From def075db88ae87104c38437a57e9327b078bb804 Mon Sep 17 00:00:00 2001
From: ripley <ripley@00db46b3-68df-0310-9c12-caf00c1e9a41>
Date: Wed, 6 Mar 2019 12:29:44 +0000
Subject: [PATCH] change the default HTTP user agent when libcurl is used

git-svn-id: https://svn.r-project.org/R/trunk@76204 00db46b3-68df-0310-9c12-caf00c1e9a41
---
 doc/NEWS.Rd                     |  5 +++++
 src/library/base/man/options.Rd | 13 ++++++++-----
 src/modules/internet/libcurl.c  | 19 +++++++++++++++++++
 3 files changed, 32 insertions(+), 5 deletions(-)

diff --git a/doc/NEWS.Rd b/doc/NEWS.Rd
index b1cab3d97ee..e6fe6126be0 100644
--- a/doc/NEWS.Rd
+++ b/doc/NEWS.Rd
@@ -302,6 +302,11 @@
       \item \code{stopifnot()} has been simplified thanks to Suharto
       Anggono's proposals to become considerably faster for cheap
       expressions.
+
+      \item The default \sQuote{user agent} has been changed when
+      accessing \samp{http://} and \samp{https://} sites using
+      \samp{libcurl}.  (A site was found which caused \samp{libcurl} to
+      infinite-loop with the previous default.)
     }
   }
 
diff --git a/src/library/base/man/options.Rd b/src/library/base/man/options.Rd
index 58f0e88672b..34b396a2e79 100644
--- a/src/library/base/man/options.Rd
+++ b/src/library/base/man/options.Rd
@@ -1,6 +1,6 @@
 % File src/library/base/man/options.Rd
 % Part of the R package, https://www.R-project.org
-% Copyright 1995-2018 R Core Team
+% Copyright 1995-2019 R Core Team
 % Distributed under GPL 2 or later
 
 \name{options}
@@ -632,10 +632,13 @@ getOption(x, default = NULL)
     \item{\code{help_type}:}{default for an argument of
     \code{\link{help}}, used also as the help type by \code{\link{?}}.}
 
-    \item{\code{HTTPUserAgent}:}{string used as the user agent in HTTP(S)
-      requests.  If \code{NULL}, requests will be made without a
-      user agent header.
-      The default is \code{R (<version> <platform> <arch> <os>)} .}
+    \item{\code{HTTPUserAgent}:}{string used as the \sQuote{user agent} in
+      HTTP(S) requests by \code{\link{download.file}}, \code{\link{url}}
+      and \code{\link{curlGetHeaders}}, or \code{NULL} when requests will
+      be made without a user agent header.  The default is
+      \code{R (<version> <platform> <arch> <os>)}
+      except when \samp{libcurl} is used when it is
+      \code{libcurl/7.<xx>.<y>} for the \samp{libcurl} version in use.}
 
     \item{\code{install.lock}:}{logical: should per-directory package
       locking be used by \code{\link{install.packages}}?  Most useful
diff --git a/src/modules/internet/libcurl.c b/src/modules/internet/libcurl.c
index 2b4d8e73d1a..407a71f5150 100644
--- a/src/modules/internet/libcurl.c
+++ b/src/modules/internet/libcurl.c
@@ -236,6 +236,7 @@ static void curlCommon(CURL *hnd, int redirect, int verify)
 	curl_easy_setopt(hnd, CURLOPT_SSL_VERIFYHOST, 0L);
 	curl_easy_setopt(hnd, CURLOPT_SSL_VERIFYPEER, 0L);
     }
+#if 0
     // for consistency, but all utils:::makeUserAgent does is look up an option.
     SEXP sMakeUserAgent = install("makeUserAgent");
     SEXP agentFun = PROTECT(lang2(sMakeUserAgent, ScalarLogical(0)));
@@ -246,6 +247,24 @@ static void curlCommon(CURL *hnd, int redirect, int verify)
     if(TYPEOF(sua) != NILSXP)
 	curl_easy_setopt(hnd, CURLOPT_USERAGENT, CHAR(STRING_ELT(sua, 0)));
     UNPROTECT(2);
+#else
+    int Default = 1;
+    SEXP sua = GetOption1(install("HTTPUserAgent")); // set in utils startup
+    if (TYPEOF(sua) == STRSXP && LENGTH(sua) == 1 ) {
+	const char *p = CHAR(STRING_ELT(sua, 0));
+	if (p[0] && p[1] && p[2] && p[0] == 'R' && p[1] == ' ' && p[2] == '(') {
+	} else {
+	    Default = 0;
+	    curl_easy_setopt(hnd, CURLOPT_USERAGENT, p);
+	}
+    }
+    if (Default) {
+	char buf[20];
+	curl_version_info_data *d = curl_version_info(CURLVERSION_NOW);
+	snprintf(buf, 20, "libcurl/%s", d->version);
+	curl_easy_setopt(hnd, CURLOPT_USERAGENT, buf);
+    }
+#endif
     int timeout0 = asInteger(GetOption1(install("timeout")));
     long timeout = timeout0 = NA_INTEGER ? 0 : 1000L * timeout0;
     curl_easy_setopt(hnd, CURLOPT_CONNECTTIMEOUT_MS, timeout);