Better bounding boxes for pdf usetex, including descents.

jkseppan · jkseppan · commit 63c74adbdceb · 2007-09-10T20:31:01.000Z
svn path=/trunk/matplotlib/; revision=3827
diff --git a/lib/matplotlib/backends/backend_pdf.py b/lib/matplotlib/backends/backend_pdf.py
@@ -1427,14 +1427,14 @@ def draw_mathtext(self, gc, x, y, s, prop, angle):
         # Pop off the global transformation
         self.file.output(Op.grestore)
 
-    def _draw_tex(self, gc, x, y, s, prop, angle):
+    def draw_tex(self, gc, x, y, s, prop, angle):
         # Rename to draw_tex to enable
 
         texmanager = self.get_texmanager()
         fontsize = prop.get_size_in_points()
         dvifile = texmanager.make_dvi(s, fontsize)
         dvi = dviread.Dvi(dvifile, 72)
-        text, boxes = iter(dvi).next()
+        page = iter(dvi).next()
         dvi.close()
 
         if angle == 0:          # avoid rounding errors in common case
@@ -1448,7 +1448,7 @@ def mytrans(x1, y1, x=x, y=y, a=angle / 180.0 * pi):
         # Gather font information and do some setup for combining
         # characters into strings.
         oldfontnum, seq = None, []
-        for x1, y1, fontnum, glyph, width in text:
+        for x1, y1, fontnum, glyph, width in page.text:
             if fontnum != oldfontnum:
                 texname, fontsize = dvi.fontinfo(fontnum)
                 fontinfo = self.tex_font_mapping(texname)
@@ -1462,8 +1462,8 @@ def mytrans(x1, y1, x=x, y=y, a=angle / 180.0 * pi):
         seq += [('end',)]
 
         # Find consecutive text strings with constant x coordinate and
-        # combine into one string (if needed kern would be less than
-        # 0.1 points) or several strings interspersed with kerns.
+        # combine into a sequence of strings and kerns, or just one
+        # string (if any kerns would be less than 0.1 points).
         i, curx = 0, 0
         while i < len(seq)-1:
             elt, next = seq[i:i+2]
@@ -1503,7 +1503,7 @@ def mytrans(x1, y1, x=x, y=y, a=angle / 180.0 * pi):
         boxgc = self.new_gc()
         boxgc.copy_properties(gc)
         boxgc.set_linewidth(0)
-        for x1, y1, h, w in boxes:
+        for x1, y1, h, w in page.boxes:
             (x1, y1), (x2, y2), (x3, y3), (x4, y4) = \
                 mytrans(x1, y1), mytrans(x1+w, y1), \
                 mytrans(x1+w, y1+h), mytrans(x1, y1+h)
@@ -1653,14 +1653,9 @@ def get_text_width_height_descent(self, s, prop, ismath):
             fontsize = prop.get_size_in_points()
             dvifile = texmanager.make_dvi(s, fontsize)
             dvi = dviread.Dvi(dvifile, 72)
-            text, boxes = iter(dvi).next()
-            # TODO: better bounding box -- this is not quite right:
-            l = min(p[0] for p in text+boxes)
-            r = max(p[0] for p in text+boxes) + fontsize
-            b = min(p[1] for p in text+boxes)
-            t = max(p[1] for p in text+boxes) + fontsize
-            # (not to even mention finding the baseline)
-            return r-l, t-b, t-b
+            page = iter(dvi).next()
+            dvi.close()
+            return page.width, page.height, page.descent
         if ismath:
             w, h, d, glyphs, rects, used_characters = \
                 self.mathtext_parser.parse(s, 72, prop)
diff --git a/lib/matplotlib/dviread.py b/lib/matplotlib/dviread.py
@@ -6,19 +6,21 @@
 Interface:
 
     dvi = Dvi(filename, 72)
-    for text, boxes in dvi:          # iterate over pages
-        text, boxes = dvi.output(72)
-        for x,y,font,glyph,width in text:
+    for page in dvi:          # iterate over pages
+        w, h, d = page.width, page.height, page.descent
+        for x,y,font,glyph,width in page.text:
             fontname, pointsize = dvi.fontinfo(font)
             ...
-        for x,y,height,width in boxes:
+        for x,y,height,width in page.boxes:
             ...
 """
 
-# TODO: support for TeX virtual fonts (*.vf) which are a dvi-like format
+# TODO: support TeX virtual fonts (*.vf) which are a sort of
+#       subroutine collections for dvi files
 
 import matplotlib
 import matplotlib.cbook as mpl_cbook
+import numpy as npy
 import os
 import struct
 
@@ -74,20 +76,33 @@ def close(self):
     def _output(self):
         """
         Output the text and boxes belonging to the most recent page.
-        text, boxes = dvi._output()
+        page = dvi._output()
         """
-        t0 = self.text[0]
-        minx, miny, maxx, maxy = t0[0], t0[1], t0[0], t0[1]
+        minx, miny, maxx, maxy = npy.inf, npy.inf, -npy.inf, -npy.inf
+        maxy_pure = -npy.inf
         for elt in self.text + self.boxes:
-            x,y = elt[:2]
-            if x < minx: minx = x
-            if y < miny: miny = y
-            if x > maxx: maxx = x
-            if y > maxy: maxy = y
+            if len(elt) == 4:   # box
+                x,y,h,w = elt
+                e = 0           # zero depth
+            else:               # glyph
+                x,y,f,g,w = elt
+                font = self.fonts[f]
+                h = (font.scale * font.tfm.height[g]) >> 20
+                e = (font.scale * font.tfm.depth[g]) >> 20
+            minx = min(minx, x)
+            miny = min(miny, y - h)
+            maxx = max(maxx, x + w)
+            maxy = max(maxy, y + e)
+            maxy_pure = max(maxy_pure, y)
+
         d = self.dpi / (72.27 * 2**16) # from TeX's "scaled points" to dpi units
         text =  [ ((x-minx)*d, (maxy-y)*d, f, g, w*d) for (x,y,f,g,w) in self.text ]
         boxes = [ ((x-minx)*d, (maxy-y)*d, h*d, w*d) for (x,y,h,w) in self.boxes ]
-        return text, boxes
+
+        return mpl_cbook.Bunch(text=text, boxes=boxes, 
+                               width=(maxx-minx)*d, 
+                               height=(maxy_pure-miny)*d, 
+                               descent=(maxy-maxy_pure)*d)
 
     def fontinfo(self, f):
         """
@@ -361,28 +376,36 @@ class Tfm(object):
       width[i]: width of character #i, needs to be scaled 
         by the factor specified in the dvi file
         (this is a dict because indexing may not start from 0)
+      height[i], depth[i]: height and depth of character #i
     """
 
     def __init__(self, filename):
         file = open(filename, 'rb')
 
         try:
             header1 = file.read(24)
-            lh, bc, ec, nw = \
-                struct.unpack('!4H', header1[2:10])
+            lh, bc, ec, nw, nh, nd = \
+                struct.unpack('!6H', header1[2:14])
             header2 = file.read(4*lh)
             self.checksum, self.design_size = \
                 struct.unpack('!2I', header2[:8])
             # there is also encoding information etc.
             char_info = file.read(4*(ec-bc+1))
             widths = file.read(4*nw)
+            heights = file.read(4*nh)
+            depths = file.read(4*nd)
         finally:
             file.close()
 
-        widths = struct.unpack('!%dI' % nw, widths)
-        self.width = {}
+        self.width, self.height, self.depth = {}, {}, {}
+        widths, heights, depths = \
+            [ struct.unpack('!%dI' % n, x) 
+              for n,x in [(nw, widths), (nh, heights), (nd, depths)] ]
         for i in range(ec-bc):
             self.width[bc+i] = widths[ord(char_info[4*i])]
+            self.height[bc+i] = heights[ord(char_info[4*i+1]) >> 4]
+            self.depth[bc+i] = depths[ord(char_info[4*i+1]) & 0xf]
+
 
 class PsfontsMap(object):
     """
