string split: allowing for quoted strings with shlex

Author: martinkilbinger

cs_util/args.py

@@ -9,6 +9,7 @@
 
 import sys
 import os
+import shlex
 
 from optparse import OptionParser
 
@@ -103,6 +104,9 @@ def my_string_split(string, num=-1, verbose=False, stop=False, sep=None):
     the first in the list [space, underscore] that occurs in the string.
     (Thus, if both occur, use space.)
 
+    Handles quoted strings: entries containing spaces can be enclosed
+    in double quotes, e.g., 'value1 "entry with spaces" value3'.
+
     Parameters
     ----------
     string : str
@@ -131,6 +135,19 @@ def my_string_split(string, num=-1, verbose=False, stop=False, sep=None):
     if string is None:
         return None
 
+    # Handle quoted strings with shlex
+    if '"' in string or "'" in string:
+        try:
+            res = shlex.split(string)
+            if num != -1 and num != len(res) and stop:
+                raise ValueError(
+                    f"String has {len(res)} elements, required is {num}"
+                )
+            return res
+        except ValueError:
+            # Fall through to regular splitting if shlex fails
+            pass
+
     if sep is None:
         has_space = string.find(" ")
         has_underscore = string.find("_")
@@ -162,7 +179,7 @@ def my_string_split(string, num=-1, verbose=False, stop=False, sep=None):
 
     if num != -1 and num != len(res) and stop:
         raise ValueError(
-            f"String '{len(res)}' has length {num}, required is {num}"
+            f"String has {len(res)} elements, required is {num}"
         )
 
     return res

cs_util/cat.py

@@ -253,19 +253,36 @@ def read_dndz(file_path):
 
     Returns
     -------
-    list :
+    np.array
         redshift bin centers
-    list :
+    np.array
         number densities
-    list :
-        redshift bin edges
+    np.array
+        redshift bin edges; one less than centers and density arrays
 
     """
-    dat = ascii.read(file_path, format="commented_header")
+    try:
+        # Expecting header line "# z dn_dz"
+        dat = ascii.read(file_path, format="commented_header")
+        missing = [col for col in ("z", "dn_dz") if col not in dat.dtype.names]
+        if missing:
+            raise ValueError(
+                f"Missing columns in dndz path {file_path}: {missing}"
+            )
+    except:
+        # No header line
+        dat = ascii.read(file_path)
+        dat.rename_column("col1", "z")
+        dat.rename_column("col2", "dn_dz")
+
+    # Remove last n(z) value which should be zero, to match bin centers
+    tolerance = 1e-5
+    if dat["dn_dz"][-1] / sum(dat["dn_dz"]) > tolerance:
+        raise ValueError("dn_dz at last z-edge = {dat['dn_dz'][-1]}, no zero")
 
-    # Remove last n(z) value which is zero, to match bin centers
     nz = dat["dn_dz"][:-1]
     z_edges = dat["z"]
+
     z_centers = bin_edges2centers(z_edges)
 
     return z_centers, nz, z_edges