[FIX] Update values of oxidation states (#202)

* update values of oxidation states

* update number of rows

* update reference

* add utils to update oxidation states

docs/source/references.bib

@@ -831,9 +831,9 @@ @misc{NIST-CH-WB
 @misc{enwiki:1102394064,
   author = {{Wikipedia contributors}},
   title  = {Oxidation state --- {Wikipedia}{,} The Free Encyclopedia},
-  year   = {2022},
-  url    = {https://en.wikipedia.org/w/index.php?title=Oxidation_state&oldid=1102394064},
-  note   = {[Online; accessed 28-September-2022]}
+  year   = {2024},
+  url    = {https://en.wikipedia.org/w/index.php?title=Oxidation_state&oldid=1253639950},
+  note   = {[Online; accessed 4-November-2024]}
 }
 @book{haynes2016crc,
   title     = {CRC Handbook of Chemistry and Physics},

mendeleev/interfaces/wiki_oxidation_states.py

@@ -0,0 +1,112 @@
+import requests
+import pandas as pd
+import polars as pl
+from bs4 import BeautifulSoup
+from bs4.element import Tag
+
+from mendeleev.models import OxidationState
+from mendeleev.db import get_session
+
+
+def fetch_oxidation_states_table() -> Tag:
+    "Fetch the table with oxidation states from wikipedia"
+
+    url = "https://en.wikipedia.org/wiki/Oxidation_state"
+    response = requests.get(url)
+    if response.status_code == 200:
+        html_content = response.text
+        soup = BeautifulSoup(html_content, "html.parser")
+        table_selector = "table.wikitable.sortable"
+
+        if table := soup.select_one(table_selector):
+            print("Table found and extracted!")
+            return table
+        else:
+            print("Table not found with the specified CSS selector.")
+    else:
+        print(f"Failed to retrieve page. Status code: {response.status_code}")
+
+
+def parse_oxidation_states_table(table: Tag) -> list[dict]:
+    """
+    Parse oxidation states from a table into a DataFrame.
+
+    Args:
+        table (bs4.element.Tag): The BeautifulSoup Tag object for the table.
+
+    Returns:
+        pd.DataFrame: A DataFrame with Z values as the first column and oxidation states from -5 to 9 as other columns.
+    """
+    parsed_rows = []
+
+    for row in table.find_all("tr")[4:]:  # Skip the header rows
+        # Extract the atomic number (Z value)
+        z_value = row.find_all("td")[0].text.strip()
+        if not z_value.isdigit():
+            continue
+        z_value = int(z_value)
+
+        oxidation_states = {"Z": z_value}
+        for i in range(-5, 10):
+            oxidation_states[i] = None
+
+        # iterate over oxidation state cells, starting from the 4th cell to the 18th
+        for i, cell in enumerate(row.find_all("td")[3:18], start=-5):
+            if cell.text.strip():
+                if cell.find("b"):
+                    oxidation_states[i] = True
+                else:
+                    oxidation_states[i] = False
+        parsed_rows.append(oxidation_states)
+    return parsed_rows
+
+
+def create_pandas(data: list[dict]) -> pd.DataFrame:
+    "Create a pandas dataframe"
+    ox = pd.DataFrame(data)
+    ox = (
+        ox.melt(id_vars=["Z"], value_vars=list(range(-5, 10)))
+        .dropna(axis="rows", subset=["value"])
+        .sort_values(by="Z")
+        .reset_index(drop=True)
+    )
+    ox.loc[:, "category"] = ox.value.map({True: "main", False: "extended"})
+    return ox.rename(
+        columns={"Z": "atomic_number", "variable": "oxidation_state"}
+    ).drop(columns=["value"], axis="columns")
+
+
+def create_polars(data: list[dict]) -> pl.DataFrame:
+    "Create a polars dataframe"
+    parsed_str = [{str(k): v for k, v in d.items()} for d in data]
+    ox = pl.from_dicts(parsed_str)
+    return (
+        ox.melt(id_vars=["Z"])
+        .drop_nulls(subset=["value"])
+        .rename(
+            {"Z": "atomic_number", "variable": "oxidation_state", "value": "category"}
+        )
+        .cast({"oxidation_state": pl.Int32})
+        .with_columns(pl.col("category").map_dict({True: "main", False: "extended"}))
+    )
+
+
+if __name__ == "__main__":
+    table = fetch_oxidation_states_table()
+    parsed = parse_oxidation_states_table(table)
+    ox = create_pandas(parsed)
+
+    update = False
+    if update:
+        session = get_session(read_only=False)
+        objects = [
+            OxidationState(
+                atomic_number=row["atomic_number"],
+                oxidation_state=row["oxidation_state"],
+                category=row["category"],
+            )
+            for i, row in ox.iterrows()
+        ]
+        session.add_all(objects)
+        session.commit()
+        session.close()

mendeleev/models.py

@@ -428,13 +428,15 @@ def oxidation_states(self, category: str = "main") -> List[int]:
             )
 
         if category == "all":
-            return [o.oxidation_state for o in self._oxidation_states]
+            return sorted([o.oxidation_state for o in self._oxidation_states])
         else:
-            return [
-                o.oxidation_state
-                for o in self._oxidation_states
-                if o.category == category
-            ]
+            return sorted(
+                [
+                    o.oxidation_state
+                    for o in self._oxidation_states
+                    if o.category == category
+                ]
+            )
 
     def zeff(
         self, n: int = None, o: str = None, method: str = "slater", alle: bool = False

tests/test_fetch.py

@@ -16,7 +16,7 @@
     ("ionizationenergies", 5847),
     ("groups", 18),
     ("series", 10),
-    ("oxidationstates", 579),
+    ("oxidationstates", 601),
     ("phasetransitions", 108),
 ]