Showder data frame changes and some other formatting things

bolides/astro_utils.py

@@ -7,6 +7,8 @@
 from astropy.coordinates import ICRS, SkyCoord
 from astropy.time import Time
 from scipy.special import comb
+from pytz import timezone
+utc = timezone('UTC')
 
 def get_phase(datetime):
     """Get lunar phase (0.01=new moon just happened, 0.99=new moon about to happen)"""

bolides/bdf.py

@@ -530,6 +530,9 @@ def filter_shower(self, shower=None, years=None, padding=1, sdf=None, exclude=Fa
         `~BolideDataFrame`
             The filtered `~BolideDataFrame`
         """
+
+        if not pd.api.types.is_datetime64_any_dtype(self.datetime):
+            self.datetime = pd.to_datetime(self.datetime, errors='coerce')
         if years is None:
             years = list(range(min(self.datetime).year-1, max(self.datetime).year+1))
         elif type(years) is int:
@@ -542,7 +545,7 @@ def filter_shower(self, shower=None, years=None, padding=1, sdf=None, exclude=Fa
                 from . import ShowerDataFrame
                 sdf = ShowerDataFrame()
                 self._showers = sdf
-
+        
         dates = sdf.get_dates(shower, years).datetime
         date_padding = timedelta(days=padding)
         date_ranges = [[d-date_padding, d+date_padding] for d in dates]
@@ -553,6 +556,53 @@ def filter_shower(self, shower=None, years=None, padding=1, sdf=None, exclude=Fa
         else:
             good_locs = counts == np.zeros(len(counts))
         return self[good_locs]
+
+    def filter_out_all_showers(self, shower=None, exclude=True, sdf=None):
+        """Filter out all bolides that are part of any meteor shower.
+
+        Parameters
+        ----------
+        sdf: ShowerDataFrame
+            Optionally provide a ShowerDataFrame to use for filtering.
+            If not provided, the ShowerDataFrame is pulled from the IAU Meteor Data Center.
+        shower: str
+            The meteor shower to filter by. If None, all showers are used.
+            Can enter either IAU number, IAU 3-letter code, or full shower name.
+            Refer to the IAU Meteor Data center at https://www.ta3.sk/IAUC22DB/MDC2022/.
+        exclude: bool
+            Whether or not to exclude bolides around the given showers or keep only bolides in showers. Default is to exclude.
+
+        Returns
+        -------
+        `~BolideDataFrame`
+            The filtered `~BolideDataFrame`
+        """
+        years = list(range(min(self.datetime).year-1, max(self.datetime).year+1))
+        if sdf is None:
+            if hasattr(self, '_showers'):
+                sdf = self._showers
+            else:
+                from . import ShowerDataFrame
+                sdf = ShowerDataFrame()
+                self._showers = sdf
+
+
+        start = sdf.get_start_dates(shower, years).datetime
+        end = sdf.get_end_dates(shower, years).datetime
+        date_ranges = [
+            [date_start, date_end]
+            for date_start, date_end in zip(start, end)
+            if not (pd.isna(date_start) or pd.isna(date_end))
+        ]
+        print(date_ranges)
+        counts = np.sum(np.array([list(self.datetime.between(d[0], d[1])) for d in date_ranges]), axis=0)
+        if not exclude:
+
+            good_locs = counts != np.zeros(len(counts))
+        else:
+            # keep bolides that are not part of any shower
+            good_locs = counts == np.zeros(len(counts))
+        return self[good_locs]
 
     def plot_detections(self, category=None, *args, **kwargs):
         """Plot detections of bolides.

bolides/plotting.py

@@ -163,7 +163,7 @@ def plot_scatter(
                 if s is not None and hasattr(s, '__getitem__'):
                     kwargs['s'] = s[idx]
                 ax.scatter(x[idx], y[idx], color=scalarMap.to_rgba(num), label=label, **kwargs)
-            plt.legend()
+            plt.legend(fontsize=20)
 
         if coastlines:
             ax.coastlines()  # plot coastlines
@@ -343,10 +343,8 @@ def plot_density(
         filled_c = ax.contourf(x, y, z*mask, levels=levels[1:],
                                transform=ccrs.PlateCarree(), **kwargs)
 
-        # make lines invisible
-        for c in filled_c.collections:
-            c.set_edgecolor('none')
-            c.set_linewidth(0.000000000001)
+        filled_c.set_edgecolor('none')
+        filled_c.set_linewidth(0.000000000001)
 
         if coastlines:
             ax.coastlines()  # plot coastlines
@@ -356,8 +354,10 @@ def plot_density(
         if boundary:
             add_boundary(ax, boundary, boundary_style)
 
-        plt.colorbar(filled_c, alpha=kwargs['alpha'],
-                     label='bolide density (km$^{-2}$)')
+        cbar = plt.colorbar(filled_c, alpha=kwargs['alpha'])
+
+        cbar.ax.tick_params(labelsize=15)
+        cbar.set_label(label='bolide density (km$^{-2}$)', size=20, weight='bold')
 
         if title is not None:
             plt.title(title)

bolides/sdf.py

@@ -1,6 +1,7 @@
 import pandas as pd
 import requests
 import numpy as np
+import warnings
 from . import ROOT_PATH
 
 
@@ -14,7 +15,7 @@ class ShowerDataFrame(pd.DataFrame):
         Specifies the source for the initialized. Can be:
 
         - ``'established', 'all', 'working'``: initialize from different sets
-          of data offered at the IAU Meteor Data Center, https://www.ta3.sk/IAUC22DB/MDC2007/
+          of data offered at the IAU Meteor Data Center, https://www.ta3.sk/IAUC22DB/MDC2022/
         - ``'csv'``: initialize from a .csv file
 
     file : str
@@ -32,7 +33,7 @@ def __init__(self, *args, **kwargs):
         if source == 'csv':
             df = pd.read_csv(file, index_col=0, keep_default_na=False, na_values='')
         else:
-            url = 'https://www.ta3.sk/IAUC22DB/MDC2007/Etc/stream'+source+'data.txt'
+            url = 'https://www.ta3.sk/IAUC22DB/MDC2022/Etc/stream'+source+'data2022.txt'
             r = requests.get(url)
             start_line = 0
             for num, line in enumerate(r.text.splitlines()):
@@ -41,12 +42,13 @@ def __init__(self, *args, **kwargs):
                     break
             column_line = r.text.splitlines()[start_line-3]
             import re
-            columns = re.split(r" {2,}", column_line)[1:-1]
+            columns = re.split(r" {2,}", column_line)[1:-3]
             data_lines = r.text.splitlines()[start_line:]
             data = '\n'.join(data_lines)
             import io
             csv_io = io.StringIO(data)
             df = pd.read_csv(csv_io, sep="|", header=None)
+            df = df.iloc[:, :len(columns)]
             df.columns = columns
             for col in df.columns:
                 if df[col].dtype == 'O':
@@ -135,7 +137,7 @@ def plot_orbits(self, date='2000-01-01T12:00:00',
                     orb = Orbit.from_classical(Sun, a, ecc, inc, raan, argp, nu, plane=Planes.EARTH_ECLIPTIC)
                 except ValueError:
                     continue
-                plotter.plot(orb, label=row['shower name'])
+                plotter.plot(orb, label=row['shower name-designation'])
 
         if use_3d:
             fig = plotter._figure
@@ -203,6 +205,24 @@ def plot_orbits(self, date='2000-01-01T12:00:00',
         return plotter
 
     def get_dates(self, showers, years):
+        """
+        Get the dates from solar longitude.
+
+        Parameters
+        ----------
+        showers : str or list of str
+            The meteor shower(s) to get the dates for.
+        years : int or list of int
+            The year(s) to get the dates for.
+
+        Returns
+        -------
+        pd.DataFrame
+            A DataFrame with the shower Codes, names, IAUNos, and the corresponding dates
+            for the specified meteor showers and years.
+        """
+        if showers is None:
+            showers = self['shower name-designation'].unique()
         if type(years) is int:
             years = [years]
         if type(showers) in [str, int]:
@@ -214,28 +234,167 @@ def get_dates(self, showers, years):
         elif len(showers[0]) == 3:
             col = 'Code'
         else:
-            col = 'shower name'
+            col = 'shower name-designation'
         sdf = self[self[col].isin(showers)]
         import warnings
         if len(sdf) == 0:
             warnings.warn('No showers with '+col+'in'+str(showers)+'.')
 
         sdf = sdf[sdf.activity == 'annual']
-        sdf = sdf.dropna(subset=['LaSun'])
+        sdf = sdf.dropna(subset=['LoS'])
         num = len(sdf)
         sdf = pd.concat([sdf]*len(years), ignore_index=True)
         years = np.repeat(years, num)
 
-        lons = sdf.LaSun
+        lons = sdf.LoS
         from .astro_utils import sol_lon_to_datetime
         dts = []
         for lon, year in zip(lons, years):
             dts.append(sol_lon_to_datetime(lon, year))
         return pd.DataFrame({'Code': sdf.Code,
-                             'shower name': sdf['shower name'],
+                             'shower name': sdf['shower name-designation'],
+                             'IAUNo': sdf.IAUNo,
+                             'datetime': dts})
+
+    def get_start_dates(self, showers, years):
+        """Get the start dates of the meteor showers from solar longitude.
+
+        Parameters
+        ----------
+        showers : str or list of str
+            The meteor shower(s) to get the start dates for.
+        years : int or list of int
+            The year(s) to get the start dates for.
+
+        Returns
+        -------
+        pd.Series
+            A series with the start dates of the meteor showers.
+        """
+        if showers is None:
+            showers = self['shower name-designation'].unique()
+        if type(years) is int:
+            years = [years]
+        if type(showers) in [str, int]:
+            showers = [showers]
+        showers = [str(s) for s in showers]
+        if showers[0].isdigit():
+            col = 'IAUNo'
+            showers = [int(s) for s in showers]
+        elif len(showers[0]) == 3:
+            col = 'Code'
+        else:
+            col = 'shower name-designation'
+        sdf = self[self[col].isin(showers)]
+        import warnings
+        if len(sdf) == 0:
+            warnings.warn('No showers with '+col+'in'+str(showers)+'.')
+
+        sdf = sdf[sdf.activity == 'annual']
+        sdf = sdf.dropna(subset=['LoSb'])
+        num = len(sdf)
+        sdf = pd.concat([sdf]*len(years), ignore_index=True)
+        years = np.repeat(years, num)
+
+        lons = sdf.LoSb
+        from .astro_utils import sol_lon_to_datetime
+        dts = []
+        for lon, year in zip(lons, years):
+            if lon=='' or lon is None or pd.isna(lon):
+                dts.append(np.nan)
+            else:
+                dts.append(sol_lon_to_datetime(float(lon), year))
+        return pd.DataFrame({'Code': sdf.Code,
+                             'shower name': sdf['shower name-designation'],
+                             'IAUNo': sdf.IAUNo,
+                             'datetime': dts})
+
+    def get_end_dates(self, showers, years):
+        """Get the end dates of the meteor showers from solar longitude.
+
+        Parameters
+        ----------
+        showers : str or list of str
+            The meteor shower(s) to get the start dates for.
+        years : int or list of int
+            The year(s) to get the start dates for.
+
+        Returns
+        -------
+        pd.Series
+            A series with the start dates of the meteor showers.
+        """
+        if showers is None:
+            showers = self['shower name-designation'].unique()
+        if type(years) is int:
+            years = [years]
+        if type(showers) in [str, int]:
+            showers = [showers]
+        showers = [str(s) for s in showers]
+        if showers[0].isdigit():
+            col = 'IAUNo'
+            showers = [int(s) for s in showers]
+        elif len(showers[0]) == 3:
+            col = 'Code'
+        else:
+            col = 'shower name-designation'
+        sdf = self[self[col].isin(showers)]
+        import warnings
+        if len(sdf) == 0:
+            warnings.warn('No showers with '+col+'in'+str(showers)+'.')
+
+        sdf = sdf[sdf.activity == 'annual']
+        sdf = sdf.dropna(subset=['LoSe'])
+        num = len(sdf)
+        sdf = pd.concat([sdf]*len(years), ignore_index=True)
+        years = np.repeat(years, num)
+
+        lons = pd.to_numeric(sdf.LoSe, errors='coerce')
+        from .astro_utils import sol_lon_to_datetime
+        dts = []
+        for lon, year in zip(lons, years):
+            if lon=='' or lon is None or pd.isna(lon):
+                dts.append(np.nan)
+            else:
+                dts.append(sol_lon_to_datetime(float(lon), year))
+        return pd.DataFrame({'Code': sdf.Code,
+                             'shower name': sdf['shower name-designation'],
                              'IAUNo': sdf.IAUNo,
                              'datetime': dts})
 
+    def fastest_showers(self, n=5):
+        """
+        Return the top n shower Codes by mean Vg.
+
+        Parameters
+        -----------
+        n : int
+            The number of top showers to return based on their mean Vg. Maximum is 122.
+
+        Returns
+        -------
+        pd.DataFrame
+            A DataFrame with the shower Codes and their mean Vg, sorted by mean Vg in descending order.
+        """
+        # Ensure Vg is numeric, coerce errors to NaN
+        sdf = self.copy()
+        sdf['Vg'] = pd.to_numeric(sdf['Vg'], errors='coerce')
+        sdf_sorted = sdf.sort_values('Code')
+
+        mean_vgs = []
+        codes = []
+        for code, group in sdf_sorted.groupby('Code'):
+            codes.append(code)
+            mean_vgs.append(group['Vg'].mean())
+
+        result = pd.DataFrame({'Code': codes, 'mean_Vg': mean_vgs})
+
+        if len(result) < n:
+            warnings.warn('Maximum number of showers is ' + str(len(result)))
+
+        result = result.sort_values('mean_Vg', ascending=False).head(n).reset_index(drop=True)
+        return result
+
     def __setattr__(self, attr, val):
         from warnings import filterwarnings
         filterwarnings("ignore", message="Pandas doesn't allow columns to be created via a new attribute name")

notebooks/tutorial.ipynb

@@ -4423,7 +4423,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.4"
+   "version": "3.13.2"
   }
  },
  "nbformat": 4,