https://github.com/mkearney/attrbl

A tidy approach to attributes
https://github.com/mkearney/attrbl

attr attributes dataframes mkearney-r-package r rstats tibble tidy tidy-data tidyverse

Last synced: about 1 month ago
JSON representation

A tidy approach to attributes

• Host: GitHub
• URL: https://github.com/mkearney/attrbl
• Owner: mkearney
• Created: 2018-02-10T20:14:34.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2018-02-22T00:06:12.000Z (over 6 years ago)
• Last Synced: 2024-06-11T08:49:22.184Z (3 months ago)
• Topics: attr, attributes, dataframes, mkearney-r-package, r, rstats, tibble, tidy, tidy-data, tidyverse
• Language: R
• Homepage:
• Size: 23.4 KB
• Stars: 4
• Watchers: 5
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.Rmd

Awesome Lists containing this project

• jimsghstars - mkearney/attrbl - A tidy approach to attributes (R)

README

        
---

title: "attrbl"

output: github_document

---

```{r setup, include=FALSE}

knitr::opts_chunk$set(echo = TRUE, collapse = FALSE, comment = "#>")

library(attrbl)

library(magrittr)

```

A tibble-like package for working with attributes and data frames.

***Warning: this package is in early development***

## Install

Install the dev version from Github

```{r, eval = FALSE}

if (!requireNamespace("devtools", quietly = TRUE)) {

  install.packages("devtools")

}

devtools::install_github("mkearney/attrbl")

```

## Use case

Let's say you have a list of two data frames. And each data frame contains a data frame attribute "users" with around 100 observations.

```{r, cache=TRUE}

## tweets data with users data attribute

rts <- list(

  rtweet::search_tweets("statistics", verbose = FALSE),

  rtweet::search_tweets("data science", verbose = FALSE))

## each object contains "users" data attribute with around 100 rows

rts %>%

  lapply(rtweet::users_data) %>%

  lapply(nrow)

```

When you bind the data frames using `do.call(..., rbind)`, it returns a "users" attribute. But it only has around 100 rows (it should be closer to 200). It completely **drops** the second "users" attribute!

```{r}

## base R's method

rts %>%

  do.call("rbind", .) %>%

  attr("users") %>%

  nrow()

```

When you bind the data frames using `dplyr::bind_rows(...)`, it **doesn't return** a "users" attribute at all!

```{r}

## dplyr's bind_rows method

rts %>%

  dplyr::bind_rows() %>%

  attr("users")

```

But when you bind the data frames using `attrbl::do_call_rbind(...)`, it not only **keeps** all "users" attributes. It binds them together for you!

```{r}

## attrbl's do_call_rbind method

rts %>%

  lapply(as_attrbl) %>%

  do_call_rbind() %>%

  attr("users") %>%

  nrow()

```

## Usage

### Bind data frames without losing attributes

When a `data.frame` is merged with `dplyr::bind_rows(...)`, it **loses** its attribute(s).

```{r}

## list of data frames

mtcars2 <- list(mtcars, mtcars)

## dplyr method

mtcars2 %>%

  dplyr::bind_rows() %>%

  attr("row.names")

```

When an `attrbl` is merged with `attrbl::do_call_rbind`, it **keeps** its attribute(s).

```{r}

## attrbl method

mtcars2 %>%

  lapply(as_attrbl) %>%

  do_call_rbind() %>%

  attr("row.names")

```

### Select attributes with `atselect`

Use `atselect` to select attributes the tidy way (no quotes required).

```{r}

atselect(mtcars, row.names, class)

```

### Add attributes with `add_attr`

Use `add_attr` to add one or more attributes to a data frame.

```{r}

mtcars %>%

  add_attr(seed = quote(set.seed(1234)), timestamp = Sys.time()) %>%

  attributes()

```