Merge remote-tracking branch 'upstream/master' into upstream-merge

13 files changed, 335 insertions, 113 deletions

diff --git a/src/libfetchers/cache.cc b/src/libfetchers/cache.cc
index 34ff6f85b..0c8ecac9d 100644
--- a/src/libfetchers/cache.cc
+++ b/src/libfetchers/cache.cc
@@ -52,13 +52,13 @@ struct CacheImpl : Cache
         const Attrs & inAttrs,
         const Attrs & infoAttrs,
         const StorePath & storePath,
-        bool immutable) override
+        bool locked) override
     {
         _state.lock()->add.use()
             (attrsToJSON(inAttrs).dump())
             (attrsToJSON(infoAttrs).dump())
             (store->printStorePath(storePath))
-            (immutable)
+            (locked)
             (time(0)).exec();
     }
 
@@ -91,7 +91,7 @@ struct CacheImpl : Cache
 
         auto infoJSON = stmt.getStr(0);
         auto storePath = store->parseStorePath(stmt.getStr(1));
-        auto immutable = stmt.getInt(2) != 0;
+        auto locked = stmt.getInt(2) != 0;
         auto timestamp = stmt.getInt(3);
 
         store->addTempRoot(storePath);
@@ -105,7 +105,7 @@ struct CacheImpl : Cache
             inAttrsJSON, infoJSON, store->printStorePath(storePath));
 
         return Result {
-            .expired = !immutable && (settings.tarballTtl.get() == 0 || timestamp + settings.tarballTtl < time(0)),
+            .expired = !locked && (settings.tarballTtl.get() == 0 || timestamp + settings.tarballTtl < time(0)),
             .infoAttrs = jsonToAttrs(nlohmann::json::parse(infoJSON)),
             .storePath = std::move(storePath)
         };
diff --git a/src/libfetchers/cache.hh b/src/libfetchers/cache.hh
index 3db4f081c..3763ee2a6 100644
--- a/src/libfetchers/cache.hh
+++ b/src/libfetchers/cache.hh
@@ -13,7 +13,7 @@ struct Cache
         const Attrs & inAttrs,
         const Attrs & infoAttrs,
         const StorePath & storePath,
-        bool immutable) = 0;
+        bool locked) = 0;
 
     virtual std::optional<std::pair<Attrs, StorePath>> lookup(
         ref<Store> store,
diff --git a/src/libfetchers/fetch-settings.cc b/src/libfetchers/fetch-settings.cc
new file mode 100644
index 000000000..e7d5244dc
--- /dev/null
+++ b/src/libfetchers/fetch-settings.cc
@@ -0,0 +1,13 @@
+#include "fetch-settings.hh"
+
+namespace nix {
+
+FetchSettings::FetchSettings()
+{
+}
+
+FetchSettings fetchSettings;
+
+static GlobalConfig::Register rFetchSettings(&fetchSettings);
+
+}
diff --git a/src/libfetchers/fetch-settings.hh b/src/libfetchers/fetch-settings.hh
new file mode 100644
index 000000000..04c9feda0
--- /dev/null
+++ b/src/libfetchers/fetch-settings.hh
@@ -0,0 +1,93 @@
+#pragma once
+
+#include "types.hh"
+#include "config.hh"
+#include "util.hh"
+
+#include <map>
+#include <limits>
+
+#include <sys/types.h>
+
+namespace nix {
+
+struct FetchSettings : public Config
+{
+    FetchSettings();
+
+    Setting<StringMap> accessTokens{this, {}, "access-tokens",
+        R"(
+          Access tokens used to access protected GitHub, GitLab, or
+          other locations requiring token-based authentication.
+
+          Access tokens are specified as a string made up of
+          space-separated `host=token` values.  The specific token
+          used is selected by matching the `host` portion against the
+          "host" specification of the input. The actual use of the
+          `token` value is determined by the type of resource being
+          accessed:
+
+          * Github: the token value is the OAUTH-TOKEN string obtained
+            as the Personal Access Token from the Github server (see
+            https://docs.github.com/en/developers/apps/building-oauth-apps/authorizing-oauth-apps).
+
+          * Gitlab: the token value is either the OAuth2 token or the
+            Personal Access Token (these are different types tokens
+            for gitlab, see
+            https://docs.gitlab.com/12.10/ee/api/README.html#authentication).
+            The `token` value should be `type:tokenstring` where
+            `type` is either `OAuth2` or `PAT` to indicate which type
+            of token is being specified.
+
+          Example `~/.config/nix/nix.conf`:
+
+          ```
+          access-tokens = github.com=23ac...b289 gitlab.mycompany.com=PAT:A123Bp_Cd..EfG gitlab.com=OAuth2:1jklw3jk
+          ```
+
+          Example `~/code/flake.nix`:
+
+          ```nix
+          input.foo = {
+            type = "gitlab";
+            host = "gitlab.mycompany.com";
+            owner = "mycompany";
+            repo = "pro";
+          };
+          ```
+
+          This example specifies three tokens, one each for accessing
+          github.com, gitlab.mycompany.com, and sourceforge.net.
+
+          The `input.foo` uses the "gitlab" fetcher, which might
+          requires specifying the token type along with the token
+          value.
+          )"};
+
+    Setting<bool> allowDirty{this, true, "allow-dirty",
+        "Whether to allow dirty Git/Mercurial trees."};
+
+    Setting<bool> warnDirty{this, true, "warn-dirty",
+        "Whether to warn about dirty Git/Mercurial trees."};
+
+    Setting<std::string> flakeRegistry{this, "https://github.com/NixOS/flake-registry/raw/master/flake-registry.json", "flake-registry",
+        "Path or URI of the global flake registry."};
+
+    Setting<bool> useRegistries{this, true, "use-registries",
+        "Whether to use flake registries to resolve flake references."};
+
+    Setting<bool> acceptFlakeConfig{this, false, "accept-flake-config",
+        "Whether to accept nix configuration from a flake without prompting."};
+
+    Setting<std::string> commitLockFileSummary{
+        this, "", "commit-lockfile-summary",
+        R"(
+          The commit summary to use when committing changed flake lock files. If
+          empty, the summary is generated based on the action performed.
+        )"};
+};
+
+// FIXME: don't use a global variable.
+extern FetchSettings fetchSettings;
+
+}
diff --git a/src/libfetchers/fetchers.cc b/src/libfetchers/fetchers.cc
index e158d914b..976f40d3b 100644
--- a/src/libfetchers/fetchers.cc
+++ b/src/libfetchers/fetchers.cc
@@ -24,11 +24,11 @@ static void fixupInput(Input & input)
     input.getType();
     input.getRef();
     if (input.getRev())
-        input.immutable = true;
+        input.locked = true;
     input.getRevCount();
     input.getLastModified();
     if (input.getNarHash())
-        input.immutable = true;
+        input.locked = true;
 }
 
 Input Input::fromURL(const ParsedURL & url)
@@ -124,15 +124,13 @@ std::pair<Tree, Input> Input::fetch(ref<Store> store) const
             debug("using substituted/cached input '%s' in '%s'",
                 to_string(), store->printStorePath(storePath));
 
-            auto actualPath = store->toRealPath(storePath);
-
-            return {fetchers::Tree(std::move(actualPath), std::move(storePath)), *this};
+            return {Tree { .actualPath = store->toRealPath(storePath), .storePath = std::move(storePath) }, *this};
         } catch (Error & e) {
             debug("substitution of input '%s' failed: %s", to_string(), e.what());
         }
     }
 
-    auto [tree, input] = [&]() -> std::pair<Tree, Input> {
+    auto [storePath, input] = [&]() -> std::pair<StorePath, Input> {
         try {
             return scheme->fetch(store, *this);
         } catch (Error & e) {
@@ -141,8 +139,10 @@ std::pair<Tree, Input> Input::fetch(ref<Store> store) const
         }
     }();
 
-    if (tree.actualPath == "")
-        tree.actualPath = store->toRealPath(tree.storePath);
+    Tree tree {
+        .actualPath = store->toRealPath(storePath),
+        .storePath = storePath,
+    };
 
     auto narHash = store->queryPathInfo(tree.storePath)->narHash;
     input.attrs.insert_or_assign("narHash", narHash.to_string(SRI, true));
@@ -165,7 +165,7 @@ std::pair<Tree, Input> Input::fetch(ref<Store> store) const
                 input.to_string(), *prevRevCount);
     }
 
-    input.immutable = true;
+    input.locked = true;
 
     assert(input.hasAllInfo());
 
@@ -209,7 +209,7 @@ StorePath Input::computeStorePath(Store & store) const
 {
     auto narHash = getNarHash();
     if (!narHash)
-        throw Error("cannot compute store path for mutable input '%s'", to_string());
+        throw Error("cannot compute store path for unlocked input '%s'", to_string());
     return store.makeFixedOutputPath(FileIngestionMethod::Recursive, *narHash, getName());
 }
 
diff --git a/src/libfetchers/fetchers.hh b/src/libfetchers/fetchers.hh
index c43b047a7..bc9a76b0b 100644
--- a/src/libfetchers/fetchers.hh
+++ b/src/libfetchers/fetchers.hh
@@ -16,7 +16,6 @@ struct Tree
 {
     Path actualPath;
     StorePath storePath;
-    Tree(Path && actualPath, StorePath && storePath) : actualPath(actualPath), storePath(std::move(storePath)) {}
 };
 
 struct InputScheme;
@@ -35,7 +34,7 @@ struct Input
 
     std::shared_ptr<InputScheme> scheme; // note: can be null
     Attrs attrs;
-    bool immutable = false;
+    bool locked = false;
     bool direct = true;
 
     /* path of the parent of this input, used for relative path resolution */
@@ -60,9 +59,9 @@ public:
        one that goes through a registry. */
     bool isDirect() const { return direct; }
 
-    /* Check whether this is an "immutable" input, that is,
+    /* Check whether this is a "locked" input, that is,
        one that contains a commit hash or content hash. */
-    bool isImmutable() const { return immutable; }
+    bool isLocked() const { return locked; }
 
     bool hasAllInfo() const;
 
@@ -70,6 +69,8 @@ public:
 
     bool contains(const Input & other) const;
 
+    /* Fetch the input into the Nix store, returning the location in
+       the Nix store and the locked input. */
     std::pair<Tree, Input> fetch(ref<Store> store) const;
 
     Input applyOverrides(
@@ -131,7 +132,7 @@ struct InputScheme
 
     virtual void markChangedFile(const Input & input, std::string_view file, std::optional<std::string> commitMsg);
 
-    virtual std::pair<Tree, Input> fetch(ref<Store> store, const Input & input) = 0;
+    virtual std::pair<StorePath, Input> fetch(ref<Store> store, const Input & input) = 0;
 };
 
 void registerInputScheme(std::shared_ptr<InputScheme> && fetcher);
@@ -147,14 +148,14 @@ DownloadFileResult downloadFile(
     ref<Store> store,
     const std::string & url,
     const std::string & name,
-    bool immutable,
+    bool locked,
     const Headers & headers = {});
 
 std::pair<Tree, time_t> downloadTarball(
     ref<Store> store,
     const std::string & url,
     const std::string & name,
-    bool immutable,
+    bool locked,
     const Headers & headers = {});
 
 }
diff --git a/src/libfetchers/git.cc b/src/libfetchers/git.cc
index 544d2ffbf..f8433bc28 100644
--- a/src/libfetchers/git.cc
+++ b/src/libfetchers/git.cc
@@ -6,6 +6,8 @@
 #include "url-parts.hh"
 #include "pathlocks.hh"
 
+#include "fetch-settings.hh"
+
 #include <sys/time.h>
 #include <sys/wait.h>
 
@@ -172,7 +174,7 @@ struct GitInputScheme : InputScheme
         return {isLocal, isLocal ? url.path : url.base};
     }
 
-    std::pair<Tree, Input> fetch(ref<Store> store, const Input & _input) override
+    std::pair<StorePath, Input> fetch(ref<Store> store, const Input & _input) override
     {
         Input input(_input);
 
@@ -187,7 +189,7 @@ struct GitInputScheme : InputScheme
         if (submodules) cacheType += "-submodules";
         if (allRefs) cacheType += "-all-refs";
 
-        auto getImmutableAttrs = [&]()
+        auto getLockedAttrs = [&]()
         {
             return Attrs({
                 {"type", cacheType},
@@ -197,21 +199,18 @@ struct GitInputScheme : InputScheme
         };
 
         auto makeResult = [&](const Attrs & infoAttrs, StorePath && storePath)
-            -> std::pair<Tree, Input>
+            -> std::pair<StorePath, Input>
         {
             assert(input.getRev());
             assert(!_input.getRev() || _input.getRev() == input.getRev());
             if (!shallow)
                 input.attrs.insert_or_assign("revCount", getIntAttr(infoAttrs, "revCount"));
             input.attrs.insert_or_assign("lastModified", getIntAttr(infoAttrs, "lastModified"));
-            return {
-                Tree(store->toRealPath(storePath), std::move(storePath)),
-                input
-            };
+            return {std::move(storePath), input};
         };
 
         if (input.getRev()) {
-            if (auto res = getCache()->lookup(store, getImmutableAttrs()))
+            if (auto res = getCache()->lookup(store, getLockedAttrs()))
                 return makeResult(res->first, std::move(res->second));
         }
 
@@ -223,22 +222,46 @@ struct GitInputScheme : InputScheme
         if (!input.getRef() && !input.getRev() && isLocal) {
             bool clean = false;
 
-            /* Check whether this repo has any commits. There are
-               probably better ways to do this. */
-            auto gitDir = actualUrl + "/.git";
-            auto commonGitDir = chomp(runProgram(
-                "git",
-                true,
-                { "-C", actualUrl, "rev-parse", "--git-common-dir" }
-            ));
-            if (commonGitDir != ".git")
-                    gitDir = commonGitDir;
-
-            bool haveCommits = !readDirectory(gitDir + "/refs/heads").empty();
+            auto env = getEnv();
+            // Set LC_ALL to C: because we rely on the error messages from git rev-parse to determine what went wrong
+            // that way unknown errors can lead to a failure instead of continuing through the wrong code path
+            env["LC_ALL"] = "C";
+
+            /* Check whether HEAD points to something that looks like a commit,
+               since that is the refrence we want to use later on. */
+            auto result = runProgram(RunOptions {
+                .program = "git",
+                .args = { "-C", actualUrl, "--git-dir=.git", "rev-parse", "--verify", "--no-revs", "HEAD^{commit}" },
+                .environment = env,
+                .mergeStderrToStdout = true
+            });
+            auto exitCode = WEXITSTATUS(result.first);
+            auto errorMessage = result.second;
+
+            if (errorMessage.find("fatal: not a git repository") != std::string::npos) {
+                throw Error("'%s' is not a Git repository", actualUrl);
+            } else if (errorMessage.find("fatal: Needed a single revision") != std::string::npos) {
+                // indicates that the repo does not have any commits
+                // we want to proceed and will consider it dirty later
+            } else if (exitCode != 0) {
+                // any other errors should lead to a failure
+                throw Error("getting the HEAD of the Git tree '%s' failed with exit code %d:\n%s", actualUrl, exitCode, errorMessage);
+            }
 
+            bool hasHead = exitCode == 0;
             try {
-                if (haveCommits) {
-                    runProgram("git", true, { "-C", actualUrl, "diff-index", "--quiet", "HEAD", "--" });
+                if (hasHead) {
+                    // Using git diff is preferrable over lower-level operations here,
+                    // because its conceptually simpler and we only need the exit code anyways.
+                    auto gitDiffOpts = Strings({ "-C", actualUrl, "diff", "HEAD", "--quiet"});
+                    if (!submodules) {
+                        // Changes in submodules should only make the tree dirty
+                        // when those submodules will be copied as well.
+                        gitDiffOpts.emplace_back("--ignore-submodules");
+                    }
+                    gitDiffOpts.emplace_back("--");
+                    runProgram("git", true, gitDiffOpts);
+
                     clean = true;
                 }
             } catch (ExecError & e) {
@@ -249,10 +272,10 @@ struct GitInputScheme : InputScheme
 
                 /* This is an unclean working tree. So copy all tracked files. */
 
-                if (!settings.allowDirty)
+                if (!fetchSettings.allowDirty)
                     throw Error("Git tree '%s' is dirty", actualUrl);
 
-                if (settings.warnDirty)
+                if (fetchSettings.warnDirty)
                     warn("Git tree '%s' is dirty", actualUrl);
 
                 auto gitOpts = Strings({ "-C", actualUrl, "ls-files", "-z" });
@@ -262,9 +285,11 @@ struct GitInputScheme : InputScheme
                 auto files = tokenizeString<std::set<std::string>>(
                     runProgram("git", true, gitOpts), "\0"s);
 
+                Path actualPath(absPath(actualUrl));
+
                 PathFilter filter = [&](const Path & p) -> bool {
-                    assert(hasPrefix(p, actualUrl));
-                    std::string file(p, actualUrl.size() + 1);
+                    assert(hasPrefix(p, actualPath));
+                    std::string file(p, actualPath.size() + 1);
 
                     auto st = lstat(p);
 
@@ -277,24 +302,21 @@ struct GitInputScheme : InputScheme
                     return files.count(file);
                 };
 
-                auto storePath = store->addToStore(input.getName(), actualUrl, FileIngestionMethod::Recursive, htSHA256, filter);
+                auto storePath = store->addToStore(input.getName(), actualPath, FileIngestionMethod::Recursive, htSHA256, filter);
 
                 // FIXME: maybe we should use the timestamp of the last
                 // modified dirty file?
                 input.attrs.insert_or_assign(
                     "lastModified",
-                    haveCommits ? std::stoull(runProgram("git", true, { "-C", actualUrl, "log", "-1", "--format=%ct", "--no-show-signature", "HEAD" })) : 0);
+                    hasHead ? std::stoull(runProgram("git", true, { "-C", actualPath, "log", "-1", "--format=%ct", "--no-show-signature", "HEAD" })) : 0);
 
-                return {
-                    Tree(store->toRealPath(storePath), std::move(storePath)),
-                    input
-                };
+                return {std::move(storePath), input};
             }
         }
 
         if (!input.getRef()) input.attrs.insert_or_assign("ref", isLocal ? readHead(actualUrl) : "master");
 
-        Attrs mutableAttrs({
+        Attrs unlockedAttrs({
             {"type", cacheType},
             {"name", name},
             {"url", actualUrl},
@@ -313,7 +335,7 @@ struct GitInputScheme : InputScheme
 
         } else {
 
-            if (auto res = getCache()->lookup(store, mutableAttrs)) {
+            if (auto res = getCache()->lookup(store, unlockedAttrs)) {
                 auto rev2 = Hash::parseAny(getStrAttr(res->first, "rev"), htSHA1);
                 if (!input.getRev() || input.getRev() == rev2) {
                     input.attrs.insert_or_assign("rev", rev2.gitRev());
@@ -410,7 +432,7 @@ struct GitInputScheme : InputScheme
 
         /* Now that we know the ref, check again whether we have it in
            the store. */
-        if (auto res = getCache()->lookup(store, getImmutableAttrs()))
+        if (auto res = getCache()->lookup(store, getLockedAttrs()))
             return makeResult(res->first, std::move(res->second));
 
         Path tmpDir = createTempDir();
@@ -482,14 +504,14 @@ struct GitInputScheme : InputScheme
         if (!_input.getRev())
             getCache()->add(
                 store,
-                mutableAttrs,
+                unlockedAttrs,
                 infoAttrs,
                 storePath,
                 false);
 
         getCache()->add(
             store,
-            getImmutableAttrs(),
+            getLockedAttrs(),
             infoAttrs,
             storePath,
             true);
diff --git a/src/libfetchers/github.cc b/src/libfetchers/github.cc
index 1c539b80e..58b6e7c04 100644
--- a/src/libfetchers/github.cc
+++ b/src/libfetchers/github.cc
@@ -1,13 +1,16 @@
 #include "filetransfer.hh"
 #include "cache.hh"
-#include "fetchers.hh"
 #include "globals.hh"
 #include "store-api.hh"
 #include "types.hh"
 #include "url-parts.hh"
 
+#include "fetchers.hh"
+#include "fetch-settings.hh"
+
 #include <optional>
 #include <nlohmann/json.hpp>
+#include <fstream>
 
 namespace nix::fetchers {
 
@@ -17,7 +20,7 @@ struct DownloadUrl
     Headers headers;
 };
 
-// A github or gitlab host
+// A github, gitlab, or sourcehut host
 const static std::string hostRegexS = "[a-zA-Z0-9.]*"; // FIXME: check
 std::regex hostRegex(hostRegexS, std::regex::ECMAScript);
 
@@ -156,7 +159,7 @@ struct GitArchiveInputScheme : InputScheme
 
     std::optional<std::string> getAccessToken(const std::string & host) const
     {
-        auto tokens = settings.accessTokens.get();
+        auto tokens = fetchSettings.accessTokens.get();
         if (auto token = get(tokens, host))
             return *token;
         return {};
@@ -180,7 +183,7 @@ struct GitArchiveInputScheme : InputScheme
 
     virtual DownloadUrl getDownloadUrl(const Input & input) const = 0;
 
-    std::pair<Tree, Input> fetch(ref<Store> store, const Input & _input) override
+    std::pair<StorePath, Input> fetch(ref<Store> store, const Input & _input) override
     {
         Input input(_input);
 
@@ -192,17 +195,14 @@ struct GitArchiveInputScheme : InputScheme
         input.attrs.erase("ref");
         input.attrs.insert_or_assign("rev", rev->gitRev());
 
-        Attrs immutableAttrs({
+        Attrs lockedAttrs({
             {"type", "git-tarball"},
             {"rev", rev->gitRev()},
         });
 
-        if (auto res = getCache()->lookup(store, immutableAttrs)) {
+        if (auto res = getCache()->lookup(store, lockedAttrs)) {
             input.attrs.insert_or_assign("lastModified", getIntAttr(res->first, "lastModified"));
-            return {
-                Tree(store->toRealPath(res->second), std::move(res->second)),
-                input
-            };
+            return {std::move(res->second), input};
         }
 
         auto url = getDownloadUrl(input);
@@ -213,7 +213,7 @@ struct GitArchiveInputScheme : InputScheme
 
         getCache()->add(
             store,
-            immutableAttrs,
+            lockedAttrs,
             {
                 {"rev", rev->gitRev()},
                 {"lastModified", uint64_t(lastModified)}
@@ -221,7 +221,7 @@ struct GitArchiveInputScheme : InputScheme
             tree.storePath,
             true);
 
-        return {std::move(tree), input};
+        return {std::move(tree.storePath), input};
     }
 };
 
@@ -348,7 +348,97 @@ struct GitLabInputScheme : GitArchiveInputScheme
     }
 };
 
+struct SourceHutInputScheme : GitArchiveInputScheme
+{
+    std::string type() override { return "sourcehut"; }
+
+    std::optional<std::pair<std::string, std::string>> accessHeaderFromToken(const std::string & token) const override
+    {
+        // SourceHut supports both PAT and OAuth2. See
+        // https://man.sr.ht/meta.sr.ht/oauth.md
+        return std::pair<std::string, std::string>("Authorization", fmt("Bearer %s", token));
+        // Note: This currently serves no purpose, as this kind of authorization
+        // does not allow for downloading tarballs on sourcehut private repos.
+        // Once it is implemented, however, should work as expected.
+    }
+
+    Hash getRevFromRef(nix::ref<Store> store, const Input & input) const override
+    {
+        // TODO: In the future, when the sourcehut graphql API is implemented for mercurial
+        // and with anonymous access, this method should use it instead.
+
+        auto ref = *input.getRef();
+
+        auto host = maybeGetStrAttr(input.attrs, "host").value_or("git.sr.ht");
+        auto base_url = fmt("https://%s/%s/%s",
+            host, getStrAttr(input.attrs, "owner"), getStrAttr(input.attrs, "repo"));
+
+        Headers headers = makeHeadersWithAuthTokens(host);
+
+        std::string ref_uri;
+        if (ref == "HEAD") {
+            auto file = store->toRealPath(
+                downloadFile(store, fmt("%s/HEAD", base_url), "source", false, headers).storePath);
+            std::ifstream is(file);
+            std::string line;
+            getline(is, line);
+
+            auto ref_index = line.find("ref: ");
+            if (ref_index == std::string::npos) {
+                throw BadURL("in '%d', couldn't resolve HEAD ref '%d'", input.to_string(), ref);
+            }
+
+            ref_uri = line.substr(ref_index+5, line.length()-1);
+        } else
+            ref_uri = fmt("refs/(heads|tags)/%s", ref);
+
+        auto file = store->toRealPath(
+            downloadFile(store, fmt("%s/info/refs", base_url), "source", false, headers).storePath);
+        std::ifstream is(file);
+
+        std::string line;
+        std::string id;
+        while(getline(is, line)) {
+            // Append $ to avoid partial name matches
+            std::regex pattern(fmt("%s$", ref_uri));
+
+            if (std::regex_search(line, pattern)) {
+                id = line.substr(0, line.find('\t'));
+                break;
+            }
+        }
+
+        if(id.empty())
+            throw BadURL("in '%d', couldn't find ref '%d'", input.to_string(), ref);
+
+        auto rev = Hash::parseAny(id, htSHA1);
+        debug("HEAD revision for '%s' is %s", fmt("%s/%s", base_url, ref), rev.gitRev());
+        return rev;
+    }
+
+    DownloadUrl getDownloadUrl(const Input & input) const override
+    {
+        auto host = maybeGetStrAttr(input.attrs, "host").value_or("git.sr.ht");
+        auto url = fmt("https://%s/%s/%s/archive/%s.tar.gz",
+            host, getStrAttr(input.attrs, "owner"), getStrAttr(input.attrs, "repo"),
+            input.getRev()->to_string(Base16, false));
+
+        Headers headers = makeHeadersWithAuthTokens(host);
+        return DownloadUrl { url, headers };
+    }
+
+    void clone(const Input & input, const Path & destDir) override
+    {
+        auto host = maybeGetStrAttr(input.attrs, "host").value_or("git.sr.ht");
+        Input::fromURL(fmt("git+https://%s/%s/%s",
+                host, getStrAttr(input.attrs, "owner"), getStrAttr(input.attrs, "repo")))
+            .applyOverrides(input.getRef(), input.getRev())
+            .clone(destDir);
+    }
+};
+
 static auto rGitHubInputScheme = OnStartup([] { registerInputScheme(std::make_unique<GitHubInputScheme>()); });
 static auto rGitLabInputScheme = OnStartup([] { registerInputScheme(std::make_unique<GitLabInputScheme>()); });
+static auto rSourceHutInputScheme = OnStartup([] { registerInputScheme(std::make_unique<SourceHutInputScheme>()); });
 
 }
diff --git a/src/libfetchers/indirect.cc b/src/libfetchers/indirect.cc
index 10e59919a..9288fc6cf 100644
--- a/src/libfetchers/indirect.cc
+++ b/src/libfetchers/indirect.cc
@@ -94,7 +94,7 @@ struct IndirectInputScheme : InputScheme
         return input;
     }
 
-    std::pair<Tree, Input> fetch(ref<Store> store, const Input & input) override
+    std::pair<StorePath, Input> fetch(ref<Store> store, const Input & input) override
     {
         throw Error("indirect input '%s' cannot be fetched directly", input.to_string());
     }
diff --git a/src/libfetchers/mercurial.cc b/src/libfetchers/mercurial.cc
index d52d4641b..8b82e9daa 100644
--- a/src/libfetchers/mercurial.cc
+++ b/src/libfetchers/mercurial.cc
@@ -5,6 +5,8 @@
 #include "store-api.hh"
 #include "url-parts.hh"
 
+#include "fetch-settings.hh"
+
 #include <sys/time.h>
 
 using namespace std::string_literals;
@@ -26,7 +28,7 @@ static RunOptions hgOptions(const Strings & args)
 }
 
 // runProgram wrapper that uses hgOptions instead of stock RunOptions.
-static string runHg(const Strings & args, const std::optional<std::string> & input = {})
+static std::string runHg(const Strings & args, const std::optional<std::string> & input = {})
 {
     RunOptions opts = hgOptions(args);
     opts.input = input;
@@ -143,7 +145,7 @@ struct MercurialInputScheme : InputScheme
         return {isLocal, isLocal ? url.path : url.base};
     }
 
-    std::pair<Tree, Input> fetch(ref<Store> store, const Input & _input) override
+    std::pair<StorePath, Input> fetch(ref<Store> store, const Input & _input) override
     {
         Input input(_input);
 
@@ -165,10 +167,10 @@ struct MercurialInputScheme : InputScheme
                 /* This is an unclean working tree. So copy all tracked
                    files. */
 
-                if (!settings.allowDirty)
+                if (!fetchSettings.allowDirty)
                     throw Error("Mercurial tree '%s' is unclean", actualUrl);
 
-                if (settings.warnDirty)
+                if (fetchSettings.warnDirty)
                     warn("Mercurial tree '%s' is unclean", actualUrl);
 
                 input.attrs.insert_or_assign("ref", chomp(runHg({ "branch", "-R", actualUrl })));
@@ -193,16 +195,13 @@ struct MercurialInputScheme : InputScheme
 
                 auto storePath = store->addToStore(input.getName(), actualUrl, FileIngestionMethod::Recursive, htSHA256, filter);
 
-                return {
-                    Tree(store->toRealPath(storePath), std::move(storePath)),
-                    input
-                };
+                return {std::move(storePath), input};
             }
         }
 
         if (!input.getRef()) input.attrs.insert_or_assign("ref", "default");
 
-        auto getImmutableAttrs = [&]()
+        auto getLockedAttrs = [&]()
         {
             return Attrs({
                 {"type", "hg"},
@@ -212,32 +211,29 @@ struct MercurialInputScheme : InputScheme
         };
 
         auto makeResult = [&](const Attrs & infoAttrs, StorePath && storePath)
-            -> std::pair<Tree, Input>
+            -> std::pair<StorePath, Input>
         {
             assert(input.getRev());
             assert(!_input.getRev() || _input.getRev() == input.getRev());
             input.attrs.insert_or_assign("revCount", getIntAttr(infoAttrs, "revCount"));
-            return {
-                Tree(store->toRealPath(storePath), std::move(storePath)),
-                input
-            };
+            return {std::move(storePath), input};
         };
 
         if (input.getRev()) {
-            if (auto res = getCache()->lookup(store, getImmutableAttrs()))
+            if (auto res = getCache()->lookup(store, getLockedAttrs()))
                 return makeResult(res->first, std::move(res->second));
         }
 
         auto revOrRef = input.getRev() ? input.getRev()->gitRev() : *input.getRef();
 
-        Attrs mutableAttrs({
+        Attrs unlockedAttrs({
             {"type", "hg"},
             {"name", name},
             {"url", actualUrl},
             {"ref", *input.getRef()},
         });
 
-        if (auto res = getCache()->lookup(store, mutableAttrs)) {
+        if (auto res = getCache()->lookup(store, unlockedAttrs)) {
             auto rev2 = Hash::parseAny(getStrAttr(res->first, "rev"), htSHA1);
             if (!input.getRev() || input.getRev() == rev2) {
                 input.attrs.insert_or_assign("rev", rev2.gitRev());
@@ -260,7 +256,7 @@ struct MercurialInputScheme : InputScheme
                     runHg({ "pull", "-R", cacheDir, "--", actualUrl });
                 }
                 catch (ExecError & e) {
-                    string transJournal = cacheDir + "/.hg/store/journal";
+                    auto transJournal = cacheDir + "/.hg/store/journal";
                     /* hg throws "abandoned transaction" error only if this file exists */
                     if (pathExists(transJournal)) {
                         runHg({ "recover", "-R", cacheDir });
@@ -283,7 +279,7 @@ struct MercurialInputScheme : InputScheme
         auto revCount = std::stoull(tokens[1]);
         input.attrs.insert_or_assign("ref", tokens[2]);
 
-        if (auto res = getCache()->lookup(store, getImmutableAttrs()))
+        if (auto res = getCache()->lookup(store, getLockedAttrs()))
             return makeResult(res->first, std::move(res->second));
 
         Path tmpDir = createTempDir();
@@ -303,14 +299,14 @@ struct MercurialInputScheme : InputScheme
         if (!_input.getRev())
             getCache()->add(
                 store,
-                mutableAttrs,
+                unlockedAttrs,
                 infoAttrs,
                 storePath,
                 false);
 
         getCache()->add(
             store,
-            getImmutableAttrs(),
+            getLockedAttrs(),
             infoAttrs,
             storePath,
             true);
diff --git a/src/libfetchers/path.cc b/src/libfetchers/path.cc
index 07e543c53..f0ef97da5 100644
--- a/src/libfetchers/path.cc
+++ b/src/libfetchers/path.cc
@@ -1,5 +1,6 @@
 #include "fetchers.hh"
 #include "store-api.hh"
+#include "archive.hh"
 
 namespace nix::fetchers {
 
@@ -80,8 +81,9 @@ struct PathInputScheme : InputScheme
         // nothing to do
     }
 
-    std::pair<Tree, Input> fetch(ref<Store> store, const Input & input) override
+    std::pair<StorePath, Input> fetch(ref<Store> store, const Input & _input) override
     {
+        Input input(_input);
         std::string absPath;
         auto path = getStrAttr(input.attrs, "path");
 
@@ -111,14 +113,17 @@ struct PathInputScheme : InputScheme
         if (storePath)
             store->addTempRoot(*storePath);
 
-        if (!storePath || storePath->name() != "source" || !store->isValidPath(*storePath))
+        time_t mtime = 0;
+        if (!storePath || storePath->name() != "source" || !store->isValidPath(*storePath)) {
             // FIXME: try to substitute storePath.
-            storePath = store->addToStore("source", absPath);
-
-        return {
-            Tree(store->toRealPath(*storePath), std::move(*storePath)),
-            input
-        };
+            auto src = sinkToSource([&](Sink & sink) {
+                mtime = dumpPathAndGetMtime(absPath, sink, defaultPathFilter);
+            });
+            storePath = store->addToStoreFromDump(*src, "source");
+        }
+        input.attrs.insert_or_assign("lastModified", uint64_t(mtime));
+
+        return {std::move(*storePath), input};
     }
 };
 
diff --git a/src/libfetchers/registry.cc b/src/libfetchers/registry.cc
index f35359d4b..acd1ff866 100644
--- a/src/libfetchers/registry.cc
+++ b/src/libfetchers/registry.cc
@@ -5,6 +5,8 @@
 #include "store-api.hh"
 #include "local-fs-store.hh"
 
+#include "fetch-settings.hh"
+
 #include <nlohmann/json.hpp>
 
 namespace nix::fetchers {
@@ -150,7 +152,7 @@ void overrideRegistry(
 static std::shared_ptr<Registry> getGlobalRegistry(ref<Store> store)
 {
     static auto reg = [&]() {
-        auto path = settings.flakeRegistry.get();
+        auto path = fetchSettings.flakeRegistry.get();
 
         if (!hasPrefix(path, "/")) {
             auto storePath = downloadFile(store, path, "flake-registry.json", false).storePath;
diff --git a/src/libfetchers/tarball.cc b/src/libfetchers/tarball.cc
index c933475ca..dde0ad761 100644
--- a/src/libfetchers/tarball.cc
+++ b/src/libfetchers/tarball.cc
@@ -13,7 +13,7 @@ DownloadFileResult downloadFile(
     ref<Store> store,
     const std::string & url,
     const std::string & name,
-    bool immutable,
+    bool locked,
     const Headers & headers)
 {
     // FIXME: check store
@@ -88,7 +88,7 @@ DownloadFileResult downloadFile(
         inAttrs,
         infoAttrs,
         *storePath,
-        immutable);
+        locked);
 
     if (url != res.effectiveUri)
         getCache()->add(
@@ -100,7 +100,7 @@ DownloadFileResult downloadFile(
             },
             infoAttrs,
             *storePath,
-            immutable);
+            locked);
 
     return {
         .storePath = std::move(*storePath),
@@ -113,7 +113,7 @@ std::pair<Tree, time_t> downloadTarball(
     ref<Store> store,
     const std::string & url,
     const std::string & name,
-    bool immutable,
+    bool locked,
     const Headers & headers)
 {
     Attrs inAttrs({
@@ -126,11 +126,11 @@ std::pair<Tree, time_t> downloadTarball(
 
     if (cached && !cached->expired)
         return {
-            Tree(store->toRealPath(cached->storePath), std::move(cached->storePath)),
+            Tree { .actualPath = store->toRealPath(cached->storePath), .storePath = std::move(cached->storePath) },
             getIntAttr(cached->infoAttrs, "lastModified")
         };
 
-    auto res = downloadFile(store, url, name, immutable, headers);
+    auto res = downloadFile(store, url, name, locked, headers);
 
     std::optional<StorePath> unpackedStorePath;
     time_t lastModified;
@@ -160,10 +160,10 @@ std::pair<Tree, time_t> downloadTarball(
         inAttrs,
         infoAttrs,
         *unpackedStorePath,
-        immutable);
+        locked);
 
     return {
-        Tree(store->toRealPath(*unpackedStorePath), std::move(*unpackedStorePath)),
+        Tree { .actualPath = store->toRealPath(*unpackedStorePath), .storePath = std::move(*unpackedStorePath) },
         lastModified,
     };
 }
@@ -202,7 +202,7 @@ struct TarballInputScheme : InputScheme
 
         Input input;
         input.attrs = attrs;
-        //input.immutable = (bool) maybeGetStrAttr(input.attrs, "hash");
+        //input.locked = (bool) maybeGetStrAttr(input.attrs, "hash");
         return input;
     }
 
@@ -225,10 +225,10 @@ struct TarballInputScheme : InputScheme
         return true;
     }
 
-    std::pair<Tree, Input> fetch(ref<Store> store, const Input & input) override
+    std::pair<StorePath, Input> fetch(ref<Store> store, const Input & input) override
     {
         auto tree = downloadTarball(store, getStrAttr(input.attrs, "url"), input.getName(), false).first;
-        return {std::move(tree), input};
+        return {std::move(tree.storePath), input};
     }
 };