add project document and sync with live.kladjes.nl

converters/include-files.lua

@@ -0,0 +1,127 @@
+--- include-files.lua – filter to include Markdown files
+---
+--- Copyright: © 2019–2021 Albert Krewinkel
+--- License:   MIT – see LICENSE file for details
+
+-- Module pandoc.path is required and was added in version 2.12
+PANDOC_VERSION:must_be_at_least '2.12'
+
+local List = require 'pandoc.List'
+local path = require 'pandoc.path'
+local system = require 'pandoc.system'
+
+--- Get include auto mode
+local include_auto = false
+function get_vars (meta)
+  if meta['include-auto'] then
+    include_auto = true
+  end
+end
+
+--- Keep last heading level found
+local last_heading_level = 0
+function update_last_level(header)
+  last_heading_level = header.level
+end
+
+--- Update contents of included file
+local function update_contents(blocks, shift_by, include_path)
+  local update_contents_filter = {
+    -- Shift headings in block list by given number
+    Header = function (header)
+      if shift_by then
+        header.level = header.level + shift_by
+      end
+      return header
+    end,
+    -- If link paths are relative then prepend include file path
+    Link = function (link)
+        if path.is_relative(link.target) and string.sub(path.filename(link.target), 1, 1) ~= '#' then
+            link.target = path.normalize(path.join({include_path, link.target}))
+        end
+        return link
+    end,
+    -- If image paths are relative then prepend include file path
+    Image = function (image)
+      if path.is_relative(image.src) then
+        image.src = path.normalize(path.join({include_path, image.src}))
+      end
+      return image
+    end,
+    -- Update path for include-code-files.lua filter style CodeBlocks
+    CodeBlock = function (cb)
+      if cb.attributes.include and path.is_relative(cb.attributes.include) then
+        cb.attributes.include =
+          path.normalize(path.join({include_path, cb.attributes.include}))
+        end
+      return cb
+    end
+  }
+
+  return pandoc.walk_block(pandoc.Div(blocks), update_contents_filter).content
+end
+
+--- Filter function for code blocks
+local transclude
+function transclude (cb)
+  -- ignore code blocks which are not of class "include".
+  if not cb.classes:includes 'include' then
+    return
+  end
+
+  -- Markdown is used if this is nil.
+  local format = cb.attributes['format']
+
+  -- Attributes shift headings
+  local shift_heading_level_by = 0
+  local shift_input = cb.attributes['shift-heading-level-by']
+  if shift_input then
+    shift_heading_level_by = tonumber(shift_input)
+  else
+    if include_auto then
+      -- Auto shift headings
+      shift_heading_level_by = last_heading_level
+    end
+  end
+
+  --- keep track of level before recusion
+  local buffer_last_heading_level = last_heading_level
+
+  local blocks = List:new()
+  for line in cb.text:gmatch('[^\n]+') do
+    if line:sub(1,2) ~= '//' then
+      local fh = io.open(line)
+      if not fh then
+        io.stderr:write("Cannot open file " .. line .. " | Skipping includes\n")
+      else
+        -- read file as the given format with global reader options
+        local contents = pandoc.read(
+          fh:read '*a',
+          format,
+          PANDOC_READER_OPTIONS
+        ).blocks
+        last_heading_level = 0
+        -- recursive transclusion
+        contents = system.with_working_directory(
+            path.directory(line),
+            function ()
+              return pandoc.walk_block(
+                pandoc.Div(contents),
+                { Header = update_last_level, CodeBlock = transclude }
+              )
+            end).content
+        --- reset to level before recursion
+        last_heading_level = buffer_last_heading_level
+        blocks:extend(update_contents(contents, shift_heading_level_by,
+                                      path.directory(line)))
+        fh:close()
+      end
+    end
+  end
+  return blocks
+end
+
+return {
+  { Meta = get_vars },
+  { Header = update_last_level, CodeBlock = transclude }
+}

converters/mdToLatex.sh

@@ -7,21 +7,47 @@ TEMP_MD_FILE="$BUILD_DIR/$(basename "$MD_FILE")"
 TEMP_TEX_FILE="$BUILD_DIR/$(basename "$MD_FILE" | sed -e 's|md$|latex|')"
 mkdir -p "$(dirname "$TEMP_MD_FILE")"
 
-title="$(grep '^# ' "$MD_FILE" | sed 's|^# ||')"
+cp "$MD_FILE" "$TEMP_MD_FILE"
 
-cat "$MD_FILE" | sed \
+for line in $(grep '^!\[.*\](.*\.md)$' "$TEMP_MD_FILE" | sed 's/ /%20;/g')
+do
+    src=$(echo "$line" | sed -e 's/^.*(//' -e 's/).*$//' -e 's/%20;/ /g')
+    echo "include found: markdown/$src"
+
+    cp "markdown/$src" "$BUILD_DIR/$src"
+    sed -i "$BUILD_DIR/$src" \
+        -e 's|\[toc\]||' \
+        -e 's|^\[parent\].*$||' \
+        -e 's|^# |\\newpage\n# |' \
+        -e 's|^## |\\newpage\n# |' \
+        -e 's|\[\([^]]*\)\](#\([^)]*\))|[\1](#\L\2)|'
+
+    sed -i "$TEMP_MD_FILE" \
+        -e "s/^\!\[.*\]($src)\$/\`\`\`\\{.include\\}\n${src}\n\`\`\`/"
+done
+
+title="$(grep '^# ' "$MD_FILE" | head -l 1 | sed 's|^# ||')"
+
+sed -i "$TEMP_MD_FILE" \
     -e 's|\[toc\]|\\tableofcontents|' \
     -e 's|^\[parent\].*$||' \
     -e 's|^# .*$||' \
     -e 's|^#||' \
     -e 's|^# |\\newpage\n# |' \
-    -e 's|\[\([^]]*\)\](#\([^)]*\))|[\1](#\L\2)|' \
-    >"$TEMP_MD_FILE"
+    -e 's|\[\([^]]*\)\](#\([^)]*\))|[\1](#\L\2)|'
 
 cd "$BUILD_DIR"
-pandoc --to=latex --template "${BASE_DIR}/converters/template.latex" -o "$TEX_FILE" "$(basename "$TEMP_MD_FILE")"
+pandoc --lua-filter=../../converters/include-files.lua --to=latex --from=markdown+abbreviations --template "${BASE_DIR}/converters/template.latex" -o "$TEX_FILE" --data-dir="../../markdown" "$(basename "$TEMP_MD_FILE")"
 cd "$BASE_DIR"
 
+# for line in $(grep '^!\[.*\](.*\.md)$' "$TEMP_MD_FILE" | sed 's/ /%20;/g')
+# do
+#     src=$(echo "$line" | sed -e 's/^.*(//' -e 's/).*$//' -e 's/%20;/ /g')
+
+#     sed -i "$TEMP_MD_FILE" \
+#         -e "s/^!\[.*\]($src)\$/\\include{$src}/"
+# done
+
 sed --in-place \
     -e "s|?title?|$title|" \
     "$TEX_FILE"