Add fetchMercurial primop

E.g.

  $ nix eval '(fetchMercurial https://www.mercurial-scm.org/repo/hello)'
  { branch = "default"; outPath = "/nix/store/alvb9y1kfz42bjishqmyy3pphnrh1pfa-source"; rev = "82e55d328c8ca4ee16520036c0aaace03a5beb65"; revCount = 1; shortRev = "82e55d328c8c"; }

  $ nix eval '(fetchMercurial { url = https://www.mercurial-scm.org/repo/hello; rev = "0a04b987be5ae354b710cefeba0e2d9de7ad41a9"; })'
  { branch = "default"; outPath = "/nix/store/alvb9y1kfz42bjishqmyy3pphnrh1pfa-source"; rev = "0a04b987be5ae354b710cefeba0e2d9de7ad41a9"; revCount = 0; shortRev = "0a04b987be5a"; }

  $ nix eval '(fetchMercurial /tmp/unclean-hg-tree)'
  { branch = "default"; outPath = "/nix/store/cm750cdw1x8wfpm3jq7mz09r30l9r024-source"; rev = "0000000000000000000000000000000000000000"; revCount = 0; shortRev = "000000000000"; }

release.nix

@@ -76,7 +76,7 @@ let
           [ curl
             bzip2 xz brotli
             openssl pkgconfig sqlite boehmgc
-
+            mercurial
           ]
           ++ lib.optional stdenv.isLinux libseccomp
           ++ lib.optional (stdenv.isLinux || stdenv.isDarwin) libsodium

shell.nix

@@ -23,6 +23,9 @@ with import ./release-common.nix { inherit pkgs; };
       # For nix-perl
       perl
       perlPackages.DBDSQLite
+
+      # Tests
+      mercurial
     ]
     ++ lib.optional stdenv.isLinux libseccomp;
 

src/libexpr/primops/fetchMercurial.cc

@@ -0,0 +1,188 @@
+#include "primops.hh"
+#include "eval-inline.hh"
+#include "download.hh"
+#include "store-api.hh"
+#include "pathlocks.hh"
+
+#include <sys/time.h>
+
+#include <regex>
+
+#include <nlohmann/json.hpp>
+
+using namespace std::string_literals;
+
+namespace nix {
+
+struct HgInfo
+{
+    Path storePath;
+    std::string branch;
+    std::string rev;
+    uint64_t revCount = 0;
+};
+
+HgInfo exportMercurial(ref<Store> store, const std::string & uri,
+    std::string rev, const std::string & name)
+{
+    if (rev == "" && hasPrefix(uri, "/") && pathExists(uri + "/.hg")) {
+
+        bool clean = runProgram("hg", true, { "status", "-R", uri, "--modified", "--added", "--removed" }) == "";
+
+        if (!clean) {
+
+            /* This is an unclean working tree. So copy all tracked
+               files. */
+
+            printTalkative("copying unclean Mercurial working tree '%s'", uri);
+
+            HgInfo hgInfo;
+            hgInfo.rev = "0000000000000000000000000000000000000000";
+            hgInfo.branch = chomp(runProgram("hg", true, { "branch", "-R", uri }));
+
+            auto files = tokenizeString<std::set<std::string>>(
+                runProgram("hg", true, { "status", "-R", uri, "--clean", "--modified", "--added", "--no-status", "--print0" }), "\0"s);
+
+            PathFilter filter = [&](const Path & p) -> bool {
+                assert(hasPrefix(p, uri));
+                auto st = lstat(p);
+                std::string file(p, uri.size() + 1);
+                if (file == ".hg") return false;
+                // FIXME: filter out directories with no tracked files.
+                if (S_ISDIR(st.st_mode)) return true;
+                return files.count(file);
+            };
+
+            hgInfo.storePath = store->addToStore("source", uri, true, htSHA256, filter);
+
+            return hgInfo;
+        }
+    }
+
+    if (rev == "") rev = "default";
+
+    Path cacheDir = fmt("%s/nix/hg/%s", getCacheDir(), hashString(htSHA256, uri).to_string(Base32, false));
+
+    Path stampFile = fmt("%s/.hg/%s.stamp", cacheDir, hashString(htSHA512, rev).to_string(Base32, false));
+
+    /* If we haven't pulled this repo less than ‘tarball-ttl’ seconds,
+       do so now. FIXME: don't do this if "rev" is a hash and we
+       fetched it previously */
+    time_t now = time(0);
+    struct stat st;
+    if (stat(stampFile.c_str(), &st) != 0 ||
+        st.st_mtime < now - settings.tarballTtl)
+    {
+        Activity act(*logger, lvlTalkative, actUnknown, fmt("fetching Mercurial repository '%s'", uri));
+
+        if (pathExists(cacheDir)) {
+            runProgram("hg", true, { "pull", "-R", cacheDir, "--", uri });
+        } else {
+            createDirs(dirOf(cacheDir));
+            runProgram("hg", true, { "clone", "--noupdate", "--", uri, cacheDir });
+        }
+
+        writeFile(stampFile, "");
+    }
+
+    auto tokens = tokenizeString<std::vector<std::string>>(
+        runProgram("hg", true, { "log", "-R", cacheDir, "-r", rev, "--template", "{node} {rev} {branch}" }));
+    assert(tokens.size() == 3);
+
+    HgInfo hgInfo;
+    hgInfo.rev = tokens[0];
+    hgInfo.revCount = std::stoull(tokens[1]);
+    hgInfo.branch = tokens[2];
+
+    std::string storeLinkName = hashString(htSHA512, name + std::string("\0"s) + hgInfo.rev).to_string(Base32, false);
+    Path storeLink = fmt("%s/.hg/%s.link", cacheDir, storeLinkName);
+
+    try {
+        auto json = nlohmann::json::parse(readFile(storeLink));
+
+        assert(json["name"] == name && json["rev"] == hgInfo.rev);
+
+        hgInfo.storePath = json["storePath"];
+
+        if (store->isValidPath(hgInfo.storePath)) {
+            printTalkative("using cached Mercurial store path '%s'", hgInfo.storePath);
+            return hgInfo;
+        }
+
+    } catch (SysError & e) {
+        if (e.errNo != ENOENT) throw;
+    }
+
+    Path tmpDir = createTempDir();
+    AutoDelete delTmpDir(tmpDir, true);
+
+    runProgram("hg", true, { "archive", "-R", cacheDir, "-r", rev, tmpDir });
+
+    deletePath(tmpDir + "/.hg_archival.txt");
+
+    hgInfo.storePath = store->addToStore(name, tmpDir);
+
+    nlohmann::json json;
+    json["storePath"] = hgInfo.storePath;
+    json["uri"] = uri;
+    json["name"] = name;
+    json["branch"] = hgInfo.branch;
+    json["rev"] = hgInfo.rev;
+    json["revCount"] = hgInfo.revCount;
+
+    writeFile(storeLink, json.dump());
+
+    return hgInfo;
+}
+
+static void prim_fetchMercurial(EvalState & state, const Pos & pos, Value * * args, Value & v)
+{
+    std::string url;
+    std::string rev;
+    std::string name = "source";
+    PathSet context;
+
+    state.forceValue(*args[0]);
+
+    if (args[0]->type == tAttrs) {
+
+        state.forceAttrs(*args[0], pos);
+
+        for (auto & attr : *args[0]->attrs) {
+            string n(attr.name);
+            if (n == "url")
+                url = state.coerceToString(*attr.pos, *attr.value, context, false, false);
+            else if (n == "rev")
+                rev = state.forceStringNoCtx(*attr.value, *attr.pos);
+            else if (n == "name")
+                name = state.forceStringNoCtx(*attr.value, *attr.pos);
+            else
+                throw EvalError("unsupported argument '%s' to 'fetchGit', at %s", attr.name, *attr.pos);
+        }
+
+        if (url.empty())
+            throw EvalError(format("'url' argument required, at %1%") % pos);
+
+    } else
+        url = state.coerceToString(pos, *args[0], context, false, false);
+
+    if (!isUri(url)) url = absPath(url);
+
+    // FIXME: git externals probably can be used to bypass the URI
+    // whitelist. Ah well.
+    state.checkURI(url);
+
+    auto hgInfo = exportMercurial(state.store, url, rev, name);
+
+    state.mkAttrs(v, 8);
+    mkString(*state.allocAttr(v, state.sOutPath), hgInfo.storePath, PathSet({hgInfo.storePath}));
+    mkString(*state.allocAttr(v, state.symbols.create("branch")), hgInfo.branch);
+    mkString(*state.allocAttr(v, state.symbols.create("rev")), hgInfo.rev);
+    mkString(*state.allocAttr(v, state.symbols.create("shortRev")), std::string(hgInfo.rev, 0, 12));
+    mkInt(*state.allocAttr(v, state.symbols.create("revCount")), hgInfo.revCount);
+    v.attrs->sort();
+}
+
+static RegisterPrimOp r("fetchMercurial", 1, prim_fetchMercurial);
+
+}

src/libexpr/primops/fetchgit.cc

@@ -106,10 +106,9 @@ GitInfo exportGit(ref<Store> store, const std::string & uri,
 
     std::string storeLinkName = hashString(htSHA512, name + std::string("\0"s) + gitInfo.rev).to_string(Base32, false);
     Path storeLink = cacheDir + "/" + storeLinkName + ".link";
-    PathLocks storeLinkLock({storeLink}, fmt("waiting for lock on '%1%'...", storeLink));
+    PathLocks storeLinkLock({storeLink}, fmt("waiting for lock on '%1%'...", storeLink)); // FIXME: broken
 
     try {
-        // FIXME: doesn't handle empty lines
         auto json = nlohmann::json::parse(readFile(storeLink));
 
         assert(json["name"] == name && json["rev"] == gitInfo.rev);

src/libstore/download.cc

@@ -707,7 +707,7 @@ bool isUri(const string & s)
     size_t pos = s.find("://");
     if (pos == string::npos) return false;
     string scheme(s, 0, pos);
-    return scheme == "http" || scheme == "https" || scheme == "file" || scheme == "channel" || scheme == "git" || scheme == "s3";
+    return scheme == "http" || scheme == "https" || scheme == "file" || scheme == "channel" || scheme == "git" || scheme == "s3" || scheme == "ssh";
 }
 
 

tests/fetchMercurial.sh

@@ -0,0 +1,73 @@
+source common.sh
+
+if [[ -z $(type -p hg) ]]; then
+    echo "Mercurial not installed; skipping Mercurial tests"
+    exit 0
+fi
+
+clearStore
+
+repo=$TEST_ROOT/hg
+
+rm -rfv $repo ${repo}-tmp $TEST_HOME/.cache/nix/hg
+
+hg init $repo
+echo '[ui]' >> $repo/.hg/hgrc
+echo 'username = Foobar <foobar@example.org>' >> $repo/.hg/hgrc
+
+echo utrecht > $repo/hello
+hg add --cwd $repo hello
+hg commit --cwd $repo -m 'Bla1'
+rev1=$(hg log --cwd $repo -r tip --template '{node}')
+
+echo world > $repo/hello
+hg commit --cwd $repo -m 'Bla2'
+rev2=$(hg log --cwd $repo -r tip --template '{node}')
+
+hg log --cwd $repo
+
+hg log --cwd $repo -r tip --template '{node}\n'
+
+path=$(nix eval --raw "(builtins.fetchMercurial file://$repo).outPath")
+[[ $(cat $path/hello) = world ]]
+
+# Fetch again. This should be cached.
+mv $repo ${repo}-tmp
+path2=$(nix eval --raw "(builtins.fetchMercurial file://$repo).outPath")
+[[ $path = $path2 ]]
+
+[[ $(nix eval --raw "(builtins.fetchMercurial file://$repo).branch") = default ]]
+[[ $(nix eval "(builtins.fetchMercurial file://$repo).revCount") = 1 ]]
+[[ $(nix eval --raw "(builtins.fetchMercurial file://$repo).rev") = $rev2 ]]
+
+# But with TTL 0, it should fail.
+(! nix eval --tarball-ttl 0 --raw "(builtins.fetchMercurial file://$repo)")
+
+mv ${repo}-tmp $repo
+
+# Using a clean working tree should produce the same result.
+path2=$(nix eval --raw "(builtins.fetchMercurial $repo).outPath")
+[[ $path = $path2 ]]
+
+# Using an unclean tree should yield the tracked but uncommitted changes.
+echo foo > $repo/foo
+echo bar > $repo/bar
+hg add --cwd $repo foo
+hg rm --cwd $repo hello
+
+path2=$(nix eval --raw "(builtins.fetchMercurial $repo).outPath")
+[ ! -e $path2/hello ]
+[ ! -e $path2/bar ]
+[[ $(cat $path2/foo) = foo ]]
+
+[[ $(nix eval --raw "(builtins.fetchMercurial $repo).rev") = 0000000000000000000000000000000000000000 ]]
+
+# ... unless we're using an explicit rev.
+path3=$(nix eval --raw "(builtins.fetchMercurial { url = $repo; rev = \"default\"; }).outPath")
+[[ $path = $path3 ]]
+
+# Committing should not affect the store path.
+hg commit --cwd $repo -m 'Bla3'
+
+path4=$(nix eval --tarball-ttl 0 --raw "(builtins.fetchMercurial file://$repo).outPath")
+[[ $path2 = $path4 ]]

tests/local.mk

@@ -15,7 +15,8 @@ nix_tests = \
   linux-sandbox.sh \
   build-remote.sh \
   nar-index.sh \
-  structured-attrs.sh
+  structured-attrs.sh \
+  fetchMercurial.sh
   # parallel.sh
 
 install-tests += $(foreach x, $(nix_tests), tests/$(x))