Update tools.py

* Improved date logic

pydbsmgr/lightest.py

@@ -17,10 +17,20 @@ def clean_frame(self) -> DataFrame:
         for column_index, datatype in enumerate(table.dtypes):
             if datatype == "object" or datatype == "datetime64[ns]":
                 x = (table[cols[column_index]].values)[0]
+                datetype_column = True
                 if isinstance(x, str):
-                    if (x.find("/") != -1 or x.find("-")) != -1 and not (
-                        x.find("//") or x.find("\\")
-                    ) != -1:
+                    if (
+                        x == ""
+                        or x.find("/") != -1
+                        or x.find("-") != -1
+                        or x == np.datetime64("NaT")
+                    ):
+                        datetype_column = (
+                            (table[cols[column_index]].apply(check_if_contains_dates))
+                            .isin([True])
+                            .any()
+                        )
+                    if not (x.find("//") or x.find("\\")) != -1 and datetype_column:
                         with concurrent.futures.ThreadPoolExecutor() as executor:
                             table[cols[column_index]] = list(
                                 executor.map(clean_and_convert_to, table[cols[column_index]])

pydbsmgr/main.py

@@ -22,6 +22,17 @@
 ########################################################################################
 
 
+def check_if_contains_dates(input_string: str) -> bool:
+    """Check if a string contains date."""
+    if input_string == "":
+        return False
+    else:
+        if re.search(r"\d{4}(-|/)\d{1,2}(-|/)\d{1,2}", str(input_string)):
+            return True
+        else:
+            return False
+
+
 def remove_numeric_char(input_string: str) -> str:
     """Remove all numeric characters from a string.
 

pydbsmgr/utils/tools.py

@@ -15,7 +15,7 @@
 from pandas.errors import IntCastingNaNError
 from pyarrow import Table
 
-from pydbsmgr.main import is_number_regex
+from pydbsmgr.main import is_number_regex, check_if_contains_dates
 
 
 class ColumnsCheck:
@@ -280,10 +280,18 @@ def _check_datetime(self) -> None:
             col = cols[column_index]
             if datatype == "object":
                 x = (df_[col].values)[0]
+                datetype_column = True
                 if isinstance(x, str):
-                    if (x.find("/") != -1 or x.find("-")) != -1 and not (
-                        x.find("//") or x.find("\\")
-                    ) != -1:
+                    if (
+                        x == ""
+                        or x.find("/") != -1
+                        or x.find("-") != -1
+                        or x == np.datetime64("NaT")
+                    ):
+                        datetype_column = (
+                            (df_[col].apply(check_if_contains_dates)).isin([True]).any()
+                        )
+                    if not (x.find("//") or x.find("\\")) != -1 and datetype_column:
                         try:
                             with concurrent.futures.ThreadPoolExecutor() as executor:
                                 df_[col] = list(

setup.py

@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="pydbsmgr",
-    version="0.5.7",
+    version="0.5.8",
     author="J. A. Moreno-Guerra",
     author_email="[email protected]",
     description="Testing installation of Package",