wangyuchen opened a new issue #7058: URL: https://github.com/apache/arrow/issues/7058
https://github.com/apache/arrow/blob/d1f757339c24ff06e1d39f878323878be64e0369/r/tools/linuxlibs.R#L39-L47 I think there might be a bug in this utility function. In the installation script for Linux, it uses this function to try to download things and capture if download is successful. However, a `tempfile()` is always created *before* the attempt to download, and the tempfile path is passed to the `to_file` argument. This behaves as expected when `libcurl` method is used for downloading files. But with `wget` or `curl`, when the download is unsuccessful, the file would still exist, so `file.exist()` will still return `TRUE`, as if the download is successful. This potentially can guide the installation on the wrong path. Reprex: ``` r base_url <- "https://dl.bintray.com/ursalabs/arrow-r/libarrow/bin/centos-7/"; wrong_file <- "arrow-999999.zip" quietly <- FALSE try_download <- function(from_url, to_file) { try( suppressWarnings( download.file(from_url, to_file, quiet = quietly) ), silent = quietly ) file.exists(to_file) } options(download.file.method = "curl") try_download(paste0(base_url, wrong_file), tempfile()) #> [1] TRUE options(download.file.method = "wget") try_download(paste0(base_url, wrong_file), tempfile()) #> Error in download.file(from_url, to_file, quiet = quietly) : #> 'wget' call had nonzero exit status #> [1] TRUE options(download.file.method = "libcurl") try_download(paste0(base_url, wrong_file), tempfile()) #> Error in download.file(from_url, to_file, quiet = quietly) : #> cannot open URL 'https://dl.bintray.com/ursalabs/arrow-r/libarrow/bin/centos-7/arrow-999999.zip' #> [1] FALSE ``` <sup>Created on 2020-04-28 by the [reprex package](https://reprex.tidyverse.org) (v0.3.0)</sup> ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
