Articles

Articles about Buckaroo can be found here. It was more expedient to use Read The Docs for a common blog task than dealing with medium paywalls or yet another devops struggle.

• Auto Cleaning
  • Better typing
  • Heuristic cleaning
    • Using Autocleaning
  • Changing individual coercions
  • Turning off auto cleaning
    • How Autotyping Works
  • Why three functions?
  • How do I add special replacement functionality?
• BuckarooCompare — Diff Your DataFrames
  • Quick start
  • How the diff works
  • The eqs summary
  • Using it with the server
  • Multi-column joins
  • Use cases
  • Integration with datacompy
  • Limitations
• The Dastardly DataFrame Dataset
  • Why this matters
  • The Dastardly DataFrames
    • Infinity and NaN
    • Really Big Numbers
    • Column Named “index”
    • Named Index
    • MultiIndex Columns
    • MultiIndex on Rows
    • Three-Level MultiIndex
    • MultiIndex on Both Axes
    • Weird Types (Pandas)
    • Weird Types (Polars)
  • Full dtype coverage
  • How this demo was built
  • Try it yourself
• Data Flow through Buckaroo
  • Customization points of Buckaroo
  • Full Flow
    • Glossary
  • Quick Start to extending Buckaroo
• use
• Don’t use df.head(), try buckaroo instead
• Buckaroo Embedding Guide
  • Why embed
  • Choose your embedding mode
    • Data size guidelines
  • Generate a static embed
  • Customizing appearance
    • Column config overrides
    • Tooltips
    • Analysis classes
    • Pinned rows
  • Integration patterns
    • Static HTML file
    • React component
    • Sphinx / ReadTheDocs
  • What’s included in the bundle
• Histograms
  • Interpreting Buckaroo histograms by example
    • Common patterns
    • Categorical patterns
    • Numeric patterns
  • Other research
• Pluggable analysis framework
  • Analysis In Action
  • Features of the Pluggable Analysis Framework
  • Order of Operations
• Related projects
  • Other jupyter widgets
  • Other exploratory data analysis tools
  • Other data tables
  • Other js data tables
  • Why list potential competitors
• Roadmap
  • Priorities
  • Release Plans
    • 0.4 Series
    • 0.5 series
    • 0.6 series
    • 0.7 series
• So You Want to Write a DataFrame Viewer
  • Use-case questions
  • Processing: server-side or browser-based
  • Serializing data
  • The table viewer component
  • The notebook environment
  • Other questions
  • Conclusion
  • Comparison of open source DataFrame viewers
• Static Embedding & the Incredible Shrinking Widget
  • How it works
  • Two embedding modes
  • Bundle size
  • What we did to get here
  • What’s next: CDN-hosted viewer
  • For larger data: Parquet range queries
  • Try it
• How Types and Data Move from Engine to Browser
  • Column renaming: why everything becomes a, b, c
  • Cleaning before serialization
  • Parquet encoding and base64 transport
  • hyparquet: decoding Parquet in the browser
  • Displayers and formatters: the last mile
  • The full pipeline
  • Why it ended up this way
• Why Buckaroo Uses Depot for CI
  • The problem with GitHub Actions
  • What Buckaroo’s CI does
  • LLMs changed the equation
  • The benchmark
  • The results
  • What’s actually happening
  • What Depot actually gave me
  • Before and after
  • Testing against dependency versions
  • The scariest part
  • What I’d tell other open source maintainers

• Index

• Module Index

• Search Page

• Buckaroo on github