Merge pull request #607 from nu-radio/hotfix/readRNOGDataMattak-file-…

…support

Fix support for relative path descriptions to root files in readRNOGDataMattak

NuRadioReco/modules/RNO_G/channelBlockOffsetFitter.py

@@ -217,7 +217,7 @@ def fit_block_offsets(
     # obtain guesses for block offsets
     a_guess = np.mean(np.split(filtered_trace, n_blocks), axis=1)
     if mode == 'approximate':
-        block_offsets = a_guess
+        block_offsets = a_guess + np.mean(trace)
     elif mode == 'fit':
         # self._offset_guess[channel_id] = a_guess
         # we can get rid of one parameter through a global shift

NuRadioReco/modules/io/RNO_G/readRNOGDataMattak.py

@@ -22,7 +22,7 @@
 import random
 
 
-def create_random_directory_path(prefix="/tmp/", n=7):
+def _create_random_directory_path(prefix="/tmp/", n=7):
     """
     Produces a path for a temporary directory with a n letter random suffix
 
@@ -48,7 +48,7 @@ def create_random_directory_path(prefix="/tmp/", n=7):
     return path
 
 
-def baseline_correction(wfs, n_bins=128, func=np.median, return_offsets=False):
+def _baseline_correction(wfs, n_bins=128, func=np.median, return_offsets=False):
     """
     Simple baseline correction function. 
     
@@ -150,7 +150,7 @@ def get_time_offset(trigger_type):
         raise KeyError(f"Unknown trigger type: {trigger_type}. Known are: {known_trigger_types}. Abort ....")
 
 
-def all_files_in_directory(mattak_dir):
+def _all_files_in_directory(mattak_dir):
     """
     Checks if all Mattak root files are in a directory.
     Ignoring runinfo.root because (asaik) not all runs have those and information is currently not read by Mattak.
@@ -186,18 +186,41 @@ class readRNOGData:
 
     def __init__(self, run_table_path=None, load_run_table=True, log_level=logging.INFO):
         """
+        Reader for RNO-G ``.root`` files
+
+        This class provides read access to RNO-G ``.root`` files and converts them
+        to NuRadioMC :class:`Events <NuRadioReco.framework.event.Event>`. Requires ``mattak``
+        (https://github.com/RNO-G/mattak) to be installed.
+
         Parameters
         ----------
 
         run_table_path: str | None
-            Path to a run_table.cvs file. If None, the run table is queried from the DB. (Default: None)
+            Path to a run_table.csv file. If None, the run table is queried from the DB. (Default: None)
 
         load_run_table: bool
             If True, try to load the run_table from run_table_path. Otherwise, skip this.
 
         log_level: enum
             Set verbosity level of logger. If logging.DEBUG, set mattak to verbose (unless specified in mattak_kwargs).
             (Default: logging.INFO)
+
+        Examples
+        --------
+
+        .. code-block::
+
+            reader = readRNOGDataMattak.readRNOGData() # initialize reader
+            reader.begin('/path/to/root_file_or_folder')
+
+            evt = reader.get_event_by_index(0) # returns the first event in the file
+            # OR
+            evt = reader.get_event(run_nr=1100, event_id=679) # returns the event with run_number 1100 and event_id 679
+            # OR
+            for evt in reader.run(): # loop over all events in file
+                # perform some analysis
+                pass
+
         """
         self.logger = logging.getLogger('NuRadioReco.readRNOGData')
         self.logger.setLevel(log_level)
@@ -222,6 +245,18 @@ def __init__(self, run_table_path=None, load_run_table=True, log_level=logging.I
                     self.logger.warn("Import of run table failed. Runs can not be filtered.! \n"
                             "You can get the interface from GitHub: [email protected]:RNO-G/rnog-runtable.git")
             else:
+                # some users may mistakenly try to pass the .root files to __init__
+                # we check for this and raise a (hopefully) helpful error message
+                user_passed_root_file_msg = (
+                    "The optional argument run_table_path expects a csv file, "
+                    "but you passed a list of files or a .root file. Note that "
+                    "the .root files to read in should be passed to the `begin` method of this class"
+                )
+                if isinstance(run_table_path, (list, np.ndarray)):
+                    raise TypeError(user_passed_root_file_msg)
+                elif os.path.isdir(run_table_path) or run_table_path.endswith('.root'):
+                    raise ValueError(user_passed_root_file_msg)
+
                 import pandas
                 self.__run_table = pandas.read_csv(run_table_path)
 
@@ -353,14 +388,14 @@ def begin(self,
         self._datasets = []
         self.__n_events_per_dataset = []
 
+        if not isinstance(dirs_files, (list, np.ndarray)):
+            dirs_files = [dirs_files]
+
         self.logger.info(f"Parse through / read-in {len(dirs_files)} directory(ies) / file(s).")
 
         self.__skipped_runs = 0
         self.__n_runs = 0
 
-        if not isinstance(dirs_files, (list, np.ndarray)):
-            dirs_files = [dirs_files]
-
         # Set verbose for mattak
         if "verbose" in mattak_kwargs:
             verbose = mattak_kwargs.pop("verbose")
@@ -375,7 +410,7 @@ def begin(self,
 
             if os.path.isdir(dir_file):
 
-                if not all_files_in_directory(dir_file):
+                if not _all_files_in_directory(dir_file):
                     self.logger.error(f"Incomplete directory: {dir_file}. Skip ...")
                     continue
             else:
@@ -384,7 +419,7 @@ def begin(self,
                 # it is not a combined file). To work around this: Create a tmp directory under `/tmp/`, link the file you want to
                 # read into this directory with the the name `combined.root`, use this path to read the run.
 
-                path = create_random_directory_path()
+                path = _create_random_directory_path()
                 self.logger.debug(f"Create temporary directory: {path}")
                 if os.path.exists(path):
                     raise ValueError(f"Temporary directory {path} already exists.")
@@ -393,7 +428,7 @@ def begin(self,
                 self.__temporary_dirs.append(path)  # for housekeeping
 
                 self.logger.debug(f"Create symlink for {dir_file}")
-                os.symlink(dir_file, os.path.join(path, "combined.root"))
+                os.symlink(os.path.abspath(dir_file), os.path.join(path, "combined.root"))
 
                 dir_file = path  # set path to e.g. /tmp/NuRadioReco_XXXXXXX/combined.root
 
@@ -721,10 +756,10 @@ def run(self):
         """
         Loop over all events.
 
-        Returns
-        -------
+        Yields
+        ------
 
-        evt: generator(NuRadioReco.framework.event)
+        evt: `NuRadioReco.framework.event.Event`
         """
         event_idx = -1
         for dataset in self._datasets:
@@ -774,7 +809,7 @@ def get_event_by_index(self, event_index):
         Returns
         -------
 
-        evt: NuRadioReco.framework.event
+        evt: `NuRadioReco.framework.event.Event`
         """
 
         self.logger.debug(f"Processing event number {event_index} out of total {self._n_events_total}")
@@ -816,7 +851,7 @@ def get_event(self, run_nr, event_id):
         Returns
         -------
 
-        evt: NuRadioReco.framework.event
+        evt: `NuRadioReco.framework.event.Event`
         """
 
         self.logger.debug(f"Processing event {event_id}")

README.md

@@ -35,9 +35,10 @@ Also please visit https://nu-radio.github.io/NuRadioMC/Introduction/pages/contri
 ## Publications builing up on NuRadioMC/Reco
 NuRadioMC is used in an increasing number of studies. To get an overview for what NuRadioMC can be used for, please have a look at the following publications or see [here](https://inspirehep.net/literature?sort=mostrecent&size=25&page=1&q=refersto%3Arecid%3A1738571%20or%20refersto%3Arecid%3A1725583):
 
+* V. Valera, M. Bustamante, O. Mena, "Joint measurement of the ultra-high-energy neutrino spectrum and cross section", [arXiv:2308.07709](https://arxiv.org/abs/2308.07709)
 * IceCube-Gen2 collaboration, [IceCube-Gen2 Technical Design Report](https://icecube-gen2.wisc.edu/science/publications/TDR)
 * ARIANNA collaboration (A. Anker et al.), "Developing New Analysis Tools for Near Surface Radio-based Neutrino Detectors", [arXiv:2307.07188](https://arxiv.org/abs/2307.07188)
-* L. Pyras, C. Glaser S. Hallmann and A. Nelles, "Atmospheric muons at PeV energies in radio neutrino detectors", [arXiv:2307.04736](https://arxiv.org/abs/2307.04736)
+* L. Pyras, C. Glaser S. Hallmann and A. Nelles, "Atmospheric muons at PeV energies in radio neutrino detectors", JCAP 10 (2023) 043, [arXiv:2307.04736](https://arxiv.org/abs/2307.04736)
 * I. Plaisier, S. Bouma, A. Nelles, "Reconstructing the arrival direction of neutrinos in deep in-ice radio detectors", [arXiv:2302.00054](https://arxiv.org/abs/2302.00054)
 * S. Bouma, A. Nelles for the IceCube-Gen2 collaboration, "Direction reconstruction performance for IceCube-Gen2 Radio", [PoS(ICRC2023)1045](https://pos.sissa.it/444/1045/pdf)
 * F. Schlüter and S. Toscano for the IceCube-Gen2 collaboration, "Estimating the coincidence rate between the optical and radio array of IceCube-Gen2", [PoS(ICRC2023)1022](https://pos.sissa.it/444/1022/pdf)

changelog.txt

@@ -2,6 +2,10 @@ Changelog - to keep track of all relevant changes
 
 please update the categories "new features" and "bugfixes" before a pull request merge!
 
+version 2.2.1
+bugfixes:
+- readRNOGDataMattak: fix bug where .root files would not be found if they are passed as relative paths (instead of absolute paths or folder)
+
 version 2.2.0
 new features:
 - expand values stored in SimplePhasedTrigger

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "NuRadioMC"
-version = "2.2.0"
+version = "2.2.1"
 authors = ["Christian Glaser et al."]
 homepage = "https://github.com/nu-radio/NuRadioMC"
 documentation = "https://nu-radio.github.io/NuRadioMC/main.html"