From 86282b1ab7f2d5befb550c0ee7953aa930b547d0 Mon Sep 17 00:00:00 2001
From: phga <toerd@posteo.de>
Date: Wed, 20 Oct 2021 18:11:14 +0200
Subject: [PATCH] feat: exception checking

---
 melli/excel2sql.py | 34 +++++++++++++++++++---------------
 1 file changed, 19 insertions(+), 15 deletions(-)

diff --git a/melli/excel2sql.py b/melli/excel2sql.py
index a325b7b..b170497 100755
--- a/melli/excel2sql.py
+++ b/melli/excel2sql.py
@@ -1,23 +1,20 @@
 #!/bin/python3
 import pandas as pd
-import numpy as np
 from argparse import ArgumentParser
 
+# Wraps strings and timestamps into single quotes
 def wrap_types(val):
-    if np.isnan(val):
-        print("NaN FOUND")
-        val = "NULL"
-    if type(val) in [str, pd._libs.tslibs.timestamps.Timestamp]:
+    if type(val) in [str, pd._libs.tslibs.timestamps.Timestamp] and val != "NULL":
         return f"'{val}'"
     else:
         return val
 
 p = ArgumentParser()
-# action store_true is for flags
 p.add_argument("-i", "--input", nargs="+", help="Path to input excel file")
 p.add_argument("-o", "--output", nargs="+", help="Path to output sql file")
 args = p.parse_args()
 
+# Manual fallback if argument is not specified
 if not args.input:
     in_file_path = input("Please enter path to input excel file: ")
 else:
@@ -28,20 +25,26 @@ if not args.output:
 else:
     out_file_path = args.output[0]
 
+# Simple input argument checks
+if out_file_path.endswith(("xlsx", "xls")):
+    print("Aborting task. Possible overwrite of source file detected.")
+    print(f"Specified output file '{out_file_path}' ends with .xls/.xlsx")
+    exit(1)
+
+if not in_file_path.endswith(("xlsx", "xls")):
+    print("Aborting task. No input file in excel format specified.")
+    exit(1)
 
 print(f'Input: {in_file_path}\nOutput: {out_file_path}')
 
 in_file = pd.ExcelFile(in_file_path)
-print(in_file.parse("Datei", skiprows=1).head())
 
 with open(out_file_path, "w") as of:
     for sheet in in_file.sheet_names:
-        curr_meta = in_file.parse(sheet, nrows=1, header=None)
-        curr_sheet = in_file.parse(sheet, header=1)
-        curr_meta = curr_meta[0][0].split("#")
-        db = curr_meta[0]
-        table_prefix = curr_meta[1]
-        table_suffix = sheet
+        print(f'Processing sheet "{sheet}"...')
+        db_table = in_file.parse(sheet, nrows=1, header=None)[0][0]
+        curr_sheet = in_file.parse(sheet, header=1).fillna("NULL")
+        count = 0
         # One line is one test subject/object
         for line in curr_sheet.iterrows():
             indices = line[1].index
@@ -49,7 +52,6 @@ with open(out_file_path, "w") as of:
             cols = ""
             vals = ""
             for i in range(len(indices)):
-                # print(type(values[i]), ": ", values[i])
                 if i != 0:
                     cols += f", {indices[i]}"
                     vals += f", {wrap_types(values[i])}"
@@ -57,4 +59,6 @@ with open(out_file_path, "w") as of:
                     cols += f"{indices[i]}"
                     vals += f"{wrap_types(values[i])}"
             # INSERT INTO $DB.$TABLE ($COL1, $COL2, ...) VALUES ($VAL1, $VAL2, ...)
-            of.write(f'INSERT INTO "{db}"."{table_prefix}.{table_suffix}" ({cols}) VALUES ({vals})\n')
+            of.write(f'INSERT INTO {db_table} ({cols}) VALUES ({vals})\n')
+            count += 1
+        print(f'Created {count} INSERT statements for sheet "{sheet}"\n')