Skip to content

Segment extraction module

The segment extraction module is responsible for extracting translatable segments from source objects.

StreamFieldSegmentExtractor

A helper class to help traverse StreamField values and extract segments.

Source code in wagtail_localize/segments/extract.py
 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
class StreamFieldSegmentExtractor:
    """
    A helper class to help traverse StreamField values and extract segments.
    """

    def __init__(self, field, include_overridables=False):
        """
        Initialises a StreamFieldSegmentExtractor.

        Args:
            field (StreamField): The StreamField to extract segments from.
            include_overridables (boolean, optional): Set this to True to extract overridable segments too.
        """
        self.field = field
        self.include_overridables = include_overridables

    def handle_block(self, block_type, block_value, raw_value=None):
        # Need to check if the app is installed before importing EmbedBlock
        # See: https://github.com/wagtail/wagtail-localize/issues/309
        if apps.is_installed("wagtail.embeds"):
            from wagtail.embeds.blocks import EmbedBlock

            if isinstance(block_type, EmbedBlock):
                if self.include_overridables:
                    return [OverridableSegmentValue("", block_value.url)]
                else:
                    return []

        if hasattr(block_type, "get_translatable_segments"):
            return block_type.get_translatable_segments(block_value)

        elif isinstance(block_type, (blocks.URLBlock, blocks.EmailBlock)):
            if self.include_overridables:
                return [OverridableSegmentValue("", block_value)]
            else:
                return []

        elif isinstance(
            block_type, (blocks.CharBlock, blocks.TextBlock, blocks.BlockQuoteBlock)
        ):
            return [StringSegmentValue("", block_value)]

        elif isinstance(block_type, blocks.RichTextBlock):
            template, strings = extract_strings(block_value.source)

            # Find all unique href values
            hrefs = set()
            for _string, attrs in strings:
                for tag_attrs in attrs.values():
                    if "href" in tag_attrs:
                        hrefs.add(tag_attrs["href"])

            ret = (
                [TemplateSegmentValue("", "html", template, len(strings))]
                + [
                    StringSegmentValue("", string, attrs=attrs)
                    for string, attrs in strings
                ]
                + [
                    OverridableSegmentValue(quote_path_component(href), href)
                    for href in sorted(hrefs)
                ]
            )
            return ret

        elif isinstance(block_type, blocks.ChooserBlock):
            return self.handle_related_object_block(block_value)

        elif isinstance(block_type, blocks.StructBlock):
            return self.handle_struct_block(block_value, raw_value=raw_value)

        elif isinstance(block_type, blocks.ListBlock):
            return self.handle_list_block(block_value, raw_value=raw_value)

        elif isinstance(block_type, blocks.StreamBlock):
            return self.handle_stream_block(block_value)

        # Ignore everything else
        return []

    def handle_related_object_block(self, related_object):
        if related_object is None:
            return []

        # All FKs to translatable models should be translatable.
        # With the exception of pages that are special because we can localize them at runtime easily.
        # TODO: Perhaps we need a special type for pages where it links to the translation if availabe,
        # but falls back to the source if it isn't translated yet?
        # Note: This exact same decision was made for regular foreign keys in fields.py
        if isinstance(related_object, TranslatableMixin) and not isinstance(
            related_object, Page
        ):
            return [RelatedObjectSegmentValue.from_instance("", related_object)]
        else:
            return [OverridableSegmentValue("", related_object.pk)]

    def handle_struct_block(self, struct_block, raw_value=None):
        segments = []

        for field_name, block_value in struct_block.items():
            block_type = struct_block.block.child_blocks[field_name]
            try:
                block_raw_value = raw_value["value"].get(field_name)
            except (KeyError, TypeError):
                # e.g. raw_value is None, or is that from chooser
                block_raw_value = None
            segments.extend(
                segment.wrap(field_name)
                for segment in self.handle_block(
                    block_type, block_value, raw_value=block_raw_value
                )
            )

        return segments

    def handle_list_block(self, list_block, raw_value=None):
        segments = []
        if raw_value is not None:
            # Wagtail 2.16 changes ListBlock values to be ListValue objects (i.e. {'value': '', 'id': ''})
            # and will automatically convert from the simple list format used before. However that requires
            # the block to be saved. bound_blocks will return ListValue objects, so we need to check that the
            # stored value is the new format before extracting segments, otherwise the block ids will continue
            # to change.
            has_block_format = False
            if (
                isinstance(raw_value, dict)
                and "value" in raw_value
                and len(raw_value["value"]) > 0
            ):
                has_block_format = list_block.list_block._item_is_in_block_format(
                    raw_value["value"][0]
                )
            elif isinstance(raw_value, list) and len(raw_value) > 0:
                has_block_format = list_block.list_block._item_is_in_block_format(
                    raw_value[0]
                )
            if has_block_format:
                for index, block in enumerate(list_block.bound_blocks):
                    # pass on the relevant sub-block raw value, should it be a ListBlock
                    try:
                        if isinstance(raw_value, list):
                            block_raw_value = raw_value[index]
                        elif isinstance(raw_value, dict):
                            block_raw_value = raw_value["value"][index]
                        else:
                            block_raw_value = None
                    except (IndexError, KeyError):
                        block_raw_value = None

                    segments.extend(
                        segment.wrap(block.id)
                        for segment in self.handle_block(
                            block.block, block.value, raw_value=block_raw_value
                        )
                    )
        return segments

    def handle_stream_block(self, stream_block):
        segments = []

        for index, block in enumerate(stream_block):
            raw_data = stream_block.raw_data[index]
            segments.extend(
                segment.wrap(block.id)
                for segment in self.handle_block(
                    block.block, block.value, raw_value=raw_data
                )
            )

        return segments

__init__(field, include_overridables=False)

Initialises a StreamFieldSegmentExtractor.

Parameters:

Name Type Description Default
field StreamField

The StreamField to extract segments from.

required
include_overridables boolean

Set this to True to extract overridable segments too.

False
Source code in wagtail_localize/segments/extract.py
32
33
34
35
36
37
38
39
40
41
def __init__(self, field, include_overridables=False):
    """
    Initialises a StreamFieldSegmentExtractor.

    Args:
        field (StreamField): The StreamField to extract segments from.
        include_overridables (boolean, optional): Set this to True to extract overridable segments too.
    """
    self.field = field
    self.include_overridables = include_overridables

extract_segments(instance)

Extracts segments from the given model instance.

Parameters:

Name Type Description Default
instance Model

The model instance to extract segments from.

required

Returns:

Type Description

list[StringSegmentValue, TemplateSegmentValue, RelatedObjectSegmentValue, or OverridableSegmentValue]: The segment values that have been extracted.

Source code in wagtail_localize/segments/extract.py
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
def extract_segments(instance):
    """
    Extracts segments from the given model instance.

    Args:
        instance (Model): The model instance to extract segments from.

    Returns:
        list[StringSegmentValue, TemplateSegmentValue, RelatedObjectSegmentValue, or OverridableSegmentValue]: The
            segment values that have been extracted.
    """
    segments = []

    for translatable_field in get_translatable_fields(instance.__class__):
        field = translatable_field.get_field(instance.__class__)
        is_translatable = translatable_field.is_translated(instance)
        is_synchronized = translatable_field.is_synchronized(instance)
        is_overridable = translatable_field.is_overridable(instance)
        extract_overridables = is_synchronized and is_overridable

        if hasattr(field, "get_translatable_segments"):
            if is_translatable:
                segments.extend(
                    segment.wrap(field.name)
                    for segment in field.get_translatable_segments(
                        field.value_from_object(instance)
                    )
                )

        elif isinstance(field, StreamField):
            if is_translatable:
                segments.extend(
                    segment.wrap(field.name)
                    for segment in StreamFieldSegmentExtractor(
                        field, include_overridables=extract_overridables
                    ).handle_stream_block(field.value_from_object(instance))
                )

        elif isinstance(field, RichTextField):
            value = field.value_from_object(instance)

            if is_translatable:
                template, strings = extract_strings(value)

                # Find all unique href values
                hrefs = set()
                for _string, attrs in strings:
                    for tag_attrs in attrs.values():
                        if "href" in tag_attrs:
                            hrefs.add(tag_attrs["href"])

                field_segments = (
                    [TemplateSegmentValue("", "html", template, len(strings))]
                    + [
                        StringSegmentValue("", string, attrs=attrs)
                        for string, attrs in strings
                    ]
                    + [
                        OverridableSegmentValue(quote_path_component(href), href)
                        for href in sorted(hrefs)
                    ]
                )

                segments.extend(segment.wrap(field.name) for segment in field_segments)

            if extract_overridables:
                pass  # TODO: Extract images and links

        elif isinstance(field, (models.TextField, models.CharField)):
            if not field.choices:
                value = field.value_from_object(instance)

                if value is None:
                    continue

                if is_translatable:
                    segments.append(StringSegmentValue(field.name, value))

                elif extract_overridables:
                    segments.append(OverridableSegmentValue(field.name, value))

        elif isinstance(field, models.ForeignKey):
            if is_translatable:
                if not issubclass(field.related_model, TranslatableMixin):
                    raise ImproperlyConfigured(
                        "The foreign key `{}.{}.{}` was registered as a translatable "
                        "field but the model it points to `{}.{}` is not translatable".format(
                            field.model._meta.app_label,
                            field.model.__name__,
                            field.name,
                            field.related_model._meta.app_label,
                            field.related_model.__name__,
                        )
                    )

                related_instance = getattr(instance, field.name)

                if related_instance:
                    segments.append(
                        RelatedObjectSegmentValue.from_instance(
                            field.name, related_instance
                        )
                    )

            elif extract_overridables:
                related_instance = getattr(instance, field.name)

                if related_instance:
                    segments.append(
                        OverridableSegmentValue(field.name, related_instance.pk)
                    )
        elif (
            isinstance(field, models.ManyToOneRel)
            and isinstance(field.remote_field, ParentalKey)
            and issubclass(field.related_model, TranslatableMixin)
        ):
            manager = getattr(instance, field.name)

            if is_translatable:
                for child_instance in manager.all():
                    segments.extend(
                        segment.wrap(str(child_instance.translation_key)).wrap(
                            field.name
                        )
                        for segment in extract_segments(child_instance)
                    )

            elif extract_overridables:
                pass  # TODO

    class Counter:
        def __init__(self):
            self.value = 0

        def next(self):
            self.value += 1
            return self.value

    counter = Counter()

    return [
        segment.with_order(counter.next())
        for segment in segments
        if not segment.is_empty()
    ]

quote_path_component(text)

Puts quotes around the path compoenents, and escapes any special characters.

Source code in wagtail_localize/segments/extract.py
20
21
22
23
24
def quote_path_component(text):
    """
    Puts quotes around the path compoenents, and escapes any special characters.
    """
    return "'" + text.replace("\\", "\\\\").replace("'", "\\'") + "'"