ft: Create ForecastRepository class to handle different type of access/storage. Models have a flag to re-stage.

refac: Renamed ModelTree as ForecastRegistry and PathTree as ExperimentRegistry. Moved .dir() method to ForecastRegistry. Decoupled the get_source from the directory creation, which is now inside the model class. Removed dataclasses from Registries. Removed __call__ override for registry classes and replaced them with get_path(). Created an abc BaseFileRegistry, from which ExperimentRegistry and ForecastRegistry inherits. Removed query_gcmt, as it was included in pycsep. Renamed iter_attr() from Experiment.as_dict() to parse_nested_dict, and moved to utils.py.
sty: used black and pydocstringformatter in main modules.
build: Added pydocstringformatter options to .toml.
tests: updated all unit tests with the new classes.
fix: conda manager prints output live to shell.

Author: pabloitu

docs/reference/api_reference.rst

@@ -111,11 +111,32 @@ Accessors
 .. autosummary::
    :toctree: generated
 
-    query_gcmt
     from_zenodo
     from_git
 
 
+Environments
+------------
+
+.. :currentmodule:: floatcsep.environments
+
+.. automodule:: floatcsep.environments
+
+.. autosummary::
+   :toctree: generated
+
+    CondaManager
+    CondaManager.create_environment
+    CondaManager.env_exists
+    CondaManager.install_dependencies
+    CondaManager.run_command
+
+    VenvManager
+    CondaManager.create_environment
+    CondaManager.env_exists
+    CondaManager.install_dependencies
+    CondaManager.run_command
+
 
 Extras
 ------

floatcsep/accessors.py

@@ -14,48 +14,18 @@
 TIMEOUT = 180
 
 
-def query_gcmt(
-    start_time,
-    end_time,
-    min_magnitude=5.0,
-    max_depth=None,
-    catalog_id=None,
-    min_latitude=None,
-    max_latitude=None,
-    min_longitude=None,
-    max_longitude=None,
-):
-
-    eventlist = _query_gcmt(
-        start_time=start_time,
-        end_time=end_time,
-        min_magnitude=min_magnitude,
-        min_latitude=min_latitude,
-        max_latitude=max_latitude,
-        min_longitude=min_longitude,
-        max_longitude=max_longitude,
-        max_depth=max_depth,
-    )
-
-    catalog = CSEPCatalog(
-        data=eventlist, name="gCMT", catalog_id=catalog_id, date_accessed=utc_now_datetime()
-    )
-    return catalog
-
-
 def from_zenodo(record_id, folder, force=False):
     """
     Download data from a Zenodo repository.
-    Downloads if file does not exist, checksum has changed in local respect to
-    url or force
+
+    Downloads if file does not exist, checksum has changed in local respect to url or force
 
     Args:
         record_id: corresponding to the Zenodo repository
         folder: where the repository files will be downloaded
         force: force download even if file exists and checksum passes
 
     Returns:
-
     """
     # Grab the urls and filenames and checksums
     r = requests.get(f"https://zenodo.org/api/records/{record_id}", timeout=3)
@@ -87,8 +57,7 @@ def from_zenodo(record_id, folder, force=False):
 
 def from_git(url, path, branch=None, depth=1, **kwargs):
     """
-
-    Clones a shallow repository from a git url
+    Clones a shallow repository from a git url.
 
     Args:
         url (str): url of the repository
@@ -115,185 +84,13 @@ def from_git(url, path, branch=None, depth=1, **kwargs):
     return repo
 
 
-def _query_gcmt(
-    start_time,
-    end_time,
-    min_magnitude=3.50,
-    min_latitude=None,
-    max_latitude=None,
-    min_longitude=None,
-    max_longitude=None,
-    max_depth=1000,
-    extra_gcmt_params=None,
-):
-    """
-    Return GCMT eventlist from IRIS web service.
-    For details see "https://service.iris.edu/fdsnws/event/1/"
-    Args:
-        start_time (datetime.datetime): start time of catalog query
-        end_time (datetime.datetime): end time of catalog query
-        min_magnitude (float): minimum magnitude of query
-        min_latitude (float): minimum latitude of query
-        max_latitude (float): maximum latitude of query
-        min_longitude (float): minimum longitude of query
-        max_longitude (float): maximum longitude of query
-        max_depth (float): maximum depth of query
-        extra_gcmt_params (dict): additional parameters to pass to IRIS search
-         function
-
-    Returns:
-        eventlist
-    """
-    extra_gcmt_params = extra_gcmt_params or {}
-
-    eventlist = gcmt_search(
-        minmagnitude=min_magnitude,
-        minlatitude=min_latitude,
-        maxlatitude=max_latitude,
-        minlongitude=min_longitude,
-        maxlongitude=max_longitude,
-        starttime=start_time.isoformat(),
-        endtime=end_time.isoformat(),
-        maxdepth=max_depth,
-        **extra_gcmt_params,
-    )
-
-    return eventlist
-
-
-def gcmt_search(
-    format="text",
-    starttime=None,
-    endtime=None,
-    updatedafter=None,
-    minlatitude=None,
-    maxlatitude=None,
-    minlongitude=None,
-    maxlongitude=None,
-    latitude=None,
-    longitude=None,
-    maxradius=None,
-    catalog="GCMT",
-    contributor=None,
-    maxdepth=1000,
-    maxmagnitude=10.0,
-    mindepth=-100,
-    minmagnitude=0,
-    offset=1,
-    orderby="time-asc",
-    host=None,
-    verbose=False,
-):
-    """Search the IRIS database for events matching input criteria.
-    This search function is a wrapper around the ComCat Web API described here:
-    https://service.iris.edu/fdsnws/event/1/
-
-    This function returns a list of SummaryEvent objects, described elsewhere in this package.
-    Args:
-        starttime (datetime):
-            Python datetime - Limit to events on or after the specified start time.
-        endtime (datetime):
-            Python datetime - Limit to events on or before the specified end time.
-        updatedafter (datetime):
-           Python datetime - Limit to events updated after the specified time.
-        minlatitude (float):
-            Limit to events with a latitude larger than the specified minimum.
-        maxlatitude (float):
-            Limit to events with a latitude smaller than the specified maximum.
-        minlongitude (float):
-            Limit to events with a longitude larger than the specified minimum.
-        maxlongitude (float):
-            Limit to events with a longitude smaller than the specified maximum.
-        latitude (float):
-            Specify the latitude to be used for a radius search.
-        longitude (float):
-            Specify the longitude to be used for a radius search.
-        maxradius (float):
-            Limit to events within the specified maximum number of degrees
-            from the geographic point defined by the latitude and longitude parameters.
-        catalog (str):
-            Limit to events from a specified catalog.
-        contributor (str):
-            Limit to events contributed by a specified contributor.
-        maxdepth (float):
-            Limit to events with depth less than the specified maximum.
-        maxmagnitude (float):
-            Limit to events with a magnitude smaller than the specified maximum.
-        mindepth (float):
-            Limit to events with depth more than the specified minimum.
-        minmagnitude (float):
-            Limit to events with a magnitude larger than the specified minimum.
-        offset (int):
-            Return results starting at the event count specified, starting at 1.
-        orderby (str):
-            Order the results. The allowed values are:
-            - time order by origin descending time
-            - time-asc order by origin ascending time
-            - magnitude order by descending magnitude
-            - magnitude-asc order by ascending magnitude
-        host (str):
-            Replace default ComCat host (earthquake.usgs.gov) with a custom host.
-    Returns:
-        list: List of SummaryEvent() objects.
-    """
-
-    # getting the inputargs must be the first line of the method!
-    inputargs = locals().copy()
-    newargs = {}
-
-    for key, value in inputargs.items():
-        if value is True:
-            newargs[key] = "true"
-            continue
-        if value is False:
-            newargs[key] = "false"
-            continue
-        if value is None:
-            continue
-        newargs[key] = value
-
-    del newargs["verbose"]
-
-    events = _search_gcmt(**newargs)
-
-    return events
-
-
-def _search_gcmt(**_newargs):
-    """
-    Performs de-query at ISC API and returns event list and access date
-
-    """
-    paramstr = urlencode(_newargs)
-    url = HOST_CATALOG + paramstr
-    fh = request.urlopen(url, timeout=TIMEOUT)
-    data = fh.read().decode("utf8").split("\n")
-    fh.close()
-    eventlist = []
-    for line in data[1:]:
-        line_ = line.split("|")
-        if len(line_) != 1:
-            id_ = line_[0]
-            time_ = datetime.fromisoformat(line_[1])
-            dt = datetime_to_utc_epoch(time_)
-            lat = float(line_[2])
-            lon = float(line_[3])
-            depth = float(line_[4])
-            mag = float(line_[10])
-            eventlist.append((id_, dt, lat, lon, depth, mag))
-
-    return eventlist
-
-
 def _download_file(url: str, filename: str) -> None:
     """
-
-    Downloads files (from zenodo)
+    Downloads files (from zenodo).
 
     Args:
         url (str): the url where the file is located
         filename (str): the filename required.
-
     """
     progress_bar_length = 72
     block_size = 1024
@@ -331,9 +128,7 @@ def _download_file(url: str, filename: str) -> None:
 
 
 def _check_hash(filename, checksum):
-    """
-    Checks if existing file hash matches checksum from url
-    """
+    """Checks if existing file hash matches checksum from url."""
     algorithm, value = checksum.split(":")
     if not os.path.exists(filename):
         return value, "invalid"

floatcsep/cmd/main.py

@@ -52,7 +52,7 @@ def reproduce(config, **kwargs):
 
     log.info(f"floatCSEP v{__version__} | Reproduce")
 
-    reproduced_exp = Experiment.from_yml(config, reprdir="reproduced", **kwargs)
+    reproduced_exp = Experiment.from_yml(config, repr_dir="reproduced", **kwargs)
     reproduced_exp.stage_models()
     reproduced_exp.set_tasks()
     reproduced_exp.run()