All connections are in use: execution paused

Question

I use the getYahooData() function in the TTR package quite intensively.

I have this piece of code:

 for(i in 1:nrow(symbol)){ tryCatch(prices <- getYahooData(symbol$symbol[i], from, to, freq="daily", type="price"), warning=function(e) continue <- 0) if (continue==0) next } 

This loop is long, I get this error:

Error in file (file, "rt"): all connections are used. Calls: tryCatch ... doTryCatch → getYahooData → getYahooData → read.table → file Execution paused

What can I do?

UPDATE:

If I use closeAllConnections (), I get:

  I get: *** caught segfault *** address (nil), cause 'memory not mapped' Traceback: 1: getConnection(set[i]) 2: close(getConnection(set[i])) 3: closeAllConnections() aborting ... 

Answer 1

First: never use your life to continue building. It's useless. tryCatch() will continue if you define a handler for an error or warning. He will use this instead of "default" error=function(e) stop(e) . This will stop your function. If you define a handler (either warning= or error= ), your script will not be stopped, so there is no need to continue.

This says: The correct use of tryCatch in this case would be:

 for(i in 1:nrow(symbol)){ tryCatch(prices <- getYahooData(symbol$symbol[i], from, to, freq="daily", type="price"), error = function(e){}) } 

or, if you use it in a script and want to go to the next loop when an error occurs, you can simply use:

 for(i in 1:nrow(symbol)){ prices <- try(getYahooData(symbol$symbol[i], from, to, freq="daily", type="price"), silent=TRUE) if(inherits(prices,"try-error")) { next } # only true if an error occurs ... # rest of calculations } 

If you used this tryCatch method or tried, you would not have the problems that you report here.

Now I can reproduce your case if I use non-existent characters. Your erroneous use of the tryCatch() function is causing problems. read.table returns an error ( Error in file(file, "rt") : cannot open the connection ). This is a mistake , not a warning. You will receive an additional warning that the 404 file was not found.

When a warning is issued along with an error, the function of the warning handler is first processed. This is because before a function can be stopped, a warning must be raised. Thus, it will not handle the error received , which means that on.exit(close(file)) in read.table() will not be called. Therefore, the connection is not closed correctly and is still considered open, although it can no longer be found using R (showAllConnections () shows nothing). Since the error is not addressed, something is wrong in registering the connection. Since the connection cannot be opened, on.exit(close(...)) will have no effect. showConnections() does not show the connection, but for some reason R is still thinking about it. Consequently, all hell breaks and you break your R.

thanks for the corrections @Tommy

A simple code example to illustrate this:

 myfun <- function(x){ if(x>1) warning("aWarning") stop("aStop") x } tryCatch(myfun(0.5), warning=function(w)print("warning"), error=function(e) print("stop")) [1] "stop" tryCatch(myfun(1.5), warning=function(w)print("warning"), error=function(e) print("stop")) [1] "warning" 

In short:

• Check the characters used. They are probably mistaken.
• never use a warning handler if you expect errors.

And as an extra: your loop will return the result of the last call, as you overwrite prices every time you go through the loop, in case you used the correct characters.

Edit: in case you want to continue the action

Answer 2

Close some connections? It may be as simple as inserting closeAllConnections() at the end of this loop body.

Answer 3

This is really a mistake in the R source code regarding how connections are logged. I posted some comments and a patch on the R Bugzilla website here: http://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14660 . Yoris’s recommendations sound. However, closeAllConnections () will also work when the error is fixed. (PS this is my first StackOverflow post. Forgive me if I violated etiquette.)

Answer 4

Somewhere in my last dig, someone mentioned that using a link to a URL in a function like read.table.url () or url () caused an incorrect native HTTP HTTP R connector. What works best to close connections when an error occurs is to explicitly call RCurl inside the read.table () function call or the equivalent function call. For example, this created problems when accumulating a large number of HTTP errors:

 result <- try ( DF <- read.table(con <- url(url), col.names=colNames), silent=TRUE ) 

I saw good results associated with modifying RCurl by calling the getURL function:

 result <- try ( DF <- read.table(textConnection(getURL(url)), col.names=colNames), silent=TRUE ) 

This is when starting R v2.15.3.