Add geodata filter

docs/Subcommand/Geodata.md

@@ -46,6 +46,10 @@ Specified the output format.  See `gridlabd geodata help` for details.
 
 Set the RAW output field separator.
 
+## `--filter=CALLS`
+
+The `--filter=CALLS` option applies the pandas dataframe calls to the resulting data before returning the result.  The CALLS may be a comma separate value list of valid dataframe function calls, e.g., `dropna()`, `drop(0)`, or `gt(0)`.  This may be used in conjunction with `--select` to remove rows that fail to meet certain criteria. Multiple filters are semicolon-delimited
+
 ## `-j|--join {inner,outer,left,right}`
 
 Controls how the dataset is merged with the input path. If [-j|--join {inner,outer,left,right}] is specified, the input is joined with the dataset in the manner specified, i.e., "left" uses the input index, "right" uses the dataset's index, "outer" forms the union of both indexes and sorts it, and "inner" forms the intersection of both indexes, preserving the order of the input index.
@@ -74,6 +78,10 @@ Set the RAW output record separator.
 
 Disables error output.
 
+## `--select=COLUMNS`
+
+Selects rows with columns that evaluate to `True`. Multiple colums are comma-delimited.
+
 ## `--show_config`
 
 Displays the current configuration values.

gldcore/scripts/gridlabd-geodata

@@ -17,6 +17,7 @@ OPTIONS
   [-d|--debug]                enable debug output, including error traceback
   [-f|--format FORMAT]        change output format
   [--fieldsep STRING]         set the RAW output field separator
+  [--filter=SPECS]            apply dataframe functions as filter
   [-j|--join TYPE]            control how dataset joins with input path
   [-k|--key KEY]              change method for generating keys
   [-o|--output CSVOUT]        output to CSVOUT
@@ -25,6 +26,7 @@ OPTIONS
                               merged
   [--recordsep STRING]        change the RAW output record separator
   [-s|--silent]               disable error output
+  [--select=COLUMNS]          select only rows that have true values for column(s)
   [--show_config]             display the current configuration
   [--show_options]            display the current options
   [-T|--threadcount THREADS]  change maximum thread count
@@ -355,6 +357,7 @@ import haversine
 import geopandas
 import matplotlib.pyplot as plt
 import uuid
+import folium
 
 NAME = "geodata"
 VERSION = "0.0.0"
@@ -450,6 +453,8 @@ OPTIONS = {
         "force_ascii" : True,
         "date_unit" : "s",
     },
+    "filter" : "",
+    "select" : "",
 }
 
 E_OK = 0
@@ -942,6 +947,20 @@ def write_csv(data):
         except KeyError as err:
             error("write_csv(): invalid field specified",E_INVALID)
 
+def filter(data):
+    """Filter implementation"""
+    if OPTIONS['filter']:
+        for action in OPTIONS['filter'].split(';'):
+            data = eval(f"data.{action}")
+        if OPTIONS['select']:
+            for column in OPTIONS['select'].split(','):
+                data = data[data[column]]
+        result = data
+        debug(f"filter(data='{data}') --> {dataframe_to_table(result)}")
+        return result
+    else:
+        return data
+
 def write_raw(data):
     """Write raw output"""
     specs = CONFIG['output_format'].split(":")
@@ -1361,6 +1380,7 @@ def main(argc,argv):
         error(f"directive '{DIRECTIVE[0]}' is not valid",E_SYNTAX)
     data = globals()[DIRECTIVE[0]](args)
     if type(data) is pandas.DataFrame:
+        data = filter(data)
         if CONFIG['output_format'] == "CSV" or CONFIG['output_format'].startswith("CSV:"):
             write_csv(data)
         elif CONFIG['output_format'] == "RAW" or CONFIG['output_format'].startswith("RAW:"):