In skranz/armd: Academic RMarkdown

There are different levels of precomputation. Different objects will be available in different situations:

Environemnts:

precompilation:

task.env (non chunks) <- pre.env <- init.env <- parent.env(global.env)

runtime:

task.env (for chunks) <- init.env <- parent.env(global.env)

• init_compute blocks and extra.code.file
• Objects stored in init.env
• Will compute before any chunk is run
• Will be available at compile time and during runtime of the problemset

• Only in chunks and blocks that come after the precompute statement

• precompute chunks

• Know and write into pre.env

• preknit chunks

• Know pre.env but don't write into it, unless we also have precompute=TRUE

Interactive chunks:

• If prerun = TRUE in settings, all chunks will already be run at compile time. By default, we then also have knit.prerun = TRUE and store.prerun = TRUE.

• All chunks know objects from init.env

Task Lines

• Each chunk belongs to a task.line and may have additional task.in.
• Knows all objects from earlier chunk in task.line and from all chunks in the task.in lines

• All earlier chunks in task.line must be solved and also chunks from additional task.in

• Default lines:

• A new task.line starts by default in a new exercise / section
• A note will by default have an own task.line with an in.line equal to the surrounding section / exercise
• An optional chunk will have a separate task.line but uses the surrounding task.line as task.in
• A non-chunk task by default has no task.in and starts a single task.line

skranz/armd documentation built on Sept. 4, 2020, 12:22 p.m.