keepcosmos/readability

This system is an Elixir library designed to extract readable article content and metadata from HTML webpages. It provides functionality to identify main text, titles, authors, and publication dates by cleaning and scoring HTML nodes. The implementation includes HTTP request handling, MIME type detection, and normalization of relative paths and tags.

This commit introduces the core Readability library, providing functions to summarize webpages, extract titles, authors, and publication dates, and isolate readable article content from HTML. The module includes configuration options for content scoring and cleaning, supports HTTP requests via HTTPoison with customizable options, and handles MIME type detection for markup responses.

lib · high confidence

The library now provides a complete Elixir implementation for extracting readable content from HTML. It includes an article builder that cleans and scores HTML nodes to identify the main content, along with dedicated finders for extracting the title (prioritizing h1 tags), authors, and publication dates. The system also handles HTML normalization, such as converting relative image paths to absolute URLs and stripping unnecessary tags, to ensure consistent output.

lib/readability · high confidence

The repository has been initialized with the core project structure, including the addition of a \.formatter.exs\ file to standardize code formatting via \mix format\, a \.travis.yml\ configuration to enable continuous integration for Elixir versions 1.3.4, 1.4.1, and 1.5.0, and an updated \.gitignore\ to exclude build artifacts and dependencies. The project license was standardized to Apache License 2.0 (moved to \LICENSE.md\), and the \README.md\ was expanded to provide comprehensive installation instructions, usage examples for extracting article content, and configuration options for the readability tool.

(repo-wide) · high confidence

Added BBC and BuzzFeed HTML fixtures for testing; Added test coverage for readability components; Added test suite for readability parsing and HTTP handling.

Written by watchdog.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

• First survey — no prior run to compare against. CAI 56.

• Code Health 99
• Architecture 69
• Maturity 36
• Readiness 64
• Security 96

• 115 commits — 107 feature/other, 8 fixes

• (root) — 56 commits
• (repo) — 25 commits
• lib/readability — 22 commits
• lib/readability.ex — 6 commits
• test/fixtures — 3 commits
• test/readability — 2 commits
• config/config.exs — 1 commit

• fix: Clean up and fix warnings
• fix: Fix Issue 65: incorrect node return format for doctype when using html5ever (#66)
• fix: Fix detection of title suffix
• fix: Fix merging of title tag matches and raising on no matches
• fix: Fix: Readability.published_at and Readability.title works as expected (#60)
• fix: Fixes crash when html has an xml version tag by stripping it out
• fix: fix some bug and update deps
• fix: fixed incorrectly replacing <font> by <1span> during normalization
• change: Add Elixir 1.6 formatter config file and formatted the codebase
• change: Add httpoison_options to CHANGELOG
• change: Add published_at to the extracted attributes. (#58)
• change: Added excoveralls
• change: Allow HTTPoison options to be specified in config
• change: Avoid deprecated way to identify regex
• change: Bump Floki and HTTPoison versions
• change: Bump version
• change: Bumped Floki dependency to 0.14.0
• change: Changed default title selector to h1
• change: Convert relative img paths into absolute
• change: Credo warnings
• …and 95 more

Written by watchdog.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

keepcosmos/readability was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.