import json
from goodtables import validate
from goodtables.cli import _print_report
from pathlib import Path
root = Path(__file__).parents[1]
metadata = json.load(open(root / "datapackage.json", "r"))
paths = {i["name"]: i["path"] for i in metadata["resources"] if "schema" in i}
schemas = {
i["name"]: i["schema"]
for i in metadata["resources"] if "schema" in i
}
for name, path in paths.items():
report = validate(root / path,
schema=schemas[name],
table_limit=20,
row_limit=600000)
_print_report(report)
print("\nData summary:\n")
print("Emissions sum w/o EU28: {:d} GgCO₂-equiv.".format(int(
export.Emissions.sum() - export.Emissions.loc['EUU'].sum())))
print("Percentage sum: {}".format(
export.Percentage.sum() - export.loc['EUU'].Percentage))
print("Count signatures: {}".format(export.Signature.count()))
print("Count ratified: {}".format(
export["Ratification"].count()))
ratified = export["Ratification"].notnull()
percentage_sum = (export[ratified].Percentage.sum() -
export.loc["EUU"].Percentage)
print("Sum of percentages with ratification w/o EU: {}".format(
percentage_sum))
def to_int(x):
if pd.isnull(x):
return ""
else:
return str(int(x))
export.Emissions = export.Emissions.apply(to_int)
export.Year = export.Year.apply(to_int)
export.to_csv(outfile, encoding="UTF-8")
report = validate(root / "datapackage.json")
if report["error-count"] > 0:
_print_report(report)