Fix the parsing of the sourcehut refs file

Since a26be9f3b89be2ee90c6358250b9889b37f95cf8, the same parser is used
to parse the result of sourcehut’s `HEAD` endpoint (coming from [git
dumb protocol]) and the output of `git ls-remote`. However, they are very
slightly different (the former doesn’t specify the current reference
since it’s implied to be `HEAD`).

Unify both, and make the parser a bit more robust and understandable (by
making it more typed and adding tests for it)

[git dumb protocol]: https://git-scm.com/book/en/v2/Git-Internals-Transfer-Protocols#_the_dumb_protocol

3 files changed, 98 insertions, 0 deletions

diff --git a/src/libutil/git.cc b/src/libutil/git.cc
new file mode 100644
index 000000000..f35c2fdb7
--- /dev/null
+++ b/src/libutil/git.cc
@@ -0,0 +1,25 @@
+#include "git.hh"
+
+#include <regex>
+
+namespace nix {
+namespace git {
+
+std::optional<LsRemoteRefLine> parseLsRemoteLine(std::string_view line)
+{
+    const static std::regex line_regex("^(ref: *)?([^\\s]+)(?:\\t+(.*))?$");
+    std::match_results<std::string_view::const_iterator> match;
+    if (!std::regex_match(line.cbegin(), line.cend(), match, line_regex))
+        return std::nullopt;
+
+    return LsRemoteRefLine {
+        .kind = match[1].length() == 0
+            ? LsRemoteRefLine::Kind::Object
+            : LsRemoteRefLine::Kind::Symbolic,
+        .target = match[2],
+        .reference = match[3].length() == 0 ? std::nullopt : std::optional<std::string>{ match[3] }
+    };
+}
+
+}
+}
diff --git a/src/libutil/git.hh b/src/libutil/git.hh
new file mode 100644
index 000000000..cb13ef0e5
--- /dev/null
+++ b/src/libutil/git.hh
@@ -0,0 +1,40 @@
+#pragma once
+
+#include <string>
+#include <string_view>
+#include <optional>
+
+namespace nix {
+
+namespace git {
+
+// A line from the output of `git ls-remote --symref`.
+//
+// These can be of two kinds:
+//
+// - Symbolic references of the form
+//
+//      ref: {target}	{reference}
+//
+//    where {target} is itself a reference and {reference} is optional
+//
+// - Object references of the form
+//
+//      {target}	{reference}
+//
+//    where {target} is a commit id and {reference} is mandatory
+struct LsRemoteRefLine {
+    enum struct Kind {
+        Symbolic,
+        Object
+    };
+    Kind kind;
+    std::string target;
+    std::optional<std::string> reference;
+};
+
+std::optional<LsRemoteRefLine> parseLsRemoteLine(std::string_view line);
+
+}
+
+}
diff --git a/src/libutil/tests/git.cc b/src/libutil/tests/git.cc
new file mode 100644
index 000000000..5b5715fc2
--- /dev/null
+++ b/src/libutil/tests/git.cc
@@ -0,0 +1,33 @@
+#include "git.hh"
+#include <gtest/gtest.h>
+
+namespace nix {
+
+    TEST(GitLsRemote, parseSymrefLineWithReference) {
+        auto line = "ref: refs/head/main	HEAD";
+        auto res = git::parseLsRemoteLine(line);
+        ASSERT_TRUE(res.has_value());
+        ASSERT_EQ(res->kind, git::LsRemoteRefLine::Kind::Symbolic);
+        ASSERT_EQ(res->target, "refs/head/main");
+        ASSERT_EQ(res->reference, "HEAD");
+    }
+
+    TEST(GitLsRemote, parseSymrefLineWithNoReference) {
+        auto line = "ref: refs/head/main";
+        auto res = git::parseLsRemoteLine(line);
+        ASSERT_TRUE(res.has_value());
+        ASSERT_EQ(res->kind, git::LsRemoteRefLine::Kind::Symbolic);
+        ASSERT_EQ(res->target, "refs/head/main");
+        ASSERT_EQ(res->reference, std::nullopt);
+    }
+
+    TEST(GitLsRemote, parseObjectRefLine) {
+        auto line = "abc123	refs/head/main";
+        auto res = git::parseLsRemoteLine(line);
+        ASSERT_TRUE(res.has_value());
+        ASSERT_EQ(res->kind, git::LsRemoteRefLine::Kind::Object);
+        ASSERT_EQ(res->target, "abc123");
+        ASSERT_EQ(res->reference, "refs/head/main");
+    }
+}
+