diff options
Diffstat (limited to 'tex/context/base/math-fbk.lua')
| -rw-r--r-- | tex/context/base/math-fbk.lua | 99 |
1 files changed, 65 insertions, 34 deletions
diff --git a/tex/context/base/math-fbk.lua b/tex/context/base/math-fbk.lua index bd9a1d315..76dd1ad9b 100644 --- a/tex/context/base/math-fbk.lua +++ b/tex/context/base/math-fbk.lua @@ -20,7 +20,6 @@ local virtualcharacters = { } local identifiers = fonts.hashes.identifiers local lastmathids = fonts.hashes.lastmathids -local tounicode16 = fonts.mappings.tounicode16 -- we need a trick (todo): if we define scriptscript, script and text in -- that order we could use their id's .. i.e. we could always add a font @@ -133,10 +132,8 @@ function fallbacks.apply(target,original) else -- something else end - if trace_fallbacks then - if characters[k] then - report_fallbacks("extending math font %a with %U",target.properties.fullname,k) - end + if trace_fallbacks and characters[k] then + report_fallbacks("extending math font %a with %U",target.properties.fullname,k) end end end @@ -182,22 +179,22 @@ end -- virtualcharacters[0x208B] = 0x002B virtualcharacters[0x207A] = function(data) - data.replacement = 0x2212 + data.replacement = 0x002B return raised(data) end virtualcharacters[0x207B] = function(data) - data.replacement = 0x002B + data.replacement = 0x2212 return raised(data) end virtualcharacters[0x208A] = function(data) - data.replacement = 0x2212 + data.replacement = 0x002B return raised(data,true) end virtualcharacters[0x208B] = function(data) - data.replacement = 0x002B + data.replacement = 0x2212 return raised(data,true) end @@ -332,11 +329,12 @@ end -- we could move the defs from math-act here -local function accent_to_extensible(target,newchr,original,oldchr,height,depth,swap,offset) +local function accent_to_extensible(target,newchr,original,oldchr,height,depth,swap,offset,unicode) local characters = target.characters - local addprivate = fonts.helpers.addprivate local olddata = characters[oldchr] - if olddata and not olddata.commands then + -- brrr ... pagella has only next + if olddata and not olddata.commands and olddata.width > 0 then + local addprivate = fonts.helpers.addprivate if swap then swap = characters[swap] height = swap.depth @@ -351,6 +349,7 @@ local function accent_to_extensible(target,newchr,original,oldchr,height,depth,s width = olddata.width, height = height, depth = depth, + unicode = unicode, } local glyphdata = newdata local nextglyph = olddata.next @@ -399,9 +398,9 @@ local function accent_to_extensible(target,newchr,original,oldchr,height,depth,s end end end - return glyphdata + return glyphdata, true else - return olddata + return olddata, false end end @@ -416,7 +415,7 @@ virtualcharacters[0x203E] = function(data) -- could be FE33E instead height = target.parameters.xheight/4 depth = height end - return accent_to_extensible(target,0x203E,data.original,0x0305,height,depth) + return accent_to_extensible(target,0x203E,data.original,0x0305,height,depth,nil,nil,0x203E) end virtualcharacters[0xFE33E] = virtualcharacters[0x203E] -- convenient @@ -427,38 +426,40 @@ local function smashed(data,unicode,swap,private) local original = data.original local chardata = target.characters[unicode] if chardata and chardata.height > target.parameters.xheight then - return accent_to_extensible(target,private,original,unicode,0,0,swap) + return accent_to_extensible(target,private,original,unicode,0,0,swap,nil,unicode) else return original.characters[unicode] end end -addextra(0xFE3DE, { description="EXTENSIBLE OF 0x03DE", unicodeslot=0xFE3DE, mathextensible = "r", mathstretch = "h" } ) -addextra(0xFE3DC, { description="EXTENSIBLE OF 0x03DC", unicodeslot=0xFE3DC, mathextensible = "r", mathstretch = "h" } ) -addextra(0xFE3B4, { description="EXTENSIBLE OF 0x03B4", unicodeslot=0xFE3B4, mathextensible = "r", mathstretch = "h" } ) +addextra(0xFE3DE, { description="EXTENSIBLE OF 0x03DE", unicodeslot=0xFE3DE, mathextensible = "r", mathstretch = "h", mathclass = "topaccent" } ) +addextra(0xFE3DC, { description="EXTENSIBLE OF 0x03DC", unicodeslot=0xFE3DC, mathextensible = "r", mathstretch = "h", mathclass = "topaccent" } ) +addextra(0xFE3B4, { description="EXTENSIBLE OF 0x03B4", unicodeslot=0xFE3B4, mathextensible = "r", mathstretch = "h", mathclass = "topaccent" } ) virtualcharacters[0xFE3DE] = function(data) return smashed(data,0x23DE,0x23DF,0xFE3DE) end virtualcharacters[0xFE3DC] = function(data) return smashed(data,0x23DC,0x23DD,0xFE3DC) end virtualcharacters[0xFE3B4] = function(data) return smashed(data,0x23B4,0x23B5,0xFE3B4) end -addextra(0xFE3DF, { description="EXTENSIBLE OF 0x03DF", unicodeslot=0xFE3DF, mathextensible = "r", mathstretch = "h" } ) -addextra(0xFE3DD, { description="EXTENSIBLE OF 0x03DD", unicodeslot=0xFE3DD, mathextensible = "r", mathstretch = "h" } ) -addextra(0xFE3B5, { description="EXTENSIBLE OF 0x03B5", unicodeslot=0xFE3B5, mathextensible = "r", mathstretch = "h" } ) +addextra(0xFE3DF, { description="EXTENSIBLE OF 0x03DF", unicodeslot=0xFE3DF, mathextensible = "r", mathstretch = "h", mathclass = "botaccent" } ) +addextra(0xFE3DD, { description="EXTENSIBLE OF 0x03DD", unicodeslot=0xFE3DD, mathextensible = "r", mathstretch = "h", mathclass = "botaccent" } ) +addextra(0xFE3B5, { description="EXTENSIBLE OF 0x03B5", unicodeslot=0xFE3B5, mathextensible = "r", mathstretch = "h", mathclass = "botaccent" } ) -virtualcharacters[0xFE3DF] = function(data) return data.original.characters[0x23DF] end -virtualcharacters[0xFE3DD] = function(data) return data.original.characters[0x23DD] end -virtualcharacters[0xFE3B5] = function(data) return data.original.characters[0x23B5] end +virtualcharacters[0xFE3DF] = function(data) local c = data.target.characters[0x23DF] if c then c.unicode = 0x23DF return c end end +virtualcharacters[0xFE3DD] = function(data) local c = data.target.characters[0x23DD] if c then c.unicode = 0x23DD return c end end +virtualcharacters[0xFE3B5] = function(data) local c = data.target.characters[0x23B5] if c then c.unicode = 0x23B5 return c end end -- todo: add some more .. numbers might change -addextra(0xFE302, { description="EXTENSIBLE OF 0x0302", unicodeslot=0xFE302, mathstretch = "h" } ) -addextra(0xFE303, { description="EXTENSIBLE OF 0x0303", unicodeslot=0xFE303, mathstretch = "h" } ) +addextra(0xFE302, { description="EXTENSIBLE OF 0x0302", unicodeslot=0xFE302, mathstretch = "h", mathclass = "topaccent" } ) +addextra(0xFE303, { description="EXTENSIBLE OF 0x0303", unicodeslot=0xFE303, mathstretch = "h", mathclass = "topaccent" } ) local function smashed(data,unicode,private) local target = data.target local height = target.parameters.xheight / 2 - local c = accent_to_extensible(target,private,data.original,unicode,height,0,nil,-height) - c.top_accent = nil + local c, done = accent_to_extensible(target,private,data.original,unicode,height,0,nil,-height,unicode) + if done then + c.top_accent = nil -- or maybe also all the others + end return c end @@ -466,15 +467,21 @@ virtualcharacters[0xFE302] = function(data) return smashed(data,0x0302,0xFE302) virtualcharacters[0xFE303] = function(data) return smashed(data,0x0303,0xFE303) end -- another crazy hack .. doesn't work as we define scrscr first .. we now have smaller --- primes so we have smaller primes for the moment, big ones will become an option +-- primes so we have smaller primes for the moment, big ones will become an option .. +-- these primes in fonts are a real mess .. kind of a dead end, so don't wonder about +-- the values below + +-- todo: check tounicodes local function smashed(data,unicode,optional) local oldchar = data.characters[unicode] if oldchar then - local height = 1.2 * data.target.parameters.xheight + local xheight = data.target.parameters.xheight + local height = 1.2 * xheight + local shift = oldchar.height - height local newchar = { commands = { - { "down", oldchar.height - height }, + { "down", shift }, { "char", unicode }, }, height = height, @@ -486,6 +493,30 @@ local function smashed(data,unicode,optional) end end +-- -- relocate all but less flexible so not used .. instead some noad hackery plus +-- -- the above +-- +-- local function smashed(data,unicode,optional) +-- local oldchar = data.characters[unicode] +-- if oldchar then +-- local xheight = data.target.parameters.xheight +-- local height = oldchar.height +-- local shift = oldchar.height < 1.5*xheight and -(1.8*xheight-height) or 0 +-- local newchar = { +-- commands = { +-- { "down", shift }, +-- { "char", unicode }, +-- }, +-- unicode = unicode, +-- height = height, +-- width = oldchar.width, +-- } +-- return newchar +-- elseif not optional then +-- report_fallbacks("missing %U prime in font %a",unicode,data.target.properties.fullname) +-- end +-- end + addextra(0xFE932, { description="SMASHED PRIME 0x02032", unicodeslot=0xFE932 } ) addextra(0xFE933, { description="SMASHED PRIME 0x02033", unicodeslot=0xFE933 } ) addextra(0xFE934, { description="SMASHED PRIME 0x02034", unicodeslot=0xFE934 } ) @@ -511,7 +542,7 @@ addextra(0xFE940, { category = "mn", description="SMALL ANNUITY SYMBOL", unicode local function actuarian(data) local characters = data.target.characters local parameters = data.target.parameters - local basechar = characters[0x0078] -- x (0x0058 X) + local basechar = characters[0x0078] -- x (0x0058 X) or 0x1D431 local linewidth = parameters.xheight / 10 local basewidth = basechar.width local baseheight = basechar.height @@ -519,7 +550,7 @@ local function actuarian(data) -- todo: add alttext -- compromise: lm has large hooks e.g. \actuarial{a} width = basewidth + 4 * linewidth, - tounicode = tounicode16(0x20E7), + unicode = 0x20E7, commands = { { "right", 2 * linewidth }, { "down", - baseheight - 3 * linewidth }, |