(lazy)Loading Cached Chunks into an Interactive R Session

[This article was first published on Peter E. DeWitt, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

If you cache code chunks when using knitr to generate reproducible documents then you’ve likely had the issue arrise of needing to load the results of cached chunks into an active interactive R session. The functions lazyload_cache_dir and lazyload_cache_labels have been added to the developmental version of qwraps2 to make loading cached objects into an interactive R session easy.

Let’s make some trivial cached chunks in a file report.Rmd

  ---
  title:  "A Report"
  output: html_document
  ---
  
  ```{r first-chunk, cache = TRUE}
  fit <- lm(mpg ~ wt + hp, data = mtcars)
  x <- pi
  ```
  
  ```{r second-chunk, cache = TRUE}
  fit <- lm(mpg ~ wt + hp + am, data = mtcars)
  xx <- exp(1)
  ```

The resulting project directory looks like this:

    .
    ├── report_cache
    │   └── html
    │       ├── first-chunk_bf368425c25f0c3d95cac85aff007ad1.RData
    │       ├── first-chunk_bf368425c25f0c3d95cac85aff007ad1.rdb
    │       ├── first-chunk_bf368425c25f0c3d95cac85aff007ad1.rdx
    │       ├── __packages
    │       ├── second-chunk_2c7d6b477306be1d4d4ed451f2f1b52a.RData
    │       ├── second-chunk_2c7d6b477306be1d4d4ed451f2f1b52a.rdb
    │       └── second-chunk_2c7d6b477306be1d4d4ed451f2f1b52a.rdx
    ├── report.html
    └── report.Rmd

Now, let’s assume you need to come back to this project at a later date. You want to get the objects created in the cached chunks into a new interactive R session.

If you use lazyload_cache_dir all the cached chunks will be load.

rm(list = ls())
ls()
## character(0)
qwraps2::lazyload_cache_dir(path = "report_cache/html")
## Lazyloading: report_cache/html/first-chunk
## Lazyloading: report_cache/html/second-chunk
ls()
## [1] "fit" "x"   "xx"
fit
## 
## Call:
## lm(formula = mpg ~ wt + hp + am, data = mtcars)
## 
## Coefficients:
## (Intercept)           wt           hp           am  
##    34.00288     -2.87858     -0.03748      2.08371
x
## [1] 3.141593
xx
## [1] 2.718282

This could be a problem. There was a fit created in first-chunk and another object fit created in second-chunk. The fit object in the active workspace is from second-chunk. What if you want the fit from first-chunk? Use lazyload_cache_labels.

rm(list = ls())
ls()
## character(0)
qwraps2::lazyload_cache_labels("first-chunk", path = "report_cache/html")
## Lazyloading report_cache/html/first-chunk_bf368425c25f0c3d95cac85aff007ad1
ls()
## [1] "fit" "x"
fit
## 
## Call:
## lm(formula = mpg ~ wt + hp, data = mtcars)
## 
## Coefficients:
## (Intercept)           wt           hp  
##    37.22727     -3.87783     -0.03177
x
## [1] 3.141593
xx
## Error in eval(expr, envir, enclos): object 'xx' not found

Now, what if you want fit from first-chunk and xx which was created in second-chunk?

rm(list = ls())
ls()
## character(0)
qwraps2::lazyload_cache_dir(path = "report_cache/html")
## Lazyloading: report_cache/html/first-chunk
## Lazyloading: report_cache/html/second-chunk
qwraps2::lazyload_cache_labels("first-chunk", path = "report_cache/html")
## Lazyloading report_cache/html/first-chunk_bf368425c25f0c3d95cac85aff007ad1
ls()
## [1] "fit" "x"   "xx"
fit
## 
## Call:
## lm(formula = mpg ~ wt + hp, data = mtcars)
## 
## Coefficients:
## (Intercept)           wt           hp  
##    37.22727     -3.87783     -0.03177
x
## [1] 3.141593
xx
## [1] 2.718282

A better solution would be to use unique object names.

Lastly, say you only want the fit from first-chunk and no other objects. The optional filter argument of lazyload_cache_labels can be used. The filter argument is passed to lazyLoad. “filter: An optional function which when called on a character vector of object names returns a logical vector: only objects for which this is true will be loaded.”

rm(list = ls())
ls()
## character(0)
qwraps2::lazyload_cache_labels("first-chunk",
                               path = "report_cache/html",
                               filter = function(x) x == "fit")
## Lazyloading report_cache/html/first-chunk_bf368425c25f0c3d95cac85aff007ad1
ls()
## [1] "fit"
fit
## 
## Call:
## lm(formula = mpg ~ wt + hp, data = mtcars)
## 
## Coefficients:
## (Intercept)           wt           hp  
##    37.22727     -3.87783     -0.03177

The lazyload_cache_dir and lazyload_cache_labels functions are currently in the developmental version of qwarps2. You can get the code and install the package from https://github.com/dewittpe/qwraps2. I’m not sure when the next version of qwraps2 will be pushed to CRAN, so the developmental version will have to do for now.

If you find any bugs or have suggestions on how to improve the functions please create and issue.

Note, the behavior of knitr::load_cache is very different from the behavior of qwraps2::lazyload_cache_dir and qwraps2::lazyload_cache_labels. knitr::load_cache is to be used to load cached values within a .Rmd prior to the chunk being called. See the documentation and example 114-load-cache.Rmd from the knitr-examples repository for the use of knitr::load_cache.

To leave a comment for the author, please follow the link and comment on their blog: Peter E. DeWitt.

R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)