To use this package you need to work in RStudio. The goal of
quartopost
is to create and open a Quarto blog post with the
appropriate YAML front matter in RStudio.
quartopost()
displays a dialog window where you can enter the data for
the YAML header of a new blog post.
After clicking the “Done” button the function generates the core skeleton of a Quarto post. This includes:
- creating the directory (named with the date and title in kebab notation)
- (optionally) copying images from your hard disk into the new created folder
- creating the
index.qmd
file with the YAML header populated from the data of the dialog window - opening the blog post file in RStudio for editing.
You can choose from your categories already created or add new
categories. With the package comes also an RStudio Addin so you can bind
the quartopost()
command with a shortcut.
You can install the development version of quartopost
from
GitHub with:
# install.packages("devtools")
devtools::install_github("petzi53/quartopost")
library(quartopost)
if (interactive()) {
quartopost()
}
To create Quarto posts even easier, you can configure quartopost
:
- Choose “Tools -> Modify Keyboard Shortcuts …” from the RStudio menu.
- Filter the search box for “quarto”.
- Click into the column “Shortcut” of the “Create Quarto Post” addin.
- Provide your keyboard shortcut.
If you are the only blog author on your machine, you can add your name to the author field automatically.
- Open
.Rprofile
. The easiest way it to installusethis
` and open.Rprofile
withusethis::edit_r_profile()
. - Add the following line into
.Rprofile
options(servr.daemon = TRUE,
quartopost.author = "<your name>"
)
- Restart RStudio.
If you want to prevent the confirmation question before quartopost()
creates the folder with the post, the YAML header and (optionally)
copies your chosen image, add another line into .Rprofile
:
quartopost.verbose = FALSE
You can specify if you want the new post with draft: true
(Standard)
or draft: false
.
Not all fields are always needed. The default is to get the full list of
all YAML fields provided by the dialog window, even if some of the
fields are empty (show_empty_fields = true
). With
show_empty_fields = false
only those fields with contents are
displayed in the YAML front matter.
If you are going to add all lines to .RProfile
, then your code
snippets should look like:
options(servr.daemon = TRUE,
quartopost.author = "Peter Baumgartner", # default = ""
quartopost.verbose = FALSE, # default = TRUE
quartopost.draft = FALSE, # default = TRUE
quartopost.show_empty_fields = FALSE # default = TRUE
)
Do not forget to restart RStudio after you changed .Rprofile
!
I am not very experienced in R programming, so I am very grateful that I could take help and suggestions from different sources:
I started with quartopost()
after a question or feature request at
the Posit
forum.
I got some helpful links that I tried out and used as precursors for my
implementation.
I took the new_post() script from The Mockup
blog as
an advanced organizer for the general structure of my approach. It has a
very extensive and understandable explanation. I learned about the kebab
notation I had never heard about until this blog post. For German
umlauts – and special characters in other languages -, I added the line
stringi::stri_trans_general((title), "latin-ascii")
(learned from
a StackOverflow answer).
I also used some ideas from the newpost package, particularly the use of RStudio Addins and the request that users choose from categories already used.
My knowledge of R programming is limited, but I am a complete newbie when it comes to Shiny apps. (I just started reading the fantastic Mastering Shiny book by Hadley Wickham.) I took the general structure of the Shiny window from the blogdown “New Post” RStudio addin by Yihui Xie. The lack of a similar function for Quarto was generally my main incentive to start with the quartopost project in the first place.
I don’t want to release my package to CRAN, but I tried to follow the advises of R Packages.
I am sure that my code is not efficient and that there are many ways to improve my package. I am very interested to learn better programming in R / Shiny / (JavaScript). If you find bugs or more efficient ways to solve my intended goal please do not hesitate to raise an issue or provide a pull request.