/ Docs

Quick Start

Open In Colab

프로그래밍이 처음이라면? Python 설치부터 안내하는 초보자 가이드 (블로그)를 먼저 보세요. 5분이면 첫 분석까지 끝납니다.

설치 없이 바로 해보고 싶다면? HF Spaces 데모에서 브라우저만으로 체험하거나, 위 Colab 배지를 클릭하세요.

Everything in DartLab starts with c.sections. The rest (show, trace, diff, BS, ratios) are all views on top of sections.

Installation

uv add dartlab

For the AI interface:

uv add "dartlab[ai]"
uv run dartlab ai

Auto Data Download

When you create a Company for the first time, required data is downloaded automatically. No setup needed.

[dartlab] 005930 (DART docs data) → first use: auto-downloading from HuggingFace...
[dartlab] ✓ DART docs data download complete (542KB)
[dartlab] 005930 (finance data) → first use: auto-downloading from HuggingFace...
[dartlab] ✓ finance data download complete (38KB)

From the second run onward, local cache is used for instant loading.

sections = The Entire Company

import dartlab

c = dartlab.Company("005930")  # Samsung Electronics
c.sections

sections is a Polars DataFrame. Every disclosure section is laid out as a topic × period matrix.

chapter │ topic            │ blockType │ 2025Q4 │ 2024Q4 │ 2024Q3 │ …
I       │ companyOverview  │ text      │ "…"    │ "…"    │ "…"    │
I       │ companyOverview  │ table     │ "…"    │ "…"    │ null   │
II      │ businessOverview │ text      │ "…"    │ "…"    │ "…"    │

This alone shows the entire disclosure structure of a company — topic list, period range, text/table breakdown.

c.topics              # topic summary DataFrame (source, blocks, periods)
c.sections.periods()  # list of periods
c.sections.ordered()  # sorted newest first

Views on Top of sections

When you want to drill into a single topic instead of viewing all sections:

c.show("overview")       # block index
c.show("companyOverview", 0)    # actual data for block 0
c.show("BS")                    # balance sheet (finance source)
c.show("dividend")              # dividend (report source)

Detect when text changed:

c.diff()                                    # overall change rates
c.diff("businessOverview")                  # topic history
c.diff("businessOverview", "2024", "2025")  # line-by-line comparison

Check which source was selected:

c.trace("BS")               # finance source
c.trace("overview")  # docs source

Financial Statements and Ratios

Financial statements are also views on top of sections via the finance engine:

c.BS                    # balance sheet (newest first)
c.IS                    # income statement
c.CF                    # cash flow statement
c.ratios                # financial ratios time-series DataFrame
c.finance.ratios        # latest single-period RatioResult
c.finance.ratioSeries   # ratio time-series (raw)

Insights

Grading across 7 areas (performance, profitability, stability, cash flow, governance, risk, opportunity):

c.insights                      # 10-area analysis
c.insights.grades()             # → {"performance": "A", "profitability": "B", …}
c.insights.performance.grade    # → "A"
c.insights.performance.details  # → ["Revenue high growth +8.3%", …]
c.insights.anomalies            # → outliers, risk signals

Same Interface for EDGAR

us = dartlab.Company("AAPL")
us.sections
us.show("10-K::item1Business")
us.BS
us.ratios

Same sections, same show, same diff. Only the topic names follow SEC form conventions.

Source Namespaces

When you need to go deeper:

c.docs.sections          # pure docs source
c.finance.BS             # finance engine directly
c.report.extract("배당")  # report engine directly

But for most analysis, c.sections and c.show() are all you need.

Try It Now

Explore interactively without writing code:

uv add dartlab marimo
marimo edit startMarimo/dartCompany.py    # Korean companies (DART)
marimo edit startMarimo/edgarCompany.py   # US companies (EDGAR)

Or open the Colab quickstart notebook directly in your browser.

Next Steps