search

MichaelChirico / r-bugs

A ⚠️read-only⚠️mirror of https://bugs.r-project.org/
20 stars 0 forks source link

[BUGZILLA #472] R-crash using cut-and-paste #614

Closed MichaelChirico closed 4 years ago

MichaelChirico commented 4 years ago

From: "Trenkler, Dietrich" <dtrenkler@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>-osnabrueck.de> Dear R-Team,

sorry for bugging (:-)), it's me again. In the FAQ coming along with the R distribution it reads:

If R executes an illegal instruction, or dies with an operating system error message that indicates a problem in the program (as opposed to something like "disk full"), then it is certainly a bug.

This encourages me to describe a crash I encountered a few minutes ago.

I had written a function in an editor, copied it to the clip-board and tried to establish it in my workspace (cut and paste). Suddenly I got the following message form Windows NT (sorry it's german):

Die Anweisung "0x7762d4be" verweist auf Speicher in "0x0247c000". Der Vorgang "read" konnte nicht auf dem Speicher durchgeführt werden. Klicken Sie auf "OK um die Anwendung zu beenden.

Rgui.exe-Fehler in Anwendung

There are very technical terms which I hope will be translated such that it makes sense to you:

The instruction "0x7762d4be" points to memory in "0x0247c000". "read" could not be carried out in memory. Click on "OK to stop the program.

Clicking on OK caused R to close (and of course the most recent work was lost)

--please do not edit the information below--

Version: platform = Windows arch = x86 os = Win32 system = x86, Win32 status = major = 1 minor = 0.0 year = 2000 month = February day = 29 language = R

Windows NT 4.0 (build 1381) Service Pack 3

Search Path: .GlobalEnv, Autoloads, package:base

Regards,

D.Trenkler

==================================================================== Dietrich Trenkler (trenkler@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>-osnabrueck.de) Statistik / Empirische Wirtschaftsforschung Universitaet Osnabrueck <CENSORING FROM DETECTED PHONE NUMBER ONWARDS; SEE BUGZILLA>

METADATA

MichaelChirico commented 4 years ago

From: Prof Brian D Ripley <ripley@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>> On Tue, 7 Mar 2000 dtrenkler@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>-osnabrueck.de wrote:

Dear R-Team,

sorry for bugging (:-)), it's me again. In the FAQ coming along with the
R distribution it reads:

If R executes an illegal instruction, or dies with an operating system
error message that indicates a problem in the program (as opposed to
something like "disk full"), then it is certainly a bug.

This encourages me to describe a crash I encountered a few minutes ago.

I had written a function in an  editor, copied it to the clip-board and
tried to establish it in my workspace (cut and paste). Suddenly I got
the following message form Windows NT (sorry it's german):

[...]

How big was the function you were trying to paste in? There are distinct limits to such cut-and-paste, not all of which are under our control. To quote more of the FAQ:

Please include an example that reproduces the problem, preferably the simplest one you have found.

Without such an example, there is nothing we can do to help.

As a general work-around, you will find it much easier to use source() to load in sizeable functions than cut-and-paste.

-- Brian D. Ripley, ripley@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::> Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ <CENSORING FROM DETECTED PHONE NUMBER ONWARDS; SEE BUGZILLA>

METADATA

MichaelChirico commented 4 years ago

From: Uwe Ligges <ligges@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>-dortmund.de> Hi!

Prof Brian D Ripley schrieb:


On Tue, 7 Mar 2000 dtrenkler@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>-osnabrueck.de wrote:

> Dear R-Team,
>
> sorry for bugging (:-)), it's me again. In the FAQ coming along with the
> R distribution it reads:
>
> If R executes an illegal instruction, or dies with an operating system
> error message that indicates a problem in the program (as opposed to
> something like "disk full"), then it is certainly a bug.
>
> This encourages me to describe a crash I encountered a few minutes ago.
>
> I had written a function in an  editor, copied it to the clip-board and
> tried to establish it in my workspace (cut and paste). Suddenly I got
> the following message form Windows NT (sorry it's german):

...

Happens here too (R 1.0.0, Windows NT 4.0, SP5). Sometimes with Dr. Watson, sometimes as described above by D.Trenkler, sometimes R terminates without any message.

How big was the function you were trying to paste in?  There are
distinct limits to such cut-and-paste, not all of which are under
our control. To quote more of the FAQ:

Please include an example that reproduces the problem, preferably the
simplest one you have found.

Without such an example, there is nothing we can do to help.

If you paste the following function about four times (makes no sense, but reproduces the error without pasting a really big function), the error occurs: -------------

"scoredach" <- function (x) { v1 <- x[1] u11<- x[2] u12<- x[3] v2 <- x[4] u21<- x[5] u22<- x[6] v3 <- x[7] u31<- x[8] u32<- x[9] w1 <- x[10] w2 <- x[11] w3 <- x[12] z11<- x[13] z21<- x[14] z31<- x[15] z12<- x[16] z22<- x[17] z32<- x[18] scoredach <- v1/(1 + exp(-(u11 xseq + u12 yseq))) + v2/(1 + exp(-(u21 xseq + u22 yseq))) + v3/(1 + exp(-(u31 xseq + u32 yseq)))

return(scoredach)}

---------------------

As a general work-around, you will find it much easier to use source() to
load in sizeable functions than cut-and-paste.

Regards, Uwe Ligges

METADATA

MichaelChirico commented 4 years ago

From: "Trenkler, Dietrich" <dtrenkler@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>-osnabrueck.de>

-----Ursprüngliche Nachricht-----
Von:  Prof Brian D Ripley 
Gesendet am:  Tuesday, March 07, 2000 11:30 AM
An:   dtrenkler@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>-osnabrueck.de
Cc:   r-devel@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>; R-bugs@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>
Betreff:      Re: [Rd] R-crash using cut-and-paste (PR#472)

On Tue, 7 Mar 2000 dtrenkler@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>-osnabrueck.de wrote:

> Dear R-Team,
> 
> sorry for bugging (:-)), it's me again. In the FAQ coming along with
the
> R distribution it reads:
> 
> If R executes an illegal instruction, or dies with an operating
system
> error message that indicates a problem in the program (as opposed to
> something like "disk full"), then it is certainly a bug.
> 
> This encourages me to describe a crash I encountered a few minutes
ago.
> 
> I had written a function in an  editor, copied it to the clip-board
and
> tried to establish it in my workspace (cut and paste). Suddenly I
got
> the following message form Windows NT (sorry it's german):

[...]

How big was the function you were trying to paste in?  
[Dietrich Trenkler]  About 36 lines. Here you are:

tefo_function(form,symb="%",dir="F:/DIDITEX/compstatr/ss00/r2latex/"){ tf_paste(c(dir,form,".r2l"),collapse="") cat(c("Reading ",tf,".\n"),collapse="") ss_scan(file=tf,what="",sep="\n",quiet=TRUE) sssss a(1:length(sss))[ss==as.vector("\end{RFunction}")] ssssss[1:a-1] b(1:length(sss))[ss==as.vector("\begin{RFunction}")]

collect.expressions_function(ss){
ss_ss[grep("%",ss)]
ss_paste(ss,collapse="")
ss_substring(ss,1:nchar(ss),1:nchar(ss))
ss_as.vector(ss[!((ss=="%")|(ss==" "))])
ss_paste(c("list(",ss[-1],")"),collapse="")
ss
}
a_collect.expressions(ss[-(1:a)])
a_paste(c("a0_",a),collapse="")
sss_c(sss[-(1:b)],a)
sss_c(sss,"a0_lapply(a0,signif,digits=digits)")
a_paste(c("ss_paste(c('",tf,,"'),collapse='')"),collapse="")
sss_c(sss,a)
sss_c(sss,"ss_scan(file=ss,what='',sep='\\n',quiet=TRUE)")

sssc(sss,"a(1:length(ss))[ss==as.vector('\\end{RFunction}')]") sss_c(sss,"ss_ss[-(1:a)]") sss_c(sss,"ss_ss[-grep('%',ss)]") sss_c(sss,"ss_paste(ss,collapse='\n',sep='\n')") sss_c(sss,paste("for(i in 1:length(a0)){ss_sub('",symb,"',as.character(a0),ss);a0_a0[-1]}",sep="") ) sss_c(sss,"if(!file=='')")

sss_c(sss,"write(paste(ss,collapse='\n',sep='\n'),file=paste(c(dir,fil e,'.tex'),collapse=''))") sss_c(sss,"else return(cat(ss,'\n'))") sss_c(sss,"}")

eval(parse(text=paste(sss,collapse="\n",sep="\n")),envir=.GlobalEnv) cat("Function ",form," established.\n") }

There are 
distinct limits to such cut-and-paste, not all of which are under 
our control. 
[Dietrich Trenkler]  This seems to be true. When I copied this

function to the clipboard R crashed several times (sometimes without any messages). There seems to be some interferences with the clipboard...

As a general work-around, you will find it much easier to use source()
to
load in sizeable functions than cut-and-paste.
[Dietrich Trenkler]  Henceforth I will obey to your advice.

-- 
Brian D. Ripley,                  ripley@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
<CENSORING FROM DETECTED PHONE NUMBER ONWARDS; SEE BUGZILLA>

-----------

#### METADATA
 - Comment author - Jitterbug compatibility account
 - Timestamp - 2000-03-07 17:06:34 UTC
MichaelChirico commented 4 years ago

NOTES: The problem occurred with a current history with > 512 lines and less than 16K bytes. It will be fixed in 1.0.1 and by an upgrade DLL for 1.0.0.

METADATA

MichaelChirico commented 4 years ago

Audit (from Jitterbug): Sun Mar 12 13:32:14 2000 ripley changed notes Sun Mar 12 13:32:14 2000 ripley moved from incoming to Windows-fixed

METADATA