Master list of most-requested issues

[MichaelChirico]

Redirecting from post at /orgs/.../discussions. Inspired by happening to notice an issue getting more support (through GitHub reactions) and borrowing the idea to scrape the GH API from this nice article over at tidyverse. Here are all open issues with at least 5 positive reactions:

• Delete rows by reference #635 - [
• nafill, setnafill for character, factor and other types #3992 |nafill
• Support for on-the-fly columns in on #1639 |on
• Create an efficient unnest function #2146 | NEW
• Need long-vector support #3957 | NEW
• Set up rowwise() and colwise() functions for use in .SD #1063 |NEW
• [R-Forge #1611] Allow 2 column i matrix to return a list() (or vector if same type) #657 | i
• rolling functions, rolling aggregates, sliding window, moving average #2778 | froll
• Shouldn't fcase() recycle? #4258 | NEW
• rbindlist could use an "ignore.attributes" argument #3911 | rbindlist
• Implement DT[, across(.SD, fun1, fun2, fun3), by=group] #4970 | NEW
• Add psum? #3467 | NEW
• add a 'having' parameter to [.data.table #788 | [
• Upcoming versions of base R eliminate the need for DT() functionality - consider eliminating? #5621 | DT()
• Metadata in data.tables #4804 | NEW
• support OpenMP with system macOS toolchain #4348 | INSTALLATION
• rolling funs / shift could support logical window #3241 | froll
• uniqueN() is very slow compared to length(unique()) #3739 |uniqueN
• join vignette #2181 | JOINS
• allow cross join in [.data.table #1717 | JOINS
• [R-Forge #2461] Faster version of Reduce(merge, list(DT1,DT2,DT3,...)) called mergelist (a la rbindlist) #599 | JOINS
• foverlaps() should be easier to do point-to-interval join #3721 | JOINS
• progress indicator/bar for "by" operations #3060 | NEW
• collect more statistics about the data #2879 | NEW
• Expose setlevels() and document #2219 | NEW
• join column naming with on syntax #1700 | JOINS
• FR: auto-infer LHS of := when absent #1543 | :=
• Should be able to refer to i's .SD during a join. #935 | JOINS
• Expand nomatch to accept any value (control fills, rolls, omits during joins) #857 | JOINS
• Be less conservative about ALTREP #4697 | ALTREP
• output from merge #4677 | JOINS
• Internal error: DT passed to assign has not been allocated enough column slots #4100 | INTERNAL
• nafill new type: approx #4066 | nafill
• GForce optimisation could be more smart #3815 | GForce
• allow.cartesian could allow .N on 2^31+ rows to finish #3009 | JOINS
• Timeline / interest for not yet implemented foverlaps() functionality #3002 | JOINS
• let j = c(prefix = lapply(.SD, f)) work when optimized #2311 | GForce
• implement guniqueN #1120 | GForce
• Allow locking or protecting a data.table from any modifications #1086 | NEW
• Passing named lists to .SDcols / .SD #5020 | .SD
• fpaste: fwrite output as a character vector #4572 | NEW
• New Argument: "join". Objective: ANSI SQL Syntax #3946 | JOINS
• Share data.table among R sessions by reference #3104 | NEW

---

Fulfilled already

• fwrite ignores options(scipen = 999) #2020 |fwrite| - fwrite ignores options(scipen = 999)
• feature request : transpose(keep.rownames = T)? #1886 |transpose| - feature request : transpose(keep.rownames = T)?
• Support for writing to gzip (.gz) with fwrite #2016 - Support for writing to gzip (.gz) with fwrite
• patterns in .SDcols #1878 - patterns in .SDcols
• [FR] gforce mean and sum in parallel #3042 - [FR] grouping in parallel
• shift should accept negative n #1708 - shift should accept negative n
• fwrite UTF8 #1770 |fwrite| - fwrite UTF8
• Vignette for programmatic data.table (eval, get, etc) #4247 | DOCUMENTATION | - Vignette for programmatic data.table (eval, get, etc)
• melt.data.table should offer variable to match on the name, rather than the number #3396 | RESHAPING | - melt.data.table should offer variable to match on the name, rather than the number
• data.table spark/databases interface #1828 |NEW| - [Request] data.table spark interface
• compatibility with sf library #2273 |NEW| - compatibility with sf library
• Additional arguments for setcolorder: before, after #4358 |setcolorder| - Additional arguments for setcolorder: before, after
• improvement on programmatically substituting columns in expressions #2655 | META | - improvement on programmatically substituting columns in expressions
• [R-Forge #5072] Merging using character and factor results in by column becoming NA #499 | JOINS | - [R-Forge #2461] Faster version of Reduce(merge, list(DT1,DT2,DT3,...)) called mergelist (a la rbindlist)
• "not in" operator %!in% #4152 | NEW | - "not in" operator %!in%
• GForce optimize weighted.mean #3977 | GForce | - GForce optimize weighted.mean
• GForce should be able to work with := as well. #1414 |GForce| - GForce should be able to work with := as well.
• Add an alias for functional version of := #3795 |:=| - Add an alias for functional version of :=
• names(.SD) := ... should work #795 | .SD
• File-backed data.tables #1336 |NEW
• What about data.table powered by GPU along with CPU? #3670 | INTERNAL

Separately for fread since there are many:

• Automatic detection of dec=',' in Europe #2431
• [R-Forge #2605] add filtering option to fread so it can load less than all rows #583
• Implement comment.char argument in fread #856
• when will sep2 in fread be implemented? #1162
• fread option to skip lines with format errors (with warning) #711
• [R-Forge #4931] Support file connections for fread #561
• fread tries to map memory for the entire file when using nrows #4329
• fread fails with uneven number of columns when max columns in final row (with fill=TRUE and col.names set) #2691
• fread not reading in column of all character "NA" as character with na.strings = "" #2714
• fread problem with different number of columns #1812
• [Request] fread to support thousand separator #1636
• Chunked fread #1721

Please note that upvoting/reacting to this issue will make the universe collapse, probably.

Code to reproduce:

library(gh)
library(data.table)
# reactions API is "in development" so requires special header:
react_header = 'application/vnd.github.squirrel-girl-preview+json'

repo = 'Rdatatable/data.table'
issue_query_fmt = "/repos/%s/issues?state=open&per_page=100&page=%d"
issues = list()
page = 1L
repeat {
  cat(sprintf("Reading page %d\n", page))
  resp <- gh(sprintf(issue_query_fmt, repo, page), .send_headers = c(Accept = react_header))
  if (!length(resp)) break
  issues = c(issues, resp)
  page = page + 1L
}
    
# filter pull requests
issues = issues[sapply(issues, function(issue) is.null(issue$pull_request))]
names(issues) = as.character(sapply(issues, `[[`, 'number'))

reactions = rbindlist(lapply(issues, `[[`, 'reactions'))

reactions[ , issue := gsub('.*issues/([0-9]+)/reactions', '\\1', url)]
reactions[ , url := NULL]
reactions[ , positive := total_count - `-1` - confused]
reactions[ , setdiff(names(reactions), c('issue', 'positive')) := NULL]

reactions[ , title := sapply(issues[issue], `[[`, 'title')]

# don't include this issue...
reactions[order(-positive)][positive >= 5 & issue != '3189']