Spaces:
Running
Running
Create csvloader.py
Browse files- app/csvloader.py +144 -0
app/csvloader.py
ADDED
|
@@ -0,0 +1,144 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# ==============================
|
| 2 |
+
# Imports
|
| 3 |
+
# ==============================
|
| 4 |
+
import requests
|
| 5 |
+
import zipfile
|
| 6 |
+
from io import BytesIO, StringIO
|
| 7 |
+
from datetime import datetime as dt
|
| 8 |
+
from typing import Dict, Union
|
| 9 |
+
|
| 10 |
+
import pandas as pd
|
| 11 |
+
|
| 12 |
+
from persist import exists, load, save
|
| 13 |
+
|
| 14 |
+
|
| 15 |
+
# ==============================
|
| 16 |
+
# Raw CSV Loader (NO parsing)
|
| 17 |
+
# ==============================
|
| 18 |
+
def load_csv(url: str) -> Union[str, Dict[str, str]]:
|
| 19 |
+
"""
|
| 20 |
+
Pure transport loader
|
| 21 |
+
- .csv -> raw CSV text (str)
|
| 22 |
+
- .zip -> {filename: raw CSV text}
|
| 23 |
+
|
| 24 |
+
NO parsing
|
| 25 |
+
NO cleaning
|
| 26 |
+
NO assumptions
|
| 27 |
+
"""
|
| 28 |
+
r = requests.get(url)
|
| 29 |
+
r.raise_for_status()
|
| 30 |
+
|
| 31 |
+
if url.lower().endswith(".zip"):
|
| 32 |
+
z = zipfile.ZipFile(BytesIO(r.content))
|
| 33 |
+
out: Dict[str, str] = {}
|
| 34 |
+
|
| 35 |
+
for name in z.namelist():
|
| 36 |
+
if name.lower().endswith(".csv"):
|
| 37 |
+
with z.open(name) as f:
|
| 38 |
+
out[name] = f.read().decode("utf-8", errors="ignore")
|
| 39 |
+
return out
|
| 40 |
+
|
| 41 |
+
return r.text
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
# ==============================
|
| 45 |
+
# NSE High-Low HTML Formatter
|
| 46 |
+
# ==============================
|
| 47 |
+
def _highlow_html_formatter(df: pd.DataFrame, date_str: str) -> str:
|
| 48 |
+
metric = "PERCENT_CHANGE"
|
| 49 |
+
df_html = df.copy()
|
| 50 |
+
|
| 51 |
+
top_up = df[metric].nlargest(3).index if metric in df else []
|
| 52 |
+
top_dn = df[metric].nsmallest(3).index if metric in df else []
|
| 53 |
+
|
| 54 |
+
for idx, row in df_html.iterrows():
|
| 55 |
+
for col in df_html.columns:
|
| 56 |
+
val = row[col]
|
| 57 |
+
style = ""
|
| 58 |
+
|
| 59 |
+
if isinstance(val, (int, float)):
|
| 60 |
+
txt = f"{val:.2f}"
|
| 61 |
+
if val > 0:
|
| 62 |
+
style = "pos"
|
| 63 |
+
elif val < 0:
|
| 64 |
+
style = "neg"
|
| 65 |
+
|
| 66 |
+
if col == metric:
|
| 67 |
+
if idx in top_up:
|
| 68 |
+
style += " top-up"
|
| 69 |
+
elif idx in top_dn:
|
| 70 |
+
style += " top-down"
|
| 71 |
+
|
| 72 |
+
df_html.at[idx, col] = f'<span class="{style.strip()}">{txt}</span>'
|
| 73 |
+
else:
|
| 74 |
+
df_html.at[idx, col] = str(val)
|
| 75 |
+
|
| 76 |
+
return f"""
|
| 77 |
+
<!DOCTYPE html>
|
| 78 |
+
<html>
|
| 79 |
+
<head>
|
| 80 |
+
<meta charset="UTF-8">
|
| 81 |
+
<title>NSE High-Low {date_str}</title>
|
| 82 |
+
<style>
|
| 83 |
+
body {{ font-family: Arial; background:#f5f5f5; padding:12px; }}
|
| 84 |
+
table {{ border-collapse: collapse; width:100%; background:white; }}
|
| 85 |
+
th, td {{ border:1px solid #bbb; padding:6px; font-size:13px; }}
|
| 86 |
+
th {{ background:#222; color:white; }}
|
| 87 |
+
.pos {{ color:green; font-weight:bold; }}
|
| 88 |
+
.neg {{ color:red; font-weight:bold; }}
|
| 89 |
+
.top-up {{ background:#b6f2b6; }}
|
| 90 |
+
.top-down {{ background:#f2b6b6; }}
|
| 91 |
+
</style>
|
| 92 |
+
</head>
|
| 93 |
+
<body>
|
| 94 |
+
<h2>NSE Index High / Low — {date_str}</h2>
|
| 95 |
+
{df_html.to_html(index=False, escape=False)}
|
| 96 |
+
</body>
|
| 97 |
+
</html>
|
| 98 |
+
"""
|
| 99 |
+
|
| 100 |
+
|
| 101 |
+
# ==============================
|
| 102 |
+
# NSE High-Low Master Function
|
| 103 |
+
# ==============================
|
| 104 |
+
def nse_highlow(date_str: str | None = None) -> str:
|
| 105 |
+
"""
|
| 106 |
+
Master NSE High-Low function
|
| 107 |
+
|
| 108 |
+
Responsibilities:
|
| 109 |
+
- Knows NSE CSV structure
|
| 110 |
+
- Header starts at row index 2 (skip 0 & 1)
|
| 111 |
+
- Uses raw CSV loader
|
| 112 |
+
- Builds HTML
|
| 113 |
+
- Persists ONLY HTML
|
| 114 |
+
"""
|
| 115 |
+
if not date_str:
|
| 116 |
+
date_str = dt.now().strftime("%d-%m-%Y")
|
| 117 |
+
|
| 118 |
+
cache_key = f"highlow_{date_str}"
|
| 119 |
+
|
| 120 |
+
if exists(cache_key, "html"):
|
| 121 |
+
return load(cache_key, "html")
|
| 122 |
+
|
| 123 |
+
d = dt.strptime(date_str, "%d-%m-%Y")
|
| 124 |
+
url = (
|
| 125 |
+
"https://archives.nseindia.com/content/indices/"
|
| 126 |
+
f"ind_close_all_{d.strftime('%d%m%Y')}.csv"
|
| 127 |
+
)
|
| 128 |
+
|
| 129 |
+
# 1️⃣ Load raw CSV text
|
| 130 |
+
csv_text = load_csv(url)
|
| 131 |
+
|
| 132 |
+
# 2️⃣ NSE-specific parsing (header row = 2)
|
| 133 |
+
df = pd.read_csv(
|
| 134 |
+
StringIO(csv_text),
|
| 135 |
+
header=0
|
| 136 |
+
)
|
| 137 |
+
|
| 138 |
+
# 3️⃣ Build HTML
|
| 139 |
+
html = _highlow_html_formatter(df, date_str)
|
| 140 |
+
|
| 141 |
+
# 4️⃣ Persist HTML only
|
| 142 |
+
save(cache_key, html, "html")
|
| 143 |
+
|
| 144 |
+
return html
|