + break + else + s = s.next + end + else + break + end + end + end + end + local l2 = ligatures.ligature + if l2 then + if chainindex then + stop = last + end + if trace_ligatures then + if start == stop then + logprocess("%s: replacing character %s by ligature %s case 3",cref(kind,chainname,chainlookupname,lookupname,chainindex),gref(startchar),gref(l2)) + else + logprocess("%s: replacing character %s upto %s by ligature %s case 4",cref(kind,chainname,chainlookupname,lookupname,chainindex),gref(startchar),gref(stop.char),gref(l2)) + end + end + head, start = toligature(kind,lookupname,head,start,stop,l2,currentlookup.flags[1],discfound) + return head, start, true, nofreplacements + elseif trace_bugs then + if start == stop then + logwarning("%s: replacing character %s by ligature fails",cref(kind,chainname,chainlookupname,lookupname,chainindex),gref(startchar)) + else + logwarning("%s: replacing character %s upto %s by ligature fails",cref(kind,chainname,chainlookupname,lookupname,chainindex),gref(startchar),gref(stop.char)) + end + end + end + end + return head, start, false, 0 +end + +chainmores.gsub_ligature = chainprocs.gsub_ligature + +function chainprocs.gpos_mark2base(head,start,stop,kind,chainname,currentcontext,lookuphash,currentlookup,chainlookupname) + local markchar = start.char + if marks[markchar] then + local subtables = currentlookup.subtables + local lookupname = subtables[1] + local markanchors = lookuphash[lookupname] + if markanchors then + markanchors = markanchors[markchar] + end + if markanchors then + local base = start.prev -- [glyph] [start=mark] + if base and base.id == glyph_code and base.font == currentfont and base.subtype<256 then + local basechar = base.char + if marks[basechar] then + while true do + base = base.prev + if base and base.id == glyph_code and base.font == currentfont and base.subtype<256 then + basechar = base.char + if not marks[basechar] then + break + end + else + if trace_bugs then + logwarning("%s: no base for mark %s",pref(kind,lookupname),gref(markchar)) + end + return head, start, false + end + end + end + local baseanchors = descriptions[basechar].anchors + if baseanchors then + local baseanchors = baseanchors['basechar'] + if baseanchors then + local al = anchorlookups[lookupname] + for anchor,ba in next, baseanchors do + if al[anchor] then + local ma = markanchors[anchor] + if ma then + local dx, dy, bound = setmark(start,base,tfmdata.parameters.factor,rlmode,ba,ma) + if trace_marks then + logprocess("%s, anchor %s, bound %s: anchoring mark %s to basechar %s => (%p,%p)", + cref(kind,chainname,chainlookupname,lookupname),anchor,bound,gref(markchar),gref(basechar),dx,dy) + end + return head, start, true + end + end + end + if trace_bugs then + logwarning("%s, no matching anchors for mark %s and base %s",cref(kind,chainname,chainlookupname,lookupname),gref(markchar),gref(basechar)) + end + end + end + elseif trace_bugs then + logwarning("%s: prev node is no char",cref(kind,chainname,chainlookupname,lookupname)) + end + elseif trace_bugs then + logwarning("%s: mark %s has no anchors",cref(kind,chainname,chainlookupname,lookupname),gref(markchar)) + end + elseif trace_bugs then + logwarning("%s: mark %s is no mark",cref(kind,chainname,chainlookupname),gref(markchar)) + end + return head, start, false +end + +function chainprocs.gpos_mark2ligature(head,start,stop,kind,chainname,currentcontext,lookuphash,currentlookup,chainlookupname) + local markchar = start.char + if marks[markchar] then + local subtables = currentlookup.subtables + local lookupname = subtables[1] + local markanchors = lookuphash[lookupname] + if markanchors then + markanchors = markanchors[markchar] + end + if markanchors then + local base = start.prev -- [glyph] [optional marks] [start=mark] + if base and base.id == glyph_code and base.font == currentfont and base.subtype<256 then + local basechar = base.char + if marks[basechar] then + while true do + base = base.prev + if base and base.id == glyph_code and base.font == currentfont and base.subtype<256 then + basechar = base.char + if not marks[basechar] then + break + end + else + if trace_bugs then + logwarning("%s: no base for mark %s",cref(kind,chainname,chainlookupname,lookupname),markchar) + end + return head, start, false + end + end + end + -- todo: like marks a ligatures hash + local index = start[a_ligacomp] + local baseanchors = descriptions[basechar].anchors + if baseanchors then + local baseanchors = baseanchors['baselig'] + if baseanchors then + local al = anchorlookups[lookupname] + for anchor,ba in next, baseanchors do + if al[anchor] then + local ma = markanchors[anchor] + if ma then + ba = ba[index] + if ba then + local dx, dy, bound = setmark(start,base,tfmdata.parameters.factor,rlmode,ba,ma) -- index + if trace_marks then + logprocess("%s, anchor %s, bound %s: anchoring mark %s to baselig %s at index %s => (%p,%p)", + cref(kind,chainname,chainlookupname,lookupname),anchor,a or bound,gref(markchar),gref(basechar),index,dx,dy) + end + return head, start, true + end + end + end + end + if trace_bugs then + logwarning("%s: no matching anchors for mark %s and baselig %s",cref(kind,chainname,chainlookupname,lookupname),gref(markchar),gref(basechar)) + end + end + end + elseif trace_bugs then + logwarning("feature %s, lookup %s: prev node is no char",kind,lookupname) + end + elseif trace_bugs then + logwarning("%s: mark %s has no anchors",cref(kind,chainname,chainlookupname,lookupname),gref(markchar)) + end + elseif trace_bugs then + logwarning("%s: mark %s is no mark",cref(kind,chainname,chainlookupname),gref(markchar)) + end + return head, start, false +end + +function chainprocs.gpos_mark2mark(head,start,stop,kind,chainname,currentcontext,lookuphash,currentlookup,chainlookupname) + local markchar = start.char + if marks[markchar] then + -- local alreadydone = markonce and start[a_markmark] + -- if not alreadydone then + -- local markanchors = descriptions[markchar].anchors markanchors = markanchors and markanchors.mark + local subtables = currentlookup.subtables + local lookupname = subtables[1] + local markanchors = lookuphash[lookupname] + if markanchors then + markanchors = markanchors[markchar] + end + if markanchors then + local base = start.prev -- [glyph] [basemark] [start=mark] + local slc = start[a_ligacomp] + if slc then -- a rather messy loop ... needs checking with husayni + while base do + local blc = base[a_ligacomp] + if blc and blc ~= slc then + base = base.prev + else + break + end + end + end + if base and base.id == glyph_code and base.font == currentfont and base.subtype<256 then -- subtype test can go + local basechar = base.char + local baseanchors = descriptions[basechar].anchors + if baseanchors then + baseanchors = baseanchors['basemark'] + if baseanchors then + local al = anchorlookups[lookupname] + for anchor,ba in next, baseanchors do + if al[anchor] then + local ma = markanchors[anchor] + if ma then + local dx, dy, bound = setmark(start,base,tfmdata.parameters.factor,rlmode,ba,ma,true) + if trace_marks then + logprocess("%s, anchor %s, bound %s: anchoring mark %s to basemark %s => (%p,%p)", + cref(kind,chainname,chainlookupname,lookupname),anchor,bound,gref(markchar),gref(basechar),dx,dy) + end + return head, start, true + end + end + end + if trace_bugs then + logwarning("%s: no matching anchors for mark %s and basemark %s",gref(kind,chainname,chainlookupname,lookupname),gref(markchar),gref(basechar)) + end + end + end + elseif trace_bugs then + logwarning("%s: prev node is no mark",cref(kind,chainname,chainlookupname,lookupname)) + end + elseif trace_bugs then + logwarning("%s: mark %s has no anchors",cref(kind,chainname,chainlookupname,lookupname),gref(markchar)) + end + -- elseif trace_marks and trace_details then + -- logprocess("%s, mark %s is already bound (n=%s), ignoring mark2mark",pref(kind,lookupname),gref(markchar),alreadydone) + -- end + elseif trace_bugs then + logwarning("%s: mark %s is no mark",cref(kind,chainname,chainlookupname),gref(markchar)) + end + return head, start, false +end + +function chainprocs.gpos_cursive(head,start,stop,kind,chainname,currentcontext,lookuphash,currentlookup,chainlookupname) + local alreadydone = cursonce and start[a_cursbase] + if not alreadydone then + local startchar = start.char + local subtables = currentlookup.subtables + local lookupname = subtables[1] + local exitanchors = lookuphash[lookupname] + if exitanchors then + exitanchors = exitanchors[startchar] + end + if exitanchors then + local done = false + if marks[startchar] then + if trace_cursive then + logprocess("%s: ignoring cursive for mark %s",pref(kind,lookupname),gref(startchar)) + end + else + local nxt = start.next + while not done and nxt and nxt.id == glyph_code and nxt.font == currentfont and nxt.subtype<256 do + local nextchar = nxt.char + if marks[nextchar] then + -- should not happen (maybe warning) + nxt = nxt.next + else + local entryanchors = descriptions[nextchar] + if entryanchors then + entryanchors = entryanchors.anchors + if entryanchors then + entryanchors = entryanchors['centry'] + if entryanchors then + local al = anchorlookups[lookupname] + for anchor, entry in next, entryanchors do + if al[anchor] then + local exit = exitanchors[anchor] + if exit then + local dx, dy, bound = setcursive(start,nxt,tfmdata.parameters.factor,rlmode,exit,entry,characters[startchar],characters[nextchar]) + if trace_cursive then + logprocess("%s: moving %s to %s cursive (%p,%p) using anchor %s and bound %s in rlmode %s",pref(kind,lookupname),gref(startchar),gref(nextchar),dx,dy,anchor,bound,rlmode) + end + done = true + break + end + end + end + end + end + elseif trace_bugs then + -- logwarning("%s: char %s is missing in font",pref(kind,lookupname),gref(startchar)) + onetimemessage(currentfont,startchar,"no entry anchors",report_fonts) + end + break + end + end + end + return head, start, done + else + if trace_cursive and trace_details then + logprocess("%s, cursive %s is already done",pref(kind,lookupname),gref(start.char),alreadydone) + end + return head, start, false + end + end + return head, start, false +end + +function chainprocs.gpos_single(head,start,stop,kind,chainname,currentcontext,lookuphash,currentlookup,chainlookupname,chainindex,sequence) + -- untested .. needs checking for the new model + local startchar = start.char + local subtables = currentlookup.subtables + local lookupname = subtables[1] + local kerns = lookuphash[lookupname] + if kerns then + kerns = kerns[startchar] -- needed ? + if kerns then + local dx, dy, w, h = setpair(start,tfmdata.parameters.factor,rlmode,sequence.flags[4],kerns,characters[startchar]) + if trace_kerns then + logprocess("%s: shifting single %s by (%p,%p) and correction (%p,%p)",cref(kind,chainname,chainlookupname),gref(startchar),dx,dy,w,h) + end + end + end + return head, start, false +end + +chainmores.gpos_single = chainprocs.gpos_single -- okay? + +-- when machines become faster i will make a shared function + +function chainprocs.gpos_pair(head,start,stop,kind,chainname,currentcontext,lookuphash,currentlookup,chainlookupname,chainindex,sequence) + local snext = start.next + if snext then + local startchar = start.char + local subtables = currentlookup.subtables + local lookupname = subtables[1] + local kerns = lookuphash[lookupname] + if kerns then + kerns = kerns[startchar] + if kerns then + local lookuptype = lookuptypes[lookupname] + local prev, done = start, false + local factor = tfmdata.parameters.factor + while snext and snext.id == glyph_code and snext.font == currentfont and snext.subtype<256 do + local nextchar = snext.char + local krn = kerns[nextchar] + if not krn and marks[nextchar] then + prev = snext + snext = snext.next + else + if not krn then + -- skip + elseif type(krn) == "table" then + if lookuptype == "pair" then + local a, b = krn[2], krn[3] + if a and #a > 0 then + local startchar = start.char + local x, y, w, h = setpair(start,factor,rlmode,sequence.flags[4],a,characters[startchar]) + if trace_kerns then + logprocess("%s: shifting first of pair %s and %s by (%p,%p) and correction (%p,%p)",cref(kind,chainname,chainlookupname),gref(startchar),gref(nextchar),x,y,w,h) + end + end + if b and #b > 0 then + local startchar = start.char + local x, y, w, h = setpair(snext,factor,rlmode,sequence.flags[4],b,characters[nextchar]) + if trace_kerns then + logprocess("%s: shifting second of pair %s and %s by (%p,%p) and correction (%p,%p)",cref(kind,chainname,chainlookupname),gref(startchar),gref(nextchar),x,y,w,h) + end + end + else + report_process("%s: check this out (old kern stuff)",cref(kind,chainname,chainlookupname)) + local a, b = krn[2], krn[6] + if a and a ~= 0 then + local k = setkern(snext,factor,rlmode,a) + if trace_kerns then + logprocess("%s: inserting first kern %s between %s and %s",cref(kind,chainname,chainlookupname),k,gref(prev.char),gref(nextchar)) + end + end + if b and b ~= 0 then + logwarning("%s: ignoring second kern xoff %s",cref(kind,chainname,chainlookupname),b*factor) + end + end + done = true + elseif krn ~= 0 then + local k = setkern(snext,factor,rlmode,krn) + if trace_kerns then + logprocess("%s: inserting kern %s between %s and %s",cref(kind,chainname,chainlookupname),k,gref(prev.char),gref(nextchar)) + end + done = true + end + break + end + end + return head, start, done + end + end + end + return head, start, false +end + +chainmores.gpos_pair = chainprocs.gpos_pair -- okay? + +-- what pointer to return, spec says stop +-- to be discussed ... is bidi changer a space? +-- elseif char == zwnj and sequence[n][32] then -- brrr + +-- somehow l or f is global +-- we don't need to pass the currentcontext, saves a bit +-- make a slow variant then can be activated but with more tracing + +local function show_skip(kind,chainname,char,ck,class) + if ck[9] then + logwarning("%s: skipping char %s, class %a, rule %a, lookuptype %a, %a => %a",cref(kind,chainname),gref(char),class,ck[1],ck[2],ck[9],ck[10]) + else + logwarning("%s: skipping char %s, class %a, rule %a, lookuptype %a",cref(kind,chainname),gref(char),class,ck[1],ck[2]) + end +end + +local function normal_handle_contextchain(head,start,kind,chainname,contexts,sequence,lookuphash) + -- local rule, lookuptype, sequence, f, l, lookups = ck[1], ck[2] ,ck[3], ck[4], ck[5], ck[6] + local flags = sequence.flags + local done = false + local skipmark = flags[1] + local skipligature = flags[2] + local skipbase = flags[3] + local someskip = skipmark or skipligature or skipbase -- could be stored in flags for a fast test (hm, flags could be false !) + local markclass = sequence.markclass -- todo, first we need a proper test + local skipped = false + for k=1,#contexts do + local match = true + local current = start + local last = start + local ck = contexts[k] + local seq = ck[3] + local s = #seq + -- f..l = mid string + if s == 1 then + -- never happens + match = current.id == glyph_code and current.font == currentfont and current.subtype<256 and seq[1][current.char] + else + -- maybe we need a better space check (maybe check for glue or category or combination) + -- we cannot optimize for n=2 because there can be disc nodes + local f, l = ck[4], ck[5] + -- current match + if f == 1 and f == l then -- current only + -- already a hit + -- match = true + else -- before/current/after | before/current | current/after + -- no need to test first hit (to be optimized) + if f == l then -- new, else last out of sync (f is > 1) + -- match = true + else + local n = f + 1 + last = last.next + while n <= l do + if last then + local id = last.id + if id == glyph_code then + if last.font == currentfont and last.subtype<256 then + local char = last.char + local ccd = descriptions[char] + if ccd then + local class = ccd.class + if class == skipmark or class == skipligature or class == skipbase or (markclass and class == "mark" and not markclass[char]) then + skipped = true + if trace_skips then + show_skip(kind,chainname,char,ck,class) + end + last = last.next + elseif seq[n][char] then + if n < l then + last = last.next + end + n = n + 1 + else + match = false + break + end + else + match = false + break + end + else + match = false + break + end + elseif id == disc_code then + last = last.next + else + match = false + break + end + else + match = false + break + end + end + end + end + -- before + if match and f > 1 then + local prev = start.prev + if prev then + local n = f-1 + while n >= 1 do + if prev then + local id = prev.id + if id == glyph_code then + if prev.font == currentfont and prev.subtype<256 then -- normal char + local char = prev.char + local ccd = descriptions[char] + if ccd then + local class = ccd.class + if class == skipmark or class == skipligature or class == skipbase or (markclass and class == "mark" and not markclass[char]) then + skipped = true + if trace_skips then + show_skip(kind,chainname,char,ck,class) + end + elseif seq[n][char] then + n = n -1 + else + match = false + break + end + else + match = false + break + end + else + match = false + break + end + elseif id == disc_code then + -- skip 'm + elseif seq[n][32] then + n = n -1 + else + match = false + break + end + prev = prev.prev + elseif seq[n][32] then -- somewhat special, as zapfino can have many preceding spaces + n = n -1 + else + match = false + break + end + end + elseif f == 2 then + match = seq[1][32] + else + for n=f-1,1 do + if not seq[n][32] then + match = false + break + end + end + end + end + -- after + if match and s > l then + local current = last and last.next + if current then + -- removed optimization for s-l == 1, we have to deal with marks anyway + local n = l + 1 + while n <= s do + if current then + local id = current.id + if id == glyph_code then + if current.font == currentfont and current.subtype<256 then -- normal char + local char = current.char + local ccd = descriptions[char] + if ccd then + local class = ccd.class + if class == skipmark or class == skipligature or class == skipbase or (markclass and class == "mark" and not markclass[char]) then + skipped = true + if trace_skips then + show_skip(kind,chainname,char,ck,class) + end + elseif seq[n][char] then + n = n + 1 + else + match = false + break + end + else + match = false + break + end + else + match = false + break + end + elseif id == disc_code then + -- skip 'm + elseif seq[n][32] then -- brrr + n = n + 1 + else + match = false + break + end + current = current.next + elseif seq[n][32] then + n = n + 1 + else + match = false + break + end + end + elseif s-l == 1 then + match = seq[s][32] + else + for n=l+1,s do + if not seq[n][32] then + match = false + break + end + end + end + end + end + if match then + -- ck == currentcontext + if trace_contexts then + local rule, lookuptype, f, l = ck[1], ck[2], ck[4], ck[5] + local char = start.char + if ck[9] then + logwarning("%s: rule %s matches at char %s for (%s,%s,%s) chars, lookuptype %a, %a => %a", + cref(kind,chainname),rule,gref(char),f-1,l-f+1,s-l,lookuptype,ck[9],ck[10]) + else + logwarning("%s: rule %s matches at char %s for (%s,%s,%s) chars, lookuptype %a", + cref(kind,chainname),rule,gref(char),f-1,l-f+1,s-l,lookuptype) + end + end + local chainlookups = ck[6] + if chainlookups then + local nofchainlookups = #chainlookups + -- we can speed this up if needed + if nofchainlookups == 1 then + local chainlookupname = chainlookups[1] + local chainlookup = lookuptable[chainlookupname] + if chainlookup then + local cp = chainprocs[chainlookup.type] + if cp then + local ok + head, start, ok = cp(head,start,last,kind,chainname,ck,lookuphash,chainlookup,chainlookupname,nil,sequence) + if ok then + done = true + end + else + logprocess("%s: %s is not yet supported",cref(kind,chainname,chainlookupname),chainlookup.type) + end + else -- shouldn't happen + logprocess("%s is not yet supported",cref(kind,chainname,chainlookupname)) + end + else + local i = 1 + repeat + if skipped then + while true do + local char = start.char + local ccd = descriptions[char] + if ccd then + local class = ccd.class + if class == skipmark or class == skipligature or class == skipbase or (markclass and class == "mark" and not markclass[char]) then + start = start.next + else + break + end + else + break + end + end + end + local chainlookupname = chainlookups[i] + local chainlookup = lookuptable[chainlookupname] + if not chainlookup then + -- okay, n matches, < n replacements + i = i + 1 + else + local cp = chainmores[chainlookup.type] + if not cp then + -- actually an error + logprocess("%s: %s is not yet supported",cref(kind,chainname,chainlookupname),chainlookup.type) + i = i + 1 + else + local ok, n + head, start, ok, n = cp(head,start,last,kind,chainname,ck,lookuphash,chainlookup,chainlookupname,i,sequence) + -- messy since last can be changed ! + if ok then + done = true + -- skip next one(s) if ligature + i = i + (n or 1) + else + i = i + 1 + end + end + end + if start then + start = start.next + else + -- weird + end + until i > nofchainlookups + end + else + local replacements = ck[7] + if replacements then + head, start, done = chainprocs.reversesub(head,start,last,kind,chainname,ck,lookuphash,replacements) -- sequence + else + done = true -- can be meant to be skipped + if trace_contexts then + logprocess("%s: skipping match",cref(kind,chainname)) + end + end + end + end + end + return head, start, done +end + +-- Because we want to keep this elsewhere (an because speed is less an issue) we +-- pass the font id so that the verbose variant can access the relevant helper tables. + +local verbose_handle_contextchain = function(font,...) + logwarning("no verbose handler installed, reverting to 'normal'") + otf.setcontextchain() + return normal_handle_contextchain(...) +end + +otf.chainhandlers = { + normal = normal_handle_contextchain, + verbose = verbose_handle_contextchain, +} + +function otf.setcontextchain(method) + if not method or method == "normal" or not otf.chainhandlers[method] then + if handlers.contextchain then -- no need for a message while making the format + logwarning("installing normal contextchain handler") + end + handlers.contextchain = normal_handle_contextchain + else + logwarning("installing contextchain handler %a",method) + local handler = otf.chainhandlers[method] + handlers.contextchain = function(...) + return handler(currentfont,...) -- hm, get rid of ... + end + end + handlers.gsub_context = handlers.contextchain + handlers.gsub_contextchain = handlers.contextchain + handlers.gsub_reversecontextchain = handlers.contextchain + handlers.gpos_contextchain = handlers.contextchain + handlers.gpos_context = handlers.contextchain +end + +otf.setcontextchain() + +local missing = { } -- we only report once + +local function logprocess(...) + if trace_steps then + registermessage(...) + end + report_process(...) +end + +local logwarning = report_process + +local function report_missing_cache(typ,lookup) + local f = missing[currentfont] if not f then f = { } missing[currentfont] = f end + local t = f[typ] if not t then t = { } f[typ] = t end + if not t[lookup] then + t[lookup] = true + logwarning("missing cache for lookup %a, type %a, font %a, name %a",lookup,typ,currentfont,tfmdata.properties.fullname) + end +end + +local resolved = { } -- we only resolve a font,script,language pair once + +-- todo: pass all these 'locals' in a table + +local lookuphashes = { } + +setmetatableindex(lookuphashes, function(t,font) + local lookuphash = fontdata[font].resources.lookuphash + if not lookuphash or not next(lookuphash) then + lookuphash = false + end + t[font] = lookuphash + return lookuphash +end) + +-- fonts.hashes.lookups = lookuphashes + +local autofeatures = fonts.analyzers.features -- was: constants + +local function initialize(sequence,script,language,enabled) + local features = sequence.features + if features then + for kind, scripts in next, features do + local valid = enabled[kind] + if valid then + local languages = scripts[script] or scripts[wildcard] + if languages and (languages[language] or languages[wildcard]) then + return { valid, autofeatures[kind] or false, sequence.chain or 0, kind, sequence } + end + end + end + end + return false +end + +function otf.dataset(tfmdata,font) -- generic variant, overloaded in context + local shared = tfmdata.shared + local properties = tfmdata.properties + local language = properties.language or "dflt" + local script = properties.script or "dflt" + local enabled = shared.features + local res = resolved[font] + if not res then + res = { } + resolved[font] = res + end + local rs = res[script] + if not rs then + rs = { } + res[script] = rs + end + local rl = rs[language] + if not rl then + rl = { + -- indexed but we can also add specific data by key + } + rs[language] = rl + local sequences = tfmdata.resources.sequences +-- setmetatableindex(rl, function(t,k) +-- if type(k) == "number" then +-- local v = enabled and initialize(sequences[k],script,language,enabled) +-- t[k] = v +-- return v +-- end +-- end) +for s=1,#sequences do + local v = enabled and initialize(sequences[s],script,language,enabled) + if v then + rl[#rl+1] = v + end +end + end + return rl +end + +-- elseif id == glue_code then +-- if p[5] then -- chain +-- local pc = pp[32] +-- if pc then +-- start, ok = start, false -- p[1](start,kind,p[2],pc,p[3],p[4]) +-- if ok then +-- done = true +-- end +-- if start then start = start.next end +-- else +-- start = start.next +-- end +-- else +-- start = start.next +-- end + +-- there will be a new direction parser (pre-parsed etc) + +-- less bytecode: 290 -> 254 +-- +-- attr = attr or false +-- +-- local a = getattr(start,0) +-- if (a == attr and (not attribute or getattr(start,a_state) == attribute)) or (not attribute or getattr(start,a_state) == attribute) then +-- -- the action +-- end + +local function featuresprocessor(head,font,attr) + + local lookuphash = lookuphashes[font] -- we can also check sequences here + + if not lookuphash then + return head, false + end + + if trace_steps then + checkstep(head) + end + + tfmdata = fontdata[font] + descriptions = tfmdata.descriptions + characters = tfmdata.characters + resources = tfmdata.resources + + marks = resources.marks + anchorlookups = resources.lookup_to_anchor + lookuptable = resources.lookups + lookuptypes = resources.lookuptypes + + currentfont = font + rlmode = 0 + + local sequences = resources.sequences + local done = false + local datasets = otf.dataset(tfmdata,font,attr) + + local dirstack = { } -- could move outside function + + -- We could work on sub start-stop ranges instead but I wonder if there is that + -- much speed gain (experiments showed that it made not much sense) and we need + -- to keep track of directions anyway. Also at some point I want to play with + -- font interactions and then we do need the full sweeps. + + -- Keeping track of the headnode is needed for devanagari (I generalized it a bit + -- so that multiple cases are also covered.) + + for s=1,#datasets do + local dataset = datasets[s] + featurevalue = dataset[1] -- todo: pass to function instead of using a global + + local sequence = dataset[5] -- sequences[s] -- also dataset[5] + local rlparmode = 0 + local topstack = 0 + local success = false + local attribute = dataset[2] + local chain = dataset[3] -- sequence.chain or 0 + local typ = sequence.type + local subtables = sequence.subtables + if chain < 0 then + -- this is a limited case, no special treatments like 'init' etc + local handler = handlers[typ] + -- we need to get rid of this slide! probably no longer needed in latest luatex + local start = find_node_tail(head) -- slow (we can store tail because there's always a skip at the end): todo + while start do + local id = start.id + if id == glyph_code then + if start.font == font and start.subtype<256 then + local a = start[0] + if a then + a = a == attr + else + a = true + end + if a then + for i=1,#subtables do + local lookupname = subtables[i] + local lookupcache = lookuphash[lookupname] + if lookupcache then + local lookupmatch = lookupcache[start.char] + if lookupmatch then + head, start, success = handler(head,start,dataset[4],lookupname,lookupmatch,sequence,lookuphash,i) + if success then + break + end + end + else + report_missing_cache(typ,lookupname) + end + end + if start then start = start.prev end + else + start = start.prev + end + else + start = start.prev + end + else + start = start.prev + end + end + else + local handler = handlers[typ] + local ns = #subtables + local start = head -- local ? + rlmode = 0 -- to be checked ? + if ns == 1 then -- happens often + local lookupname = subtables[1] + local lookupcache = lookuphash[lookupname] + if not lookupcache then -- also check for empty cache + report_missing_cache(typ,lookupname) + else + + local function subrun(start) + -- mostly for gsub, gpos would demand a more clever approach + local head = start + local done = false + while start do + local id = start.id + if id == glyph_code and start.font == font and start.subtype <256 then + local a = start[0] + if a then + a = (a == attr) and (not attribute or start[a_state] == attribute) + else + a = not attribute or start[a_state] == attribute + end + if a then + local lookupmatch = lookupcache[start.char] + if lookupmatch then + -- sequence kan weg + local ok + head, start, ok = handler(head,start,dataset[4],lookupname,lookupmatch,sequence,lookuphash,1) + if ok then + done = true + end + end + if start then start = start.next end + else + start = start.next + end + else + start = start.next + end + end + if done then + success = true + return head + end + end + + local function kerndisc(disc) -- we can assume that prev and next are glyphs + local prev = disc.prev + local next = disc.next + if prev and next then + prev.next = next + -- next.prev = prev + local a = prev[0] + if a then + a = (a == attr) and (not attribute or prev[a_state] == attribute) + else + a = not attribute or prev[a_state] == attribute + end + if a then + local lookupmatch = lookupcache[prev.char] + if lookupmatch then + -- sequence kan weg + local h, d, ok = handler(head,prev,dataset[4],lookupname,lookupmatch,sequence,lookuphash,1) + if ok then + done = true + success = true + end + end + end + prev.next = disc + -- next.prev = disc + end + return next + end + + while start do + local id = start.id + if id == glyph_code then + if start.font == font and start.subtype<256 then + local a = start[0] + if a then + a = (a == attr) and (not attribute or start[a_state] == attribute) + else + a = not attribute or start[a_state] == attribute + end + if a then + local lookupmatch = lookupcache[start.char] + if lookupmatch then + -- sequence kan weg + local ok + head, start, ok = handler(head,start,dataset[4],lookupname,lookupmatch,sequence,lookuphash,1) + if ok then + success = true + end + end + if start then start = start.next end + else + start = start.next + end + else + start = start.next + end + elseif id == disc_code then + -- mostly for gsub + if start.subtype == discretionary_code then + local pre = start.pre + if pre then + local new = subrun(pre) + if new then start.pre = new end + end + local post = start.post + if post then + local new = subrun(post) + if new then start.post = new end + end + local replace = start.replace + if replace then + local new = subrun(replace) + if new then start.replace = new end + end +elseif typ == "gpos_single" or typ == "gpos_pair" then + kerndisc(start) + end + start = start.next + elseif id == whatsit_code then -- will be function + local subtype = start.subtype + if subtype == dir_code then + local dir = start.dir + if dir == "+TRT" or dir == "+TLT" then + topstack = topstack + 1 + dirstack[topstack] = dir + elseif dir == "-TRT" or dir == "-TLT" then + topstack = topstack - 1 + end + local newdir = dirstack[topstack] + if newdir == "+TRT" then + rlmode = -1 + elseif newdir == "+TLT" then + rlmode = 1 + else + rlmode = rlparmode + end + if trace_directions then + report_process("directions after txtdir %a: parmode %a, txtmode %a, # stack %a, new dir %a",dir,rlparmode,rlmode,topstack,newdir) + end + elseif subtype == localpar_code then + local dir = start.dir + if dir == "TRT" then + rlparmode = -1 + elseif dir == "TLT" then + rlparmode = 1 + else + rlparmode = 0 + end + -- one might wonder if the par dir should be looked at, so we might as well drop the next line + rlmode = rlparmode + if trace_directions then + report_process("directions after pardir %a: parmode %a, txtmode %a",dir,rlparmode,rlmode) + end + end + start = start.next + elseif id == math_code then + start = end_of_math(start).next + else + start = start.next + end + end + end + else + + local function subrun(start) + -- mostly for gsub, gpos would demand a more clever approach + local head = start + local done = false + while start do + local id = start.id + if id == glyph_code and start.id == font and start.subtype <256 then + local a = start[0] + if a then + a = (a == attr) and (not attribute or start[a_state] == attribute) + else + a = not attribute or start[a_state] == attribute + end + if a then + for i=1,ns do + local lookupname = subtables[i] + local lookupcache = lookuphash[lookupname] + if lookupcache then + local lookupmatch = lookupcache[start.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[4],lookupname,lookupmatch,sequence,lookuphash,i) + if ok then + done = true + break + elseif not start then + -- don't ask why ... shouldn't happen + break + end + end + else + report_missing_cache(typ,lookupname) + end + end + if start then start = start.next end + else + start = start.next + end + else + start = start.next + end + end + if done then + success = true + return head + end + end + + local function kerndisc(disc) -- we can assume that prev and next are glyphs + local prev = disc.prev + local next = disc.next + if prev and next then + prev.next = next + -- next.prev = prev + local a = prev[0] + if a then + a = (a == attr) and (not attribute or prev[a_state] == attribute) + else + a = not attribute or prev[a_state] == attribute + end + if a then + for i=1,ns do + local lookupname = subtables[i] + local lookupcache = lookuphash[lookupname] + if lookupcache then + local lookupmatch = lookupcache[prev.char] + if lookupmatch then + -- we could move all code inline but that makes things even more unreadable + local h, d, ok = handler(head,prev,dataset[4],lookupname,lookupmatch,sequence,lookuphash,i) + if ok then + done = true + break + end + end + else + report_missing_cache(typ,lookupname) + end + end + end + prev.next = disc + -- next.prev = disc + end + return next + end + + while start do + local id = start.id + if id == glyph_code then + if start.font == font and start.subtype<256 then + local a = start[0] + if a then + a = (a == attr) and (not attribute or start[a_state] == attribute) + else + a = not attribute or start[a_state] == attribute + end + if a then + for i=1,ns do + local lookupname = subtables[i] + local lookupcache = lookuphash[lookupname] + if lookupcache then + local lookupmatch = lookupcache[start.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[4],lookupname,lookupmatch,sequence,lookuphash,i) + if ok then + success = true + break + elseif not start then + -- don't ask why ... shouldn't happen + break + end + end + else + report_missing_cache(typ,lookupname) + end + end + if start then start = start.next end + else + start = start.next + end + else + start = start.next + end + elseif id == disc_code then + -- mostly for gsub + if start.subtype == discretionary_code then + local pre = start.pre + if pre then + local new = subrun(pre) + if new then start.pre = new end + end + local post = start.post + if post then + local new = subrun(post) + if new then start.post = new end + end + local replace = start.replace + if replace then + local new = subrun(replace) + if new then start.replace = new end + end +elseif typ == "gpos_single" or typ == "gpos_pair" then + kerndisc(start) + end + start = start.next + elseif id == whatsit_code then + local subtype = start.subtype + if subtype == dir_code then + local dir = start.dir + if dir == "+TRT" or dir == "+TLT" then + topstack = topstack + 1 + dirstack[topstack] = dir + elseif dir == "-TRT" or dir == "-TLT" then + topstack = topstack - 1 + end + local newdir = dirstack[topstack] + if newdir == "+TRT" then + rlmode = -1 + elseif newdir == "+TLT" then + rlmode = 1 + else + rlmode = rlparmode + end + if trace_directions then + report_process("directions after txtdir %a: parmode %a, txtmode %a, # stack %a, new dir %a",dir,rlparmode,rlmode,topstack,newdir) + end + elseif subtype == localpar_code then + local dir = start.dir + if dir == "TRT" then + rlparmode = -1 + elseif dir == "TLT" then + rlparmode = 1 + else + rlparmode = 0 + end + rlmode = rlparmode + if trace_directions then + report_process("directions after pardir %a: parmode %a, txtmode %a",dir,rlparmode,rlmode) + end + end + start = start.next + elseif id == math_code then + start = end_of_math(start).next + else + start = start.next + end + end + end + end + if success then + done = true + end + if trace_steps then -- ? + registerstep(head) + end + end + return head, done +end + +local function generic(lookupdata,lookupname,unicode,lookuphash) + local target = lookuphash[lookupname] + if target then + target[unicode] = lookupdata + else + lookuphash[lookupname] = { [unicode] = lookupdata } + end +end + +local action = { + + substitution = generic, + multiple = generic, + alternate = generic, + position = generic, + + ligature = function(lookupdata,lookupname,unicode,lookuphash) + local target = lookuphash[lookupname] + if not target then + target = { } + lookuphash[lookupname] = target + end + for i=1,#lookupdata do + local li = lookupdata[i] + local tu = target[li] + if not tu then + tu = { } + target[li] = tu + end + target = tu + end + target.ligature = unicode + end, + + pair = function(lookupdata,lookupname,unicode,lookuphash) + local target = lookuphash[lookupname] + if not target then + target = { } + lookuphash[lookupname] = target + end + local others = target[unicode] + local paired = lookupdata[1] + if others then + others[paired] = lookupdata + else + others = { [paired] = lookupdata } + target[unicode] = others + end + end, + +} + +local function prepare_lookups(tfmdata) + + local rawdata = tfmdata.shared.rawdata + local resources = rawdata.resources + local lookuphash = resources.lookuphash + local anchor_to_lookup = resources.anchor_to_lookup + local lookup_to_anchor = resources.lookup_to_anchor + local lookuptypes = resources.lookuptypes + local characters = tfmdata.characters + local descriptions = tfmdata.descriptions + + -- we cannot free the entries in the descriptions as sometimes we access + -- then directly (for instance anchors) ... selectively freeing does save + -- much memory as it's only a reference to a table and the slot in the + -- description hash is not freed anyway + + for unicode, character in next, characters do -- we cannot loop over descriptions ! + + local description = descriptions[unicode] + + if description then + + local lookups = description.slookups + if lookups then + for lookupname, lookupdata in next, lookups do + action[lookuptypes[lookupname]](lookupdata,lookupname,unicode,lookuphash) + end + end + + local lookups = description.mlookups + if lookups then + for lookupname, lookuplist in next, lookups do + local lookuptype = lookuptypes[lookupname] + for l=1,#lookuplist do + local lookupdata = lookuplist[l] + action[lookuptype](lookupdata,lookupname,unicode,lookuphash) + end + end + end + + local list = description.kerns + if list then + for lookup, krn in next, list do -- ref to glyph, saves lookup + local target = lookuphash[lookup] + if target then + target[unicode] = krn + else + lookuphash[lookup] = { [unicode] = krn } + end + end + end + + local list = description.anchors + if list then + for typ, anchors in next, list do -- types + if typ == "mark" or typ == "cexit" then -- or entry? + for name, anchor in next, anchors do + local lookups = anchor_to_lookup[name] + if lookups then + for lookup, _ in next, lookups do + local target = lookuphash[lookup] + if target then + target[unicode] = anchors + else + lookuphash[lookup] = { [unicode] = anchors } + end + end + end + end + end + end + end + + end + + end + +end + +local function split(replacement,original) + local result = { } + for i=1,#replacement do + result[original[i]] = replacement[i] + end + return result +end + +local valid = { + coverage = { chainsub = true, chainpos = true, contextsub = true }, + reversecoverage = { reversesub = true }, + glyphs = { chainsub = true, chainpos = true }, +} + +local function prepare_contextchains(tfmdata) + local rawdata = tfmdata.shared.rawdata + local resources = rawdata.resources + local lookuphash = resources.lookuphash + local lookups = rawdata.lookups + if lookups then + for lookupname, lookupdata in next, rawdata.lookups do + local lookuptype = lookupdata.type + if lookuptype then + local rules = lookupdata.rules + if rules then + local format = lookupdata.format + local validformat = valid[format] + if not validformat then + report_prepare("unsupported format %a",format) + elseif not validformat[lookuptype] then + -- todo: dejavu-serif has one (but i need to see what use it has) + report_prepare("unsupported format %a, lookuptype %a, lookupname %a",format,lookuptype,lookupname) + else + local contexts = lookuphash[lookupname] + if not contexts then + contexts = { } + lookuphash[lookupname] = contexts + end + local t, nt = { }, 0 + for nofrules=1,#rules do + local rule = rules[nofrules] + local current = rule.current + local before = rule.before + local after = rule.after + local replacements = rule.replacements + local sequence = { } + local nofsequences = 0 + -- Eventually we can store start, stop and sequence in the cached file + -- but then less sharing takes place so best not do that without a lot + -- of profiling so let's forget about it. + if before then + for n=1,#before do + nofsequences = nofsequences + 1 + sequence[nofsequences] = before[n] + end + end + local start = nofsequences + 1 + for n=1,#current do + nofsequences = nofsequences + 1 + sequence[nofsequences] = current[n] + end + local stop = nofsequences + if after then + for n=1,#after do + nofsequences = nofsequences + 1 + sequence[nofsequences] = after[n] + end + end + if sequence[1] then + -- Replacements only happen with reverse lookups as they are single only. We + -- could pack them into current (replacement value instead of true) and then + -- use sequence[start] instead but it's somewhat ugly. + nt = nt + 1 + t[nt] = { nofrules, lookuptype, sequence, start, stop, rule.lookups, replacements } + for unic, _ in next, sequence[start] do + local cu = contexts[unic] + if not cu then + contexts[unic] = t + end + end + end + end + end + else + -- no rules + end + else + report_prepare("missing lookuptype for lookupname %a",lookupname) + end + end + end +end + +-- we can consider lookuphash == false (initialized but empty) vs lookuphash == table + +local function featuresinitializer(tfmdata,value) + if true then -- value then + -- beware we need to use the topmost properties table + local rawdata = tfmdata.shared.rawdata + local properties = rawdata.properties + if not properties.initialized then + local starttime = trace_preparing and os.clock() + local resources = rawdata.resources + resources.lookuphash = resources.lookuphash or { } + prepare_contextchains(tfmdata) + prepare_lookups(tfmdata) + properties.initialized = true + if trace_preparing then + report_prepare("preparation time is %0.3f seconds for %a",os.clock()-starttime,tfmdata.properties.fullname) + end + end + end +end + +registerotffeature { + name = "features", + description = "features", + default = true, + initializers = { + position = 1, + node = featuresinitializer, + }, + processors = { + node = featuresprocessor, + } +} + +-- This can be used for extra handlers, but should be used with care! + +otf.handlers = handlers |