CTAN sync 202302250301

3 files changed, 275 insertions, 31 deletions

diff --git a/support/make4ht/domfilters/make4ht-inlinecss.lua b/support/make4ht/domfilters/make4ht-inlinecss.lua
new file mode 100644
index 0000000000..b086a2fc42
--- /dev/null
+++ b/support/make4ht/domfilters/make4ht-inlinecss.lua
@@ -0,0 +1,83 @@
+local cssquery  = require "luaxml-cssquery"
+
+local log = logging.new("inlinecss")
+
+local cssrules = {}
+local cssobj   = cssquery()
+
+local function parse_rule(line)
+  -- parse CSS selector and attributes
+  -- they are always on one line in the CSS file produced by TeX4ht
+  local selector, values = line:match("%s*(.-)%s*(%b{})")
+  if values then
+    values = values:sub(2,-2)
+  end
+  return selector, values
+end
+
+local function join_values(old, new)
+  -- correctly joins two attribute lists, depending on the ending
+  local separator = ";"
+  if not old then return new end
+  -- if old already ends with ;, then don't use semicolon as a separator
+  if old:match(";%s*$") then separator = "" end
+  return old .. separator .. new
+end
+
+local function parse_css(filename)
+  local css_file = io.open(filename, "r")
+  if not css_file then return nil, "cannot load css file: " .. (filename or "") end
+  local newlines = {}
+  for line in css_file:lines() do
+    -- match lines that contain # or =, as these can be id or attribute selectors
+    if line:match("[%#%=].-{") then
+      -- update attributes for the current selector
+      local selector, value = parse_rule(line)
+      local oldvalue = cssrules[selector] 
+      cssrules[selector] = join_values(oldvalue, value)
+    else
+      newlines[#newlines+1] = line
+    end
+  end
+  -- we need to add css rules
+  for selector, value in pairs(cssrules) do
+    cssobj:add_selector(selector, function(dom) end, {value=value})
+  end
+  css_file:close()
+  -- write new version of the CSS file, without rules for ids and attributes
+  local css_file = io.open(filename, "w")
+  css_file:write(table.concat(newlines, "\n"))
+  css_file:close()
+  return true
+end
+
+local processed = false
+
+-- process the HTML file and insert inline CSS for id and attribute selectors
+return function(dom, par)
+  if not processed then 
+    -- process the CSS file before everything else, but only once
+    processed = true
+    local css_file = par.input .. ".css"
+    local status, msg = parse_css(css_file)
+    if not status then log:warning(msg) end
+  end
+  -- loop over all elements in the current page
+  dom:traverse_elements(function(curr)
+    -- use CSS object to match if the current element
+    -- is matched by id attribute selector
+    local matched = cssobj:match_querylist(curr)
+    if #matched > 0 then
+      -- join possible already existing style attribute with values from the CSS file
+      local values = curr:get_attribute("style")
+      -- join values of all matched rules
+      for _,rule in ipairs(matched) do
+        values = join_values(values, rule.params.value)
+      end
+      curr:set_attribute("style", values)
+    end
+
+  end)
+  return dom
+end
+
diff --git a/support/make4ht/domfilters/make4ht-joincharacters.lua b/support/make4ht/domfilters/make4ht-joincharacters.lua
index 34737f921a..7f70cca3db 100644
--- a/support/make4ht/domfilters/make4ht-joincharacters.lua
+++ b/support/make4ht/domfilters/make4ht-joincharacters.lua
@@ -107,7 +107,7 @@ local function join_characters(obj,par)
           -- we must create a new node
           el:add_child_node(el:create_text_node(s))
           next_el:remove_node()
-          real_next = nil
+          -- real_next = nil
         else
           real_next = nil
         end
diff --git a/support/make4ht/domfilters/make4ht-mathmlfixes.lua b/support/make4ht/domfilters/make4ht-mathmlfixes.lua
index 12754bedb9..7fa74f4632 100644
--- a/support/make4ht/domfilters/make4ht-mathmlfixes.lua
+++ b/support/make4ht/domfilters/make4ht-mathmlfixes.lua
@@ -5,17 +5,53 @@ local token = {"mi", "mn", "mo", "mtext", "mspace", "ms"}
 local token_elements = {}
 for _, tok in ipairs(token) do token_elements[tok] = true end
 
+-- helper functions to support MathML elements with prefixes (<mml:mi> etc).
+--
+local function get_element_name(el)
+  -- return element name and xmlns prefix
+  local name = el:get_element_name()
+  if name:match(":") then
+    local prefix, real_name =  name:match("([^%:]+):?(.+)")
+    return real_name, prefix
+  else
+    return name
+  end
+end
+
+local function get_attribute(el, attr_name)
+  -- attributes can have the prefix, but sometimes they don't have it
+  -- so we need to catch both cases
+  local _, prefix = get_element_name(el)
+  prefix = prefix or ""
+  return el:get_attribute(attr_name) or el:get_attribute(prefix .. ":" .. attr_name)
+end
+
+local function get_new_element_name(name, prefix)
+  return prefix and prefix .. ":" .. name or name
+end
+
+local function update_element_name(el, name, prefix)
+  local newname = get_new_element_name(name, prefix)
+  el._name = newname
+end
+
+local function create_element(el, name, prefix)
+  return el:create_element(newname)
+end
+
 local function is_token_element(el)
-  return token_elements[el:get_element_name()]
+  local name, prefix = get_element_name(el)
+  return token_elements[name], prefix
 end
 
 local function fix_token_elements(el)
   -- find token elements that are children of other token elements
   if is_token_element(el) then
     local parent = el:get_parent()
-    if is_token_element(parent) then
+    local is_parent_token, prefix = is_token_element(parent)
+    if is_parent_token then
       -- change top element in nested token elements to mstyle
-      parent._name = "mstyle"
+      update_element_name(parent, "mstyle", prefix)
     end
   end
 end
@@ -23,7 +59,8 @@ end
 local function fix_nested_mstyle(el)
   -- the <mstyle> element can be child of token elements
   -- we must exterminate it
-  if el:get_element_name() == "mstyle" then
+  local el_name = get_element_name(el)
+  if el_name == "mstyle" then
     local parent = el:get_parent()
     if is_token_element(parent) then
       -- if parent doesn't have the mathvariant attribute copy it from <mstyle>
@@ -44,13 +81,13 @@ local function fix_mathvariant(el)
     -- find if element has <mstyle> parent, and its value of mathvariant
     if not x:is_element() then
       return nil
-    elseif x:get_element_name() == "mstyle" then 
+    elseif get_element_name(x) == "mstyle" then 
       return x:get_attribute("mathvariant")
     else
       return find_mstyle(x:get_parent())
     end
   end
-  if el:get_element_name() == "mi" then
+  if get_element_name(el) == "mi" then
     -- process only <mi> that have mathvariant set
     local oldmathvariant = el:get_attribute("mathvariant")
     if oldmathvariant then
@@ -62,6 +99,74 @@ local function fix_mathvariant(el)
   end
 end
 
+local function contains_only_text(el)
+  -- detect if element contains only text
+  local elements = 0
+  local text     = 0
+  local children = el:get_children() or {}
+  for _ , child in ipairs(children) do
+    if child:is_text() then text = text + 1
+    elseif child:is_element() then elements = elements + 1
+    end
+  end
+  return text > 0 and elements == 0
+end
+
+-- check if <mstyle> element contains direct text. in that case, add
+-- <mtext>
+local function fix_missing_mtext(el)
+  if el:get_element_name() == "mstyle" and contains_only_text(el) then
+    -- add child <mtext>
+    log:debug("mstyle contains only text: " .. el:get_text())
+    -- copy the current mode, change it's element name to mtext and add it as a child of <mstyle>
+    local copy = el:copy_node()
+    copy._name = "mtext"
+    copy._parent = el
+    el._children = {copy}
+  end
+end
+
+local function is_radical(el)
+  local radicals = {msup=true, msub=true, msubsup=true}
+  return radicals[el:get_element_name()]
+end
+
+local function get_mrow_child(el)
+  local get_first = function(x) 
+    local children = x:get_children() 
+    return children[1]
+  end
+  local first = get_first(el)
+  -- either return first child, and if the child is <mrow>, return it's first child
+  if first and first:is_element() then
+    if first:get_element_name() == "mrow" then
+      return get_first(first), first
+    else
+      return first
+    end
+  end
+end
+
+local function fix_radicals(el)
+  if is_radical(el) then
+    local first_child, mrow = get_mrow_child(el)
+    -- if the first child is only one character long, it is possible that there is a problem
+    if first_child and string.len(first_child:get_text()) == 1 then
+      local name = first_child:get_element_name() 
+      local siblings = el:get_siblings()
+      local pos = el:find_element_pos()
+      -- it doesn't make sense to do any further processing if the element is at the beginning
+      if pos == 1 then return end
+      if name == "mo" then
+        for i = pos, 1,-1 do
+        end
+
+      end
+    end
+
+  end
+end
+
 -- put <mrow> as child of <math> if it already isn't here
 local allowed_top_mrow = {
   math=true
@@ -73,24 +178,25 @@ local function top_mrow(math)
   -- don't process elements that already are mrow
   local parent = math:get_parent()
   local parent_name
-  if parent then parent_name = parent:get_element_name() end
-  local current_name = math:get_element_name()
+  if parent then parent_name = get_element_name(parent) end
+  local current_name, prefix = get_element_name(math)
   if #children < 2 or not allowed_top_mrow[current_name] or current_name == "mrow" or parent_name == "mrow" then return nil end
   local mrow_count = 0
   for _,v in ipairs(children) do
     if v:is_element() and is_token_element(v) then
       put_mrow = true
       -- break
-    elseif v:is_element() and v:get_element_name() == "mrow" then
+    elseif v:is_element() and get_element_name(v) == "mrow" then
       mrow_count = mrow_count + 1
     end
   end
-  if not put_mrow and math:get_element_name() == "math" and mrow_count == 0 then
+  if not put_mrow and get_element_name(math) == "math" and mrow_count == 0 then
     -- put at least one <mrow> to each <math>
     put_mrow = true
   end
   if put_mrow then
-    local mrow = math:create_element("mrow")
+    local newname = get_new_element_name("mrow", prefix)
+    local mrow = math:create_element(newname)
     for _, el in ipairs(children) do
       mrow:add_child_node(el)
     end
@@ -106,7 +212,9 @@ local function get_fence(el, attr, form)
   local char = el:get_attribute(attr)
   local mo 
   if char then
-    mo = el:create_element("mo", {fence="true", form = form})
+    local name, prefix = get_element_name(el)
+    local newname = get_new_element_name("mo", prefix)
+    mo = el:create_element(newname, {fence="true", form = form})
     mo:add_child_node(mo:create_text_node(char))
   end
   return mo
@@ -116,13 +224,15 @@ end
 local function fix_mfenced(el)
   -- TeX4ht uses in some cases <mfenced> element which is deprecated in MathML.
   -- Firefox doesn't support it already.
-  if el:get_element_name() == "mfenced" then
+  local name, prefix = get_element_name(el)
+  if name == "mfenced" then
     -- we must replace it by <mrow><mo>start</mo><mfenced children...><mo>end</mo></mrow>
     local open = get_fence(el, "open", "prefix")
     local close = get_fence(el, "close", "postfix")
     -- there can be also separator attribute, but it is not used in TeX4ht
     -- change <mfenced> to <mrow> and remove all attributes
-    el._name = "mrow"
+    local newname = get_new_element_name("mrow", prefix)
+    el._name = newname
     el._attr = {}
     -- open must be first child, close needs to be last
     if open then el:add_child_node(open, 1) end
@@ -131,7 +241,7 @@ local function fix_mfenced(el)
 end
 
 local function is_fence(el)
-  return el:get_element_name() == "mo" and el:get_attribute("fence") == "true"
+  return get_element_name(el) == "mo" and el:get_attribute("fence") == "true"
 end
 
 local function fix_mo_to_mfenced(el)
@@ -155,25 +265,27 @@ local function fix_mo_to_mfenced(el)
       end 
     end
     -- convert parent <mrow> to <mfenced>
-    parent._name = "mfenced"
+    local _, prefix = get_element_name(parent)
+    local newname = get_new_element_name("mfenced", prefix)
+    parent._name = newname
     parent._attr = {open = open, close = close}
   end
 end
 
 local function fix_numbers(el)
   -- convert <mn>1</mn><mo>.</mo><mn>3</mn> to <mn>1.3</mn>
-  if el:get_element_name() == "mn" then
+  if get_element_name(el) == "mn" then
     local n = el:get_sibling_node(1)
     -- test if next  element is <mo class="MathClass-punc">.</mo>
     if n and n:is_element() 
-         and n:get_element_name() == "mo" 
-         and n:get_attribute("class") == "MathClass-punc" 
+         and get_element_name(n) == "mo" 
+         and get_attribute(n, "class") == "MathClass-punc" 
          and n:get_text() == "." 
     then
       -- get next element and test if it is <mn>
       local x = el:get_sibling_node(2)
       if x and x:is_element() 
-           and x:get_element_name() == "mn" 
+           and get_element_name(x) == "mn" 
       then
         -- join numbers and set it as text content of the current element
         local newnumber = el:get_text() .. "." .. x:get_text()
@@ -194,7 +306,7 @@ local function just_operators(list)
   -- count <mo> and return true if list contains just them
   local mo = 0
   for _, x in ipairs(list) do
-    if x:get_element_name() == "mo" then mo = mo + 1 end
+    if get_element_name(x) == "mo" then mo = mo + 1 end
   end
   return mo
 end
@@ -205,18 +317,22 @@ local function fix_operators(x)
   -- this fixes issues in LibreOffice with a^{*}
   -- I hope it doesn't introduce different issues
   -- process only <mo>
-  if x:get_element_name() ~= "mo" then return nil end
+  local el_name, prefix = get_element_name(x)
+  if el_name ~= "mo" then return nil end
 	local siblings = x:get_siblings()
 	-- test if current element list contains only <mo>
 	if just_operators(siblings) == #siblings then
 		if #siblings == 1 then
-			-- one <mo> translates to <mtext>
-			x._name = "mtext"
-      log:debug("changing one <mo> to <mtext>: " .. x:get_text())
-      -- I think we should use <mi>, but LO incorrectly renders it in <msubsup>,
-      -- even if we use the mathvariant="normal" attribute. <mtext> works, so
-      -- we use that instead.
-			-- x:set_attribute("mathvariant", "normal")
+      if not x:get_attribute("stretchy") then
+        -- one <mo> translates to <mtext>
+        local newname = get_new_element_name("mtext", prefix)
+        x._name = newname
+        log:debug("changing one <mo> to <mtext>: " .. x:get_text())
+        -- I think we should use <mi>, but LO incorrectly renders it in <msubsup>,
+        -- even if we use the mathvariant="normal" attribute. <mtext> works, so
+        -- we use that instead.
+        -- x:set_attribute("mathvariant", "normal")
+      end
 		else
 			-- multiple <mo> translate to <mtext>
 			local text = {}
@@ -231,7 +347,8 @@ local function fix_operators(x)
       log:debug("changing <mo> to <mtext>: " .. newtext)
       x:add_child_node(text_el)
       -- change <mo> to <mtext>
-      x._name = "mtext"
+      local newname = get_new_element_name("mtext", prefix)
+      x._name = newname
       -- remove subsequent <mo>
       for i = 2, #siblings do
         siblings[i]:remove_node()
@@ -240,6 +357,47 @@ local function fix_operators(x)
   end
 end
 
+local function is_last_element(el)
+  local siblings = el:get_siblings()
+  -- return true only if the current element is the last in the parent's children
+  for i = #siblings, 1, -1 do
+    local curr = siblings[i]
+    if curr == el then
+      return true
+    elseif curr:is_element() then
+      return false
+    end
+  end
+  return false
+end
+
+local function is_empty_row(el)
+  -- empty row should contain only one <mtd>
+  local count = 0
+  if el:get_text():match("^%s*$") then
+    for _, child in ipairs(el:get_children()) do
+      if child:is_element() then count = count + 1 end
+    end
+  end
+  -- if there is one or zero childrens, then it is empty row
+  return count < 2
+end
+
+
+local function delete_last_empty_mtr(el)
+  -- arrays sometimes contain last empty row, which causes rendering issues,
+  -- so we should remove them
+  local el_name, prefix = get_element_name(el)
+  if el_name == "mtr" 
+    and get_attribute(el, "class") == "array-row" 
+    and is_last_element(el)
+    and is_empty_row(el)
+  then
+    el:remove_node()
+  end
+
+end
+
 return function(dom)
   dom:traverse_elements(function(el)
     if settings.output_format ~= "odt" then
@@ -248,12 +406,15 @@ return function(dom)
     else
       fix_mo_to_mfenced(el)
     end
+    fix_radicals(el)
     fix_token_elements(el)
     fix_nested_mstyle(el)
+    fix_missing_mtext(el)
     fix_numbers(el)
     fix_operators(el)
     fix_mathvariant(el)
     top_mrow(el)
+    delete_last_empty_mtr(el)
   end)
   return dom
 end