Quickstart Guide

This guide gets you up and running with nse-xbrl-parser in 2 minutes.

1. Auto-Detect and Parse Any Filing (Recommended)

Give any XML file path, raw string, or bytes to parse (async) or parse_sync (synchronous). The library automatically detects whether the document is Financial Results, Shareholding Pattern, Corporate Governance, or Investor Complaints, and gives you direct access to typed data or universal facts:

import asyncio
from nse_xbrl_parser import parse, FilingFormat

async def main():
    # Pass a path, raw XML string, or bytes - it auto-detects everything!
    filing = await parse("filing.xml")

    print("Detected Format:", filing.format)
    # Output: FilingFormat.FINANCIAL_RESULTS, SHAREHOLDING_PATTERN, etc.

    print("Company:", filing.company_name)
    print("Scrip Code:", filing.scrip_code)

    # Access the typed model directly via .data
    if filing.format == FilingFormat.FINANCIAL_RESULTS:
        print(f"Revenue: ₹{filing.data.revenue_from_operations:,.2f}")
        print(f"Net Profit: ₹{filing.data.profit_after_tax:,.2f}")
        print(f"Basic EPS: ₹{filing.data.basic_eps:.2f}")

    # Or export everything as a clean dictionary
    summary = filing.to_dict()

asyncio.run(main())

Synchronous one-liner:

from nse_xbrl_parser import parse_sync

filing = parse_sync("filing.xml")
print(filing.format, filing.company_name, filing.data)

---

2. Universal XBRL Fact Engine

If you are working with arbitrary or custom filings, or want to query the underlying facts directly, access filing.raw_instance or use parse_file:

import asyncio
from nse_xbrl_parser import parse_file

async def main():
    instance = await parse_file("sample_filing.xml")
    
    # Check total facts parsed
    print(f"Extracted {len(instance.facts)} facts")
    
    # Lookup by tag name (Clark notation stripped)
    company = instance.get_fact_value("NameOfTheCompany")
    scrip = instance.get_fact_value("ScripCode")
    print(f"Filer: {company} (Scrip: {scrip})")

asyncio.run(main())

---

3. Working with Financial Results

Indian financial results follow standardized Ind AS / Banking / Insurance taxonomy tags. Convert the universal XBRLInstance into a typed FinancialResultsModel:

from nse_xbrl_parser import parse_xbrl_sync, FinancialResultsModel

instance = parse_xbrl_sync(xml_bytes)
model = FinancialResultsModel.from_instance(instance)

print("Quarter:", model.reporting_quarter)
print("Standalone/Consolidated:", model.nature_of_report)
print(f"Revenue: ₹{model.revenue_from_operations:,.2f}")
print(f"Total Expenses: ₹{model.total_expenses:,.2f}")
print(f"PBT: ₹{model.profit_before_tax:,.2f}")
print(f"Tax: ₹{model.tax_expense:,.2f}")
print(f"PAT (Net Profit): ₹{model.profit_after_tax:,.2f}")
print(f"EBITDA: ₹{model.ebitda:,.2f}")
print(f"Basic EPS: ₹{model.basic_eps:.2f}")

---

4. Accessing Taxonomy Information

Inspect any of the 58 NSE/BSE regulatory taxonomies:

import asyncio
from nse_xbrl_parser import TaxonomyRegistry, AsyncTaxonomyDownloader

async def main():
    registry = TaxonomyRegistry()

    # Search for Shareholding Pattern (Regulation 31)
    shp_info = registry.get_shareholding_pattern_taxonomy()
    print("Name:", shp_info.name)
    print("Download URL:", shp_info.primary_taxonomy_url)

    # Automatically download and parse its XSD schema and Label linkbase
    downloader = AsyncTaxonomyDownloader()
    schema, linkbase = await downloader.load_taxonomy_metadata(shp_info)

    print(f"Elements in schema: {len(schema.elements)}")
    label = linkbase.get_label("PublicShareholdingAsAPercentageOfTotalNumberOfShares")
    print("Human label:", label)

asyncio.run(main())