[REPL] Add line parser for all queries that accept slice criteria

[MaxAtoms]

Adds a REPL line parser to all queries that accept one or multiple slice criteria 🎉

Dataframe Shape
Single criterion (optional)

R> :query @df-shape (18@df3) file://test/testfiles/dataframe.R
Query: df-shape (3 ms)
   ╰ 18@df3: {"colnames":["id","score","level","category"],"cols":[3,4],"rows":[0,4]}
All queries together required ≈11 ms (1ms accuracy, total 12 ms)

Lineage
Single criterion

R> :query @lineage (11@sum) file://test/testfiles/example.R
Query: lineage (11 ms)
   ╰ 11@sum: {40, 0, 23, built-in:sum, ... (see JSON)}
All queries together required ≈11 ms (1ms accuracy, total 12 ms)

Location-Map
Multiple criteria (optional)

R> :query @location-map (11@sum;12@product) file://test/testfiles/example.R
Query: location-map (3 ms)
   ╰ File List:
      ╰ 0: `test/testfiles/example.R`
   ╰ Id List: {40, 48}
All queries together required ≈6 ms (1ms accuracy, total 8 ms)

Origin
Single criterion

R> :query @origin (11@sum) file://test/testfiles/example.R
Query: origin (11 ms)
   ╰ Origins for {11@sum}
        ╰ {"type":0,"id":0}, {"type":0,"id":23}

Resolve Value
Multiple criteria

R> :query @resolve-value (1@sum;4@N) file://test/testfiles/example.R
Query: resolve-value (7 ms)
   ╰ Values for {1@sum, 4@N}
        ╰ [0L, 0L], [10L, 10L]
All queries together required ≈7 ms (1ms accuracy, total 9 ms)

Static Slice
Multiple criteria, backward and forward slice

R> :query @static-slice (11@sum) file://test/testfiles/example.R
Query: static-slice (9 ms)
   ╰ Slice for {11@sum}
     ╰ Code (newline as <code>&#92;n</code>): <code>sum <- 0\nw <- 7\nN <- 10\nfor(i in 1:(N-1)) sum <- sum + i + w\nsum</code>
All queries together required ≈9 ms (1ms accuracy, total 10 ms)

R> :query @static-slice (11@sum)f file://test/testfiles/example.R
Query: static-slice (7 ms)
   ╰ Slice for {11@sum}
     ╰ Code (newline as <code>&#92;n</code>): <code>cat("Sum:", sum, "\n")</code>
All queries together required ≈7 ms (1ms accuracy, total 8 ms)

---

Example Code

test/testfiles/dataframe.R:

library(dplyr)

df1 <- data.frame( 
	id = 1:4, 
	age = c(25, 32, 35, 40),
	score = c(90, 85, 88, 92)
)
df2 <- data.frame( 
	id = c(1, 2, 4),
	category = c("A", "B", "A") 
)
df3 <- df1 %>% 
	filter(age > 30) %>% 
	mutate(level = score^2) %>% 
	left_join(df2, by = "id") %>%
	select(-age)
	
print(df3$level)

[MaxAtoms]

We look for a character sequence here, but only skip the first character of the sequence...any ideas on how to write this more clearly, @EagleoutIce? 🤔

[EagleoutIce]

what do you mean with skipping the first character here? we could also try a peek ahead that pattern matches against a regex for that matter, but in genreal i dont think this looks that bad

[MaxAtoms]

We match on the split string (e.g. ;:) and if it is present, we forward over the semicolon and start the next argument with the :. That behavior is correct for our single specific use case currently, but it might be surprising in general.

[MaxAtoms]

Todo: Use regex to allow arbitrary number of whitespace chars between the semicolon and colon