Skip to content

image

Strategy class for image/jpg.

ImageConfig

Bases: AttrDict

Configuration data model for ImageDataParseStrategy.

Source code in oteapi/strategies/parse/image.py
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
class ImageConfig(AttrDict):
    """Configuration data model for
    [`ImageDataParseStrategy`][oteapi.strategies.parse.image.ImageDataParseStrategy]."""

    # Resource config
    downloadUrl: Optional[HostlessAnyUrl] = Field(
        None, description=ResourceConfig.model_fields["downloadUrl"].description
    )
    mediaType: Optional[
        Literal[
            "image/jpg",
            "image/jpeg",
            "image/jp2",
            "image/png",
            "image/gif",
            "image/tiff",
            "image/eps",
        ]
    ] = Field(
        None,
        description=ResourceConfig.model_fields["mediaType"].description,
    )

    # Image parse strategy-specific config
    crop: Optional[Tuple[int, int, int, int]] = Field(
        None,
        description="Box cropping parameters (left, top, right, bottom).",
        # Effectively mapping 'imagecrop' to 'crop'.
        # 'imagecrop' is used by the crop filter strategy.
        validation_alias=AliasChoices("crop", "imagecrop"),
    )
    datacache_config: Optional[DataCacheConfig] = Field(
        None,
        description="Configuration options for the local data cache.",
    )
    image_key: Optional[str] = Field(
        None,
        description="Key to use when storing the image data in datacache.",
    )
    image_mode: Optional[str] = Field(
        None,
        description=(
            "Pillow mode to convert image into. See "
            "https://pillow.readthedocs.io/en/stable/handbook/concepts.html "
            "for details."
        ),
    )

crop: Optional[Tuple[int, int, int, int]] = Field(None, description='Box cropping parameters (left, top, right, bottom).', validation_alias=AliasChoices('crop', 'imagecrop')) class-attribute instance-attribute

datacache_config: Optional[DataCacheConfig] = Field(None, description='Configuration options for the local data cache.') class-attribute instance-attribute

downloadUrl: Optional[HostlessAnyUrl] = Field(None, description=ResourceConfig.model_fields['downloadUrl'].description) class-attribute instance-attribute

image_key: Optional[str] = Field(None, description='Key to use when storing the image data in datacache.') class-attribute instance-attribute

image_mode: Optional[str] = Field(None, description='Pillow mode to convert image into. See https://pillow.readthedocs.io/en/stable/handbook/concepts.html for details.') class-attribute instance-attribute

mediaType: Optional[Literal['image/jpg', 'image/jpeg', 'image/jp2', 'image/png', 'image/gif', 'image/tiff', 'image/eps']] = Field(None, description=ResourceConfig.model_fields['mediaType'].description) class-attribute instance-attribute

ImageDataParseStrategy

Parse strategy for images.

This strategy uses Pillow to read a raw image from the data cache, converts it into a NumPy array and stores the new array in the data cache.

It also supports simple cropping and image conversions.

The key to the new array and other metadata is returned. See ImageParseContent for more info.

Source code in oteapi/strategies/parse/image.py
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
@dataclass
class ImageDataParseStrategy:
    """Parse strategy for images.

    This strategy uses Pillow to read a raw image from the data cache,
    converts it into a NumPy array and stores the new array in the
    data cache.

    It also supports simple cropping and image conversions.

    The key to the new array and other metadata is returned. See
    [`ImageParseContent`][oteapi.strategies.parse.image.ImageParseContent]
    for more info.

    """

    parse_config: ImageParserConfig

    def initialize(self) -> AttrDict:
        """Initialize strategy."""
        return AttrDict()

    def get(self) -> ImageParseContent:
        """Execute the strategy."""

        config = self.parse_config.configuration

        if config.mediaType is None:
            raise ValueError("No media type provided to the image parser")

        mime_format = config.mediaType.split("/")[1]
        image_format = SupportedFormat[mime_format].value

        # Download the image
        download_config = config.model_dump()
        download_config["configuration"] = config.model_dump()
        output = create_strategy("download", download_config).get()

        if config.datacache_config and config.datacache_config.accessKey:
            cache_key = config.datacache_config.accessKey
        elif "key" in output:
            cache_key = output["key"]
        else:
            raise RuntimeError("No data cache key provided to the downloaded content")

        cache = DataCache(config.datacache_config)

        # Treat image according to filter values
        with (
            cache.getfile(cache_key, suffix=mime_format) as filename,
            Image.open(filename, formats=[image_format]) as image,
        ):
            if config.crop:
                image = image.crop(config.crop)
            if config.image_mode:
                image = image.convert(mode=config.image_mode)

            if image_format == "GIF" and image.info.get("version", b"").startswith(
                b"GIF"
            ):
                image.info.update(
                    {"version": image.info.get("version", b"")[len(b"GIF") :]}
                )

            image_key = cache.add(
                image.tobytes(),
                key=config.image_key,
            )

            if image.mode == "P":
                image_palette_key = cache.add(image.getpalette())
            else:
                image_palette_key = None

            # The returned content must be json serialisable - filter out all
            # non-json serialisable fields in image.info
            if image.info:
                image_info = {
                    key: val
                    for key, val in image.info.items()
                    if isinstance(val, (str, int, float, type(None), bool, tuple, list))
                }
            else:
                image_info = {}

            content = ImageParseContent(
                image_key=image_key,
                image_size=image.size,
                image_mode=image.mode,
                image_palette_key=image_palette_key,
                image_info=image_info,
            )

        return content

parse_config: ImageParserConfig instance-attribute

get()

Execute the strategy.

Source code in oteapi/strategies/parse/image.py
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
def get(self) -> ImageParseContent:
    """Execute the strategy."""

    config = self.parse_config.configuration

    if config.mediaType is None:
        raise ValueError("No media type provided to the image parser")

    mime_format = config.mediaType.split("/")[1]
    image_format = SupportedFormat[mime_format].value

    # Download the image
    download_config = config.model_dump()
    download_config["configuration"] = config.model_dump()
    output = create_strategy("download", download_config).get()

    if config.datacache_config and config.datacache_config.accessKey:
        cache_key = config.datacache_config.accessKey
    elif "key" in output:
        cache_key = output["key"]
    else:
        raise RuntimeError("No data cache key provided to the downloaded content")

    cache = DataCache(config.datacache_config)

    # Treat image according to filter values
    with (
        cache.getfile(cache_key, suffix=mime_format) as filename,
        Image.open(filename, formats=[image_format]) as image,
    ):
        if config.crop:
            image = image.crop(config.crop)
        if config.image_mode:
            image = image.convert(mode=config.image_mode)

        if image_format == "GIF" and image.info.get("version", b"").startswith(
            b"GIF"
        ):
            image.info.update(
                {"version": image.info.get("version", b"")[len(b"GIF") :]}
            )

        image_key = cache.add(
            image.tobytes(),
            key=config.image_key,
        )

        if image.mode == "P":
            image_palette_key = cache.add(image.getpalette())
        else:
            image_palette_key = None

        # The returned content must be json serialisable - filter out all
        # non-json serialisable fields in image.info
        if image.info:
            image_info = {
                key: val
                for key, val in image.info.items()
                if isinstance(val, (str, int, float, type(None), bool, tuple, list))
            }
        else:
            image_info = {}

        content = ImageParseContent(
            image_key=image_key,
            image_size=image.size,
            image_mode=image.mode,
            image_palette_key=image_palette_key,
            image_info=image_info,
        )

    return content

initialize()

Initialize strategy.

Source code in oteapi/strategies/parse/image.py
144
145
146
def initialize(self) -> AttrDict:
    """Initialize strategy."""
    return AttrDict()

ImageParseContent

Bases: AttrDict

Configuration model for the returned content from the Image parser.

See Pillow handbook for more details on image_mode, image_palette, and image_info.

Source code in oteapi/strategies/parse/image.py
 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
class ImageParseContent(AttrDict):
    """Configuration model for the returned content from the Image parser.

    See
    [Pillow handbook](https://pillow.readthedocs.io/en/stable/handbook/concepts.html)
    for more details on `image_mode`, `image_palette`, and `image_info`.
    """

    image_key: str = Field(
        ...,
        description="Key with which the image content is stored in the data cache.",
    )
    image_size: Tuple[int, int] = Field(
        ...,
        description="Image size (width, height).",
    )
    image_mode: str = Field(
        ...,
        description="Image mode. Examples: 'L', 'P', 'RGB', 'RGBA'...",
    )
    image_palette_key: Optional[str] = Field(
        None,
        description="Datacache key for colour palette if mode is 'P'.",
    )
    image_info: dict = Field(
        {},
        description="Additional information about the image.",
    )

image_info: dict = Field({}, description='Additional information about the image.') class-attribute instance-attribute

image_key: str = Field(..., description='Key with which the image content is stored in the data cache.') class-attribute instance-attribute

image_mode: str = Field(..., description="Image mode. Examples: 'L', 'P', 'RGB', 'RGBA'...") class-attribute instance-attribute

image_palette_key: Optional[str] = Field(None, description="Datacache key for colour palette if mode is 'P'.") class-attribute instance-attribute

image_size: Tuple[int, int] = Field(..., description='Image size (width, height).') class-attribute instance-attribute

ImageParserConfig

Bases: ParserConfig

Image parse strategy resource config.

Source code in oteapi/strategies/parse/image.py
71
72
73
74
75
76
77
78
79
80
81
class ImageParserConfig(ParserConfig):
    """Image parse strategy resource config."""

    parserType: Literal["parser/image"] = Field(
        "parser/image",
        description=ParserConfig.model_fields["parserType"].description,
    )
    configuration: ImageConfig = Field(
        ...,
        description="Image parse strategy-specific configuration.",
    )

configuration: ImageConfig = Field(..., description='Image parse strategy-specific configuration.') class-attribute instance-attribute

parserType: Literal['parser/image'] = Field('parser/image', description=ParserConfig.model_fields['parserType'].description) class-attribute instance-attribute

SupportedFormat

Bases: Enum

Supported formats for ImageDataParseStrategy.

Source code in oteapi/strategies/parse/image.py
84
85
86
87
88
89
90
91
92
93
class SupportedFormat(Enum):
    """Supported formats for `ImageDataParseStrategy`."""

    jpeg = "JPEG"
    jpg = "JPEG"
    jp2 = "JPEG2000"
    png = "PNG"
    gif = "GIF"
    tiff = "TIFF"
    eps = "EPS"

eps = 'EPS' class-attribute instance-attribute

gif = 'GIF' class-attribute instance-attribute

jp2 = 'JPEG2000' class-attribute instance-attribute

jpeg = 'JPEG' class-attribute instance-attribute

jpg = 'JPEG' class-attribute instance-attribute

png = 'PNG' class-attribute instance-attribute

tiff = 'TIFF' class-attribute instance-attribute