gmprocess 2.7.0__tar.gz → 2.8.0__tar.gz

gmprocess-2.8.0/.coveragerc

@@ -0,0 +1,13 @@
+[report]
+    precision=2
+    omit=
+        # No way to test GUI
+        src/gmprocess/gui/gui_ucla/*
+        */__init__.py
+        # This is just a helper script that is only useful for USGS staff
+        src/gmprocess/bin/bundle_datarelease.py
+        # To be removed, includes a warning that it is not tested
+        src/gmprocess/waveform_processing/nn_quality_assurance.py
+[run]
+    patch = subprocess
+    parallel = true

{gmprocess-2.7.0 → gmprocess-2.8.0}/.gitlab-ci.yml

@@ -48,6 +48,9 @@ run tests and coverage:
   rules:
     - *fork
     - *upstream
+    - if: $CI_PIPELINE_SOURCE == "merge_request_event"
+  variables:
+    VCR_RECORD_MODE: "none"
   script:
     - pip install -e .[test]
     - strec_cfg update --datafolder /home/usgs-user/ --gcmt
@@ -102,3 +105,23 @@ pages:
     - cd docs
     - bash makedocs.sh
   stage: docs
+
+regression test:
+  rules:
+    - if: $CI_PIPELINE_SOURCE == "schedule"
+      variables:
+        ENVIRONMENT: staging
+    - if: > 
+        $CI_PIPELINE_SOURCE != "schedule"
+        && $CI_PIPELINE_SOURCE != "merge_request_event"
+      when: manual
+      allow_failure: true
+  script:
+    - pip install -e .[test]
+    - pip install polars
+    - strec_cfg update --datafolder /home/usgs-user/ --gcmt
+    - mkdir /home/usgs-user/slabs
+    - cp tests/data/slab_data/*.grd /home/usgs-user/slabs/.
+    - cd tests/regression
+    - python regression_test.py --parameters ci.yml
+  stage: test

{gmprocess-2.7.0 → gmprocess-2.8.0}/CHANGELOG.md

@@ -1,5 +1,41 @@
 ## main
 
+## 2.8.0 / 2026-08-27
+
+- Enhancements
+  - Configure pytest to track subprocesses.
+  - Regression test with ci.yml config and nn00919749 event for CI monthly pipeline schedule
+  - Add extended StationXML parser.
+  - Increase use of pytest fixtures in an attempt to speed up unit tests.
+  - Vectorize the std_dev clipping check.
+  - Optimized the PRISM adaptive baseline correction method.
+  - Undo a the use of LazyLoader for situations where it is not necessary.
+  - Stop using SciPy for physical constants.
+  - Update tests to copy data into tmp_path, making tests independent and parallel-safe.
+  - Added FDSNFetcher tests.
+  - Added validation check to StationStream to look for traces with all identical values.
+  - Fixed bug in knet fetcher related to directory names in downloaded zip files.
+  - Fixed bug in check_tail when the trace ends before the window.
+- Dependencies
+  - Removed esi-utils-rupture and replaced with esi-shakelib (min version 1.2.1)
+  - numpy>=2.0
+  - ps2ff>=1.5.9
+  - Added pytest-xdist to test dependencies; set addopts = "-n auto" in pyproject.toml to enable.
+- Bugfix
+  - Fixed a crash in the SCEDC cloud channel query for windows spanning midnight UTC.
+  - Cloud download windows crossing a month or year boundary were treated as a single UTC day.
+  - The NCEDC cloud channel query returned a dataframe with no columns when the query failed.
+  - handle numpy integer and float in stream_workspace \_serializer
+  - FDSNFetcher was not handling the international date line crossing correctly and skipping GeoNet data.
+  - Removed the usage of StreamCollection objects in the global fetcher, as they are not used and were not
+    consistently returned by all fetcher classes.
+  - Fixing bug in knet reader where both horizontal channels were assigned an azimuth of 90.0.
+  - Fixing bug in knet fetcher where it was not handling multiple candidate events.
+  - Changed output units of SAC files in export_ff_sac.py to be m/s^2, since that is what FFM expects.
+- Other
+  - Remove test for nnet_qa step, issue warning that it is on longer tested, and will soon be removed; this is because I don't think anyone uses it and it is a lot of code and the unit tests take a long time.
+- CI/Pipelines
+  - Fix issue with unscheduled (unit tests) not running automatically due to scheduled regression test
 
 ## 2.7.0 / 2026-06-04
 

{gmprocess-2.7.0 → gmprocess-2.8.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: gmprocess
-Version: 2.7.0
+Version: 2.8.0
 Summary: USGS Automated Ground Motion Processing Software
 Author-email: Eric Thompson <emthompsone@usgs.gov>, Mike Hearne <mhearne@usgs.gov>, Brad Aagaard <baagaard@usgs.gov>, Bruce Worden <cbworden@contractor.usgs.gov>, John Rekoske <jrekoske@ucsd.edu>, Heather Hunsinger <hhunsinger@usgs.gov>, Gabe Ferragut <gferragut@usgs.gov>
 License: License
@@ -57,7 +57,7 @@ Requires-Dist: configobj>=5.0.6
 Requires-Dist: esi-core>=1.2.1
 Requires-Dist: esi-utils-colors>1.0
 Requires-Dist: esi-utils-io>1.0
-Requires-Dist: esi-utils-rupture<=1.0.4,>1.0
+Requires-Dist: esi-shakelib>=1.2.1
 Requires-Dist: folium>=0.12
 Requires-Dist: gmpacket>=0.1.6
 Requires-Dist: h3
@@ -65,13 +65,13 @@ Requires-Dist: h5py>=3.11
 Requires-Dist: lxml>=4.6.1
 Requires-Dist: matplotlib>=3.1.0
 Requires-Dist: numba>=0.58
-Requires-Dist: numpy>=1.26
+Requires-Dist: numpy<2.4,>=2.0
 Requires-Dist: obspy>=1.5.0
 Requires-Dist: openpyxl>=3.1.0
 Requires-Dist: openquake.engine>=3.18
 Requires-Dist: pandas>=1.3.0
 Requires-Dist: Pint>0.24.3
-Requires-Dist: ps2ff>=1.5.2
+Requires-Dist: ps2ff>=1.5.9
 Requires-Dist: pyarrow
 Requires-Dist: pyasdf>=0.7
 Requires-Dist: pydantic[email]>=2.0
@@ -93,10 +93,13 @@ Requires-Dist: flake8>=3.9; extra == "dev"
 Requires-Dist: ipython>=7.26; extra == "dev"
 Requires-Dist: pylint>=2.17; extra == "dev"
 Provides-Extra: test
+Requires-Dist: pytest-xdist>=3.8; extra == "test"
 Requires-Dist: pytest>=6.2; extra == "test"
 Requires-Dist: pytest-cov>=2.12; extra == "test"
 Requires-Dist: pytest-console-scripts>=1.4; extra == "test"
+Requires-Dist: pytest-subprocess>=1.6.0; extra == "test"
 Requires-Dist: vcrpy>=4.1; extra == "test"
+Requires-Dist: polars; extra == "test"
 Provides-Extra: doc
 Requires-Dist: altair>=4.2.2; extra == "doc"
 Requires-Dist: furo>=2023.3.27; extra == "doc"
@@ -129,10 +132,13 @@ Most of the extensions that we provide are to handle strong motion data and rela
 # Getting Help
 The developers can be reached at [gmprocess@usgs.gov](mailto:gmprocess@usgs.gov).
 
-If you would like to post an issue to ask a question, request a new feature, or report a bug, you can do so at the [github mirror](https://github.com/DOI-USGS/ghsc-esi-groundmotion-processing) of this repository. 
+If you would like to post an issue, ask a question, request a new feature, or report a bug, you can do so at here in 
+[Work Items](https://code.usgs.gov/ghsc/esi/groundmotion-processing/-/work_items). 
+This GitLab instance, however, is restricted to registered users, and registering a user is not immediate or automatic, requiring an internal approval process.
+Posting issues is easier on GitHub, and the developers of this repository monitor the
+[groundmotion-processing-github](https://github.com/DOI-USGS/groundmotion-processing-github) repository's issues to alleviate need to deal with the code.usgs.gov restrictions. 
 
-If you would like to contribute merge requests to this repository, please let us know and we can add you as a collaborator. 
-This will require that you create an account at code.usgs.gov, which is a more involved process than creating a personal account at github.com and posting to the issues on the github mirror.
+If you would like to contribute merge requests to this repository, please let us know and we can assist you in registering an account at [code.usgs.gov](code.usgs.gov). 
 
 # Suggested Citation
 If you wish to cite this software please cite:

{gmprocess-2.7.0 → gmprocess-2.8.0}/README.md

@@ -12,10 +12,13 @@ Most of the extensions that we provide are to handle strong motion data and rela
 # Getting Help
 The developers can be reached at [gmprocess@usgs.gov](mailto:gmprocess@usgs.gov).
 
-If you would like to post an issue to ask a question, request a new feature, or report a bug, you can do so at the [github mirror](https://github.com/DOI-USGS/ghsc-esi-groundmotion-processing) of this repository. 
+If you would like to post an issue, ask a question, request a new feature, or report a bug, you can do so at here in 
+[Work Items](https://code.usgs.gov/ghsc/esi/groundmotion-processing/-/work_items). 
+This GitLab instance, however, is restricted to registered users, and registering a user is not immediate or automatic, requiring an internal approval process.
+Posting issues is easier on GitHub, and the developers of this repository monitor the
+[groundmotion-processing-github](https://github.com/DOI-USGS/groundmotion-processing-github) repository's issues to alleviate need to deal with the code.usgs.gov restrictions. 
 
-If you would like to contribute merge requests to this repository, please let us know and we can add you as a collaborator. 
-This will require that you create an account at code.usgs.gov, which is a more involved process than creating a personal account at github.com and posting to the issues on the github mirror.
+If you would like to contribute merge requests to this repository, please let us know and we can assist you in registering an account at [code.usgs.gov](code.usgs.gov). 
 
 # Suggested Citation
 If you wish to cite this software please cite:

{gmprocess-2.7.0 → gmprocess-2.8.0}/pyproject.toml

@@ -23,7 +23,7 @@ dependencies = [
     "esi-core>=1.2.1",
     "esi-utils-colors>1.0",
     "esi-utils-io>1.0",
-    "esi-utils-rupture>1.0,<=1.0.4",
+    "esi-shakelib>=1.2.1",
     "folium>=0.12",
     "gmpacket>=0.1.6",
     "h3",
@@ -31,13 +31,13 @@ dependencies = [
     "lxml>=4.6.1",
     "matplotlib>=3.1.0",
     "numba>=0.58",
-    "numpy>=1.26",
+    "numpy>=2.0,<2.4",
     "obspy>=1.5.0",
     "openpyxl>=3.1.0",
     "openquake.engine>=3.18",
     "pandas>=1.3.0",
     "Pint>0.24.3",
-    "ps2ff>=1.5.2",
+    "ps2ff>=1.5.9",
     "pyarrow",
     "pyasdf>=0.7",
     "pydantic[email]>=2.0",
@@ -63,10 +63,13 @@ dev = [
     "pylint>=2.17",
 ]
 test = [
+    "pytest-xdist>=3.8",
     "pytest>=6.2",
     "pytest-cov>=2.12",
     "pytest-console-scripts>=1.4",
+    "pytest-subprocess>=1.6.0",
     "vcrpy>=4.1",
+    "polars",
 ]
 doc = [
     "altair>=4.2.2",
@@ -119,4 +122,3 @@ include = ["gmprocess*"]
 exclude = ["tests*"]
 
 [tool.setuptools_scm]
-

gmprocess-2.8.0/pytest.ini

@@ -0,0 +1,7 @@
+[pytest]
+addopts = 
+    --ignore=src/gmprocess/gui/gui_ucla/
+    -n auto
+norecursedirs = tests/regression
+filterwarnings =
+    ignore::DeprecationWarning

{gmprocess-2.7.0 → gmprocess-2.8.0}/src/gmprocess/apps/gmrecords.py

@@ -8,16 +8,15 @@ import inspect
 import io
 import logging
 import os
-from pathlib import Path
 import pkgutil
 import shutil
 import sys
 import traceback
+from pathlib import Path
 
+from gmprocess.subcommands.lazy_loader import LazyLoader
 from gmprocess.subcommands.projects import Project
 
-from ..subcommands.lazy_loader import LazyLoader
-
 VERSION = importlib.metadata.version("gmprocess")
 
 configobj = LazyLoader("configobj", globals(), "configobj")
@@ -78,7 +77,9 @@ class GMrecordsApp(object):
 
     def main(self, **kwargs):
         self.args = (
-            argparse.Namespace(**kwargs) if kwargs else self._parse_command_line()
+            argparse.Namespace(**kwargs)
+            if kwargs
+            else self._parse_command_line()
         )
         if self.args.subcommand is None:
             self.parser.print_help()
@@ -86,7 +87,9 @@ class GMrecordsApp(object):
         else:
             if self.args.datadir is None and self.args.confdir is None:
                 self.use_projects = True
-            elif self.args.datadir is not None and self.args.confdir is not None:
+            elif (
+                self.args.datadir is not None and self.args.confdir is not None
+            ):
                 self.use_projects = False
             else:
                 raise ValueError(
@@ -100,10 +103,15 @@ class GMrecordsApp(object):
                 self._initialize()
 
             subcmds_quiet = ["init", "projects", "proj", "processing_steps"]
-            if self.args.subcommand not in subcmds_quiet and not self.args.quiet:
+            if (
+                self.args.subcommand not in subcmds_quiet
+                and not self.args.quiet
+            ):
                 if self.use_projects:
                     # Print the current project information to avoid confusion
-                    proj = Project.from_config(self.projects_conf, self.project_name)
+                    proj = Project.from_config(
+                        self.projects_conf, self.project_name
+                    )
                 else:
                     proj_paths = {
                         "conf_path": self.args.confdir,
@@ -143,7 +151,9 @@ class GMrecordsApp(object):
         mod = importlib.import_module("gmprocess.subcommands")
         subcommands = {
             name: importlib.import_module(name)
-            for _, name, _ in pkgutil.iter_modules(mod.__path__, mod.__name__ + ".")
+            for _, name, _ in pkgutil.iter_modules(
+                mod.__path__, mod.__name__ + "."
+            )
         }
         self.classes = {}
         for name, module in subcommands.items():
@@ -179,14 +189,18 @@ class GMrecordsApp(object):
     @property
     def conf_path(self):
         if self.projects_path:
-            return (self.projects_path / self.current_project["conf_path"]).resolve()
+            return (
+                self.projects_path / self.current_project["conf_path"]
+            ).resolve()
         else:
             return Path(self.current_project["conf_path"]).resolve()
 
     @property
     def data_path(self):
         if self.projects_path:
-            return (self.projects_path / self.current_project["data_path"]).resolve()
+            return (
+                self.projects_path / self.current_project["data_path"]
+            ).resolve()
         else:
             return Path(self.current_project["data_path"]).resolve()
 
@@ -200,10 +214,15 @@ class GMrecordsApp(object):
 
         # get the email error notification portion of config, if present
         error_dict = {}
-        if self.conf is not None and self.conf["error_notification"] is not None:
+        if (
+            self.conf is not None
+            and self.conf["error_notification"] is not None
+        ):
             error_dict = dict(self.conf["error_notification"])
 
-        log_utils.setup_logger(self.args, log_file=log_file, error_dict=error_dict)
+        log_utils.setup_logger(
+            self.args, log_file=log_file, error_dict=error_dict
+        )
         logging.info("Logging level includes INFO.")
         logging.debug("Logging level includes DEBUG.")
         logging.info(f"PROJECTS_PATH: {self.projects_path}")
@@ -255,7 +274,9 @@ class GMrecordsApp(object):
                 str(self.projects_file), encoding="utf-8"
             )
 
-        projmod.validate_projects_config(self.projects_conf, self.projects_path)
+        projmod.validate_projects_config(
+            self.projects_conf, self.projects_path
+        )
 
         if "CALLED_FROM_PYTEST" in os.environ:
             # Check if we are running unit tests
@@ -270,12 +291,19 @@ class GMrecordsApp(object):
 
             # Otherwise use the default testing config
             else:
-                test_conf_file = (const.DATA_DIR / const.CONFIG_FILE_TEST).resolve()
+                test_conf_file = (
+                    const.DATA_DIR / const.CONFIG_FILE_TEST
+                ).resolve()
 
             # Copy into the tmp gmptest project directory
             shutil.copyfile(test_conf_file, conf_path / const.CONFIG_FILE_TEST)
 
-        subcommands_need_conf = ["download", "assemble", "autoshakemap", "autoprocess"]
+        subcommands_need_conf = [
+            "download",
+            "assemble",
+            "autoshakemap",
+            "autoprocess",
+        ]
         if self.args.func.command_name in subcommands_need_conf:
             self.conf = configmod.get_config(config_path=self.conf_path)
 
@@ -415,7 +443,9 @@ class GMrecordsApp(object):
 
         # Parsers for subcommands
         subparsers = parser.add_subparsers(
-            title="Subcommands", dest="subcommand", metavar="<command> (<aliases>)"
+            title="Subcommands",
+            dest="subcommand",
+            metavar="<command> (<aliases>)",
         )
 
         # Get subcommands and their arguments
@@ -450,3 +480,8 @@ class GMrecordsApp(object):
                 parsers[-1].add_argument(*pargs, **targ_dict)
             parsers[-1].set_defaults(func=cdict["class"])
         return parser.parse_args()
+
+
+if __name__ == "__main__":
+    app = GMrecordsApp()
+    app.main()

{gmprocess-2.7.0 → gmprocess-2.8.0}/src/gmprocess/bin/export_ff_sac.py

@@ -5,9 +5,12 @@ from pathlib import Path
 
 import matplotlib.pyplot as plt
 import numpy as np
-from gmprocess.io.asdf.stream_workspace import StreamWorkspace
 from obspy.geodetics import gps2dist_azimuth
 
+from gmprocess.io.asdf.stream_workspace import StreamWorkspace
+
+GALS_TO_MPS2 = 1 / 100.0  # convert from gals to m/s^2
+
 
 def get_parser():
     parser = argparse.ArgumentParser(
@@ -199,6 +202,9 @@ def write_sac_files(streams, passing_indices, id_selected, out_dir):
     for id_sel in id_selected:
         stream = streams[passing_indices[id_sel]]
         for trace in stream:
+            # convert to m/s^2
+            trace.data = trace.data * GALS_TO_MPS2
+            trace.stats.standard["units"] = "m/s^2"
             trace.stats.sac = make_sac_headers(
                 trace.stats.channel,
                 trace.stats.coordinates.latitude,
@@ -223,6 +229,9 @@ def write_supp_sac_files(sc, sta_ids, supp_dir):
     for stream in sc:
         if stream.passed and stream[0].stats.station in sta_ids:
             for trace in stream:
+                # convert to m/s^2
+                trace.data = trace.data * GALS_TO_MPS2
+                trace.stats.standard["units"] = "m/s^2"
                 trace.stats.sac = make_sac_headers(
                     trace.stats.channel,
                     trace.stats.coordinates.latitude,
@@ -262,7 +271,11 @@ def save_summary_plot(
     ax.set_xlabel("Longitude")
     ax.set_ylabel("Latitude")
     ax.legend()
-    fname = Path(outdir) / f"{eqid}_select_summary.png"
+    foutdir = Path(outdir)
+    if not foutdir.exists():
+        foutdir.mkdir(parents=True)
+    fname = foutdir / f"{eqid}_select_summary.png"
+
     fig.savefig(fname, dpi=300)
     plt.close(fig)
     print(f"Wrote summary plot: {fname}")
@@ -274,6 +287,7 @@ def main():
 
     sw = StreamWorkspace.open(args.workspace)
     sc = sw.get_streams(args.event_id, labels=["default"])
+    print(f"{len(sc)} streams found.")
     event = sw.get_event(args.event_id)
 
     lat, lon, dist, pgv, passing_indices = collect_stream_data(sc, event)

{gmprocess-2.7.0 → gmprocess-2.8.0}/src/gmprocess/bin/get_cali_cloud.py

@@ -74,6 +74,17 @@ SCEDC_INDEX_FILE_TEMPLATE = (
     "year_doy={year}_{jday}/{year}_{jday}_waveform_index.parquet"
 )
 
+CHANNEL_COLUMNS = [
+    "network",
+    "station",
+    "location",
+    "channel",
+    "latitude",
+    "longitude",
+    "starttime",
+    "endtime",
+]
+
 WAVEFORM_TIMEFMT = "%Y%m%dT%H%M%SZ"
 COMCAT_TIMEFMT = "%Y-%m-%dT%H:%M:%S"
 
@@ -216,7 +227,7 @@ def to_standard_waveform(provider, key_filename):
     if provider == "ncedc":
         parts = key_filename.split(".")
         station, network, channel, location = parts[0:4]
-        jday = parts[-2]
+        jday = parts[-1]
     else:
         network = key_filename[0:2]
         station = key_filename[2:7].rstrip("_")
@@ -323,25 +334,27 @@ def get_ncedc_channel_dataframe(
             level="channel",
         )
     except Exception as e:
-        channel_dataframe = pd.DataFrame([])
-        return channel_dataframe
+        logging.warning(f"Failed to download NCEDC station metadata: '{e}'")
+        # Return an empty frame that still has the expected columns, so that
+        # downstream column access does not raise.
+        return pd.DataFrame(columns=CHANNEL_COLUMNS)
     channel_rows = []
-    for network in inventory.networks:
-        for station in network.stations:
-            for channel in station.channels:
+    for inv_network in inventory.networks:
+        for inv_station in inv_network.stations:
+            for inv_channel in inv_station.channels:
                 channel_row = {
-                    "network": network.code,
-                    "station": station.code,
-                    "location": channel.location_code,
-                    "channel": channel.code,
-                    "latitude": channel.latitude,
-                    "longitude": channel.longitude,
+                    "network": inv_network.code,
+                    "station": inv_station.code,
+                    "location": inv_channel.location_code,
+                    "channel": inv_channel.code,
+                    "latitude": inv_channel.latitude,
+                    "longitude": inv_channel.longitude,
                     "starttime": starttime,
                     "endtime": endtime,
                 }
                 channel_rows.append(channel_row)
 
-    channel_dataframe = pd.DataFrame(data=channel_rows)
+    channel_dataframe = pd.DataFrame(data=channel_rows, columns=CHANNEL_COLUMNS)
     # restrict the networks in this dataframe to only those that
     # NC supports
     # supported_networks = get_provider_networks("ncedc")
@@ -376,6 +389,22 @@ def get_provider_networks(provider):
     return supported_networks
 
 
+def spans_multiple_days(starttime, endtime):
+    """Determine whether a time window crosses a UTC day boundary.
+
+    Comparing the day of the month alone would treat windows that cross a month
+    or year boundary as falling on a single day.
+
+    Args:
+        starttime (UTCDateTime): Start of the window.
+        endtime (UTCDateTime): End of the window.
+
+    Returns:
+        bool: True if the window spans more than one UTC day.
+    """
+    return (starttime.year, starttime.julday) != (endtime.year, endtime.julday)
+
+
 def get_scedc_channel_dataframe(
     network,
     station,
@@ -400,17 +429,18 @@ def get_scedc_channel_dataframe(
     bucket = PROVIDERS["scedc"]["bucket"]
     s3_client.download_fileobj(bucket, index_file1, fobj)
     dataframe = pd.read_parquet(fobj)
-    if endtime.day != starttime.day:
+    if spans_multiple_days(starttime, endtime):
         index_file2 = SCEDC_INDEX_FILE_TEMPLATE.format(
             year=endtime.year, jday=endtime.strftime("%j")
         )
         fobj = BytesIO()
         s3_client.download_fileobj(bucket, index_file2, fobj)
         dataframe2 = pd.read_parquet(fobj)
-        dataframe = pd.concat([dataframe, dataframe2])
+        dataframe = pd.concat([dataframe, dataframe2], ignore_index=True)
         dataframe.drop_duplicates(
             subset=["net", "sta", "location", "seedchan"],
             inplace=True,
+            ignore_index=True,
         )
     channel_dataframe = filter_scindex_dataframe(
         dataframe,
@@ -516,12 +546,12 @@ def get_channels_regex(oldchannel):
 
 
 def filter_by_channels(dataframe, channels):
-    indices = []
+    # Accumulate the match mask positionally. Building a DataFrame from the
+    # per-pattern boolean Series would force an index union, which fails when
+    # the caller's frame has duplicate labels (e.g. concatenated daily indices).
+    newindex = np.zeros(len(dataframe), dtype=bool)
     for channel in channels:
-        index = dataframe["seedchan"].str.contains(channel)
-        indices.append(index)
-    iframe = pd.DataFrame(indices)
-    newindex = iframe.any(axis=0)
+        newindex |= dataframe["seedchan"].str.contains(channel).to_numpy(dtype=bool)
     newframe = dataframe[newindex].copy()
     return newframe
 
@@ -574,7 +604,7 @@ def filter_scindex_dataframe(
         and maxlongitude is not None
     )
     if has_radius:
-        dataframe["distance_deg"] = vincenty(
+        dataframe["distance_deg"] = get_distance(
             latitude, longitude, dataframe["lat"], dataframe["lon"]
         )
         if not has_center:
@@ -638,7 +668,7 @@ def get_waveform_jobs(provider, dataframe, directory):
         )
         job = {"provider": provider, "key": key1, "directory": directory}
         jobs.append(job)
-        if endtime.day != starttime.day:
+        if spans_multiple_days(starttime, endtime):
             key2 = format_waveform_key(
                 provider, endtime, network, station, location, channel
             )
@@ -669,7 +699,8 @@ def get_metadata_jobs(provider, dataframe, directory):
         else:
             unmatched_rows.append(group.iloc[0])
 
-    unmatched = pd.DataFrame(data=unmatched_rows)
+    # The rows carry their source labels, which are not guaranteed to be unique.
+    unmatched = pd.DataFrame(data=unmatched_rows).reset_index(drop=True)
     return (jobs, unmatched)
 
 
@@ -768,7 +799,7 @@ def trim_waveform(job):
     filename.unlink()
 
 
-def vincenty(lat1, lon1, lat2, lon2):
+def get_distance(lat1, lon1, lat2, lon2):
     lmb1 = np.deg2rad(lon1)
     lmb2 = np.deg2rad(lon2)
     phi1 = np.deg2rad(lat1)

{gmprocess-2.7.0 → gmprocess-2.8.0}/src/gmprocess/core/scalar_event.py

@@ -20,7 +20,7 @@ class ScalarEvent(Event):
     Note that depth in obspy's Origin object is in meters, but we use km.
 
     Also, be careful because we have a similar object with the same name:
-    esi_utils_rupture.origin.Origin.
+    esi_shakelib.rupture.origin.Origin.
 
     """