chore: Avoid aliases in Stimulus

[krlmlr]

We can now handle aliases directly in the R code, this removes one layer of indirection.

Implemented by the following script:

library(conflicted)
library(tidyverse)
conflicted::conflicts_prefer(purrr::set_names)

yaml <- yaml::read_yaml("tools/stimulus/functions-R.yaml")
yaml


# Update functions-R.yaml -------------------------------------------------------------

non_name_list <- map(yaml, ~ .x[names(.x) != "NAME"])
has_non_name <- lengths(non_name_list) != 0
name_list <- map(yaml, ~ .x[names(.x) == "NAME"])
has_name <- lengths(name_list) != 0
names <- unlist(map(yaml, ~ .x["NAME"][[1]]))

search <- set_names(paste0("NAME: ", rex::escape(str_replace(names, "^igraph_", "")), "$"), names(has_name[has_name]))

text <- brio::read_lines("tools/stimulus/functions-R.yaml")
matches <- map(search, ~ str_which(text, .x))

stopifnot(lengths(matches) == 1)
stopifnot(has_name | has_non_name)

delete_three <- names(has_non_name[!has_non_name])

matches[delete_three] <- map(matches[delete_three], ~ .x - 0:2)

new_text <- text[-unlist(matches)]
brio::write_lines(new_text, "tools/stimulus/functions-R.yaml")


# Replace in code ---------------------------------------------------------------------

names_sorted <- names[order(-nchar(names))]

search <- paste0(names_sorted, "_impl")
replace <- paste0(str_replace(names(names_sorted), "^igraph_", ""), "_impl")

replace_all <- function(x) {
  reduce2(search, replace, ~ str_replace_all(..1, fixed(..2), ..3), .init = x)
}

replace_all_in_file <- function(path) {
  text <- brio::read_file(path)
  new_text <- replace_all(text)
  brio::write_file(new_text, path)
}

walk(fs::dir_ls("R", glob = "*.R"), replace_all_in_file)

[krlmlr]

@maelle: Will there be conflicts with your work? I can rerun my code very easily on top of your changes if needed.

[codecov]

Codecov Report

Merging #722 (68f77eb) into main (2f39a9d) will increase coverage by 0.02%.
The diff coverage is 100.00%.

❗ Current head 68f77eb differs from pull request most recent head f7d6cdd. Consider uploading reports for the commit f7d6cdd to get more accurate results

@@            Coverage Diff             @@
##             main     #722      +/-   ##
==========================================
+ Coverage   53.92%   53.95%   +0.02%     
==========================================
  Files         357      355       -2     
  Lines       73281    73344      +63     
==========================================
+ Hits        39519    39574      +55     
- Misses      33762    33770       +8     

Impacted Files	Coverage Δ
R/aaa-auto.R	76.44% <ø> (ø)
R/bipartite.R	85.71% <ø> (ø)
R/centrality.R	74.75% <ø> (ø)
R/centralization.R	82.60% <ø> (ø)
R/community.R	74.63% <ø> (ø)
R/conversion.R	73.91% <ø> (ø)
R/embedding.R	100.00% <ø> (ø)
R/flow.R	79.13% <ø> (ø)
R/hrg.R	34.32% <ø> (+0.21%)	⬆️
R/interface.R	83.58% <ø> (ø)
... and 8 more

... and 7 files with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[maelle]

Does this rename, or move the definition of, functions used in https://github.com/igraph/rigraph/blob/main/R/zzz-deprecate.R (second arguments of calls to deprecated())? If so, yes that would influence my script, thanks for asking.

[krlmlr]

No, R/zzz.R is not touched. Can we merge it then?

[maelle]

we decided waiting for my finalizing #716

[krlmlr]

Sorry, trying to avoid cascades of blockers here. We'll work around any problems that arise from that.