This seems to work well for “exploratory programming”, where you write random code to test out ideas. Is this comparison with R and RStudio valid? Is this an efficient way of using Julia, besides notebooks? I feel like this is exactly what the Julia VSCode extension was made to support. I mean, this is a full-fledged IDE, so tons of features. You get language server support, highlighting, and the code is automatically saved. The main benefit compared to working in the terminal is that you have a Julia file where you write and evaluate your code. I even set up the same editor layout as in RStudio: code to the left, plots to the right, the terminal is the bottom panel. Observe results as plots or printouts.The code I want to evaluate is dumped into the Julia REPL and then executed.The Julia VSCode extension lets me do the exact same thing: However, the bottom panel is the R REPL! Shift+Enter simply dumps lines of code in there, so working in RStudio is working “in the REPL”, all the time. Recently it struck me that R is used “from the REPL” too: in RStudio, you write code in a file, Shift+Enter to run a block of code and observe results either as plots or as printouts in the bottom panel. So I quickly switched to Jupyter and Pluto notebooks for Julia development. It used to be the weirdest thing to me, since I kept using the REPL literally from the terminal. The most basic Julia workflows involve using a text editor in conjunction with the julia command line. Julia’s REPL provides rich functionality that facilitates an efficient interactive workflow. So I frequently hear that Julia is supposed to be interactive, where you work mostly in the REPL:
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |