Up our debugging game significantly by not making the rookie equals assignment object mistake

pomodoro_to_harvest.py

@@ -6,34 +6,52 @@ import sys
 # Import our local settings management.
 import settings
 
+if not hasattr(sys, 'ps1'):
+    import copy
+    debug = True
+else:
+    debug = False
+
 # This works for one file:
-# timelog = pd.read_csv(settings.pomodoro_logfile())
+timelog = pd.read_csv(settings.pomodoro_logfile())
 
 # For multiple files:
-path = settings.pomodoro_logpath()
+#path = settings.pomodoro_logpath()
-all_files = glob.glob(path + "*.csv")
+#all_files = glob.glob(path + "*.csv")
+#
+#li = []
+#
+#for filename in all_files:
+#    df = pd.read_csv(filename, index_col=None, header=0)
+#    li.append(df)
 
-li = []
+# timelog = pd.concat(li, axis=0, ignore_index=True)
 
-for filename in all_files:
+if debug:
-    df = pd.read_csv(filename, index_col=None, header=0)
+   imported = copy.deepcopy(timelog)
-    li.append(df)
-
-timelog = pd.concat(li, axis=0, ignore_index=True)
 
 timelog.drop_duplicates(inplace=True)
 
+if debug:
+   nodupes = copy.deepcopy(timelog)
+
 # Dump bad data.  The real solution here is to get rid of the damned 'Cancel'
 # button on the Pomodoro Prompt dialog, but i don't know how to do that, so we
 # need to drop the rows where the work task description is blank, which is
 # coming in as not a number for reasons i'm not entirely clear on.  Maybe
 # because it's the last row of the spreadsheet?  Anyway we cannot do anything
 # with no data in the description, so drop them at the outset.
-timelog = timelog.dropna()
+# We can allow no data in the 'intention' so define the three columns to check:
+timelog = timelog.dropna(subset=['started', 'recorded', 'description'])
 timelog = timelog.reset_index(drop=True)
 
-# For debugging, keep original description around.
+if debug:
-timelog["orig_desc"] = timelog["description"]
+   dropna = copy.deepcopy(timelog)
+
+   # For debugging, keep originals around.
+   timelog["orig_desc"] = timelog["description"]
+   timelog["orig_started"] = timelog["started"]
+   timelog["orig_recorded"] = timelog["recorded"]
 
 # Clean up description before we go to work on it.
 timelog['description'] = timelog['description'].str.strip()
@@ -44,6 +62,9 @@ timelog['description'] = timelog['description'].str.strip()
 timelog["description"] = list(timelog["description"].str.split(";"))
 timelog = timelog.explode("description").reset_index()
 
+if debug:
+    mess = copy.deepcopy(timelog)
+
 timelog["started"] = pd.to_datetime(timelog["started"]).dt.tz_convert("US/Eastern")
 timelog["recorded"] = pd.to_datetime(timelog["recorded"]).dt.tz_convert("US/Eastern")
 
@@ -112,7 +133,7 @@ for preferred, alternatives in replacement_project_names.items():
 # The fillna is essential or we drop entries with blank ('None') projects.
 tl = timelog.groupby(["date", timelog.project.fillna(""), "description"]).agg({"time": 'sum', "started": 'min', "recorded": 'max'}).reset_index()
 
-if not hasattr(sys, 'ps1'):
+if debug:
     tl.to_csv('harvest-ready.csv', index=False)
     settings.pomodoro_latest_recorded(tl.recorded.max())
 else:

settings.py

@@ -14,7 +14,7 @@ def write():
 
 if not os.path.isfile('settings.ini'):
     # Set some essential initial values.
-    # pomodoro['logfile'] = '~/Projects/agaric/python/pomodoroprompt/log/2021.csv'
+    pomodoro['logfile'] = '~/Projects/agaric/python/pomodoroprompt/log/timelog.csv'
     # This path must be absolute, for some reason using ~ for home isn  t wor::
     pomodoro['logpath'] = '/home/mlncn/Projects/agaric/python/pomodoroprompt/log/'
     write()