Merge pull request #3907 from yawor/pcfenconding

Support for different charset encodings in PcfFontFile

Tests/fonts/LICENSE.txt

@@ -4,6 +4,7 @@ NotoSansJP-Thin.otf, from https://www.google.com/get/noto/help/cjk/
 AdobeVFPrototype.ttf, from https://github.com/adobe-fonts/adobe-variable-font-prototype
 TINY5x3GX.ttf, from http://velvetyne.fr/fonts/tiny
 ArefRuqaa-Regular.ttf, from https://github.com/google/fonts/tree/master/ofl/arefruqaa
+ter-x20b.pcf, from http://terminus-font.sourceforge.net/
 
 All of the above fonts are published under the SIL Open Font License (OFL) v1.1 (http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&id=OFL), which allows you to copy, modify, and redistribute them if you need to.
 

Tests/test_font_pcf_charsets.py

@@ -0,0 +1,120 @@
+import os
+
+from PIL import FontFile, Image, ImageDraw, ImageFont, PcfFontFile
+
+from .helper import assert_image_equal, assert_image_similar, skip_unless_feature
+
+fontname = "Tests/fonts/ter-x20b.pcf"
+
+charsets = {
+    "iso8859-1": {
+        "glyph_count": 223,
+        "message": "hello, world",
+        "image1": "Tests/images/test_draw_pbm_ter_en_target.png",
+    },
+    "iso8859-2": {
+        "glyph_count": 223,
+        "message": "witaj świecie",
+        "image1": "Tests/images/test_draw_pbm_ter_pl_target.png",
+    },
+    "cp1250": {
+        "glyph_count": 250,
+        "message": "witaj świecie",
+        "image1": "Tests/images/test_draw_pbm_ter_pl_target.png",
+    },
+}
+
+
+pytestmark = skip_unless_feature("zlib")
+
+
+def save_font(request, tmp_path, encoding):
+    with open(fontname, "rb") as test_file:
+        font = PcfFontFile.PcfFontFile(test_file, encoding)
+    assert isinstance(font, FontFile.FontFile)
+    # check the number of characters in the font
+    assert len([_f for _f in font.glyph if _f]) == charsets[encoding]["glyph_count"]
+
+    tempname = str(tmp_path / "temp.pil")
+
+    def delete_tempfile():
+        try:
+            os.remove(tempname[:-4] + ".pbm")
+        except OSError:
+            pass  # report?
+
+    request.addfinalizer(delete_tempfile)
+    font.save(tempname)
+
+    with Image.open(tempname.replace(".pil", ".pbm")) as loaded:
+        with Image.open("Tests/fonts/ter-x20b-%s.pbm" % encoding) as target:
+            assert_image_equal(loaded, target)
+
+    with open(tempname, "rb") as f_loaded:
+        with open("Tests/fonts/ter-x20b-%s.pil" % encoding, "rb") as f_target:
+            assert f_loaded.read() == f_target.read()
+    return tempname
+
+
+def _test_sanity(request, tmp_path, encoding):
+    save_font(request, tmp_path, encoding)
+
+
+def test_sanity_iso8859_1(request, tmp_path):
+    _test_sanity(request, tmp_path, "iso8859-1")
+
+
+def test_sanity_iso8859_2(request, tmp_path):
+    _test_sanity(request, tmp_path, "iso8859-2")
+
+
+def test_sanity_cp1250(request, tmp_path):
+    _test_sanity(request, tmp_path, "cp1250")
+
+
+def _test_draw(request, tmp_path, encoding):
+    tempname = save_font(request, tmp_path, encoding)
+    font = ImageFont.load(tempname)
+    im = Image.new("L", (150, 30), "white")
+    draw = ImageDraw.Draw(im)
+    message = charsets[encoding]["message"].encode(encoding)
+    draw.text((0, 0), message, "black", font=font)
+    with Image.open(charsets[encoding]["image1"]) as target:
+        assert_image_similar(im, target, 0)
+
+
+def test_draw_iso8859_1(request, tmp_path):
+    _test_draw(request, tmp_path, "iso8859-1")
+
+
+def test_draw_iso8859_2(request, tmp_path):
+    _test_draw(request, tmp_path, "iso8859-2")
+
+
+def test_draw_cp1250(request, tmp_path):
+    _test_draw(request, tmp_path, "cp1250")
+
+
+def _test_textsize(request, tmp_path, encoding):
+    tempname = save_font(request, tmp_path, encoding)
+    font = ImageFont.load(tempname)
+    for i in range(255):
+        (dx, dy) = font.getsize(bytearray([i]))
+        assert dy == 20
+        assert dx in (0, 10)
+    message = charsets[encoding]["message"].encode(encoding)
+    for l in range(len(message)):
+        msg = message[: l + 1]
+        assert font.getsize(msg) == (len(msg) * 10, 20)
+
+
+def test_textsize_iso8859_1(request, tmp_path):
+    _test_textsize(request, tmp_path, "iso8859-1")
+
+
+def test_textsize_iso8859_2(request, tmp_path):
+    _test_textsize(request, tmp_path, "iso8859-2")
+
+
+def test_textsize_cp1250(request, tmp_path):
+    _test_textsize(request, tmp_path, "cp1250")

src/PIL/PcfFontFile.py

@@ -56,7 +56,9 @@ class PcfFontFile(FontFile.FontFile):
 
     name = "name"
 
-    def __init__(self, fp):
+    def __init__(self, fp, charset_encoding="iso8859-1"):
+
+        self.charset_encoding = charset_encoding
 
         magic = l32(fp.read(4))
         if magic != PCF_MAGIC:
@@ -229,12 +231,17 @@ class PcfFontFile(FontFile.FontFile):
 
         nencoding = (lastCol - firstCol + 1) * (lastRow - firstRow + 1)
 
-        for i in range(nencoding):
-            encodingOffset = i16(fp.read(2))
-            if encodingOffset != 0xFFFF:
+        encodingOffsets = [i16(fp.read(2)) for _ in range(nencoding)]
+
+        for i in range(firstCol, len(encoding)):
             try:
-                    encoding[i + firstCol] = encodingOffset
-                except IndexError:
-                    break  # only load ISO-8859-1 glyphs
+                encodingOffset = encodingOffsets[
+                    ord(bytearray([i]).decode(self.charset_encoding))
+                ]
+                if encodingOffset != 0xFFFF:
+                    encoding[i] = encodingOffset
+            except UnicodeDecodeError:
+                # character is not supported in selected encoding
+                pass
 
         return encoding