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

4 files changed, 18 insertions, 63 deletions

diff --git a/src/libfetchers/git-utils.cc b/src/libfetchers/git-utils.cc
deleted file mode 100644
index b2d6b7893..000000000
--- a/src/libfetchers/git-utils.cc
+++ /dev/null
@@ -1,27 +0,0 @@
-#include "git-utils.hh"
-
-#include <regex>
-
-std::optional<std::string> parseListReferenceHeadRef(std::string_view line)
-{
-    const static std::regex head_ref_regex("^ref: ([^\\s]+)\\t+HEAD$");
-    std::match_results<std::string_view::const_iterator> match;
-    if (std::regex_match(line.cbegin(), line.cend(), match, head_ref_regex)) {
-        return match[1];
-    } else {
-        return std::nullopt;
-    }
-}
-
-std::optional<std::string> parseListReferenceForRev(std::string_view rev, std::string_view line)
-{
-    const static std::regex rev_regex("^([^\\t]+)\\t+(.*)$");
-    std::match_results<std::string_view::const_iterator> match;
-    if (!std::regex_match(line.cbegin(), line.cend(), match, rev_regex)) {
-        return std::nullopt;
-    }
-    if (rev != match[2].str()) {
-        return std::nullopt;
-    }
-    return match[1];
-}
diff --git a/src/libfetchers/git-utils.hh b/src/libfetchers/git-utils.hh
deleted file mode 100644
index 946a68a9e..000000000
--- a/src/libfetchers/git-utils.hh
+++ /dev/null
@@ -1,23 +0,0 @@
-#pragma once
-
-#include <string>
-#include <string_view>
-#include <optional>
-
-// Parses the HEAD ref as reported by `git ls-remote --symref`
-//
-// Returns the head branch name as reported by `git ls-remote --symref`, e.g., if
-// ls-remote returns the output below, "main" is returned based on the ref line.
-//
-//   ref: refs/heads/main       HEAD
-//
-// If the repository is in 'detached head' state (HEAD is pointing to a rev
-// instead of a branch), parseListReferenceForRev("HEAD") may be used instead.
-std::optional<std::string> parseListReferenceHeadRef(std::string_view line);
-
-// Parses a reference line from `git ls-remote --symref`, e.g.,
-// parseListReferenceForRev("refs/heads/master", line) will return 6926...
-// given the line below.
-//
-// 6926beab444c33fb57b21819b6642d032016bb1e	refs/heads/master
-std::optional<std::string> parseListReferenceForRev(std::string_view rev, std::string_view line);
diff --git a/src/libfetchers/git.cc b/src/libfetchers/git.cc
index 266246fe9..d23a820a4 100644
--- a/src/libfetchers/git.cc
+++ b/src/libfetchers/git.cc
@@ -6,7 +6,7 @@
 #include "url-parts.hh"
 #include "pathlocks.hh"
 #include "util.hh"
-#include "git-utils.hh"
+#include "git.hh"
 
 #include "fetch-settings.hh"
 
@@ -72,13 +72,16 @@ std::optional<std::string> readHead(const Path & path)
 
     std::string_view line = output;
     line = line.substr(0, line.find("\n"));
-    if (const auto ref = parseListReferenceHeadRef(line); ref) {
-        debug("resolved HEAD ref '%s' for repo '%s'", *ref, path);
-        return *ref;
-    }
-    if (const auto rev = parseListReferenceForRev("HEAD", line); rev) {
-        debug("resolved HEAD rev '%s' for repo '%s'", *rev, path);
-        return *rev;
+    if (const auto parseResult = git::parseLsRemoteLine(line)) {
+        switch (parseResult->kind) {
+            case git::LsRemoteRefLine::Kind::Symbolic:
+                debug("resolved HEAD ref '%s' for repo '%s'", parseResult->target, path);
+                break;
+            case git::LsRemoteRefLine::Kind::Object:
+                debug("resolved HEAD rev '%s' for repo '%s'", parseResult->target, path);
+                break;
+        }
+        return parseResult->target;
     }
     return std::nullopt;
 }
diff --git a/src/libfetchers/github.cc b/src/libfetchers/github.cc
index 1bdf2759f..a1084c984 100644
--- a/src/libfetchers/github.cc
+++ b/src/libfetchers/github.cc
@@ -4,7 +4,7 @@
 #include "store-api.hh"
 #include "types.hh"
 #include "url-parts.hh"
-#include "git-utils.hh"
+#include "git.hh"
 #include "fetchers.hh"
 #include "fetch-settings.hh"
 
@@ -383,11 +383,11 @@ struct SourceHutInputScheme : GitArchiveInputScheme
             std::string line;
             getline(is, line);
 
-            auto r = parseListReferenceHeadRef(line);
-            if (!r) {
+            auto remoteLine = git::parseLsRemoteLine(line);
+            if (!remoteLine) {
                 throw BadURL("in '%d', couldn't resolve HEAD ref '%d'", input.to_string(), ref);
             }
-            ref_uri = *r;
+            ref_uri = remoteLine->target;
         } else {
             ref_uri = fmt("refs/(heads|tags)/%s", ref);
         }
@@ -399,7 +399,9 @@ struct SourceHutInputScheme : GitArchiveInputScheme
         std::string line;
         std::optional<std::string> id;
         while(!id && getline(is, line)) {
-            id = parseListReferenceForRev(ref_uri, line);
+            auto parsedLine = git::parseLsRemoteLine(line);
+            if (parsedLine && parsedLine->reference == ref_uri)
+                id = parsedLine->target;
         }
 
         if(!id)