📈 NEPSE All Scraper

A free, open-source data pipeline for the Nepal Stock Exchange.
Automatically scrapes prices, dividends, right shares, and floorsheet data
for 365 listed companies — committed to this repo every weekday via GitHub Actions.

🚀 Live API + Swagger UI: https://samirwagle.github.io/Nepse-All-Scraper/

---

📦 What's Inside

This repo is data-first. Every weekday after NEPSE closes, GitHub Actions scrapes the latest data and commits it directly back to the data/ folder. No database, no server — just flat CSV files you can plug into anything.

Data	Where	Updated
OHLC price history	data/company-wise/{SYMBOL}/prices.csv	Locally (run once)
Dividend history	data/company-wise/{SYMBOL}/dividend.csv	Every weekday ✅
Right share history	data/company-wise/{SYMBOL}/right-share.csv	Every weekday ✅
Full daily floorsheet	data/floorsheet_YYYY-MM-DD.csv + .json	Every weekday ✅

---

�️ Repository Layout

Nepse-All-Scraper/
│
├── .github/
│   └── workflows/
│       └── daily_scraper.yml      # GitHub Actions — runs every weekday at 6:30 PM NPT
│
├── data/
│   ├── company_list.json          # 365 priority company symbols
│   ├── company_id_mapping.json    # Symbol → ShareSansar internal ID
│   ├── floorsheet_YYYY-MM-DD.csv  # Daily floorsheet (all trades)
│   ├── floorsheet_YYYY-MM-DD.json # Same data as JSON
│   └── company-wise/
│       └── {SYMBOL}/
│           ├── prices.csv         # Full OHLC price history
│           ├── dividend.csv       # Dividend history
│           └── right-share.csv    # Right share history
│
└── scraper/
    ├── run_github_actions.py      # ← GitHub Actions entry point
    ├── run_daily.py               # ← Local price scraper CLI
    └── core/
        ├── daily.py               # Orchestrates price scraping
        ├── daily_prices.py        # Daily price summary updater
        ├── floorsheet.py          # Floorsheet scraper (merolagani.com)
        └── history.py             # OHLC price history scraper

---

🤖 Automation — GitHub Actions

The workflow .github/workflows/daily_scraper.yml runs automatically every weekday (Mon–Fri) at 6:30 PM Nepal time (12:45 UTC), right after NEPSE closes.

┌─────────────────────────────────────────────────────┐
│              GitHub Actions — Daily Run              │
├─────────────┬───────────────────────────────────────┤
│  Dividends  │  Updates dividend.csv (all 365)        │
│ Right Shares│  Updates right-share.csv (all 365)     │
│  Floorsheet │  Full day's trades from merolagani.com │
│  Commit     │  git push → data/ auto-updated in repo │
└─────────────┴───────────────────────────────────────┘

Trigger manually: GitHub → Actions → Daily Scraper → Run workflow

---

⚡ Quickstart

Prerequisites

pip install requests beautifulsoup4

Run the same scrape as GitHub Actions (dividends + right shares + floorsheet)

# All three in one go
python scraper/run_github_actions.py

# Or individually
python scraper/run_github_actions.py --dividends
python scraper/run_github_actions.py --right-shares
python scraper/run_github_actions.py --floorsheet

# Test floorsheet with limited pages (faster)
python scraper/run_github_actions.py --floorsheet --max-pages 3

Scrape full OHLC price history (local only, first-time)

# Full history for all 365 companies — takes ~2-4 hours on first run
python scraper/run_daily.py --full-scrape

# Incremental — only fetches newer records than what's already in prices.csv
python scraper/run_daily.py --incremental

# Only process newly listed companies (new IPOs)
python scraper/run_daily.py --new-only

Why are prices local-only?
Price scraping needs the existing prices.csv files to know where to stop (incremental logic). Run it locally once, push the data, then automation handles daily updates.

---

🌐 Public API (free, no auth)

Two equivalent ways to consume the data — pick whichever your client likes.

A. Swagger / JSON via GitHub Pages

https://samirwagle.github.io/Nepse-All-Scraper/                 # Swagger UI
https://samirwagle.github.io/Nepse-All-Scraper/openapi.json     # OpenAPI 3.1 spec
https://samirwagle.github.io/Nepse-All-Scraper/api/companies.json
https://samirwagle.github.io/Nepse-All-Scraper/api/latest.json
https://samirwagle.github.io/Nepse-All-Scraper/api/prices/ADBL.json
https://samirwagle.github.io/Nepse-All-Scraper/api/dividends/ADBL.json
https://samirwagle.github.io/Nepse-All-Scraper/api/right-shares/NABIL.json
https://samirwagle.github.io/Nepse-All-Scraper/api/floorsheet/index.json
https://samirwagle.github.io/Nepse-All-Scraper/api/status.json  # health snapshot

B. Raw CSV via jsDelivr CDN (faster, no rate limit)

jsDelivr auto-mirrors public GitHub repos on a global CDN. Same files, faster, no GitHub rate limits, automatic gzip:

https://cdn.jsdelivr.net/gh/SamirWagle/Nepse-All-Scraper@main/data/company_list.json
https://cdn.jsdelivr.net/gh/SamirWagle/Nepse-All-Scraper@main/data/company-wise/ADBL/prices.csv
https://cdn.jsdelivr.net/gh/SamirWagle/Nepse-All-Scraper@main/data/company-wise/ADBL/dividend.csv
https://cdn.jsdelivr.net/gh/SamirWagle/Nepse-All-Scraper@main/data/floorsheet/floorsheet_2026-05-04.csv

Symbols with / in the name (e.g. GBILD86/87) live on disk as GBILD86-87. Use the dashed form in URLs.

Quick examples

# Python
import requests
prices = requests.get("https://samirwagle.github.io/Nepse-All-Scraper/api/prices/ADBL.json").json()
print(prices["count"], prices["data"][-1])

// JavaScript / TypeScript
const r = await fetch("https://samirwagle.github.io/Nepse-All-Scraper/api/latest.json");
const latest = await r.json();
console.log(latest["NABIL"]);

# CLI
curl https://cdn.jsdelivr.net/gh/SamirWagle/Nepse-All-Scraper@main/data/company-wise/NABIL/dividend.csv

---

� Data Formats

prices.csv

date, open, high, low, ltp, percent_change, qty, turnover
2024-01-15, 1200, 1250, 1190, 1240, +1.5%, 3400, 4216000

dividend.csv

fiscal_year, bonus_share, cash_dividend, total_dividend, book_closure_date
2079/80, 10%, 5%, 15%, 2023-12-01

right-share.csv

ratio, total_units, issue_price, opening_date, closing_date, status, issue_manager
1:1, 5000000, 100, 2023-11-01, 2023-11-15, Closed, XYZ Capital

floorsheet_YYYY-MM-DD.csv

date, sn, contract_no, stock_symbol, buyer, seller, quantity, rate, amount
2024-01-15, 1, 100012345, ADBL, 21, 42, 500, 1240, 620000

---

⚙️ How Incremental Scraping Works

prices.csv                    ShareSansar AJAX
─────────────                 ──────────────────
Latest date: 2024-01-10  →   Stop fetching when
                              record date ≤ 2024-01-10

Result: Only 1-2 pages fetched instead of 70+

history.py reads the most recent date from the existing prices.csv, passes it as a stop_date to the paginator, and halts the moment it hits older data. This makes daily updates take seconds instead of hours.

---

🚀 First-Time Setup

# 1. Clone the repo
git clone https://github.com/SamirWagle/Nepse-All-Scraper.git
cd Nepse-All-Scraper

# 2. Install dependencies
pip install requests beautifulsoup4

# 3. Run the full price history scrape (one-time, takes 2-4 hours)
python scraper/run_daily.py --full-scrape

# 4. Push everything to the repo
git add data/
git commit -m "chore: initial data load"
git push

From that point on, GitHub Actions handles everything automatically every weekday. ✅

---

🗺️ Roadmap

• Phase 1 — NEPSE Scraper (prices, dividends, right shares, floorsheet)
• Phase 2 — GitHub Actions automation + incremental updates
• Phase 3 — Frontend / API layer

Want to help build Phase 3? PRs are welcome.

---

⚠️ Disclaimer

This project is for educational purposes only.
Data is sourced from publicly available websites (ShareSansar, Merolagani).
Not financial advice. Do your own research before making investment decisions.

---

Made with ❤️ for the Nepali investor community