Fix import from Paprika not importing some images

This fixes 4 issues that can arise when importing recipes from Paprika:
1.  A silent failure to import the image due to an empty (0 bytes) or incomplete JPEG image. There was nothing in the logs, just no image attached to the imported recipe.
    The problem is that the temp file contents may not have been flushed to disk before copying the temp file to the working directory as `original.jpeg`.
    This was the main problem I faced when importing from Paprika and meant that none of the images would come across in the import.
2.  The below error log when importing a recipe that has no image.
    ```
    Failed to download image for slow-cooker-beef-ragu: slow-cooker-beef-ragu
    ```
    The error log doesn't cause any issues with the import, it's just not particularly helpful and did make me look into it breifly when trying to figure out why the other images failed to import.
    The error message has been replaced by an info message saying
    ```
    Recipe 'Slow-Cooker Beef Ragu' has no image
    ```
3.  Failure to import the image for a recipe when there's another recipe with the same name, either in the same import or pre-existing.
    The failure occurs with the error log
    ```
    Failed to download image for chicken-pot-pie-2: chicken-pot-pie-2
    ```
    The way it was handling images did not take into account that the imported recipe might end up with a slug that is not the same as the slugified name.
4.  Silently failing to import images due to unrecognised file contents when converting to WEBP.
    For my issue, the unrecognised files were all due to issue 1 making them all empty, but this could also just be for other reasons such as a corrupt file or unrecognised format.
    Now when an image being imported can't be processed, the below error will be logged
    ```
    Failed to download image for chicken-and-leek-casserole: cannot identify image file '/app/data/recipes/4598f8cb-fabd-4e1e-bca7-a357d4e09f34/images/original.jpeg'
    ```
    This applies to migrations from all supported formats, not just Paprika.
This commit is contained in:
brokeh 2025-08-07 22:12:06 +10:00
commit f4b62bc539
7 changed files with 30 additions and 41 deletions

View file

@ -1,6 +1,7 @@
import contextlib import contextlib
from pathlib import Path from pathlib import Path
from PIL import UnidentifiedImageError
from pydantic import UUID4 from pydantic import UUID4
from mealie.core import root_logger from mealie.core import root_logger
@ -24,6 +25,7 @@ from mealie.services.scraper import cleaner
from .._base_service import BaseService from .._base_service import BaseService
from .utils.database_helpers import DatabaseMigrationHelpers from .utils.database_helpers import DatabaseMigrationHelpers
from .utils.migration_alias import MigrationAlias from .utils.migration_alias import MigrationAlias
from .utils.migration_helpers import import_image
class BaseMigrator(BaseService): class BaseMigrator(BaseService):
@ -269,3 +271,9 @@ class BaseMigrator(BaseService):
recipe = cleaner.clean(recipe_dict, self.translator, url=recipe_dict.get("org_url", None)) recipe = cleaner.clean(recipe_dict, self.translator, url=recipe_dict.get("org_url", None))
return recipe return recipe
def import_image(self, slug: str, src: str | Path, recipe_id: UUID4):
try:
import_image(src, recipe_id)
except UnidentifiedImageError as e:
self.logger.error(f"Failed to import image for {slug}: {e}")

View file

@ -4,7 +4,7 @@ from pathlib import Path
from ._migration_base import BaseMigrator from ._migration_base import BaseMigrator
from .utils.migration_alias import MigrationAlias from .utils.migration_alias import MigrationAlias
from .utils.migration_helpers import MigrationReaders, import_image, split_by_comma from .utils.migration_helpers import MigrationReaders, split_by_comma
class ChowdownMigrator(BaseMigrator): class ChowdownMigrator(BaseMigrator):
@ -64,4 +64,4 @@ class ChowdownMigrator(BaseMigrator):
except StopIteration: except StopIteration:
continue continue
if cd_image: if cd_image:
import_image(cd_image, recipe_id) self.import_image(slug, cd_image, recipe_id)

View file

@ -9,13 +9,7 @@ from mealie.schema.reports.reports import ReportEntryCreate
from ._migration_base import BaseMigrator from ._migration_base import BaseMigrator
from .utils.migration_alias import MigrationAlias from .utils.migration_alias import MigrationAlias
from .utils.migration_helpers import ( from .utils.migration_helpers import MigrationReaders, glob_walker, parse_iso8601_duration, split_by_comma
MigrationReaders,
glob_walker,
import_image,
parse_iso8601_duration,
split_by_comma,
)
@dataclass @dataclass
@ -103,4 +97,4 @@ class NextcloudMigrator(BaseMigrator):
if status: if status:
nc_dir = nextcloud_dirs[slug] nc_dir = nextcloud_dirs[slug]
if nc_dir.image: if nc_dir.image:
import_image(nc_dir.image, recipe_id) self.import_image(slug, nc_dir.image, recipe_id)

View file

@ -7,13 +7,10 @@ import zipfile
from gzip import GzipFile from gzip import GzipFile
from pathlib import Path from pathlib import Path
from slugify import slugify
from mealie.schema.recipe import RecipeNote from mealie.schema.recipe import RecipeNote
from ._migration_base import BaseMigrator from ._migration_base import BaseMigrator
from .utils.migration_alias import MigrationAlias from .utils.migration_alias import MigrationAlias
from .utils.migration_helpers import import_image
def paprika_recipes(file: Path): def paprika_recipes(file: Path):
@ -67,32 +64,26 @@ class PaprikaMigrator(BaseMigrator):
] ]
def _migrate(self) -> None: def _migrate(self) -> None:
recipe_image_urls = {} recipes = [r for r in paprika_recipes(self.archive) if "name" in r]
recipe_models = [self.clean_recipe_dictionary(r) for r in recipes]
results = self.import_recipes_to_database(recipe_models)
recipes = [] for (slug, recipe_id, status), recipe in zip(results, recipes, strict=True):
for recipe in paprika_recipes(self.archive): if not status:
if "name" not in recipe:
continue continue
recipe_model = self.clean_recipe_dictionary(recipe) image_data = recipe.get("photo_data")
if image_data is None:
if "photo_data" in recipe: self.logger.info(f"Recipe '{recipe['name']}' has no image")
recipe_image_urls[slugify(recipe["name"])] = recipe["photo_data"]
recipes.append(recipe_model)
results = self.import_recipes_to_database(recipes)
for slug, recipe_id, status in results:
if not status:
continue continue
try: try:
# Images are stored as base64 encoded strings, so we need to decode them before importing. # Images are stored as base64 encoded strings, so we need to decode them before importing.
image = io.BytesIO(base64.b64decode(recipe_image_urls[slug])) image = io.BytesIO(base64.b64decode(image_data))
with tempfile.NamedTemporaryFile(suffix=".jpeg") as temp_file: with tempfile.NamedTemporaryFile(suffix=".jpeg") as temp_file:
temp_file.write(image.read()) temp_file.write(image.read())
temp_file.flush()
path = Path(temp_file.name) path = Path(temp_file.name)
import_image(path, recipe_id) self.import_image(slug, path, recipe_id)
except Exception as e: except Exception as e:
self.logger.error(f"Failed to download image for {slug}: {e}") self.logger.error(f"Failed to import image for {slug}: {e}")

View file

@ -8,7 +8,7 @@ from mealie.services.scraper import cleaner
from ._migration_base import BaseMigrator from ._migration_base import BaseMigrator
from .utils.migration_alias import MigrationAlias from .utils.migration_alias import MigrationAlias
from .utils.migration_helpers import import_image, parse_iso8601_duration from .utils.migration_helpers import parse_iso8601_duration
def clean_instructions(instructions: list[str]) -> list[str]: def clean_instructions(instructions: list[str]) -> list[str]:
@ -98,7 +98,7 @@ class RecipeKeeperMigrator(BaseMigrator):
recipes = [self.clean_recipe_dictionary(x) for x in recipes_as_dicts] recipes = [self.clean_recipe_dictionary(x) for x in recipes_as_dicts]
results = self.import_recipes_to_database(recipes) results = self.import_recipes_to_database(recipes)
for (_, recipe_id, status), recipe in zip(results, recipes, strict=False): for (slug, recipe_id, status), recipe in zip(results, recipes, strict=False):
if status: if status:
try: try:
if not recipe or not recipe.image: if not recipe or not recipe.image:
@ -107,4 +107,4 @@ class RecipeKeeperMigrator(BaseMigrator):
except StopIteration: except StopIteration:
continue continue
import_image(recipe.image, recipe_id) self.import_image(slug, recipe.image, recipe_id)

View file

@ -10,7 +10,6 @@ from mealie.schema.reports.reports import ReportEntryCreate
from ._migration_base import BaseMigrator from ._migration_base import BaseMigrator
from .utils.migration_alias import MigrationAlias from .utils.migration_alias import MigrationAlias
from .utils.migration_helpers import import_image
def _build_ingredient_from_ingredient_data(ingredient_data: dict[str, Any], title: str | None = None) -> dict[str, Any]: def _build_ingredient_from_ingredient_data(ingredient_data: dict[str, Any], title: str | None = None) -> dict[str, Any]:
@ -150,4 +149,4 @@ class TandoorMigrator(BaseMigrator):
except StopIteration: except StopIteration:
continue continue
import_image(r.image, recipe_id) self.import_image(slug, r.image, recipe_id)

View file

@ -104,6 +104,7 @@ def import_image(src: str | Path, recipe_id: UUID4):
"""Read the successful migrations attribute and for each import the image """Read the successful migrations attribute and for each import the image
appropriately into the image directory. Minification is done in mass appropriately into the image directory. Minification is done in mass
after the migration occurs. after the migration occurs.
May raise an UnidentifiedImageError if the file is not a recognised format.
""" """
if isinstance(src, str): if isinstance(src, str):
@ -113,11 +114,7 @@ def import_image(src: str | Path, recipe_id: UUID4):
return return
data_service = RecipeDataService(recipe_id=recipe_id) data_service = RecipeDataService(recipe_id=recipe_id)
try:
data_service.write_image(src, src.suffix) data_service.write_image(src, src.suffix)
except UnidentifiedImageError:
return
async def scrape_image(image_url: str, recipe_id: UUID4): async def scrape_image(image_url: str, recipe_id: UUID4):