ruby_pptx 0.1.0

data/lib/ruby_pptx/templates/theme.xml

@@ -0,0 +1,321 @@
+<?xml version='1.0' encoding='UTF-8' standalone='yes'?>
+<a:theme
+    xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main"
+    name="Office Theme"
+    >
+  <a:themeElements>
+    <a:clrScheme name="Office">
+      <a:dk1>
+        <a:sysClr val="windowText" lastClr="000000"/>
+      </a:dk1>
+      <a:lt1>
+        <a:sysClr val="window" lastClr="FFFFFF"/>
+      </a:lt1>
+      <a:dk2>
+        <a:srgbClr val="1F497D"/>
+      </a:dk2>
+      <a:lt2>
+        <a:srgbClr val="EEECE1"/>
+      </a:lt2>
+      <a:accent1>
+        <a:srgbClr val="4F81BD"/>
+      </a:accent1>
+      <a:accent2>
+        <a:srgbClr val="C0504D"/>
+      </a:accent2>
+      <a:accent3>
+        <a:srgbClr val="9BBB59"/>
+      </a:accent3>
+      <a:accent4>
+        <a:srgbClr val="8064A2"/>
+      </a:accent4>
+      <a:accent5>
+        <a:srgbClr val="4BACC6"/>
+      </a:accent5>
+      <a:accent6>
+        <a:srgbClr val="F79646"/>
+      </a:accent6>
+      <a:hlink>
+        <a:srgbClr val="0000FF"/>
+      </a:hlink>
+      <a:folHlink>
+        <a:srgbClr val="800080"/>
+      </a:folHlink>
+    </a:clrScheme>
+    <a:fontScheme name="Office">
+      <a:majorFont>
+        <a:latin typeface="Calibri"/>
+        <a:ea typeface=""/>
+        <a:cs typeface=""/>
+        <a:font script="Jpan" typeface="ＭＳ Ｐゴシック"/>
+        <a:font script="Hang" typeface="맑은 고딕"/>
+        <a:font script="Hans" typeface="宋体"/>
+        <a:font script="Hant" typeface="新細明體"/>
+        <a:font script="Arab" typeface="Times New Roman"/>
+        <a:font script="Hebr" typeface="Times New Roman"/>
+        <a:font script="Thai" typeface="Angsana New"/>
+        <a:font script="Ethi" typeface="Nyala"/>
+        <a:font script="Beng" typeface="Vrinda"/>
+        <a:font script="Gujr" typeface="Shruti"/>
+        <a:font script="Khmr" typeface="MoolBoran"/>
+        <a:font script="Knda" typeface="Tunga"/>
+        <a:font script="Guru" typeface="Raavi"/>
+        <a:font script="Cans" typeface="Euphemia"/>
+        <a:font script="Cher" typeface="Plantagenet Cherokee"/>
+        <a:font script="Yiii" typeface="Microsoft Yi Baiti"/>
+        <a:font script="Tibt" typeface="Microsoft Himalaya"/>
+        <a:font script="Thaa" typeface="MV Boli"/>
+        <a:font script="Deva" typeface="Mangal"/>
+        <a:font script="Telu" typeface="Gautami"/>
+        <a:font script="Taml" typeface="Latha"/>
+        <a:font script="Syrc" typeface="Estrangelo Edessa"/>
+        <a:font script="Orya" typeface="Kalinga"/>
+        <a:font script="Mlym" typeface="Kartika"/>
+        <a:font script="Laoo" typeface="DokChampa"/>
+        <a:font script="Sinh" typeface="Iskoola Pota"/>
+        <a:font script="Mong" typeface="Mongolian Baiti"/>
+        <a:font script="Viet" typeface="Times New Roman"/>
+        <a:font script="Uigh" typeface="Microsoft Uighur"/>
+        <a:font script="Geor" typeface="Sylfaen"/>
+      </a:majorFont>
+      <a:minorFont>
+        <a:latin typeface="Calibri"/>
+        <a:ea typeface=""/>
+        <a:cs typeface=""/>
+        <a:font script="Jpan" typeface="ＭＳ Ｐゴシック"/>
+        <a:font script="Hang" typeface="맑은 고딕"/>
+        <a:font script="Hans" typeface="宋体"/>
+        <a:font script="Hant" typeface="新細明體"/>
+        <a:font script="Arab" typeface="Arial"/>
+        <a:font script="Hebr" typeface="Arial"/>
+        <a:font script="Thai" typeface="Cordia New"/>
+        <a:font script="Ethi" typeface="Nyala"/>
+        <a:font script="Beng" typeface="Vrinda"/>
+        <a:font script="Gujr" typeface="Shruti"/>
+        <a:font script="Khmr" typeface="DaunPenh"/>
+        <a:font script="Knda" typeface="Tunga"/>
+        <a:font script="Guru" typeface="Raavi"/>
+        <a:font script="Cans" typeface="Euphemia"/>
+        <a:font script="Cher" typeface="Plantagenet Cherokee"/>
+        <a:font script="Yiii" typeface="Microsoft Yi Baiti"/>
+        <a:font script="Tibt" typeface="Microsoft Himalaya"/>
+        <a:font script="Thaa" typeface="MV Boli"/>
+        <a:font script="Deva" typeface="Mangal"/>
+        <a:font script="Telu" typeface="Gautami"/>
+        <a:font script="Taml" typeface="Latha"/>
+        <a:font script="Syrc" typeface="Estrangelo Edessa"/>
+        <a:font script="Orya" typeface="Kalinga"/>
+        <a:font script="Mlym" typeface="Kartika"/>
+        <a:font script="Laoo" typeface="DokChampa"/>
+        <a:font script="Sinh" typeface="Iskoola Pota"/>
+        <a:font script="Mong" typeface="Mongolian Baiti"/>
+        <a:font script="Viet" typeface="Arial"/>
+        <a:font script="Uigh" typeface="Microsoft Uighur"/>
+        <a:font script="Geor" typeface="Sylfaen"/>
+      </a:minorFont>
+    </a:fontScheme>
+    <a:fmtScheme name="Office">
+      <a:fillStyleLst>
+        <a:solidFill>
+          <a:schemeClr val="phClr"/>
+        </a:solidFill>
+        <a:gradFill rotWithShape="1">
+          <a:gsLst>
+            <a:gs pos="0">
+              <a:schemeClr val="phClr">
+                <a:tint val="50000"/>
+                <a:satMod val="300000"/>
+              </a:schemeClr>
+            </a:gs>
+            <a:gs pos="35000">
+              <a:schemeClr val="phClr">
+                <a:tint val="37000"/>
+                <a:satMod val="300000"/>
+              </a:schemeClr>
+            </a:gs>
+            <a:gs pos="100000">
+              <a:schemeClr val="phClr">
+                <a:tint val="15000"/>
+                <a:satMod val="350000"/>
+              </a:schemeClr>
+            </a:gs>
+          </a:gsLst>
+          <a:lin ang="16200000" scaled="1"/>
+        </a:gradFill>
+        <a:gradFill rotWithShape="1">
+          <a:gsLst>
+            <a:gs pos="0">
+              <a:schemeClr val="phClr">
+                <a:tint val="100000"/>
+                <a:shade val="100000"/>
+                <a:satMod val="130000"/>
+              </a:schemeClr>
+            </a:gs>
+            <a:gs pos="100000">
+              <a:schemeClr val="phClr">
+                <a:tint val="50000"/>
+                <a:shade val="100000"/>
+                <a:satMod val="350000"/>
+              </a:schemeClr>
+            </a:gs>
+          </a:gsLst>
+          <a:lin ang="16200000" scaled="0"/>
+        </a:gradFill>
+      </a:fillStyleLst>
+      <a:lnStyleLst>
+        <a:ln w="9525" cap="flat" cmpd="sng" algn="ctr">
+          <a:solidFill>
+            <a:schemeClr val="phClr">
+              <a:shade val="95000"/>
+              <a:satMod val="105000"/>
+            </a:schemeClr>
+          </a:solidFill>
+          <a:prstDash val="solid"/>
+        </a:ln>
+        <a:ln w="25400" cap="flat" cmpd="sng" algn="ctr">
+          <a:solidFill>
+            <a:schemeClr val="phClr"/>
+          </a:solidFill>
+          <a:prstDash val="solid"/>
+        </a:ln>
+        <a:ln w="38100" cap="flat" cmpd="sng" algn="ctr">
+          <a:solidFill>
+            <a:schemeClr val="phClr"/>
+          </a:solidFill>
+          <a:prstDash val="solid"/>
+        </a:ln>
+      </a:lnStyleLst>
+      <a:effectStyleLst>
+        <a:effectStyle>
+          <a:effectLst>
+            <a:outerShdw blurRad="40000" dist="20000" dir="5400000" rotWithShape="0">
+              <a:srgbClr val="000000">
+                <a:alpha val="38000"/>
+              </a:srgbClr>
+            </a:outerShdw>
+          </a:effectLst>
+        </a:effectStyle>
+        <a:effectStyle>
+          <a:effectLst>
+            <a:outerShdw blurRad="40000" dist="23000" dir="5400000" rotWithShape="0">
+              <a:srgbClr val="000000">
+                <a:alpha val="35000"/>
+              </a:srgbClr>
+            </a:outerShdw>
+          </a:effectLst>
+        </a:effectStyle>
+        <a:effectStyle>
+          <a:effectLst>
+            <a:outerShdw blurRad="40000" dist="23000" dir="5400000" rotWithShape="0">
+              <a:srgbClr val="000000">
+                <a:alpha val="35000"/>
+              </a:srgbClr>
+            </a:outerShdw>
+          </a:effectLst>
+          <a:scene3d>
+            <a:camera prst="orthographicFront">
+              <a:rot lat="0" lon="0" rev="0"/>
+            </a:camera>
+            <a:lightRig rig="threePt" dir="t">
+              <a:rot lat="0" lon="0" rev="1200000"/>
+            </a:lightRig>
+          </a:scene3d>
+          <a:sp3d>
+            <a:bevelT w="63500" h="25400"/>
+          </a:sp3d>
+        </a:effectStyle>
+      </a:effectStyleLst>
+      <a:bgFillStyleLst>
+        <a:solidFill>
+          <a:schemeClr val="phClr"/>
+        </a:solidFill>
+        <a:gradFill rotWithShape="1">
+          <a:gsLst>
+            <a:gs pos="0">
+              <a:schemeClr val="phClr">
+                <a:tint val="40000"/>
+                <a:satMod val="350000"/>
+              </a:schemeClr>
+            </a:gs>
+            <a:gs pos="40000">
+              <a:schemeClr val="phClr">
+                <a:tint val="45000"/>
+                <a:shade val="99000"/>
+                <a:satMod val="350000"/>
+              </a:schemeClr>
+            </a:gs>
+            <a:gs pos="100000">
+              <a:schemeClr val="phClr">
+                <a:shade val="20000"/>
+                <a:satMod val="255000"/>
+              </a:schemeClr>
+            </a:gs>
+          </a:gsLst>
+          <a:path path="circle">
+            <a:fillToRect l="50000" t="-80000" r="50000" b="180000"/>
+          </a:path>
+        </a:gradFill>
+        <a:gradFill rotWithShape="1">
+          <a:gsLst>
+            <a:gs pos="0">
+              <a:schemeClr val="phClr">
+                <a:tint val="80000"/>
+                <a:satMod val="300000"/>
+              </a:schemeClr>
+            </a:gs>
+            <a:gs pos="100000">
+              <a:schemeClr val="phClr">
+                <a:shade val="30000"/>
+                <a:satMod val="200000"/>
+              </a:schemeClr>
+            </a:gs>
+          </a:gsLst>
+          <a:path path="circle">
+            <a:fillToRect l="50000" t="50000" r="50000" b="50000"/>
+          </a:path>
+        </a:gradFill>
+      </a:bgFillStyleLst>
+    </a:fmtScheme>
+  </a:themeElements>
+  <a:objectDefaults>
+    <a:spDef>
+      <a:spPr/>
+      <a:bodyPr/>
+      <a:lstStyle/>
+      <a:style>
+        <a:lnRef idx="1">
+          <a:schemeClr val="accent1"/>
+        </a:lnRef>
+        <a:fillRef idx="3">
+          <a:schemeClr val="accent1"/>
+        </a:fillRef>
+        <a:effectRef idx="2">
+          <a:schemeClr val="accent1"/>
+        </a:effectRef>
+        <a:fontRef idx="minor">
+          <a:schemeClr val="lt1"/>
+        </a:fontRef>
+      </a:style>
+    </a:spDef>
+    <a:lnDef>
+      <a:spPr/>
+      <a:bodyPr/>
+      <a:lstStyle/>
+      <a:style>
+        <a:lnRef idx="2">
+          <a:schemeClr val="accent1"/>
+        </a:lnRef>
+        <a:fillRef idx="0">
+          <a:schemeClr val="accent1"/>
+        </a:fillRef>
+        <a:effectRef idx="1">
+          <a:schemeClr val="accent1"/>
+        </a:effectRef>
+        <a:fontRef idx="minor">
+          <a:schemeClr val="tx1"/>
+        </a:fontRef>
+      </a:style>
+    </a:lnDef>
+  </a:objectDefaults>
+  <a:extraClrSchemeLst/>
+</a:theme>

data/lib/ruby_pptx/text/fitter.rb

@@ -0,0 +1,72 @@
+# frozen_string_literal: true
+
+require "ruby_pptx/text/font_metrics"
+
+module Pptx
+  # Finds the largest whole point size at which a string still fits a box.
+  #
+  # The search is a port of python-pptx's: candidate sizes are tried largest
+  # first, and at each size the text is wrapped greedily onto as many lines as
+  # it needs and the total height compared against the box. Only the
+  # measurement underneath differs -- see {FontMetrics}.
+  class TextFitter
+    # @param text [String]
+    # @param extents [Array(Integer, Integer)] width and height in EMU
+    # @param max_size [Integer] largest point size to consider
+    # @param font_file [String] path to the font to measure with
+    # @return [Integer, nil] nil when the text does not fit at any size
+    def self.best_fit_font_size(text, extents:, max_size:, font_file:)
+      new(text, extents: extents, font_file: font_file).best_fit(max_size)
+    end
+
+    def initialize(text, extents:, font_file:)
+      @text = text
+      # Extents may arrive as Length; measurements are plain EMU integers.
+      @width, @height = extents.map(&:to_i)
+      @metrics = font_file.is_a?(FontMetrics) ? font_file : FontMetrics.open(font_file)
+    end
+
+    # The largest whole size in 1..max_size that fits, or nil if none does.
+    def best_fit(max_size)
+      max_size.to_i.downto(1) { |size| return size if fits?(size) }
+      nil
+    end
+
+    # Wrap +text+ at +point_size+ and report the lines it needs.
+    #
+    # Exposed because it is the interesting half of the algorithm and is worth
+    # being able to check on its own.
+    def wrap(point_size)
+      lines = []
+      remainder = @text.split
+      until remainder.empty?
+        line, remainder = break_line(remainder, point_size)
+        lines << line
+      end
+      lines
+    end
+
+    private
+
+    # The height of a line is measured from "Ty" rather than from the text, so
+    # that every line is the same height whether or not it happens to contain
+    # an ascender or a descender.
+    def fits?(point_size)
+      line_height = @metrics.text_extents("Ty", point_size)[1]
+      wrap(point_size).size * line_height <= @height
+    end
+
+    # Take as many whole words as fit, but never fewer than one -- a single
+    # word wider than the box would otherwise loop forever. Such a word makes
+    # the line overflow, and the size search then rejects that size on height.
+    def break_line(words, point_size)
+      taken = 1
+      taken += 1 while taken < words.size && fits_width?(words[0, taken + 1], point_size)
+      [words[0, taken].join(" "), words[taken..]]
+    end
+
+    def fits_width?(words, point_size)
+      @metrics.text_extents(words.join(" "), point_size)[0] <= @width
+    end
+  end
+end

data/lib/ruby_pptx/text/font_metrics.rb

@@ -0,0 +1,221 @@
+# frozen_string_literal: true
+
+module Pptx
+  # Text measurements read straight out of a TrueType or OpenType font file.
+  #
+  # This is what {TextFitter} measures with. python-pptx asks Pillow to
+  # rasterize the text and reports the ink bounding box of the result; there is
+  # no Ruby equivalent of that, so this reads the glyph outlines' own bounding
+  # boxes from the font instead. See PORTING.md for how far the two diverge --
+  # in short, well under a point size.
+  #
+  # Only the tables needed for measurement are parsed. Composite glyphs are not
+  # recursed into: their bounding box is stored on the glyph itself, which is
+  # all this needs.
+  class FontMetrics
+    EMU_PER_INCH = 914_400
+    POINTS_PER_INCH = 72.0
+
+    class << self
+      # Fonts are immutable and parsing one is not free, so they are cached.
+      def open(path)
+        real = File.realpath(path)
+        cache[real] ||= new(real)
+      end
+
+      def clear_cache = cache.clear
+
+      private
+
+      def cache = @cache ||= {}
+    end
+
+    # @param path [String] a .ttf or .otf file
+    def initialize(path)
+      @path = path
+      @data = File.binread(path).b
+      @tables = read_table_directory
+      require_tables!
+      @units_per_em = u16(@tables["head"] + 18)
+      @index_to_loc_format = u16(@tables["head"] + 50)
+      @num_glyphs = u16(@tables["maxp"] + 4)
+      @num_h_metrics = u16(@tables["hhea"] + 34)
+      @cmap = read_cmap
+      @glyph_boxes = {}
+    end
+
+    attr_reader :path, :units_per_em
+
+    # The size of +text+ rendered at +point_size+, as [width, height] in EMU.
+    #
+    # The height is the ink height, so it depends on which glyphs are present:
+    # "Ty" is taller than "ace". That is deliberate -- it is how python-pptx
+    # measures, and {TextFitter} relies on it by always measuring "Ty".
+    def text_extents(text, point_size)
+      box = ink_box(text)
+      return [0, 0] if box.nil?
+
+      # Width is measured from the pen's starting position rather than from the
+      # first mark, which is what Pillow reports and so what python-pptx fits
+      # against: it counts a leading space, and the first glyph's left side
+      # bearing. Only ink that starts left of the origin pushes the edge out.
+      # Height is pure ink, top mark to bottom mark.
+      left = [box[0], 0].min
+      [to_emu(box[2] - left, point_size), to_emu(box[3] - box[1], point_size)]
+    end
+
+    def inspect = "#<Pptx::FontMetrics #{File.basename(@path)} upem=#{@units_per_em}>"
+
+    private
+
+    # Laying the glyphs out along the baseline and unioning their boxes.
+    # Kerning is not applied; see PORTING.md.
+    def ink_box(text)
+      pen = 0
+      left = bottom = right = top = nil
+      text.each_char do |char|
+        gid = glyph_id(char.ord)
+        box = glyph_box(gid)
+        if box
+          left = min(left, pen + box[0])
+          bottom = min(bottom, box[1])
+          right = max(right, pen + box[2])
+          top = max(top, box[3])
+        end
+        pen += advance(gid)
+      end
+      left && [left, bottom, right, top]
+    end
+
+    def min(a, b) = a.nil? || b < a ? b : a
+    def max(a, b) = a.nil? || b > a ? b : a
+
+    # Font units are relative to the em square, so scaling by the point size
+    # gives points, which are 1/72 inch by definition.
+    def to_emu(units, point_size)
+      (units * point_size / @units_per_em.to_f / POINTS_PER_INCH * EMU_PER_INCH).to_i
+    end
+
+    def glyph_id(codepoint) = @cmap[codepoint] || 0
+
+    # A font may store one advance for a run of trailing glyphs, in which case
+    # hmtx is truncated and the last entry stands for all of them.
+    def advance(gid)
+      u16(@tables["hmtx"] + ([gid, @num_h_metrics - 1].min * 4))
+    end
+
+    # @return [Array(Integer, Integer, Integer, Integer), nil] xMin, yMin,
+    #   xMax, yMax in font units, or nil for a glyph with no outline
+    def glyph_box(gid)
+      return @glyph_boxes[gid] if @glyph_boxes.key?(gid)
+
+      @glyph_boxes[gid] = compute_glyph_box(gid)
+    end
+
+    def compute_glyph_box(gid)
+      return cff_glyph_box(gid) if @tables["glyf"].nil?
+      return nil if gid >= @num_glyphs
+
+      from = loca(gid)
+      # A glyph with no contours, such as the space, has a zero-length entry.
+      return nil if from == loca(gid + 1)
+
+      at = @tables["glyf"] + from
+      [s16(at + 2), s16(at + 4), s16(at + 6), s16(at + 8)]
+    end
+
+    # OpenType/CFF fonts have no glyf table, and reading their charstrings to
+    # find outlines is a disproportionate amount of machinery for a measurement
+    # that only has to be close. Fall back to the advance width and the font's
+    # overall vertical extents.
+    def cff_glyph_box(gid)
+      width = advance(gid)
+      return nil if width.zero?
+
+      ascender = s16(@tables["hhea"] + 4)
+      descender = s16(@tables["hhea"] + 6)
+      [0, descender, width, ascender]
+    end
+
+    def loca(index)
+      base = @tables["loca"]
+      @index_to_loc_format.zero? ? u16(base + (index * 2)) * 2 : u32(base + (index * 4))
+    end
+
+    def u16(at) = @data.byteslice(at, 2).unpack1("n")
+    def s16(at) = @data.byteslice(at, 2).unpack1("s>")
+    def u32(at) = @data.byteslice(at, 4).unpack1("N")
+
+    def read_table_directory
+      raise Error, "#{@path} is not a TrueType or OpenType font" unless sfnt_version?
+
+      u16(4).times.to_h do |i|
+        entry = @data.byteslice(12 + (i * 16), 16)
+        [entry.byteslice(0, 4), entry.byteslice(8, 4).unpack1("N")]
+      end
+    end
+
+    # A TrueType outline font, an OpenType/CFF font, or an Apple-flavoured one.
+    def sfnt_version?
+      %W[\x00\x01\x00\x00 OTTO true].include?(@data.byteslice(0, 4))
+    end
+
+    def require_tables!
+      missing = %w[head maxp hhea hmtx cmap] - @tables.keys
+      return if missing.empty?
+
+      raise Error, "#{@path} is missing required font tables: #{missing.join(", ")}"
+    end
+
+    # A Unicode BMP subtable is enough: presentation text outside the BMP falls
+    # back to glyph 0, which measures as .notdef rather than crashing.
+    def read_cmap
+      base = @tables["cmap"]
+      offset = nil
+      u16(base + 2).times do |i|
+        platform, encoding, sub = @data.byteslice(base + 4 + (i * 8), 8).unpack("nnN")
+        next unless unicode_bmp?(platform, encoding)
+        next unless u16(base + sub) == 4
+
+        offset = base + sub
+        break
+      end
+      offset.nil? ? {} : read_cmap_format4(offset)
+    end
+
+    def unicode_bmp?(platform, encoding)
+      (platform == 3 && encoding == 1) || (platform.zero? && encoding <= 4)
+    end
+
+    def read_cmap_format4(offset)
+      seg_x2 = u16(offset + 6)
+      ends = @data.byteslice(offset + 14, seg_x2).unpack("n*")
+      # A reserved padding word sits between the end and start arrays.
+      starts = @data.byteslice(offset + 16 + seg_x2, seg_x2).unpack("n*")
+      deltas = @data.byteslice(offset + 16 + (seg_x2 * 2), seg_x2).unpack("n*")
+      ranges_at = offset + 16 + (seg_x2 * 3)
+      ranges = @data.byteslice(ranges_at, seg_x2).unpack("n*")
+
+      map = {}
+      (seg_x2 / 2).times do |i|
+        (starts[i]..ends[i]).each do |code|
+          # The final segment is a sentinel ending at 0xFFFF.
+          next if code == 0xFFFF
+
+          gid = segment_glyph(i, code, starts, deltas, ranges, ranges_at)
+          map[code] = gid if gid
+        end
+      end
+      map
+    end
+
+    # idRangeOffset is a byte offset from its own slot in the array, which is
+    # why the segment index has to be folded back in.
+    def segment_glyph(index, code, starts, deltas, ranges, ranges_at)
+      return (code + deltas[index]) % 65_536 if ranges[index].zero?
+
+      gid = u16(ranges_at + (index * 2) + ranges[index] + ((code - starts[index]) * 2))
+      gid.zero? ? nil : (gid + deltas[index]) % 65_536
+    end
+  end
+end