type.convert arg

[tdhock]

Closes #28

TODO add tests

[MichaelChirico]

nice, really like that we get this "default with override" behavior "for free"

[MichaelChirico]

LGTM and it's definitely what I had in mind.

Glancing at the PR, I wonder if it would be a better API to instead offer "default.coercion=identity", then I can use default.coercion=type.convert but also we allow even more flexibility.

put another way, before we have (1) only custom column-wise conversion allowed, here we have (2) custom column-wise conversion, possibly defaulting to type.convert(). But maybe we could get (3) custom column-wise conversion backing up to custom default conversion.

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 100.00%. Comparing base (487f89b) to head (1f364b1).
Report is 25 commits behind head on master.

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff            @@
##            master       #29   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           18        18           
  Lines          433       437    +4     
=========================================
+ Hits           433       437    +4     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[tdhock]

thanks for sharing your additional ideas. That makes sense to me but I kept the argument name as type.convert instead of your suggested default.coercion because "coercion" was not mentioned anywhere in the code/docs, but "convert" and "conversion" were.

[tdhock]

also I noticed that I had to specify type.convert(x,as.is=TRUE) to get character output and avoid a warning, so I kept the type.convert=TRUE alias for that as a shortcut.

[tdhock]

do you have a compelling example in mind that we could add to a man page or vignette?

For the SLURM sacct example (v1-capture-first vignette) we convert everything to integer but the existing example handles that by integrating the conversion function with the pattern:

(sacct.df <- data.frame(
  Elapsed = c(
    "07:04:42", "07:04:42", "07:04:49",
    "00:00:00", "00:00:00"),
  JobID=c(
    "13937810_25",
    "13937810_25.batch",
    "13937810_25.extern",
    "14022192_[1-3]",
    "14022204_[4]"),
  stringsAsFactors=FALSE))
int.pattern <- list("[0-9]+", as.integer)
range.pattern <- list(
  "\\[",
  task1=int.pattern,
  list(
    "-",#begin optional end of range.
    taskN=int.pattern
  ), "?", #end is optional.
  "\\]")
nc::capture_first_df(sacct.df, JobID=range.pattern, nomatch.error=FALSE)

gives

> nc::capture_first_df(sacct.df, JobID=range.pattern, nomatch.error=FALSE)
    Elapsed              JobID task1 taskN
     <char>             <char> <int> <int>
1: 07:04:42        13937810_25    NA    NA
2: 07:04:42  13937810_25.batch    NA    NA
3: 07:04:49 13937810_25.extern    NA    NA
4: 00:00:00     14022192_[1-3]     1     3
5: 00:00:00       14022204_[4]     4    NA

[MichaelChirico]

do you have a compelling example in mind that we could add to a man page or vignette?

I came to this via Advent of Code, specifically this year's Day 14 problem:

https://github.com/MichaelChirico/advent-of-code/blob/dc272d9c342fa4cb35e8519f5f41cd8d3e17b54d/2024/14.R#L1-L14

https://adventofcode.com/2024/day/14

In case that requires log-in to view, here's a sample file:

https://github.com/MichaelChirico/advent-of-code/blob/main/2024/input-data/14-test

The Advent of Code problems generally provide good motivating examples for {nc}, as the problem input often takes the form of structured text files from which you need to extract data in a friendlier format.

[tdhock]

thanks for sharing the example.