Source code for message_ix_models.project.advance.data

import logging
from copy import copy

from genno import Computer, Key

from message_ix_models.tools.exo_data import (
    ExoDataSource,
    iamc_like_data_for_query,
    register_source,
)
from message_ix_models.util import (
    HAS_MESSAGE_DATA,
    iter_keys,
    package_data_path,
    private_data_path,
)

__all__ = [
    "ADVANCE",
]

log = logging.getLogger(__name__)


#: Expected location of the ADVANCE WP2 data snapshot.
LOCATION = "advance", "advance_compare_20171018-134445.csv.zip"

#: Name of the data file within the archive.
NAME = "advance_compare_20171018-134445.csv"


[docs]@register_source class ADVANCE(ExoDataSource): """Provider of exogenous data from the ADVANCE project database. To use data from this source, call :func:`.exo_data.prepare_computer` with the arguments: - `source`: "ADVANCE" - `source_kw` including: - "model": one of 12 codes including "MESSAGE". - "measure": one of 3080 codes for the "VARIABLE" dimension. - "scenario": one of 51 codes including "ADV3TRAr2_Base". - "name", optional: override :attr:`.ExoDataSource.name`. - "aggregate", optional: if :any:`True`, aggregate data from the ADVANCE native regions using ``n::groups`` (same behaviour as the base class). Otherwise, do not aggregate. Example ------- >>> keys = prepare_computer( ... context, ... computer, ... source="ADVANCE", ... source_kw=dict( ... measure="Transport|Service demand|Road|Freight", ... model="MESSAGE", ... scenario="ADV3TRAr2_Base", ... ), ... ) >>> result = computer.get(keys[0]) Load the metadata packaged with :mod:`message_ix_models` to identify usable `source_kw`: >>> import sdmx >>> from message_ix_models.util import package_data_path >>> >>> msg = sdmx.read_sdmx(package_data_path("sdmx", "ADVANCE.xml")) >>> msg <sdmx.StructureMessage> <Header> prepared: '2023-11-03T21:51:46.052879' source: en: Generated by message_ix_models 2023.9.13.dev57+ga558c0b4.d20231011 test: False Codelist (5): MODEL SCENARIO REGION VARIABLE UNIT >>> msg.codelist["MODEL].items {'AIM/CGE': <Code AIM/CGE>, 'DNE21+': <Code DNE21+>, 'GCAM': <Code GCAM>, 'GEM-E3': <Code GEM-E3>, 'IMACLIM V1.1': <Code IMACLIM V1.1>, 'IMAGE': <Code IMAGE>, 'MESSAGE': <Code MESSAGE>, 'POLES ADVANCE': <Code POLES ADVANCE>, 'REMIND': <Code REMIND>, 'TIAM-UCL': <Code TIAM-UCL>, 'WITCH': <Code WITCH>, 'iPETS V.1.5': <Code iPETS V.1.5>} """ id = "ADVANCE" def __init__(self, source, source_kw): if not source == self.id: raise ValueError(source) # Map the `measure` keyword to a string appearing in the data _kw = copy(source_kw) measure = _kw.pop("measure") self.variable = { "GDP": "GDP|PPP", "POP": "Population", }.get(measure, measure) # Store the model and scenario ID self.model = _kw.pop("model", None) self.scenario = _kw.pop("scenario", None) # Set the name of the returned quantity self.name = _kw.pop("name", "") self.aggregate = _kw.pop("aggregate", True) if len(_kw): raise ValueError(_kw) def __call__(self): # Assemble a query string query = " and ".join( [ f"SCENARIO == {self.scenario!r}", f"VARIABLE == {self.variable!r}", f"MODEL == {self.model!r}" if self.model else "True", ] ) log.debug(query) # Expected location of the ADVANCE WP2 data snapshot. parts = LOCATION if HAS_MESSAGE_DATA: path = private_data_path(*parts) else: path = package_data_path("test", *parts) log.warning(f"Reading random data from {path}") return iamc_like_data_for_query( path, query, archive_member=NAME, non_iso_3166="keep" )
[docs] def transform(self, c: "Computer", base_key: Key) -> Key: """Prepare `c` to transform raw data from `base_key`. Unlike the base class version, this implementation only adds the aggregation step if :attr:`.aggregate` is :any:`True`. """ k = iter_keys(base_key) k1 = base_key if self.aggregate: # Aggregate k1 = k() c.add(k1, "aggregate", base_key, "n::groups", keep=False) # Interpolate to the desired set of periods kw = dict(fill_value="extrapolate") k2 = k() c.add(k2, "interpolate", k1, "y::coords", kwargs=kw) return k2