New config option event_data_path

To specify the path to a local directory with waveform files organized
per event.

Author: claudiodsf

CHANGELOG.md

@@ -18,7 +18,9 @@ to the latest version.
   attribute specified in `--colorby` is used.
 - New option `--range` to manually specify the range of values for the color
   scale
-- Option `waveform_data_path` renamed to `sds_data_path`
+- Config option `waveform_data_path` renamed to `sds_data_path`
+- New config option `event_data_path` to specify the path to a local directory
+  with waveform files organized per event
 - Add missing `street` map style for `map_families`
 - Initial support for plotting families found with template scan
 - Improved reading of CSV catalog files:

requake/config/configspec.conf

@@ -17,6 +17,11 @@ fdsn_dataselect_url = string(default=http://example.com)
 ## Alternatively, you can provide the path to a local SDS waveform archive
 ## (see https://docs.obspy.org/packages/autogen/obspy.clients.filesystem.sds.html)
 sds_data_path = string(default=None)
+## Or, specify a local directory with waveform files organized per event
+## This directory should contain subdirectories named after the eventid, each
+## containing waveform files in any format supported by ObsPy
+## (e.g., SAC, miniSEED). The eventid should be the same as in the catalog.
+event_data_path = string(default=None)
 
 #### Catalog-based scan
 ### The following parameters are for a catalog-based scan:
@@ -88,9 +93,10 @@ time_chunk_overlap = float(default=60)
 min_cc_mad_ratio = float(default=50)
 
 #### Processing parameters
-## Seconds before P arrival to cut trace (only used for catalog search)
+## Seconds before P arrival to cut trace
+## (only used for catalog search and continuous data)
 cc_pre_P = float(default=5)
-## Trace length in seconds (only used for catalog search)
+## Trace length in seconds (only used for catalog search and continuous data)
 cc_trace_length = float(default=120)
 ## Frequency range for computing cross-correlation (Hz)
 cc_freq_min = float(default=2)

requake/config/rq_setup.py

@@ -160,14 +160,16 @@ def _connect_station_dataselect():
         logger.info(
             f'Connected to FDSN station server: {config.fdsn_station_url}'
         )
+    config.dataselect_client = None
     if config.sds_data_path is not None:
         _connect_sds()
-    else:
-        config.dataselect_client = FDSNClient(config.fdsn_dataselect_url)
-        logger.info(
-            'Connected to FDSN dataselect server: '
-            f'{config.fdsn_dataselect_url}'
-        )
+    if config.event_data_path is not None:
+        return
+    config.dataselect_client = FDSNClient(config.fdsn_dataselect_url)
+    logger.info(
+        'Connected to FDSN dataselect server: '
+        f'{config.fdsn_dataselect_url}'
+    )
 
 
 def _connect_sds():

requake/config/utils.py

@@ -195,36 +195,32 @@ def manage_uncaught_exception(exception):
     import scipy as sp
     import obspy
     import matplotlib
-    sys.stderr.write("""
-# BEGIN TRACEBACK #############################################################
-""")
-    sys.stderr.write('\n')
-    traceback.print_exc()
-    sys.stderr.write("""
-# END TRACEBACK ###############################################################
-""")
-    sys.stderr.write("""
-
-Congratulations, you've found a bug in Requake! 🐞
-
-Please report it on https://github.com/SeismicSource/requake/issues
-or by email to [email protected].
-
-Include the following information in your report:
-
-""")
-    sys.stderr.write(f'  Requake version: {__version__}\n')
-    sys.stderr.write(f'  Python version: {sys.version}\n')
-    sys.stderr.write(f'  NumPy version: {np.__version__}\n')
-    sys.stderr.write(f'  SciPy version: {sp.__version__}\n')
-    sys.stderr.write(f'  ObsPy version: {obspy.__version__}\n')
-    sys.stderr.write(f'  Matplotlib version: {matplotlib.__version__}\n')
-    sys.stderr.write(f'  Platform: {sys.platform}\n')
-    sys.stderr.write(f'  Command line: {" ".join(sys.argv)}\n')
-    sys.stderr.write(f'  Error message: {str(exception)}\n')
-    sys.stderr.write('\n')
+    import pygments
+    from pygments.lexers.python import PythonLexer
+    from pygments.formatters.terminal import TerminalFormatter
+    tb_str = ''.join(
+        traceback.format_exception(
+            type(exception), exception, exception.__traceback__))
+    colored_tb = pygments.highlight(tb_str, PythonLexer(), TerminalFormatter())
     sys.stderr.write(
-        'Also, please copy and paste the traceback above in your '
-        'report.\n\n')
-    sys.stderr.write('Thank you for your help!\n\n')
+        f'\n# BEGIN TRACEBACK {"#" * 62}\n'
+        f'\n{colored_tb}'
+        f'\n# END TRACEBACK {"#" * 62}\n\n'
+        "Congratulations, you've found a bug in Requake! 🐞\n\n"
+        'Please report it on https://github.com/SeismicSource/requake/issues\n'
+        'or by email to [email protected].\n\n'
+        'Include the following information in your report:\n\n'
+        f'  Requake version: {__version__}\n'
+        f'  Python version: {sys.version}\n'
+        f'  NumPy version: {np.__version__}\n'
+        f'  SciPy version: {sp.__version__}\n'
+        f'  ObsPy version: {obspy.__version__}\n'
+        f'  Matplotlib version: {matplotlib.__version__}\n'
+        f'  Platform: {sys.platform}\n'
+        f'  Command line: {" ".join(sys.argv)}\n'
+        f'  Error message: {str(exception)}\n'
+        '\n'
+        'Also, please copy and paste the traceback above in your report.\n\n'
+        'Thank you for your help!\n\n'
+    )
     sys.exit(1)

requake/scan/scan_templates.py

@@ -19,7 +19,7 @@
     FamilyNotFoundError
 )
 from ..waveforms import (
-    get_waveform, cc_waveform_pair, get_arrivals,
+    get_waveform_from_client, cc_waveform_pair, get_arrivals,
     NoWaveformError
 )
 from ..catalog import RequakeEvent, generate_evid
@@ -78,7 +78,7 @@ def _scan_family_template(template, catalog_file, t0, t1):
         tr = trace_cache[key]
     except KeyError:
         try:
-            tr = get_waveform(trace_id, t0, t1)
+            tr = get_waveform_from_client(trace_id, t0, t1)
             trace_cache[key] = tr
         except NoWaveformError as err:
             raise NoWaveformError(

requake/waveforms/__init__.py

@@ -10,7 +10,7 @@
     (https://www.gnu.org/licenses/gpl-3.0-standalone.html)
 """
 from .waveforms import (  # noqa
-    get_waveform, get_event_waveform,
+    get_waveform_from_client, get_event_waveform,
     get_waveform_pair, cc_waveform_pair,
     process_waveforms,
     align_pair, align_traces,

requake/waveforms/station_metadata.py

@@ -160,7 +160,7 @@ def _download_metadata():
     )
 
 
-def get_traceid_coords(orig_time=None):
+def get_traceid_coords_old(orig_time=None):
     """
     Get coordinates for the trace_ids specified in config file.
 
@@ -177,13 +177,85 @@ def get_traceid_coords(orig_time=None):
         _download_metadata()
     traceid_coords = {}
     for trace_id in config.catalog_trace_id:
+        net, sta, loc, chan = trace_id.split('.')
+        net = net or '@@'
+        _trace_id = f'{net}.{sta}.{loc}.{chan}'
         try:
-            coords = config.inventory.get_coordinates(trace_id, orig_time)
-        except Exception as err:
+            coords = config.inventory.get_coordinates(_trace_id, orig_time)
+        # pylint: disable=broad-except
+        except Exception:
             # note: get_coordinaets raises a generic Exception
+            # try again with empty channel code
+            try:
+                _trace_id = f'{net}.{sta}.{loc}.'
+                coords = config.inventory.get_coordinates(_trace_id, orig_time)
+            except Exception as err:
+                raise MetadataMismatchError(
+                    f'Unable to find coordinates for trace {trace_id} '
+                    f'at time {orig_time}'
+                ) from err
+        traceid_coords[trace_id] = coords
+    return traceid_coords
+
+
+def _fetch_coordinates(net, sta, loc, chan, orig_time):
+    """
+    Fetch coordinates for a trace_id at a given time.
+
+    Attempt to retrieve coordinates, first with full trace_id,
+    then without channel.
+
+    :param net: network code
+    :type net: str
+    :param sta: station code
+    :type sta: str
+    :param loc: location code
+    :type loc: str
+    :param chan: channel code
+    :type chan: str
+    :param orig_time: origin time
+    :type orig_time: obspy.UTCDateTime
+
+    :return: coordinates or None
+    :rtype: dict
+    """
+    for ch in [chan, '']:
+        _trace_id = f'{net}.{sta}.{loc}.{ch}'
+        try:
+            return config.inventory.get_coordinates(_trace_id, orig_time)
+        # pylint: disable=broad-except
+        # note: get_coordinaets raises a generic Exception
+        except Exception:
+            # Try next option if it fails
+            continue
+    # If both attempts fail, return None
+    return None
+
+
+def get_traceid_coords(orig_time=None):
+    """
+    Get coordinates for the trace_ids specified in config file.
+
+    :param orig_time: origin time
+    :type orig_time: obspy.UTCDateTime
+    :return: a dictionary with trace_id as key and coordinates as value
+    :rtype: dict
+
+    :raises MetadataMismatchError: if coordinates are not found
+    """
+    if config.inventory is None:
+        _read_station_metadata()
+    if config.inventory is None:
+        _download_metadata()
+    traceid_coords = {}
+    for trace_id in config.catalog_trace_id:
+        net, sta, loc, chan = trace_id.split('.')
+        net = net or '@@'
+        coords = _fetch_coordinates(net, sta, loc, chan, orig_time)
+        if coords is None:
             raise MetadataMismatchError(
                 f'Unable to find coordinates for trace {trace_id} '
                 f'at time {orig_time}'
-            ) from err
+            )
         traceid_coords[trace_id] = coords
     return traceid_coords