Skip to content

Commit

Permalink
wip
Browse files Browse the repository at this point in the history
  • Loading branch information
@Marigold
Marigold committed Jul 21, 2023
1 parent a33dcce commit c9c83e3
Showing 1 changed file with 77 additions and 67 deletions.
144 changes: 77 additions & 67 deletions walkthrough/snapshot.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -20,33 +20,44 @@ class Options(Enum):
DATASET_MANUAL_IMPORT = "Import dataset from local file"


class OriginForm(BaseModel):
dataset_title_owid: str
dataset_description_owid: str
dataset_description_producer: str
producer: str
citation_producer: str
dataset_url_main: str
dataset_url_download: str
date_accessed: str
date_published: str


class LicenseForm(BaseModel):
url: str
name: str


class SnapshotForm(BaseModel):
namespace: str
snapshot_version: str
short_name: str
name: str
source_name: str
source_published_by: str
publication_year: Optional[str]
publication_date: Optional[str]
url: str
source_data_url: str
file_extension: str
license_name: str
license_url: str
description: str
origin: OriginForm
license: LicenseForm
is_private: bool
dataset_manual_import: bool

def __init__(self, **data: Any) -> None:
options = data.pop("options")
data["is_private"] = Options.IS_PRIVATE.value in options
data["dataset_manual_import"] = Options.DATASET_MANUAL_IMPORT.value in options
super().__init__(**data)

@property
def version(self) -> str:
return self.snapshot_version or self.publication_year or self.publication_date # type: ignore
# TODO: extract fields for origin
__import__("ipdb").set_trace()
data["origin"] = OriginForm(**data.pop("origin"))

# TODO: extract fields for license

super().__init__(**data)


def app(run_checks: bool) -> None:
Expand Down Expand Up @@ -74,15 +85,15 @@ def app(run_checks: bool) -> None:
help_text="Institution name. Example: emdat",
),
pi.input(
"Snapshot version",
"Version",
name="snapshot_version",
placeholder=str(dt.date.today()),
required=True,
value=state.get("snapshot_version", str(dt.date.today())),
help_text="Version of the snapshot dataset (by default, the current date, or exceptionally the publication date).",
),
pi.input(
"Snapshot dataset short name",
"Short name",
name="short_name",
placeholder="testing_dataset_name",
required=True,
Expand All @@ -91,79 +102,78 @@ def app(run_checks: bool) -> None:
help_text="Underscored dataset short name. Example: natural_disasters",
),
pi.input(
"Dataset full name",
name="name",
placeholder="Testing Dataset Name (Institution, 2023)",
required=True,
value=state.get("name"),
help_text="Human-readable dataset name, followed by (Institution, Year of version). Example: Natural disasters (EMDAT, 2022)",
),
pi.input(
"Source short citation",
name="source_name",
placeholder="Testing Short Citation",
required=True,
value=state.get("source_name"),
help_text="Short source citation (to show in charts). Example: EM-DAT",
),
pi.input(
"Source full citation",
name="source_published_by",
placeholder="Testing Full Citation",
required=True,
value=state.get("source_published_by"),
help_text="Testing Full Citation, as recommended by the source. Example: EM-DAT, CRED / UCLouvain, Brussels, Belgium",
"Origin: Dataset title OWID",
name="dataset_title_owid",
placeholder="Testing Title OWID",
value=state.get("dataset_title_owid"),
help_text="Original title from the source. Example: Natural disasters",
),
pi.input(
"Publication date",
name="publication_date",
placeholder="",
value=state.get("publication_date"),
help_text="Date when the dataset was published by the source. Example: 2023-01-01",
"Origin: Producer",
name="producer",
placeholder="Testing Institution",
value=state.get("producer"),
help_text="The name of the institution (without a year) or the main authors of the paper. Example: EM-DAT",
),
# TODO: should it contain year too? it would be good to clarify
pi.input(
"Publication year",
name="publication_year",
type=pi.NUMBER,
placeholder="",
help_text="Only if the exact publication date is unknown, year when the dataset was published by the source. Example: 2023",
"Origin: Citation by the producer",
name="citation_producer",
placeholder="Testing Citation",
value=state.get("citation_producer"),
help_text="The full citation that the producer asks for. Example: EM-DAT, CRED / UCLouvain, Brussels, Belgium",
),
pi.input(
"Dataset webpage URL",
name="url",
placeholder=("https://url_of_testing_source.com/"),
required=True,
value=state.get("url"),
"Origin: Dataset main URL",
name="dataset_url_main",
placeholder="https://url_of_testing_source.com/",
value=state.get("dataset_url_main"),
help_text="URL to the main page of the project.",
),
pi.input(
"Dataset download URL",
name="source_data_url",
"Origin: Dataset download URL",
name="dataset_url_download",
placeholder="https://url_of_testing_source.com/data.csv",
value=state.get("source_data_url"),
help_text="URL to download the data file.",
value=state.get("dataset_url_download"),
help_text="Direct URL to download the dataset.",
),
pi.input(
"File extension",
name="file_extension",
placeholder="csv",
value=state.get("file_extension"),
help_text="File extension (without the '.') of the file to be downloaded. Example: csv",
"Origin: Publication date",
name="date_published",
placeholder="2023-01-01",
value=state.get("date_published"),
help_text="Date when the dataset was published, could be date or year. Example: 2023-01-01 or 2023",
),
pi.textarea(
"Origin: Dataset description OWID",
name="dataset_description_owid",
value=state.get("dataset_description_owid"),
help_text="Our description of the dataset.",
),
pi.textarea(
"Origin: Dataset description by the producer",
name="dataset_description_producer",
value=state.get("dataset_description_producer"),
help_text="The description for this dataset used by the producer.",
),
pi.input(
"License URL",
"License: URL",
name="license_url",
placeholder=("https://url_of_testing_source.com/license"),
help_text="URL to the page where the source specifies the license of the dataset.",
),
pi.input(
"License name",
"License: name",
name="license_name",
placeholder="Creative Commons BY 4.0",
help_text="Name of the dataset license. Example: 'Creative Commons BY 4.0'",
),
pi.textarea(
"Description", name="description", value=state.get("description"), help_text="Dataset description."
pi.input(
"File extension",
name="file_extension",
placeholder="csv",
value=state.get("file_extension"),
help_text="File extension (without the '.') of the file to be downloaded. Example: csv",
),
pi.checkbox(
"Other options",
Expand Down

0 comments on commit c9c83e3

Please sign in to comment.