nctoolkit 1.1.10__tar.gz → 1.2.0__tar.gz

nctoolkit-1.2.0/CONTRIBUTIONS.md

@@ -0,0 +1,39 @@
+Contributing to nctoolkit
+
+🎉 Thank you for your interest in contributing to nctoolkit! Your efforts help improve this tool for the entire community. Whether you're fixing bugs, adding new features, or enhancing documentation, your contributions are invaluable. Let's make this project even better together! 🎉
+
+How to Contribute
+
+To contribute, please follow these steps:
+
+1. Fork the Repository
+Click the "Fork" button at the top right of this page to create a copy of this repository under your GitHub account.
+
+2. Clone and Set Up the Repository
+After forking, open your terminal, clone the repository to your local machine, and navigate into the project directory:
+git clone https://github.com/your-username/nctoolkit.git
+cd nctoolkit
+
+3. Create a New Branch
+Create a new branch for your contribution to keep your changes organized and separated from the main codebase:
+git checkout -b feature/your-feature-name
+
+4. Make Your Changes
+Add your feature, fix bugs, or improve documentation. Be sure to:
+Keep your changes focused on a single feature or fix for easier review.
+Test your modifications to ensure they work as intended.
+Follow any coding style guidelines noted in the project.
+
+5. Commit Your Changes
+When you’re ready, stage and commit your changes with a clear, descriptive commit message:
+git add .
+git commit -m "Add detailed description of your changes here"
+
+6. Push to Your Fork
+Push your branch to your GitHub repository:
+git push origin feature/your-feature-name
+
+7. Submit a Pull Request (PR)
+Go to the original nctoolkit repository on GitHub.
+Click the "New Pull Request" button and select your branch.
+Include a detailed description of your changes to help reviewers understand your contribution.

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: nctoolkit
-Version: 1.1.10
+Version: 1.2.0
 Summary: A Python package for netCDF analysis and post-processing
 Home-page: https://github.com/pmlmodelling/nctoolkit
 Author: Robert Wilson
@@ -56,7 +56,6 @@ Core abilities of nctoolkit include:
   - Calculating the correlations between variables
   - Calculating vertical statistics for the likes of oceanic data
   - Calculating ensemble statistics
-  - Calculating phenological metrics
 
 Operation of the package requires the installation of Climate Data Operators (CDO). This is the computational backend for most of the methods used. No knowledge of CDO is required to use nctoolkit. A couple of methods provide users with the option of using netCDF Operators (NCO) as a backend. Guidance for how to install the backends are available `here <https://nctoolkit.readthedocs.io/en/stable/installing.html>`__.
 

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/nctoolkit/__init__.py

@@ -28,7 +28,7 @@ atexit.register(clean_all)
 
 signal.signal(signal.SIGTERM, clean_all)
 
-from nctoolkit.create_ensemble import create_ensemble
+from nctoolkit.create_ensemble import create_ensemble, glob
 from nctoolkit.session import session_files
 from nctoolkit.show import nc_variables, nc_years, nc_months, nc_times
 

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/nctoolkit/assign.py

@@ -91,6 +91,8 @@ funs = [
     "exp",
     "ln",
     "sin",
+    "min",
+    "max",
     "cos",
     "tan",
     "int",

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/nctoolkit/cleanup.py

@@ -145,12 +145,12 @@ def temp_check():
         if len(mylist) > 0:
             if len(mylist) == 1:
                 print(
-                    f"{len(mylist)} file was created by nctoolkit in prior or current "
+                    f"{len(mylist)} temporary file was created by nctoolkit in prior or current "
                     f"sessions. Consider running deep_clean!"
                 )
             else:
                 print(
-                    f"{len(mylist)} files were created by nctoolkit in prior or current"
+                    f"{len(mylist)} temporary files were created by nctoolkit in prior or current"
                     f" sessions. Consider running deep_clean!"
                 )
 
@@ -163,12 +163,12 @@ def temp_check():
         if len(mylist) > 0:
             if len(mylist) == 1:
                 print(
-                    f"{len(mylist)} file was created by nctoolkit in prior or current "
+                    f"{len(mylist)} temporary file was created by nctoolkit in prior or current "
                     f"sessions. Consider running deep_clean!"
                 )
             else:
                 print(
-                    f"{len(mylist)} files were created by nctoolkit in prior or "
+                    f"{len(mylist)} temporary files were created by nctoolkit in prior or "
                     f"current sessions. Consider running deep_clean!"
                 )
 

nctoolkit-1.2.0/nctoolkit/create_ensemble.py

@@ -0,0 +1,117 @@
+from glob import glob as globber
+import os
+
+# function to find files in directory with a specified variable
+
+
+def create_ensemble(path="", recursive=True):
+    """
+    create_ensemble: Generate an ensemble of files from a directory.
+
+    Parameters
+    -------------
+    path: str
+        The directory to search for netCDF files
+    recursive : boolean
+        True/False depending on whether you want to search the path recursively.
+        Defaults to True.
+
+    Returns
+    -------------
+
+    files : list of files
+
+    Examples
+    ------------
+
+    If you wanted to recursively find all netCDF files available in a directory "data", you would do this:
+
+    >>> import nctoolkit as nc
+    >>> nc.create_ensemble("data")
+
+    If you wanted to find the files in that directory and ignore subdirectories, you would instead do this:
+
+    >>> nc.create_ensemble("data", recursive = False)
+
+
+    """
+
+    # make sure the path exists
+
+    if os.path.exists(path) is False:
+        raise ValueError("The path provided does not exist!")
+
+    # make sure the path ends with "/" if it is not empty
+    if path != "":
+        if path.endswith("/") is False:
+            path = path + "/"
+
+    if recursive:
+        files = [f for f in globber(path + "/**/*.nc*", recursive=True)]
+    else:
+        files = [f for f in globber(path + "*.nc*")]
+
+    if len(files) == 0:
+        raise ValueError("There is no data in the target directory")
+
+    return files
+
+def glob(path="", recursive=True):
+    """
+    create_ensemble: Generate an ensemble of files from a directory.
+
+    Parameters
+    -------------
+    path: str
+        The directory to search for netCDF files
+    recursive : boolean
+        True/False depending on whether you want to search the path recursively.
+        Defaults to True.
+
+    Returns
+    -------------
+
+    files : list of files
+
+    Examples
+    ------------
+
+    If you wanted to recursively find all netCDF files available in a directory "data", you would do this:
+
+    >>> import nctoolkit as nc
+    >>> nc.create_ensemble("data")
+
+    If you wanted to find the files in that directory and ignore subdirectories, you would instead do this:
+
+    >>> nc.create_ensemble("data", recursive = False)
+
+
+    """
+
+    # make sure the path exists
+
+    try:
+        files = globber(path, recursive = recursive)
+        if [x for x in files if x.endswith(".nc")]:
+            return files
+        else:
+            raise ValueError("There are no netCDF files in the target directory")
+    except:
+
+        if os.path.exists(path) is False:
+            raise ValueError("The path provided does not exist!")
+
+        # make sure the path ends with "/" if it is not empty
+        if path != "":
+            if path.endswith("/") is False:
+                path = path + "/"
+
+        if recursive:
+            files = [f for f in globber(path + "/**/*.nc*", recursive=True)]
+        else:
+            files = [f for f in globber(path + "*.nc*")]
+
+        if len(files) == 0:
+            raise ValueError("There is no data in the target directory")
+
+        return files

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/nctoolkit/plot.py

@@ -44,8 +44,6 @@ def plot(self, vars=None, autoscale=True, out=None, coast=None, **kwargs):
         A string or list of the variables to plot
     autoscale: bool
         Set to True if you want the colorbar to be scaled to the min/max of the data. Default is True
-    vars: str , list
-        A string or list of the variables to plot
     out: str
         Name of output file if you want to save as html. Defaults to None.
     coast: bool

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/nctoolkit/runners.py

@@ -22,6 +22,12 @@ from nctoolkit.temp_file import temp_file
 
 from nctoolkit.show import nc_variables
 
+def ignore_warning(x):
+    text = re.compile("remap weights from .* not used")
+    if len(text.findall(x)) > 0:
+        return True
+    return False
+
 def ann_anomaly(
     ff, baseline, metric, window, align, precision, new_files, new_commands, nc_safe):
     """
@@ -840,10 +846,14 @@ def run_cdo(command=None, target=None, out_file=None, overwrite=False, precision
             else:
                 message = f"The following month was missing in the dataset: {sel_month}"
             warnings.warn(message=message)
+        #remap weights from /tmp/nctoolkit_rwi_brtnahgznctoolkittmpqwadyf2y.nc not used, lonlat (984x582) grid with mask (407783) not found!
+        # detect regex pattern, based on above
+        
     for ww in w:
         if platform.system() == "Linux":
             if ww.message not in session_warnings:
-                session_warnings.append(ww.message)
+                if ignore_warning(str(ww.message)) is False:
+                    session_warnings.append(ww.message)
         else:
             warnings.warn(ww.message)
             #print(session_warnings)

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/nctoolkit/setters.py

@@ -278,12 +278,13 @@ def set_units(self, unit_dict=None, **kwargs):
 
     # change the units in turn. This doesn't seem to be something you can chain?
     for i in unit_dict:
-        if not isinstance(i, str):
-            raise TypeError("key,values in unit_dict are not strings")
-        if not isinstance(unit_dict[i], str):
-            raise TypeError("key,values in unit_dict are not strings")
+        if unit_dict[i] is not None:
+            if not isinstance(i, str):
+                raise TypeError("key,values in unit_dict are not strings")
+            if not isinstance(unit_dict[i], str):
+                raise TypeError("key,values in unit_dict are not strings")
 
-        cdo_command = f'-setattribute,{i}@units="{unit_dict[i]}"'
+        cdo_command = f'-setattribute,{i}@units="{unit_dict[str(i)]}"'
         self.cdo_command(cdo_command, ensemble=False)
 
 

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/nctoolkit/static_plot.py

@@ -93,6 +93,7 @@ def pub_plot(
     out=None,
     breaks=None,
     dpi = "figure",
+    font = None,
     **kwargs,
 ):
     """
@@ -360,8 +361,8 @@ def pub_plot(
 
     # set globe to True if lon lat spread is big enough
 
-    if np.max(lon) - np.min(lon) > 358:
-        if np.max(lat) - np.min(lat) > 176:
+    if np.max(lon) - np.min(lon) > 340:
+        if np.max(lat) - np.min(lat) > 160:
             globe = True
 
     # proj=ccrs.LambertConformal(central_longitude=np.mean(lon), central_latitude=np.mean(lat), false_easting=0.0, false_northing=0.0,  cutoff=38)
@@ -388,6 +389,13 @@ def pub_plot(
 
     if projection == None:
         proj = ccrs.PlateCarree()
+
+    if isinstance(projection, ccrs.NorthPolarStereo) and size == "auto":
+        size = [8,8]
+    if isinstance(projection, ccrs.SouthPolarStereo) and size == "auto":
+        size = [8,8]
+    if isinstance(projection, ccrs.AzimuthalEquidistant) and size == "auto":
+        size  = [12, 8]
     if size == "auto":
         size = [10, 10]
         if proj == ccrs.PlateCarree() or globe:
@@ -507,9 +515,13 @@ def pub_plot(
             if r_max is not None:
                 vmax = np.nanpercentile(np.ma.filled(values, np.nan), r_max)
 
-        if robust and limits is None:
+        if robust:
             vmin = np.nanpercentile(np.ma.filled(values, np.nan), 2)
             vmax = np.nanpercentile(np.ma.filled(values, np.nan), 98)
+            limits = [0,0]
+            limits[0] = vmin
+            limits[1] = vmax
+
         if limits is None:
             if r_min is not None:
                 vmin = np.nanpercentile(np.ma.filled(values, np.nan), r_min)
@@ -684,10 +696,34 @@ def pub_plot(
         else:
             fraction = 0.046 * size[1] / size[0]
 
+        min_value = np.min(values)
+        max_value = np.max(values)
+
+        min_arrow = False
+
+        if vmin is not None:
+            if vmin > min_value:
+                min_arrow = True
+        max_arrow = False
+        if vmax is not None:
+            if vmax < max_value:
+                max_arrow = True
+        
+        if min_arrow and max_arrow:
+            extend = "both"
+        else:
+            if min_arrow:
+                extend = "min"
+            else:
+                if max_arrow:
+                    extend = "max"
+                else:
+                    extend = "neither"
+
         if l_location == "bottom":
-            cb = plt.colorbar(im, fraction=fraction, pad=0.04, location=l_location)
+            cb = plt.colorbar(im, fraction=fraction, pad=0.04, location=l_location, extend = extend)
         else:
-            cb = plt.colorbar(im, fraction=fraction, pad=0.04)
+            cb = plt.colorbar(im, fraction=fraction, pad=0.04, extend = extend)
 
         # add breaks to colorbar cb
         if breaks is not None:
@@ -742,11 +778,12 @@ def pub_plot(
                     "Unable to parse legend from dataset contents. Check long names and units"
                 )
                 label = ""
+            # label may need some slight tidying
 
             if l_location == "bottom":
                 label = "\n".join(wrap(label, 120))
             else:
-                label = "\n".join(wrap(label, 50))
+                label = "\n".join(wrap(label, 30))
 
             if l_location == "bottom":
                 cbax.set_xlabel(label)
@@ -755,10 +792,18 @@ def pub_plot(
 
         if legend_position is None:
             cb.remove()
+    
+    if font is not None:
+        for text in [ax.title, ax.xaxis.label, ax.yaxis.label, cb.ax.yaxis.label, cb.ax.xaxis.label]:
+            text.set_fontsize(font)
+        for item in ([ax.title, ax.xaxis.label, ax.yaxis.label, cb.ax.yaxis.label, cb.ax.xaxis.label] +
+                 ax.get_xticklabels() + ax.get_yticklabels()):
+            item.set_fontsize(font)
 
     if out is not None:
         print("saving as file")
         plt.savefig(out, dpi = dpi)
+        plt.savefig(out, dpi = dpi)
 
 
 def quiver_plot(ds, u=None, v=None, **kwargs):
@@ -815,4 +860,4 @@ def quiver_plot(ds, u=None, v=None, **kwargs):
     if v not in vars:
         raise ValueError("v not in dataset")
 
-    pub_plot(ds, quiver=True, u=u, v=v, **kwargs)
+    pub_plot(ds, quiver=True, u=u, v=v, **kwargs)

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/nctoolkit/subset.py

@@ -45,24 +45,30 @@ def select_levels(self, levels=None):
     levels : list
         List of the form [min_level, max_level]. Levels/depth between the two will be selected
     """
+    ds_levels = self.levels
+
     if not isinstance(levels, list):
         type(levels)
         try:
             levels = float(levels)
         except:
             raise ValueError("levels provided are not valid!")
+        levels = [float(levels), float(levels)]
 
     if isinstance(levels, list):
         try:
-            levels[0] = int(np.floor(float(levels[0])))
-            levels[1] = int(np.ceil(float(levels[1])))
+            levels[0] = float(levels[0])
+            levels[1] = float(levels[1])
         except:
             raise ValueError("levels provided are not valid!")
         if levels[0] > levels[1]:
             raise ValueError("levels have the wrong order")
-        levels = f"{levels[0]}/{levels[1]}"
-
-    cdo_command = f"-sellevel,{levels}"
+    
+    sel_levels = [str(x) for x in ds_levels if levels[0] <= x <= levels[1]]
+    if len(sel_levels) == 0:
+        raise ValueError("No levels found in the dataset")
+    sel_levels = ",".join(sel_levels)
+    cdo_command = f"-sellevel,{sel_levels}"
 
     self.cdo_command(cdo_command, ensemble=False)
 

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/nctoolkit/toxarray.py

@@ -102,7 +102,7 @@ def to_xarray(self, decode_times=True, **kwargs):
         return data
 
 
-def to_dataframe(self, decode_times=True, **kwargs):
+def to_dataframe(self, decode_times=True, drop_bnds = True, **kwargs):
     """
     to_dataframe: Convert a dataset to a pandas data frame
 
@@ -111,6 +111,10 @@ def to_dataframe(self, decode_times=True, **kwargs):
     decode_times: boolean
         Set to False if you do not want xarray to decode the times prior to
         conversion to data frame. Default is True.
+    drop_bnds: boolean
+        Set to False if you do not want to drop the bounds from the data frame.
+        Variables/coordinates with '_bnds' in their names will be dropped.
+        Default is True.
     **kwargs : kwargs
         Optional arguments to be sent to subset.
 
@@ -125,4 +129,9 @@ def to_dataframe(self, decode_times=True, **kwargs):
     >>> ds.to_dataframe()
 
     """
-    return self.to_xarray(decode_times=decode_times, **kwargs).to_dataframe()
+    df = self.to_xarray(decode_times=decode_times, **kwargs).to_dataframe()
+    df = df.reset_index()
+    if drop_bnds:
+        df = df.drop(columns=[col for col in df.columns if 'bnds' in col])
+    df = df.drop_duplicates().reset_index(drop = True)
+    return df 

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/nctoolkit/verticals.py

@@ -180,6 +180,9 @@ def vertical_interp(
     if thickness is not None:
         fixed = False
 
+    if isinstance(levels, (int, float)):
+        levels = [levels]
+
     if fixed is False:
         self.to_zlevels(
             levels=levels, thickness=thickness, depths=depths, surface=surface
@@ -191,8 +194,6 @@ def vertical_interp(
 
     # first a quick fix for the case when there is only one vertical depth
 
-    if isinstance(levels, (int, float)):
-        levels = [levels]
 
     # levels = [float(x) for x in levels]
 

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/nctoolkit.egg-info/SOURCES.txt

@@ -1,5 +1,6 @@
 CITATION.cff
 CODE_OF_CONDUCT.md
+CONTRIBUTIONS.md
 LICENSE
 MANIFEST.in
 README.md

{nctoolkit-1.1.10 → nctoolkit-1.2.0}/setup.py

@@ -21,7 +21,6 @@ Core abilities of nctoolkit include:
   - Calculating the correlations between variables
   - Calculating vertical statistics for the likes of oceanic data
   - Calculating ensemble statistics
-  - Calculating phenological metrics
 
 Operation of the package requires the installation of Climate Data Operators (CDO). This is the computational backend for most of the methods used. No knowledge of CDO is required to use nctoolkit. A couple of methods provide users with the option of using netCDF Operators (NCO) as a backend. Guidance for how to install the backends are available `here <https://nctoolkit.readthedocs.io/en/stable/installing.html>`__.
 
@@ -47,7 +46,7 @@ extras_require["complete"] = ["geoviews", "rioxarray", "cfchecker", "geocube", "
 REQUIREMENTS = [i.strip() for i in open("requirements.txt").readlines()]
 
 setup(name='nctoolkit',
-      version='1.1.10',
+      version='1.2.0',
       description=DESCRIPTION,
       description_content_type='text/plain',
       long_description=LONG_DESCRIPTION,

nctoolkit-1.1.10/nctoolkit/create_ensemble.py

@@ -1,57 +0,0 @@
-import glob
-import os
-
-# function to find files in directory with a specified variable
-
-
-def create_ensemble(path="", recursive=True):
-    """
-    create_ensemble: Generate an ensemble of files from a directory.
-
-    Parameters
-    -------------
-    path: str
-        The directory to search for netCDF files
-    recursive : boolean
-        True/False depending on whether you want to search the path recursively.
-        Defaults to True.
-
-    Returns
-    -------------
-
-    files : list of files
-
-    Examples
-    ------------
-
-    If you wanted to recursively find all netCDF files available in a directory "data", you would do this:
-
-    >>> import nctoolkit as nc
-    >>> nc.create_ensemble("data")
-
-    If you wanted to find the files in that directory and ignore subdirectories, you would instead do this:
-
-    >>> nc.create_ensemble("data", recursive = False)
-
-
-    """
-
-    # make sure the path exists
-
-    if os.path.exists(path) is False:
-        raise ValueError("The path provided does not exist!")
-
-    # make sure the path ends with "/" if it is not empty
-    if path != "":
-        if path.endswith("/") is False:
-            path = path + "/"
-
-    if recursive:
-        files = [f for f in glob.glob(path + "/**/*.nc*", recursive=True)]
-    else:
-        files = [f for f in glob.glob(path + "*.nc*")]
-
-    if len(files) == 0:
-        raise ValueError("There is no data in the target directory")
-
-    return files