/
rsconnect.Rmd
63 lines (54 loc) · 2.78 KB
/
rsconnect.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
---
title: "Using renv with Posit Connect"
output: rmarkdown::html_vignette
vignette: >
%\VignetteIndexEntry{Using renv with Posit Connect}
%\VignetteEncoding{UTF-8}
%\VignetteEngine{knitr::rmarkdown}
editor_options:
markdown:
wrap: 72
---
```{r, include = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>"
)
```
[RStudio Connect](https://posit.co/products/enterprise/connect/) is a publication
platform for deploying content built in R and Python to share with a broad
audience. R users may want to develop content (like [Shiny
applications](https://shiny.posit.co/) or [RMarkdown
documents](https://rmarkdown.rstudio.com/index.html)) using renv and then
publish that content to RStudio Connect. This is a supported pattern where
renv is used to manage the local project environment and then RStudio Connect
recreates and manages the deployment environment.
## Publishing from the RStudio IDE
The RStudio IDE includes a button for push-button deployment to RStudio
Connect: ![RStudio IDE Publish Button](ide-publish.png)
When this option is used to deploy content to RStudio Connect, a manifest file
is automatically generated and sent to RStudio Connect describing the project
environment. This manifest file will reflect the project environment create and
managed by renv. The renv generated `.Rprofile` file should **not** be
included in deployments to RStudio Connect.
## Publishing programatically
When publishing content to RStudio Connect programatically, it is necessary to
generate a manifest file describing the project environment. This can be done
with the `writeManifest()` function from the
[rsconnect](https://github.com/rstudio/rsconnect) package. When using renv,
the only thing that needs to be considered is that rsconnect should be
installed and executed from within the renv environment so that it recognizes
the local project library when generating the manifest file. As long as
rsconnect is run from within the renv created environment, it will capture
project dependencies from the local renv library. This can be accomplished by
opening the project in RStudio or by starting the R session from the project
root directory. The renv generated `.Rprofile` file should **not** be
included in deployments to RStudio Connect.
## A word about packrat
RStudio Connect uses [packrat](https://rstudio.github.io/packrat/) to restore
project environments on the RStudio Connect server. This should have no impact
on how the user develops content for RStudio Connect. It is not necessary for
the user to use packrat instead of renv when developing content, as the
environment management tool used locally has no impact on the tools RStudio
Connect uses for environment management. Therefore, there should be no concerns
with using renv to develop content that will be deployed to RStudio Connect.