diff options
Diffstat (limited to 'Master/texmf-dist/tex/context/base/mkiv/font-ota.lua')
| -rw-r--r-- | Master/texmf-dist/tex/context/base/mkiv/font-ota.lua | 442 |
1 files changed, 442 insertions, 0 deletions
diff --git a/Master/texmf-dist/tex/context/base/mkiv/font-ota.lua b/Master/texmf-dist/tex/context/base/mkiv/font-ota.lua new file mode 100644 index 00000000000..6a3804a7425 --- /dev/null +++ b/Master/texmf-dist/tex/context/base/mkiv/font-ota.lua @@ -0,0 +1,442 @@ +if not modules then modules = { } end modules ['font-ota'] = { + version = 1.001, + comment = "companion to font-otf.lua (analysing)", + author = "Hans Hagen, PRAGMA-ADE, Hasselt NL", + copyright = "PRAGMA ADE / ConTeXt Development Team", + license = "see context related readme files" +} + +-- context only + +local type = type + +if not trackers then trackers = { register = function() end } end + +----- trace_analyzing = false trackers.register("otf.analyzing", function(v) trace_analyzing = v end) + +local fonts, nodes, node = fonts, nodes, node + +local allocate = utilities.storage.allocate + +local otf = fonts.handlers.otf + +local analyzers = fonts.analyzers +local initializers = allocate() +local methods = allocate() + +analyzers.initializers = initializers +analyzers.methods = methods + +local a_state = attributes.private('state') + +local nuts = nodes.nuts +local tonut = nuts.tonut + +local getfield = nuts.getfield +local getnext = nuts.getnext +local getprev = nuts.getprev +local getprev = nuts.getprev +local getprop = nuts.getprop +local setprop = nuts.setprop +local getfont = nuts.getfont +local getsubtype = nuts.getsubtype +local getchar = nuts.getchar +local ischar = nuts.is_char + +local traverse_id = nuts.traverse_id +local traverse_node_list = nuts.traverse +local end_of_math = nuts.end_of_math + +local nodecodes = nodes.nodecodes +----- glyph_code = nodecodes.glyph +local disc_code = nodecodes.disc +local math_code = nodecodes.math + +local fontdata = fonts.hashes.identifiers +local categories = characters and characters.categories or { } -- sorry, only in context +local chardata = characters and characters.data + +local otffeatures = fonts.constructors.newfeatures("otf") +local registerotffeature = otffeatures.register + +--[[ldx-- +<p>Analyzers run per script and/or language and are needed in order to +process features right.</p> +--ldx]]-- + +-- never use these numbers directly + +local s_init = 1 local s_rphf = 7 +local s_medi = 2 local s_half = 8 +local s_fina = 3 local s_pref = 9 +local s_isol = 4 local s_blwf = 10 +local s_mark = 5 local s_pstf = 11 +local s_rest = 6 + +local states = { + init = s_init, + medi = s_medi, + med2 = s_medi, + fina = s_fina, + fin2 = s_fina, + fin3 = s_fina, + isol = s_isol, + mark = s_mark, + rest = s_rest, + rphf = s_rphf, + half = s_half, + pref = s_pref, + blwf = s_blwf, + pstf = s_pstf, +} + +local features = { + init = s_init, + medi = s_medi, + med2 = s_medi, + fina = s_fina, + fin2 = s_fina, + fin3 = s_fina, + isol = s_isol, + -- mark = s_mark, + -- rest = s_rest, + rphf = s_rphf, + half = s_half, + pref = s_pref, + blwf = s_blwf, + pstf = s_pstf, +} + +analyzers.states = states +analyzers.features = features +analyzers.useunicodemarks = false + +-- todo: analyzers per script/lang, cross font, so we need an font id hash -> script +-- e.g. latin -> hyphenate, arab -> 1/2/3 analyze -- its own namespace + +function analyzers.setstate(head,font) + local useunicodemarks = analyzers.useunicodemarks + local tfmdata = fontdata[font] + local descriptions = tfmdata.descriptions + local first, last, current, n, done = nil, nil, head, 0, false -- maybe make n boolean + current = tonut(current) + while current do + local char, id = ischar(current,font) + if char and not getprop(current,a_state) then + done = true + local d = descriptions[char] + if d then + if d.class == "mark" then + done = true + setprop(current,a_state,s_mark) + elseif useunicodemarks and categories[char] == "mn" then + done = true + setprop(current,a_state,s_mark) + elseif n == 0 then + first, last, n = current, current, 1 + setprop(current,a_state,s_init) + else + last, n = current, n+1 + setprop(current,a_state,s_medi) + end + else -- finish + if first and first == last then + setprop(last,a_state,s_isol) + elseif last then + setprop(last,a_state,s_fina) + end + first, last, n = nil, nil, 0 + end + elseif char == false then + -- other font + if first and first == last then + setprop(last,a_state,s_isol) + elseif last then + setprop(last,a_state,s_fina) + end + first, last, n = nil, nil, 0 + if id == math_code then + current = end_of_math(current) + end + elseif id == disc_code then + -- always in the middle .. it doesn't make much sense to assign a property + -- here ... we might at some point decide to flag the components when present + -- but even then it's kind of bogus + setprop(current,a_state,s_medi) + last = current + else -- finish + if first and first == last then + setprop(last,a_state,s_isol) + elseif last then + setprop(last,a_state,s_fina) + end + first, last, n = nil, nil, 0 + if id == math_code then + current = end_of_math(current) + end + end + current = getnext(current) + end + if first and first == last then + setprop(last,a_state,s_isol) + elseif last then + setprop(last,a_state,s_fina) + end + return head, done +end + +-- in the future we will use language/script attributes instead of the +-- font related value, but then we also need dynamic features which is +-- somewhat slower; and .. we need a chain of them + +local function analyzeinitializer(tfmdata,value) -- attr + local script, language = otf.scriptandlanguage(tfmdata) -- attr + local action = initializers[script] + if not action then + -- skip + elseif type(action) == "function" then + return action(tfmdata,value) + else + local action = action[language] + if action then + return action(tfmdata,value) + end + end +end + +local function analyzeprocessor(head,font,attr) + local tfmdata = fontdata[font] + local script, language = otf.scriptandlanguage(tfmdata,attr) + local action = methods[script] + if not action then + -- skip + elseif type(action) == "function" then + return action(head,font,attr) + else + action = action[language] + if action then + return action(head,font,attr) + end + end + return head, false +end + +registerotffeature { + name = "analyze", + description = "analysis of character classes", + default = true, + initializers = { + node = analyzeinitializer, + }, + processors = { + position = 1, + node = analyzeprocessor, + } +} + +-- latin + +methods.latn = analyzers.setstate +-------.dflt = analyzers.setstate % can be an option or just the default + +local arab_warned = { } + +local function warning(current,what) + local char = getchar(current) + if not arab_warned[char] then + log.report("analyze","arab: character %C has no %a class",char,what) + arab_warned[char] = true + end +end + +local mappers = { + l = s_init, -- left + d = s_medi, -- double + c = s_medi, -- joiner + r = s_fina, -- right + u = s_isol, -- nonjoiner +} + +-- we can also use this trick for devanagari + +local classifiers = characters.classifiers + +if not classifiers then + + local first_arabic, last_arabic = characters.blockrange("arabic") + local first_syriac, last_syriac = characters.blockrange("syriac") + local first_mandiac, last_mandiac = characters.blockrange("mandiac") + local first_nko, last_nko = characters.blockrange("nko") + + classifiers = table.setmetatableindex(function(t,k) + local c = chardata[k] + local v = false + if c then + local arabic = c.arabic + if arabic then + v = mappers[arabic] + if not v then + log.report("analyze","error in mapping arabic %C",k) + -- error + v = false + end + elseif k >= first_arabic and k <= last_arabic or k >= first_syriac and k <= last_syriac or + k >= first_mandiac and k <= last_mandiac or k >= first_nko and k <= last_nko then + if categories[k] == "mn" then + v = s_mark + else + v = s_rest + end + end + end + t[k] = v + return v + end) + +end + +function methods.arab(head,font,attr) + local first, last = nil, nil + local c_first, c_last = nil, nil + local current, done = head, false + current = tonut(current) + while current do + local char, id = ischar(current,font) + if char and not getprop(current,a_state) then + done = true + local classifier = classifiers[char] + if not classifier then + if last then + if c_last == s_medi or c_last == s_fina then + setprop(last,a_state,s_fina) + else + warning(last,"fina") + setprop(last,a_state,s_error) + end + first, last = nil, nil + elseif first then + if c_first == s_medi or c_first == s_fina then + setprop(first,a_state,s_isol) + else + warning(first,"isol") + setprop(first,a_state,s_error) + end + first = nil + end + elseif classifier == s_mark then + setprop(current,a_state,s_mark) + elseif classifier == s_isol then + if last then + if c_last == s_medi or c_last == s_fina then + setprop(last,a_state,s_fina) + else + warning(last,"fina") + setprop(last,a_state,s_error) + end + first, last = nil, nil + elseif first then + if c_first == s_medi or c_first == s_fina then + setprop(first,a_state,s_isol) + else + warning(first,"isol") + setprop(first,a_state,s_error) + end + first = nil + end + setprop(current,a_state,s_isol) + elseif classifier == s_medi then + if first then + last = current + c_last = classifier + setprop(current,a_state,s_medi) + else + setprop(current,a_state,s_init) + first = current + c_first = classifier + end + elseif classifier == s_fina then + if last then + if getprop(last,a_state) ~= s_init then + setprop(last,a_state,s_medi) + end + setprop(current,a_state,s_fina) + first, last = nil, nil + elseif first then + -- if getprop(first,a_state) ~= s_init then + -- -- needs checking + -- setprop(first,a_state,s_medi) + -- end + setprop(current,a_state,s_fina) + first = nil + else + setprop(current,a_state,s_isol) + end + else -- classifier == s_rest + setprop(current,a_state,s_rest) + if last then + if c_last == s_medi or c_last == s_fina then + setprop(last,a_state,s_fina) + else + warning(last,"fina") + setprop(last,a_state,s_error) + end + first, last = nil, nil + elseif first then + if c_first == s_medi or c_first == s_fina then + setprop(first,a_state,s_isol) + else + warning(first,"isol") + setprop(first,a_state,s_error) + end + first = nil + end + end + else + if last then + if c_last == s_medi or c_last == s_fina then + setprop(last,a_state,s_fina) + else + warning(last,"fina") + setprop(last,a_state,s_error) + end + first, last = nil, nil + elseif first then + if c_first == s_medi or c_first == s_fina then + setprop(first,a_state,s_isol) + else + warning(first,"isol") + setprop(first,a_state,s_error) + end + first = nil + end + if id == math_code then -- a bit duplicate as we test for glyphs twice + current = end_of_math(current) + end + end + current = getnext(current) + end + if last then + if c_last == s_medi or c_last == s_fina then + setprop(last,a_state,s_fina) + else + warning(last,"fina") + setprop(last,a_state,s_error) + end + elseif first then + if c_first == s_medi or c_first == s_fina then + setprop(first,a_state,s_isol) + else + warning(first,"isol") + setprop(first,a_state,s_error) + end + end + return head, done +end + +methods.syrc = methods.arab +methods.mand = methods.arab +methods.nko = methods.arab + +directives.register("otf.analyze.useunicodemarks",function(v) + analyzers.useunicodemarks = v +end) |