similar to: question about R_ParseVector function

Dear R developers, just a quick question: I noted that R_ParseVector () has gained an additional parameter (since SVN rev. 39999). Since R_ParseVector () is part of the public API, I'm wondering, whether this API change was intentional or not. Either way is fine with me, but in order to make sure the next release of RKWard will be compilable with R 2.5.0, I'd like to make sure,

When large calls cause errors R may stall for extended periods.? This is particularly likely to happen with `do.call`, as in this example with a 24 second stall: ??? x <- runif(1e7) ??? system.time(do.call(paste0, list(abs, x)))? # intentional error ??? ## Error in (function (..., collapse = NULL)? : ??? ##?? cannot coerce type 'builtin' to vector of type 'character' ??? ##

I have just committed some changes to R-devel (which will become R 2.5.0 next spring) to add source references to parsed R code. Here's a description of the scheme: The design is done through 2 old-style classes. "srcfile" corresponds to a source file: it contains a filename, the working directory in which that filename is to be interpreted, the last modified timestamp of the

I noticed the following file descriptor leak when I couldn't remove a package unless I shut down the R session that had loaded and used it. The function that triggered the problem printed the output of a call to parse(). Each time one prints a srcref a connection is opened and not closed. It looks like it happens in as.character.srcref's call to getSrcLines, which has some logic I

Luke, thanks for considering the issue.? I would like to try to separate the problem into two parts, as I _think_ your comments address primarily part 2 below: 1. How can we avoid significant and possibly crippling ?? stalls on error with these non-standard calls. 2. What is the best way to view these non-standard calls. I agree that issue 2. requires further thought and discussion under a

Hello, Consider this file (/tmp/test.R) : <file> f <- function( x, y = 2 ){ z <- x + y print( z ) } </file> I get this in R 2.7.2 : > p <- parse( "/tmp/test.R" ) > str( attr( p, "srcref" ) ) List of 1 $ :Class 'srcref' atomic [1:4] 1 1 4 1 .. ..- attr(*, "srcfile")=Class 'srcfile' length 4 <environment>

We also have a few other suggestions and wishes about backtrace storage and display on the one hand, and display of constructed calls on the other hand. Perhaps it would be better to open a different wishlist item for traceback() to keep the discussions focused? FWIW I think deparsing backtraces lazily is a great idea. Displaying 1 line per call by default in interactive sessions, while being

Hi Duncan ? Wonderful. Thank you. Bug or no bug, I think it would be a huge improvement for user-friendliness if R printed the last line by default *every time* a script dies. Most computer languages do so. Should I file it as a request for improvement to the R code development team? Maybe R can be improved at a very low cost to the development team and a very high benefit to newbies.

On 2025-01-18 8:27 p.m., Ivo Welch wrote: > I am afraid my errors are worse! (so are my postings. I should have > given an example.) > > ``` > x <- 1 > y <- 2 > nofunction("something stupid I am doing!") > z <- 4 > ``` > > and > > ``` >> source("where-is-my-water.R") > Error in nofunction("something stupid I am

On Sun, 19 Jan 2025, Ivo Welch wrote: > Hi Duncan ? Wonderful. Thank you. Bug or no bug, I think it would be > a huge improvement for user-friendliness if R printed the last line by > default *every time* a script dies. Most computer languages do so. > > Should I file it as a request for improvement to the R code > development team? Maybe R can be improved at a very low cost

Specifically, the srcfile attribute of the srcref attribute of the two instances of the functions contain different environments, AFAICT. Environments are compared only by exact pointer, so this forces return FALSE. Snippets from .Internal(inspect(x)) and .Internal(inspect(y)): @cca008 03 CLOSXP g0c0 [MARK,NAM(2),ATT] FORMALS: @604b58 00 NILSXP g0c0 [MARK,NAM(2)] BODY: @cc9650 06 LANGSXP

First, why does this expression have a 'srcref' element: > exprA <- substitute(function(x) a*x, list(a=2)) > print(exprA) function(x) 2 * x > str(as.list(exprA)) List of 4 $ : symbol function $ :Dotted pair list of 1 ..$ x: symbol $ : language 2 * x $ :Class 'srcref' atomic [1:8] 1 20 1 34 20 34 1 1 .. ..- attr(*, "srcfile")=Classes

Thanks for pointing out the options. Using options(show.error.locations = TRUE) works on Ivo's example, but it doesn't show a location if the error happens in a function that doesn't have srcrefs, because the known location isn't on the top of the stack. Perhaps TRUE (and maybe "top"?) should look back through the stack until it finds a known location, and

Just checking in to see whether it is intended or not that assigning new formals to a function/closure causes any attributes to be dropped: EXAMPLE: > fcn <- structure(function() {}, foo="foo", class=c("foo")) > str(fcn) function () - attr(*, "srcref")=Class 'srcref' atomic [1:8] 1 18 1 30 18 30 1 1 .. ..- attr(*, "srcfile")=Classes

I've posted a patch to bugs.r-project.org that fixes the traceback() issue. It's not specific to findFun3; you get the same problem with errors from expressions like 1 + "a" In my testing, I occasionally saw cases where show.error.locations = TRUE didn't work. I'll try to track down a reproducible case, and perhaps a patch to fix it. Duncan Murdoch On

Hello, I submit a patch for review that implements code coverage tracing in the R interpreter. It records the lines that are actually executed and their associated frequency for which srcref information is available. I perfectly understands that this patch will not make its way inside R as it is, that they are many concerns of stability, compatibility, maintenance and so on. I would like to have

Sorry, I'm not seeing the first problem now: options(show.error.locations = TRUE) works fine. Not sure what I did wrong before. I'm still seeing `traceback()` failing to report the attempt to call nofunction(). I suppose this is because a context is never set up for the failed call. Perhaps findFun3 could set up a fake context so that traceback() adds one more entry? Duncan

Dear R-devel, In a package of mine, I use removeSource on expression objects in order to make expressions that are semantically the same serialize to the same byte sequences: https://github.com/cran/depcache/blob/854d68a/R/fixup.R#L8-L34 Today I learned that expressions containing function definitions also contain the source references for the functions, not as an attribute, but as a separate

And now I've found the elusive bug in show.error.locations = TRUE as well, and posted a patch for that too. Back to the original topic of this thread: I think most users should routinely use options(show.error.locations = TRUE) e.g. in their .Rprofile. That will often speed up bug fixes, especially if my patch is incorporated. Duncan Murdoch On 2025-01-20 5:56 p.m., Duncan Murdoch

understood. but, please, consider not people like me but unwary beginners and students of R. I have used R now for decades, and even I am baffled by it. Couldn't you make R code easier to debug not only for people like me (who can indeed tweak their environments) but also for novice users? On Sun, Jan 19, 2025 at 8:46?AM <luke-tierney at uiowa.edu> wrote: > > On Sun, 19 Jan