comparescript

comparescript is an R package designed to compare variables across two R scripts — when those scripts define variables with the same names. It is built to detect whether variables (e.g., data frames, vectors, lists) are identical between the scripts evaluated dynamically.

---

✅ Current Features

• 📂 Scripts loaded and evaluated in separate environments
  Same-named variables do not overwrite.

• 🔍 Compare variables across environments
  Supports identical() comparison across a user-defined set of variables (regardless of type).

• 🤫 Silent evaluation
  Messages, warnings, and console output from user scripts are suppressed for a clean experience.

• 🚫 Script preprocess
  Strips off installation-related calls like install.packages() and google drive downloads like drive_download before evaluation. This way, the evaluation of scripts presumes the required packages are already installed. If not, an expected error is thrown from the respective library call.

---

🚧 To be considered in the future

• 📦 Runtime package handling
  Currently assumes all packages used in user scripts are pre-installed. A future version may offer dynamic checks or prompts.

• 📜 Better script diagnostics Add options to log which variables were found, skipped, or mismatched — possibly with type or structure info.

• 🔧 Partial or fuzzy variable comparison
  Add support for comparing structure or summary statistics, not just identical() equality. ✅ ADDED, needs to be checked rigorously!

• 🧹 Stronger script sanitization
  Could add more preprocessing details for avoiding unwanted side effects such as downloads, file writes etc.

---

🪄 Install

To use the package, we'll install it from GitHub. For this, use:

devtools::install_github("tolgatabanli/comparescript")
library(comparescript)

Example Workflow

Assuming we have a solution sheet (reference.R) and a student assignment (student_assignment.R), we can compare the values of the variable after a dynamic evaluation of both scripts. The assessor can also select variables to compare and assign so-called weigths, scores, for each variable. If both given, these vectors should be of same length and its elements correspond to each other in the same order.

compare_variables("reference.R", "student_assignment.R")
compare_variables("reference.R", "student_assignment.R", c("var1, var2"), c(2, 1)) # var1 gives 2 points and var2 1 points if correct.