In hrbrmstr/crux: Identify the Crux of an Article

knitr::opts_chunk$set(collapse=TRUE, fig.retina=2, message=FALSE, warning=FALSE)
options(width=120)

crux

Identify the Crux of an Article

Description

Methods are provided to retrieve HTML content and return extracted metadata and summarised plain text. Further methods are provided to classify URLs with or without making network calls. Based on https://github.com/chimbori/crux.

What's Inside The Tin

The following functions are implemented:

classify_url: Classify a URL with or without making network calls
is_ad_image: Classify a URL with or without making network calls
is_likely_archive: Classify a URL with or without making network calls
is_likely_article: Classify a URL with or without making network calls
is_likely_audio: Classify a URL with or without making network calls
is_likely_binary_doc: Classify a URL with or without making network calls
is_likely_executable: Classify a URL with or without making network calls
is_likely_image: Classify a URL with or without making network calls
is_likely_video: Classify a URL with or without making network calls
is_web_scheme: Classify a URL with or without making network calls
summarise_url: Summarise the contents at a URL to essential bits

Installation

install.packages(c("cruxjars", "crux"), repos = "https://cinc.rud.is/")

Usage

library(crux)

# current version
packageVersion("crux")

str(
  summarise_url("http://time.com/5541738/joe-biden-backtracks-pence-praise-criticism/"), 1
)

str(
  classify_url("https://www.washingtonpost.com/powerpost/house-democrats-explode-in-recriminations-as-liberals-lash-out-at-moderates/2019/02/28/c3d163fe-3b87-11e9-a06c-3ec8ed509d15_story.html")
)