Update CSV py

CSV py

@@ -1,134 +1,91 @@
 # CSV.py
+
 import pandas as pd
-from typing import List, Union, Optional
-from io import BytesIO
-import zipfile
 import requests
+import zipfile
+from io import BytesIO
 from datetime import datetime as dt
+from typing import List, Union
+
 from persist import exists, load, save
 
 
-# ==========================
-# CSV / ZIP Loader
-# ==========================
 def load_csv(
-    url_or_path: str,
+    url: str,
     header_row: int = 0,
-    drop_unnamed: bool = True,
-    text_cols: Optional[List[str]] = None
+    text_cols: List[str] | None = None
 ) -> Union[pd.DataFrame, List[pd.DataFrame]]:
     """
-    Load CSV or ZIP of CSVs.
-    - If URL/path ends with .csv → returns single DataFrame
-    - If URL/path ends with .zip → returns list of DataFrames (one per CSV inside)
-    - header_row: row index for CSV header
-    - drop_unnamed: drop unnamed columns
-    - text_cols: columns to treat as text (optional)
+    Load CSV or ZIP containing CSVs
+    - .csv -> DataFrame
+    - .zip -> List[DataFrame]
     """
     text_cols = text_cols or []
 
-    # --- ZIP case ---
-    if url_or_path.lower().endswith(".zip"):
-        if url_or_path.startswith("http"):
-            r = requests.get(url_or_path)
-            r.raise_for_status()
-            z = zipfile.ZipFile(BytesIO(r.content))
-        else:
-            z = zipfile.ZipFile(url_or_path)
-
+    def _clean_df(df: pd.DataFrame) -> pd.DataFrame:
+        df = df.loc[:, ~df.columns.str.contains("^Unnamed")]
+        df.columns = (
+            df.columns
+            .str.strip()
+            .str.replace(" ", "_")
+            .str.replace("-", "_")
+        )
+        for col in df.columns:
+            if col not in text_cols:
+                df[col] = pd.to_numeric(df[col], errors="coerce")
+        return df.dropna(how="all")
+
+    if url.lower().endswith(".zip"):
+        r = requests.get(url)
+        r.raise_for_status()
+        z = zipfile.ZipFile(BytesIO(r.content))
         dfs = []
         for name in z.namelist():
             if name.lower().endswith(".csv"):
                 with z.open(name) as f:
                     df = pd.read_csv(f, header=header_row)
-                    df = _clean_df(df, drop_unnamed, text_cols)
-                    dfs.append(df)
+                    dfs.append(_clean_df(df))
         return dfs
 
-    # --- Single CSV case ---
-    else:
-        if url_or_path.startswith("http"):
-            df = pd.read_csv(url_or_path, header=header_row)
-        else:
-            df = pd.read_csv(url_or_path, header=header_row)
-        return _clean_df(df, drop_unnamed, text_cols)
-
-
-def _clean_df(df: pd.DataFrame, drop_unnamed: bool, text_cols: List[str]) -> pd.DataFrame:
-    """Helper to clean columns, convert numeric, drop empty rows"""
-    if drop_unnamed:
-        df = df.loc[:, ~df.columns.str.contains("^Unnamed")]
-
-    df.columns = (
-        df.columns
-        .str.strip()
-        .str.replace(" ", "_")
-        .str.replace("-", "_")
-    )
+    df = pd.read_csv(url, header=header_row)
+    return _clean_df(df)
 
-    for col in df.columns:
-        if col not in text_cols:
-            df[col] = pd.to_numeric(df[col], errors="coerce")
-
-    return df.dropna(how="all")
-
-
-# ==========================
-# HTML Generator
-# ==========================
-def df_to_html(
-    df: pd.DataFrame,
-    metric_col: Optional[str] = None,
-    cache_key: Optional[str] = None
-) -> str:
-    """
-    Convert DataFrame to HTML table with colored numeric values.
-    - metric_col: highlights top 3 / bottom 3 for this column
-    - cache_key: optional persist key for caching HTML
-    """
-    if df is None or df.empty:
-        return "<p>No data available.</p>"
-
-    # --- CACHE CHECK ---
-    if cache_key and exists(cache_key, "html"):
-        return load(cache_key, "html")
 
+def _highlow_html_formatter(df: pd.DataFrame, date_str: str) -> str:
+    metric = "PERCENT_CHANGE"
     df_html = df.copy()
-    top3_up, top3_down = [], []
 
-    if metric_col and metric_col in df_html.columns:
-        col_numeric = pd.to_numeric(df_html[metric_col], errors="coerce").dropna()
-        top3_up = col_numeric.nlargest(3).index.tolist()
-        top3_down = col_numeric.nsmallest(3).index.tolist()
+    top_up = df[metric].nlargest(3).index if metric in df else []
+    top_dn = df[metric].nsmallest(3).index if metric in df else []
 
     for idx, row in df_html.iterrows():
         for col in df_html.columns:
             val = row[col]
             style = ""
             if isinstance(val, (int, float)):
-                val_fmt = f"{val:.2f}"
+                txt = f"{val:.2f}"
                 if val > 0:
                     style = "pos"
                 elif val < 0:
                     style = "neg"
-                if col == metric_col:
-                    if idx in top3_up:
+                if col == metric:
+                    if idx in top_up:
                         style += " top-up"
-                    elif idx in top3_down:
+                    elif idx in top_dn:
                         style += " top-down"
-                df_html.at[idx, col] = f'<span class="{style.strip()}">{val_fmt}</span>'
+                df_html.at[idx, col] = f'<span class="{style.strip()}">{txt}</span>'
             else:
                 df_html.at[idx, col] = str(val)
 
-    html_out = f"""
+    return f"""
 <!DOCTYPE html>
 <html>
 <head>
 <meta charset="UTF-8">
-<title>CSV Table</title>
+<title>NSE High-Low {date_str}</title>
 <style>
 body {{ font-family: Arial; background:#f5f5f5; padding:12px; }}
-table {{ border-collapse: collapse; width: 100%; background:white; }}
+table {{ border-collapse: collapse; width:100%; background:white; }}
 th, td {{ border:1px solid #bbb; padding:6px; font-size:13px; }}
 th {{ background:#222; color:white; }}
 .pos {{ color:green; font-weight:bold; }}
@@ -138,42 +95,40 @@ th {{ background:#222; color:white; }}
 </style>
 </head>
 <body>
+<h2>NSE Index High / Low — {date_str}</h2>
 {df_html.to_html(index=False, escape=False)}
 </body>
 </html>
 """
 
-    # --- SAVE CACHE ---
-    if cache_key:
-        save(cache_key, html_out, "html")
-
-    return html_out
-
 
-# ==========================
-# NSE High-Low Master
-# ==========================
-def nse_highlow(date_str: str = None) -> str:
+def nse_highlow(date_str: str | None = None) -> str:
     """
-    Master function to fetch NSE High-Low CSV and return HTML.
-    - date_str: "DD-MM-YYYY", default today
-    - uses caching (persist.py) for HTML
+    Master NSE High-Low function
+    - Uses load_csv
+    - Builds HTML
+    - Persists ONLY HTML
     """
-    if date_str is None:
+    if not date_str:
         date_str = dt.now().strftime("%d-%m-%Y")
 
-    # Cache key
     cache_key = f"highlow_{date_str}"
 
-    # --- URL for NSE High-Low CSV ---
-    dt_obj = dt.strptime(date_str, "%d-%m-%Y")
-    url_date = dt_obj.strftime("%d%m%Y")
-    url = f"https://archives.nseindia.com/content/indices/ind_close_all_{url_date}.csv"
+    if exists(cache_key, "html"):
+        return load(cache_key, "html")
 
-    # --- Load CSV using load_csv ---
-    df = load_csv(url, header_row=2, text_cols=["Index_Name", "Index_Date"])
+    d = dt.strptime(date_str, "%d-%m-%Y")
+    url = (
+        "https://archives.nseindia.com/content/indices/"
+        f"ind_close_all_{d.strftime('%d%m%Y')}.csv"
+    )
 
-    # --- Generate HTML using df_to_html ---
-    html = df_to_html(df, metric_col="PERCENT_CHANGE", cache_key=cache_key)
+    df = load_csv(
+        url=url,
+        header_row=2,
+        text_cols=["Index_Name", "Index_Date"]
+    )
 
+    html = _highlow_html_formatter(df, date_str)
+    save(cache_key, html, "html")
     return html