CRAN check results at the terminal

Description

foghorn brings the results of the CRAN checks into your R terminal. This package is intended for developers that have published packages on CRAN. The results of the checks can be summarized as a tibble (see cran_results), or as “pretty” terminal output (see summary_cran_results). The package provides functions to view the logs of the CRAN check results (see cran_details), or to visit the web pages hosted by CRAN (see visit_cran_check). You can also use foghorn to follow your package during its submission process to CRAN or to WinBuilder with, cran_incoming and winbuilder_queue.

Author(s)

Maintainer: Francois Michonneau francois.michonneau@gmail.com

Other contributors:

• Ben Bolker [contributor]

See Also

Useful links:

• https://fmichonneau.github.io/foghorn/

• https://github.com/fmichonneau/foghorn

• Report bugs at https://github.com/fmichonneau/foghorn/issues

Deprecated functions

Description

Deprecated functions provided for back compatibility.

Usage

check_cran_results(...)

Arguments

Get details about the CRAN check results for packages

Description

Given the names of packages published on CRAN, return the output of checks that return notes, warnings or errors.

Usage

cran_details(pkg, src = c("website", "crandb"), ...)

## S3 method for class 'cran_details'
summary(object, show_log = TRUE, print_ok = TRUE, ...)

summary_cran_details(
  pkg,
  src = c("website", "crandb"),
  show_log = TRUE,
  print_ok = TRUE,
  ...
)

Arguments

Details

Where does the data come from?

The data comes from the CRAN servers. They generate RDS files that contains information regarding the results of the checks for all the packages, and all the flavors. This data is then used to generate the web pages.

foghorn provides access to either of these data sources. If you choose src = "website" the data is scrapped from the CRAN website. If you only need to check a few packages, this is a good option. If you choose src = "crandb" the RDS files (about 20Mb) are downloaded first from the CRAN servers.

The option max_requests can be used to limit how many pages will be scrapped from the CRAN website. The default is set to 50, use Inf to ignore this check. Consider using src = "crandb" if you need to get data from many packages or maintainers. Note that this an approximation of the number of requests that will be performed and there will be situations where more requests than this limit will be performed to retrieve the results.

The deadline column contains the date set by CRAN to fix issues with the CRAN checks before the package gets archived. If the existence of a deadline has been checked but no date has been set by the CRAN Maintainers, the deadline column for the package will be set to "". If there is a deadline, the content will be a date stored as character.

The value of the argument show can be set using the local option foghorn_columns.

When choosing src = "crandb" you can also specify the following options:

• dest: a folder where to store the RDS files (tempdir() by default).

• protocol: either https or http.

• overwrite: when FALSE (default), if the file exists in dest then it will not be downloaded again. When TRUE the file gets downloaded every time it's needed.

Value

a tibble listing the names of the packages that have non- OK check results, the nature of the result (WARN, ERROR, FAIL, NOTE, or other issues).

See Also

Note that the tools package contains unexported functions that can be used to extract summary information from the check results. Specifically tools:::sumarize_CRAN_check_status is similar to show_cran_results.

List packages in CRAN incoming queue.

Description

Check where your package stands in the CRAN incoming queue.

Usage

cran_incoming(
  pkg = NULL,
  folders = cran_incoming_folders(),
  sort_by_date = TRUE
)

cran_incoming_folders(include_archive = FALSE)

Arguments

Details

When submitting a package to CRAN, it undergoes a series of checks before it is published and publicly available. cran_incoming() allows you to check the packages that are currently in the queue, and the folder where they are located. This information could help you track your package submission. Only the following folders are considered (approximately in order of the CRAN queue sequence): newbies, inspect, pretest, recheck, pending, waiting, publish. The folder archive is not inspected by default. The folders named after the initials of the CRAN volunteers are not inspected.

Value

cran_incoming() returns tibble with the following columns:

package

package name

version

package version

cran_folder

folder where the package was found

time

date/time package was entered in the folder

size

the size of the package tarball

cran_incoming_folders() returns a character vector of the names of the folders used as part of the CRAN submission process, archive being included optionally.

Note that if the package version is not provided, it will appear as NA in the tibble.

Disclaimer

The information provided here is only to give you an indication of where your package stands in the submission process. It can be useful to confirm that your package has been correctly uploaded to CRAN. Please consult the CRAN Repository Policy if you have any questions.

Note

The meaning of the package folders is as follows (see Hornik, Ligges and Zeileis https://journal.r-project.org/archive/2018-1/cran.pdf and Uwe Ligges mailing list comment https://stat.ethz.ch/pipermail/r-package-devel/2019q1/003631.html):

newbies

for first time submission; package will be manually inspected.

inspect

package is awaiting manual inspection; always happens for first time submissions and for packages with problems that are likely to be false positives

pretest

a human has triggered a new auto-check of the package

recheck

package has passed checks and is waiting for reverse dependency checking

pending

a CRAN team member has to do a closer inspection and needs more time

waiting

CRAN's decision is waiting for a response from the package maintainer, e.g. when issues are present that CRAN cannot check for in the incoming checks

publish

package is awaiting publication

archive

package rejected: it does not pass the checks cleanly and the problems are unlikely to be false positives

References

• Hornik, Ligges and Zeileis. "Changes on CRAN: 2017-12-01 to 2018-06-30", R Journal 10(1), July 2018. https://journal.r-project.org/archive/2018-1/cran.pdf

• Maëlle Salmon, Locke Data, Stephanie Locke, Mitchell O'Hara-Wild, Hugo Gruson. "CRAN incoming dashboard", https://lockedata.github.io/cransays/articles/dashboard.html

See Also

cran_winbuilder

Examples

## Not run: 
  ## all the packages in the CRAN incoming queue
  cran_incoming()
  ## to include all the folders including `archive`
  cran_incoming(folders = cran_incoming_folders(include_archive = TRUE)
  ## to only include a folder, e.g., `inspect`
  cran_incoming(folders = "inspect")
  ## if the package `foo` is in the queue, it will appear below
  cran_incoming(pkg = "foo")

## End(Not run)

Table of the CRAN check results

Description

Make a table that summarizes the results of the CRAN checks for a set of packages specified by a maintainer or by names.

Usage

cran_results(
  email = NULL,
  pkg = NULL,
  show = getOption("foghorn_columns", c("error", "fail", "warn", "note", "ok",
    "deadline")),
  src = c("website", "crandb"),
  max_requests = 50,
  ...
)

Arguments

Details

Given the email address of a package maintainer, and/or a vector of package names, returns a tibble that allows you to detect potential issues with your packages on CRAN.

Where does the data come from?

The data comes from the CRAN servers. They generate RDS files that contains information regarding the results of the checks for all the packages, and all the flavors. This data is then used to generate the web pages.

foghorn provides access to either of these data sources. If you choose src = "website" the data is scrapped from the CRAN website. If you only need to check a few packages, this is a good option. If you choose src = "crandb" the RDS files (about 20Mb) are downloaded first from the CRAN servers.

The option max_requests can be used to limit how many pages will be scrapped from the CRAN website. The default is set to 50, use Inf to ignore this check. Consider using src = "crandb" if you need to get data from many packages or maintainers. Note that this an approximation of the number of requests that will be performed and there will be situations where more requests than this limit will be performed to retrieve the results.

The deadline column contains the date set by CRAN to fix issues with the CRAN checks before the package gets archived. If the existence of a deadline has been checked but no date has been set by the CRAN Maintainers, the deadline column for the package will be set to "". If there is a deadline, the content will be a date stored as character.

The value of the argument show can be set using the local option foghorn_columns.

When choosing src = "crandb" you can also specify the following options:

• dest: a folder where to store the RDS files (tempdir() by default).

• protocol: either https or http.

• overwrite: when FALSE (default), if the file exists in dest then it will not be downloaded again. When TRUE the file gets downloaded every time it's needed.

Value

a data frame that tabulates the number of CRAN flavors that return errors, warnings, notes, or OK for the packages. See 'Details' section for more information about the Deadline column.

See Also

Note that the tools package contains unexported functions that can be used to extract summary information from the check results. Specifically tools:::sumarize_CRAN_check_status is similar to show_cran_results.

Examples

  ## Not run: 
    cran_results(pkg="MASS")
  
## End(Not run)

The number of CRAN flavors

Description

The CRAN flavors, the systems on which CRAN tests all packages regularly, are listed https://cran.r-project.org/web/checks/check_flavors.html. To get the correct results, foghorn needs to know how many flavors CRAN uses. This function reads the number of flavors that CRAN currently uses, and caches it (per session, in the tempdir() folder). Arguments control caching, fall back, and default values.

Usage

n_cran_flavors(
  use_cache = getOption("foghorn.use_cache", TRUE),
  force_default = getOption("foghorn.force_default", FALSE),
  n_flavors = getOption("foghorn.n_flavors", 12L)
)

Arguments

Details

The default values for the arguments are read from options. Given that n_cran_flavors function is relied on internally to provide accurate information to the user, using options allows you to control how the function behaves directly. In general, the default values should not be changed. They are provided in case you have issues connecting to the web page listing the number of flavors, or you do not want to use caching.

The options can be set:

• by session, using, for instance,options("foghorn.use_cache" = FALSE).

• permanently, by adding options("foghorn.use_cache" = FALSE) in your .Rprofile.

• for a specific call, using the withr package: withr::with_options(foghorn.use_cache = FALSE, ...).

Value

The number of CRAN check flavors (as an integer).

Summary of the CRAN check results

Description

Given the email address of a package maintainer, and/or a vector of package names, it displays at the console a summary of the check results run on the CRAN flavors. This function is designed to be included in your .Rprofile to be run (periodically) at start up.

Usage

summary_cran_results(
  email = NULL,
  pkg = NULL,
  compact = FALSE,
  print_ok = TRUE,
  ...
)

## S3 method for class 'cran_results'
summary(object, compact = FALSE, print_ok = TRUE, ...)

show_cran_results(...)

Arguments

Value

Prints the packages that return errors, warnings, and notes on the CRAN flavors. The number in parenthesis after the name of the packages indicates the number of CRAN flavors that produce these results.

Examples

## Not run: 
   summary_cran_results(email = c("user1@company1.com", "user2@company2.com"))
   summary_cran_results(email = "user1@company1.com",
                        pkg = c("pkg1", "pkg2"))

## End(Not run)

Visit the CRAN check results page

Description

Visit the page in your web browser for a given package or a maintainer's email address

Usage

visit_cran_check(pkg = NULL, email = NULL)

Arguments

Value

The URL from the CRAN check results page invisibly

Show the win-builder queue

Description

Check whether your package is in the win-builder queue.

Usage

winbuilder_queue(
  pkg = NULL,
  folders = c("R-release", "R-devel", "R-oldrelease")
)

Arguments

Details

To check whether your package has successfully been submitted to win-builder, or to check whether there is unusual delay in processing packages submitted to win-builder, winbuilder_queue allows you to inspect the packages that are in the queue to be processed by the win-builder service.

Value

A tibble with the following columns:

package

package name

version

package version

folder

the folder indicating the R version that will be used to perform the checks

time

the date and time at which the package tarball was uploaded on win-builder

size

the size of the package tarball

References

• Maëlle Salmon, 2020. "Everything you should know about WinBuilder" https://blog.r-hub.io/2020/04/01/win-builder/

• Uwe Ligges. Building and checking R source packages for Windows. https://win-builder.r-project.org/

Examples

## Not run: 
  ## Get all the packages in the win-builder queue
  winbuilder_queue()
  ## Check if the 'dplyr' package is in the win-builder queue
  winbuilder_queue(pkg = "dplyr")
  ## Check which packages are in the R-devel queue
  winbuilder_queue(folders = "R-devel")

## End(Not run)