/ Docs

7. Advanced Analysis

Open In Colab

Covers deep source namespace access including K-IFRS Notes integration, tangible asset movement schedules, affiliate analysis, and governance analysis. The standard public flow is sections -> show -> trace, but this document explains the deep access layer beneath it.

  • Notes integration (12 items)
  • Direct lookup for unregistered keywords (23 keywords)
  • Tangible asset movement schedule
  • Associates and joint ventures
  • Board of directors and audit system
  • Governance and risk analysis
  • Multi-company comparison
  • Result objects and deep access patterns

Setup

import dartlab

c = dartlab.Company("005930")

K-IFRS Notes

Access 12 K-IFRS note items through c.notes. Both English property names and Korean keys are supported.

Access via English Properties

c.notes.inventory          # 재고자산 DataFrame
c.notes.receivables        # 매출채권
c.notes.borrowings         # 차입금
c.notes.tangibleAsset      # 유형자산 변동표
c.notes.intangibleAsset    # 무형자산
c.notes.provisions         # 충당부채
c.notes.eps                # 주당이익
c.notes.lease              # 리스
c.notes.investmentProperty # 투자부동산
c.notes.affiliates         # 관계기업
c.notes.segments           # 부문정보
c.notes.costByNature       # 비용의 성격별 분류

Access via Korean Keys

c.notes["재고자산"]         # Same as c.notes.inventory
c.notes["차입금"]           # Same as c.notes.borrowings
c.notes["유형자산"]         # Same as c.notes.tangibleAsset

Listing All Keys

c.notes.keys()       # ['receivables', 'inventory', 'tangibleAsset', ...]
c.notes.keys_kr()    # ['매출채권', '재고자산', '유형자산', ...]
c.notes.keys()       # Available notes keys

Direct Lookup for Unregistered Keywords

Beyond the 12 main keywords, 23 additional keywords are supported. Keywords not registered in Notes can be accessed directly via the docs.notes namespace.

# Corporate tax
result = c.docs.notes.detail("법인세")
print(result.tableDf)

# Related parties
result = c.docs.notes.detail("특수관계자")
print(result.tableDf)

All 23 Keywords

재고자산, 주당이익, 충당부채, 차입금, 매출채권, 리스, 투자부동산, 무형자산, 법인세, 특수관계자, 약정사항, 금융자산, 공정가치, 이익잉여금, 금융부채, 기타포괄손익, 사채, 종업원급여, 퇴직급여, 확정급여, 재무위험, 우발부채, 담보

Year-by-Year Detail Tables

Notes returns only DataFrames, but year-by-year detail tables are also accessible through the raw Result object.

result = c.docs.notes.detail("재고자산")
for year, periods in result.tables.items():
    for p in periods:
        print(f"[{year}] {p.period} (패턴: {p.pattern})")
        for item in p.items[:3]:
            print(f"  {item.name}: {item.values}")

Tangible Asset Movement Schedule

Accessed through Notes. The full Result can also be obtained via report.extract().

# Via Notes
c.notes.tangibleAsset   # Opening/closing balance timeseries by category

# Full Result via report.extract()
result = c.report.extract("tangibleAsset")
print(f"Reliability: {result.reliability}")  # "high" or "low"
if result.warnings:
    print(f"Warnings: {result.warnings}")
print(result.movementDf)   # Opening/closing balance timeseries by category

Year-by-Year Movement Details

for year, movements in result.movements.items():
    for m in movements:
        print(f"[{year}] {m.period}")
        print(f"  Categories: {m.categories}")
        for row in m.rows:
            print(f"  {row}")

Associates and Joint Ventures

Accessed through Notes.

# Via Notes
c.notes.affiliates   # Movement timeseries DataFrame

# Full Result via report.extract()
result = c.report.extract("affiliates")
for year, profiles in result.profiles.items():
    for p in profiles:
        print(f"[{year}] {p.name}: {p.ratio}%, book value {p.bookValue}")

Board of Directors

c.show("board")   # alias for boardOfDirectors
# year | totalDirectors | outsideDirectors | meetingCount | avgAttendanceRate

Committee composition is also available.

result = c.report.extract("boardOfDirectors")
print(result.boardDf)       # Board timeseries
print(result.committeeDf)   # Committee composition (committeeName, composition, members)

Audit System

c.show("auditSystem")
# name | role | detail

Audit activity details are also available.

result = c.report.extract("auditSystem")
print(result.committeeDf)   # Audit committee composition
print(result.activityDf)    # Audit activities (date, agenda, result)

Internal Controls

c.show("control")   # alias for internalControl
# year | opinion | auditor | hasWeakness

Governance and Risk Overview

# Related party transactions
c.show("relatedParty")   # alias for relatedPartyTx
# year | entity | sales | purchases

# Contingent liabilities
c.show("contingent")     # alias for contingentLiability
# year | totalGuaranteeAmount

# Sanctions
c.show("sanction")
# year | date | agency | action | amount

# Risk management
c.show("risk")           # alias for riskDerivative
# currency | upImpact | downImpact

Related Party Transaction Details

result = c.report.extract("relatedPartyTx")
print(result.revenueTxDf)    # Revenue transactions
print(result.guaranteeDf)    # Debt guarantees
print(result.assetTxDf)      # Asset transactions

Contingent Liability Details

result = c.report.extract("contingentLiability")
print(result.guaranteeDf)    # Debt guarantees
print(result.lawsuitDf)      # Litigation status

Multi-Company Comparison

You can call the same deep access payload across multiple companies for comparison.

Financial Comparison

import polars as pl

codes = ["005930", "000660", "035420"]
rows = []

for code in codes:
    c = dartlab.Company(code)
    bs = c.BS
    if bs is not None:
        # Check total assets for the latest year
        cols = [col for col in bs.columns if col != "account"]
        last_year = cols[-1]
        row = bs.filter(pl.col("account") == "자산총계")
        if row.height > 0:
            rows.append({
                "기업": c.corpName,
                "자산총계": row[last_year][0],
            })

print(pl.DataFrame(rows))

Dividend Comparison

dividends = []
for code in codes:
    c = dartlab.Company(code)
    div = c.dividend
    if div is not None:
        last = div.row(-1, named=True)
        dividends.append({
            "기업": c.corpName,
            "DPS": last.get("dps"),
            "배당수익률": last.get("dividendYield"),
            "배당성향": last.get("payoutRatio")
        })

print(pl.DataFrame(dividends))

Result Object Access

Properties return a single representative DataFrame. When you need the full data from a module, use report.extract().

# Property → representative DataFrame
c.audit   # Returns opinionDf only

# report.extract() → full Result object
result = c.report.extract("audit")
result.opinionDf   # Audit opinion
result.feeDf       # Audit fees

The standard path is c.show("audit"). Only use c.report.extract() when deep access is needed.

Next Steps