File: Buffer.py

package info (click to toggle)
python-fontfeatures 1.9.0%2Bds-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid
  • size: 2,096 kB
  • sloc: python: 9,112; makefile: 22
file content (349 lines) | stat: -rw-r--r-- 12,028 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
from dataclasses import dataclass
from fontFeatures import ValueRecord
from fontFeatures.utils import categorize_glyph
from youseedee import ucd_data
import sys
import warnings


def _add_value_records(vr1, vr2):
    if vr1.xPlacement or vr2.xPlacement:
        vr1.xPlacement = (vr1.xPlacement or 0) + (vr2.xPlacement or 0)
    if vr1.yPlacement or vr2.yPlacement:
        vr1.yPlacement = (vr1.yPlacement or 0) + (vr2.yPlacement or 0)
    if vr1.xAdvance or vr2.xAdvance:
        vr1.xAdvance = (vr1.xAdvance or 0) + (vr2.xAdvance or 0)
    if vr1.yAdvance or vr2.yAdvance:
        vr1.yAdvance = (vr1.yAdvance or 0) + (vr2.yAdvance or 0)


@dataclass
class BufferItem:
    # codepoint: int
    # glyph: str
    # position: ValueRecord
    # category: str

    def __repr__(self):
        s = ""
        if self.glyph:
            s = self.glyph
            if self.category[0] == "base":
                s = s + "_"
            elif self.category[0] == "mark":
                s = s + "^"
            elif self.category[0] == "ligature":
                s = s + "(fi)"
            else:
                s = s + "?"
        else:
            s = "U+%04x" % self.codepoint
        return "BufferItem(%s)" % s

    @classmethod
    def new_unicode(klass, codepoint):
        self = klass()
        self.codepoint = codepoint
        self.glyph = None
        self.feature_masks = {}
        return self

    @classmethod
    def new_glyph(klass, glyph, font):
        self = klass()
        self.codepoint = None
        self.glyph = glyph
        self.feature_masks = {}
        self.prep_glyph(font)
        return self

    def map_to_glyph(self, font):
        if not self.glyph:
            self.glyph = font.unicode_map.get(self.codepoint)
        if not self.glyph:
            # Notdef
            self.glyph = list(font.glyphs.keys())[0]
        self.prep_glyph(font)

    def prep_glyph(self, font):
        if "pytest" in sys.modules:
            if self.glyph in font.exported_glyphs():
                self.gid = font.exported_glyphs().index(self.glyph)
            else:
                self.gid = -1  # ?
        self.substituted = False
        self.ligated = False
        self.multiplied = False
        self.recategorize(font)
        try:
            self.position = ValueRecord(xAdvance=0)
            self.position.xAdvance = font.default_master.get_glyph_layer(
                self.glyph
            ).width
        except Exception:
            if "pytest" in sys.modules:
                # We tolerate broken fonts in pytest
                pass
            else:
                raise ValueError(
                    "Could not get xAdvance for glyph %s (%i)"
                    % (self.glyph, self.codepoint)
                )

    def recategorize(self, font):
        try:
            self.category = categorize_glyph(font, self.glyph)
            if not self.category[0]:
                self.category = ("unknown", None)
        except Exception as e:
            warnings.warn("Error getting category: %s" % str(e))
            self._fallback_categorize()

    def _fallback_categorize(self):
        if not self.codepoint:
            # Now what?
            self.category = ("unknown", None)
            return
        genCat = ucd_data(self.codepoint).get("General_Category", "L")
        if genCat[0] == "M":
            self.category = ("mark", None)
        elif genCat == "Ll":
            self.category = ("ligature", None)
        elif genCat[0] == "L":
            self.category = ("base", None)
        else:
            self.category = ("unknown", None)

    def add_position(self, vr2):
        _add_value_records(self.position, vr2)


class Buffer:
    """A buffer holding either characters to be shaped or shaped glyphs."""

    itemclass = BufferItem

    def __init__(
        self, font, glyphs=[], unicodes=[], direction=None, script=None, language=None
    ):
        self.font = font
        self.direction = direction
        self.script = script
        self.language = language
        self.fallback_mark_positioning = False
        self.fallback_glyph_classes = False
        self.items = []
        self.mask = []
        self.flags = 0
        self.current_feature_mask = None
        if glyphs:
            self.store_glyphs(glyphs)
            self.clear_mask()
        elif unicodes:
            self.store_unicode(unicodes)
            self.guess_segment_properties()

    def store_glyphs(self, glyphs):
        """Initialize the buffer with list of glyphs.

        Args:
            glyphs (list): A list of glyph names.
        """
        self.items = [self.itemclass.new_glyph(g, self.font) for g in glyphs]

    def store_unicode(self, unistring):
        """Initialize the buffer with list of Unicode codepoints.

        Args:
            glyphs (list or str): A list of characters.
        """
        self.items = [self.itemclass.new_unicode(ord(char)) for char in unistring]

    def guess_segment_properties(self):
        """Try to automatically determine the script and direction properties."""
        for u in self.items:
            # Guess segment properties
            if not self.script:
                thisScript = ucd_data(u.codepoint)["Script"]
                if thisScript not in ["Common", "Unknown", "Inherited"]:
                    self.script = thisScript
        if not self.direction:
            from fontFeatures.shaperLib.Shaper import _script_direction

            self.direction = _script_direction(self.script)

    def map_to_glyphs(self):
        """Convert a buffer of codepoints to its initial glyph mapping."""
        for u in self.items:
            u.map_to_glyph(self.font)
        self.clear_mask()

    @property
    def is_all_glyphs(self):
        """Returns true if the contents are glyph items."""
        return all([x.glyph is not None for x in self.items])

    @property
    def is_all_unicodes(self):
        """Returns true if the contents are character items."""
        return all([x.codepoint is not None for x in self.items])

    def __getitem__(self, key):
        indexed = self.mask[key]
        if isinstance(indexed, range) or isinstance(indexed, slice):
            indexed = slice(indexed.start, indexed.stop, indexed.step)
        if isinstance(indexed, list):
            return [self.items[g] for g in indexed]
        return self.items[indexed]

    def __setitem__(self, key, value):
        indexed = self.mask[key]
        if len(indexed) == 1:  # Easy
            self.items[indexed[0] : indexed[0] + 1] = value
            return
        if len(value) == 1:  # Also easy
            self.items[indexed[0]] = value[0]
            for i in reversed(indexed[1:]):
                del self.items[i]
            return
        else:
            raise ValueError("Too hard :-(")

    def __len__(self):
        return len(self.mask)

    def update(self):
        """Categorises glyphs and recomputes masks.

        Called internally when the contents of the buffer changes."""
        for g in self.items:
            g.recategorize(self.font)
        self.recompute_mask()

    def clear_mask(self):
        """Clear the buffer mask."""
        self.flags = 0
        self.markFilteringSet = None
        self.markAttachmentSet = None
        self.current_feature_mask = None
        self.recompute_mask()

    def set_mask(self, flags, markFilteringSet=None, markAttachmentSet=None):
        """Apply a routine's flags and mark filtering set to the buffer."""
        self.flags = flags
        if self.flags & 0x10:
            assert markFilteringSet
        self.markFilteringSet = markFilteringSet
        if self.flags & 0xFF00:
            assert markAttachmentSet
        self.markAttachmentSet = markAttachmentSet
        self.recompute_mask()

    def recompute_mask(self):
        """Computes the mask property

        ``buffer.mask`` is a list of buffer item indices which is the current
        "view" of the buffer. For example, if ``buffer.flags == 0x8``, then the
        indices of all mark glyphs will be excluded from ``buffer.mask``.
        """
        mask = range(0, len(self.items))
        self.flags = self.flags or 0
        if self.flags & 0x2:  # IgnoreBases
            mask = list(filter(lambda ix: self.items[ix].category[0] != "base", mask))
        if self.flags & 0x4:  # IgnoreLigatures
            mask = list(
                filter(lambda ix: self.items[ix].category[0] != "ligature", mask)
            )
        if self.flags & 0x8:  # IgnoreMarks
            mask = list(filter(lambda ix: self.items[ix].category[0] != "mark", mask))
        if self.flags & 0x10:  # UseMarkFilteringSet
            mask = list(
                filter(
                    lambda ix: self.items[ix].category[0] != "mark"
                    or self.items[ix].glyph in self.markFilteringSet,
                    mask,
                )
            )
        if self.flags & 0xFF00:  # MarkAttachmentType
            mask = list(
                filter(
                    lambda ix: self.items[ix].category[0] != "mark"
                    or self.items[ix].glyph in self.markAttachmentSet,
                    mask,
                )
            )

        if self.current_feature_mask:
            feature = self.current_feature_mask
            mask = list(
                filter(
                    lambda ix: (feature not in self.items[ix].feature_masks)
                    or (not self.items[ix].feature_masks[feature]),
                    mask,
                )
            )
        self.mask = mask

    def set_feature_mask(self, feature):
        """Applies the mask for a particular feature."""
        self.current_feature_mask = feature
        self.recompute_mask()

    def move_item(self, src, dest):
        """Moves an item from src to dest."""
        self.items[dest:dest] = [self.items.pop(src)]

    def merge_clusters(self, start, end):
        """Currently unimplemented."""
        pass  # XXX

    def serialize(self, additional=None, position=True, names=True, ned=False):
        """Serialize a buffer to a string.

        Returns:
            The contents of the given buffer in a string format similar to
            that used by ``hb-shape``.

        """
        outs = []
        if additional:
            if not isinstance(additional, list):
                additional = [additional]
        else:
            additional = []
        xcursor = 0
        for ix, info in enumerate(self.items):
            if hasattr(info, "glyph") and info.glyph:
                if names:
                    outs.append("%s" % info.glyph)
                else:
                    outs.append("%s" % info.gid)
            else:
                outs.append("U+%04x" % info.codepoint)
            if ned:
                position = info.position
                if xcursor + (position.xPlacement or 0):
                    outs[-1] = outs[-1] + "@%i,%i" % (
                        xcursor + (position.xPlacement or 0),
                        position.yPlacement or 0,
                    )
                xcursor = xcursor + position.xAdvance
            elif position and hasattr(info, "position"):
                position = info.position
                if hasattr(info, "syllable_index"):
                    cluster = info.syllable_index
                else:
                    cluster = ix
                outs[-1] = outs[-1] + "=%i" % (cluster)
                if position.xPlacement or position.yPlacement:
                    outs[-1] = outs[-1] + "@%i,%i" % (
                        position.xPlacement or 0,
                        position.yPlacement or 0,
                    )
                outs[-1] = outs[-1] + "+%i" % (position.xAdvance)
            relevant = list(filter(lambda a: hasattr(info, a), additional))
            if relevant:
                outs[-1] = outs[-1] + "(%s)" % ",".join(
                    [str(getattr(info, a)) for a in relevant]
                )
        return "|".join(outs)