libfetchers/github: allow slashes in refs

Refs #4061

2 files changed, 18 insertions, 4 deletions

diff --git a/src/libfetchers/github.cc b/src/libfetchers/github.cc
index 92ff224f7..3d1cc15e2 100644
--- a/src/libfetchers/github.cc
+++ b/src/libfetchers/github.cc
@@ -37,15 +37,29 @@ struct GitArchiveInputScheme : InputScheme
         std::optional<std::string> ref;
         std::optional<std::string> host_url;
 
-        if (path.size() == 2) {
-        } else if (path.size() == 3) {
+        auto size = path.size();
+        if (size == 3) {
             if (std::regex_match(path[2], revRegex))
                 rev = Hash::parseAny(path[2], htSHA1);
             else if (std::regex_match(path[2], refRegex))
                 ref = path[2];
             else
                 throw BadURL("in URL '%s', '%s' is not a commit hash or branch/tag name", url.url, path[2]);
-        } else
+        } else if (size > 3) {
+            std::string rs;
+            for (auto i = std::next(path.begin(), 2); i != path.end(); i++) {
+                rs += *i;
+                if (std::next(i) != path.end()) {
+                    rs += "/";
+                }
+            }
+
+            if (std::regex_match(rs, refRegex)) {
+                ref = rs;
+            } else {
+                throw BadURL("in URL '%s', '%s' is not a branch/tag name", url.url, rs);
+            }
+        } else if (size < 2)
             throw BadURL("URL '%s' is invalid", url.url);
 
         for (auto &[name, value] : url.query) {
diff --git a/src/libutil/url-parts.hh b/src/libutil/url-parts.hh
index 68be15cb0..e0e2809fd 100644
--- a/src/libutil/url-parts.hh
+++ b/src/libutil/url-parts.hh
@@ -22,7 +22,7 @@ const static std::string absPathRegex = "(?:(?:/" + segmentRegex + ")*/?)";
 const static std::string pathRegex = "(?:" + segmentRegex + "(?:/" + segmentRegex + ")*/?)";
 
 // A Git ref (i.e. branch or tag name).
-const static std::string refRegexS = "[a-zA-Z0-9][a-zA-Z0-9_.-]*"; // FIXME: check
+const static std::string refRegexS = "[a-zA-Z0-9][a-zA-Z0-9_.\\/-]*"; // FIXME: check
 extern std::regex refRegex;
 
 // Instead of defining what a good Git Ref is, we define what a bad Git Ref is