pxmake

pxmake: Make PX-Files in R

https://github.com/statisticsgreenland/pxmake

Repository

pxmake: Make PX-Files in R

README.Rmd

---
output: github_document
editor_options: 
  chunk_output_type: console
---



```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)
```

# pxmake 


[![CRAN status](https://www.r-pkg.org/badges/version/pxmake)](https://cran.r-project.org/package=pxmake)
![R-CMD-check](https://github.com/StatisticsGreenland/pxmake/actions/workflows/R-CMD-check.yml/badge.svg)
[![Codecov test coverage](https://codecov.io/gh/StatisticsGreenland/pxmake/branch/main/graph/badge.svg)](https://app.codecov.io/gh/StatisticsGreenland/pxmake?branch=main)
[![Mentioned in Awesome Official Statistics ](https://awesome.re/mentioned-badge.svg)](https://github.com/SNStatComp/awesome-official-statistics-software)
[![Montly downloads](https://cranlogs.r-pkg.org/badges/last-month/pxmake)](https://cran.r-project.org/package=pxmake)


## Overview

‘pxmake’ is an R package for creating and modifying PX-files.

With pxmake you can:

- Import a PX-file, modify it, and save it as a new PX-file.
- Modify all metadata keywords in a PX-file.
- Do complex modifications to a PX-file, like adding total levels to a
  variable.
- Save a PX-file as an Excel workbook.


## Installation

``` r
# Install the latest release from CRAN
install.packages('pxmake')

# Or the development version from GitHub
# install.packages('pak')
pak::pak("StatisticsGreenland/pxmake")
```

## How to use

*(Find complete documentation on [pxmake webpage](https://statisticsgreenland.github.io/pxmake/).)*

Use `px()` to import an existing PX-file into R.

```{r include=FALSE, messages = FALSE}
library(pxmake)
set.seed(1)
```

```{r, eval = FALSE}
library(pxmake)

# Import PX-file
x <- px(input = "example.px")
```

Once imported, use one of pxmake's *modifying functions*. 

In general, modifying functions are named after the keyword they modify. It's possible to chain multiple modifying functions together in tidyverse style by using the pipe operator `%>%`.

```{r}
library(magrittr) # import pipe

# Create px object from data frame
x <- px(data.frame(year = as.character(rep(2021:2023, each = 3)), 
                   group = c('a', 'b', 'c'), 
                   value = runif(9)
                   )
        ) 

head(x$data, 4)

x %>% 
  px_timeval("year") %>%  # Set year as TIMEVAL
  px_heading("year") %>%  # Set year as HEADING
  px_stub("group") %>%    # Set group as STUB
  px_decimals("2") %>%    # Set DECIMALS to 2
  px_save("example.px") # Save as PX-file
```

```{r echo = FALSE}
unlink("example.px")
```


### Modifying functions

```{r, include = FALSE}
priority1 <- c("AXIS-VERSION", "CODEPAGE", "CODES", "CONTACT", "CONTENTS", "CONTVARIABLE", "COPYRIGHT", "DATA", "DESCRIPTION", "DOMAIN", "ELIMINATION", "LANGUAGE", "LANGUAGES", "MAP", "NOTE", "NOTEX", "PRECISION", "SHOWDECIMALS", "SOURCE", "STOCKFA", "SUBJECT-AREA", "SUBJECT-CODE", "TABLEID", "UNITS", "UPDATE-FREQUENCY", "VALUENOTE", "VALUENOTE", "VALUENOTEX", "VALUES", "VARIABLE-TYPE", "VARIABLECODE")
               
priority2 <- c("BASEPERIOD", "CELLNOTE", "CELLNOTEX", "CFPRICES", "CONFIDENTIAL")

easy <- c("AGGREGALLOWED", "AUTOPEN", "AXIS-VERSION", "CODEPAGE", "CONFIDENTIAL", "COPYRIGHT", "NEXT-UPDATE", "SHOWDECIMALS", "SUBJECT-CODE", "TABLEID", "UPDATE-FREQUENCY", "CFPRICES", "DATA", "MAP", "STOCKFA")

medium <- c("CONTACT", "CONTENTS", "DESCRIPTION", "LAST-UPDATED", "LINK", "SOURCE", "SUBJECT-AREA", "TITLE", "UNITS", "DOMAIN", "ELIMINATION", "PRECISION", "VALUENOTE", "VALUENOTEX")

hard <- c("LANGUAGE", "LANGUAGES", "CONTVARIABLE", "NOTE", "NOTEX", "BASEPERIOD", "CELLNOTE", "CELLNOTEX", "VALUES", "VARIABLE-TYPE")

dont_implement <- c("CODES", "VARIABLECODE")

modifying_functions <-
  readLines('NAMESPACE') %>%
  stringr::str_match("S3method\\((?.*),px\\)") %>%
  as.data.frame() %>%
  tidyr::drop_na() %>% 
  dplyr::pull("function")

status <- 
  px_keywords %>%
  as.data.frame() %>% 
  dplyr::select(Keyword = keyword) %>% 
  dplyr::mutate(`Function name` = paste0("px_", gsub("-", "_", tolower(Keyword))),
                implemented = `Function name` %in% modifying_functions
                ) %>% 
  dplyr::mutate(Priority = 
                  dplyr::case_when(
                    Keyword %in% priority1 ~ "**",
                    Keyword %in% priority2 ~ "*",
                    TRUE                   ~ ""
                  ),
                Complexity = 
                  dplyr::case_when(
                    Keyword %in% easy   ~ "Easy",
                    Keyword %in% medium ~ "Medium",
                    Keyword %in% hard   ~ "Hard",
                    TRUE                ~ ""
                  )
                ) %>% 
  dplyr::arrange(dplyr::across(everything()))

implemented <- dplyr::filter(status, implemented)
not_implemented <- dplyr::filter(status, !implemented & !Keyword %in% dont_implement)
wont_implement <- dplyr::filter(status, Keyword %in% dont_implement)

non_keyword_modifying_functions <- 
  dplyr::tibble(`Function name` = setdiff(modifying_functions, status %>% dplyr::pull(`Function name`))) %>% 
  as.data.frame()
```

Currently the following `r nrow(implemented)` keywords have a modifying function in pxmake:

```{r, echo = FALSE}
implemented %>%
  dplyr::select(Keyword, `Function name`) %>% 
  print(row.names = FALSE, right = FALSE)
```

In addition to the above, the following `r nrow(non_keyword_modifying_functions)` modifying functions are available:

```{r, echo = FALSE}
non_keyword_modifying_functions %>% 
  print(row.names = FALSE, right = FALSE)
```

See the help page for each modifying function for more information.


Keywords without modifying functions
These `r nrow(not_implemented)` keywords currently doesn't have a modifying function, but can be implemented.
```{r, echo = FALSE}
not_implemented %>% 
  dplyr::select(-implemented) %>% 
  print(row.names = FALSE, right = FALSE)
```

Finally these `r nrow(wont_implement)` keywords will not have a modifying function, because they are automatically determined by the data.

```{r, echo = FALSE}
wont_implement %>%
  dplyr::select(Keyword) %>% 
  print(row.names = FALSE, right = FALSE)
```


## Need help?
If you have any questions or need help, feel free to [open an issue](https://github.com/StatisticsGreenland/pxmake/issues/new) on GitHub, or contact [Lars and Johan](https://github.com/StatisticsGreenland/pxmake/graphs/contributors) via email.

## For developers

### PX specification

pxmake is based on the [PX-file format specification on Statistics Swedens homepage](https://www.scb.se/globalassets/vara-tjanster/px-programmen/px-file_format_specification_2013.pdf).

### PxJob

Some tests cases uses [PxJob](https://stat.fi/tup/tilastotietokannat/px-tuoteperhe_en.html). Install [pxjob64Win](https://github.com/StatisticsGreenland/pxjob64Win) to be able ro run these tests. This only works on Windows.

### How to create a new release

1. Run `usethis::use_release_issue()` and follow checklist.
2. Sync fork on [PxTools/pxmake](https://github.com/PxTools/pxmake).

Owner

GitHub Events

Issues and Pull Requests

Last synced: 10 months ago

Packages