libfetchers: make attribute / URL query handling consistent

The original idea was to fix lix#174, but for a user friendly solution,
I figured that we'd need more consistency:

* Invalid query params will cause an error, just like invalid
  attributes. This has the following two consequences:

  * The `?dir=`-param from flakes will be removed before the URL to be
    fetched is passed to libfetchers.

  * The tarball fetcher doesn't allow URLs with custom query params
    anymore. I think this was questionable anyways given that an
    arbitrary set of query params was silently removed from the URL you
    wanted to fetch. The correct way is to use an attribute-set
    with a key `url` that contains the tarball URL to fetch.

  * Same for the git & mercurial fetchers: in that case it doesn't even
    matter though: both fetchers added unused query params to the URL
    that's passed from the input scheme to the fetcher (`url2` in the code).
    It turns out that this was never used since the query parameters were
    erased again in `getActualUrl`.

* Validation happens for both attributes and URLs. Previously, a lot of
  fetchers validated e.g. refs/revs only when specified in a URL and
  the validity of attribute names only in `inputFromAttrs`.

  Now, all the validation is done in `inputFromAttrs` and `inputFromURL`
  constructs attributes that will be passed to `inputFromAttrs`.

* Accept all attributes as URL query parameters. That also includes
  lesser used ones such as `narHash`.

  And "output" attributes like `lastModified`: these could be declared
  already when declaring inputs as attribute rather than URL. Now the
  behavior is at least consistent.

  Personally, I think we should differentiate in the future between
  "fetched input" (basically the attr-set that ends up in the lock-file)
  and "unfetched input" earlier: both inputFrom{Attrs,URL} entrypoints
  are probably OK for unfetched inputs, but for locked/fetched inputs
  a custom entrypoint should be used. Then, the current entrypoints
  wouldn't have to allow these attributes anymore.

Change-Id: I1be1992249f7af8287cfc37891ab505ddaa2e8cd

6 files changed, 135 insertions, 94 deletions

diff --git a/src/libfetchers/fetchers.hh b/src/libfetchers/fetchers.hh
index 2bb4248be..40f2b6294 100644
--- a/src/libfetchers/fetchers.hh
+++ b/src/libfetchers/fetchers.hh
@@ -159,6 +159,37 @@ struct InputScheme
         std::optional<std::string> commitMsg) const;
 
     virtual std::pair<StorePath, Input> fetch(ref<Store> store, const Input & input) = 0;
+
+protected:
+    void emplaceURLQueryIntoAttrs(
+        const ParsedURL & parsedURL,
+        Attrs & attrs,
+        const StringSet & numericParams,
+        const StringSet & booleanParams) const
+    {
+        for (auto &[name, value] : parsedURL.query) {
+            if (name == "url") {
+                throw BadURL(
+                    "URL '%s' must not override url via query param!",
+                    parsedURL.to_string()
+                );
+            } else if (numericParams.count(name) != 0) {
+                if (auto n = string2Int<uint64_t>(value)) {
+                    attrs.insert_or_assign(name, *n);
+                } else {
+                    throw BadURL(
+                        "URL '%s' has non-numeric parameter '%s'",
+                        parsedURL.to_string(),
+                        name
+                    );
+                }
+            } else if (booleanParams.count(name) != 0) {
+                attrs.emplace(name, Explicit<bool> { value == "1" });
+            } else {
+                attrs.emplace(name, value);
+            }
+        }
+    }
 };
 
 void registerInputScheme(std::shared_ptr<InputScheme> && fetcher);
diff --git a/src/libfetchers/git.cc b/src/libfetchers/git.cc
index 2817fde23..0cb826075 100644
--- a/src/libfetchers/git.cc
+++ b/src/libfetchers/git.cc
@@ -273,18 +273,15 @@ struct GitInputScheme : InputScheme
 
         Attrs attrs;
         attrs.emplace("type", "git");
-
-        for (auto & [name, value] : url.query) {
-            if (name == "rev" || name == "ref")
-                attrs.emplace(name, value);
-            else if (name == "shallow" || name == "submodules" || name == "allRefs")
-                attrs.emplace(name, Explicit<bool> { value == "1" });
-            else
-                url2.query.emplace(name, value);
-        }
-
         attrs.emplace("url", url2.to_string());
 
+        emplaceURLQueryIntoAttrs(
+            url,
+            attrs,
+            {"lastModified", "revCount"},
+            {"shallow", "submodules", "allRefs"}
+        );
+
         return inputFromAttrs(attrs);
     }
 
diff --git a/src/libfetchers/github.cc b/src/libfetchers/github.cc
index 60fefd1f3..b971781ae 100644
--- a/src/libfetchers/github.cc
+++ b/src/libfetchers/github.cc
@@ -1,3 +1,4 @@
+#include "attrs.hh"
 #include "filetransfer.hh"
 #include "cache.hh"
 #include "globals.hh"
@@ -36,18 +37,11 @@ struct GitArchiveInputScheme : InputScheme
 
         auto path = tokenizeString<std::vector<std::string>>(url.path, "/");
 
-        std::optional<Hash> rev;
-        std::optional<std::string> ref;
-        std::optional<std::string> host_url;
+        std::optional<std::string> refOrRev;
 
         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]);
+            refOrRev = path[2];
         } else if (size > 3) {
             std::string rs;
             for (auto i = std::next(path.begin(), 2); i != path.end(); i++) {
@@ -58,61 +52,91 @@ struct GitArchiveInputScheme : InputScheme
             }
 
             if (std::regex_match(rs, refRegex)) {
-                ref = rs;
+                refOrRev = 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);
 
+        Attrs attrs;
+        attrs.emplace("type", type());
+        attrs.emplace("owner", path[0]);
+        attrs.emplace("repo", path[1]);
+
         for (auto &[name, value] : url.query) {
-            if (name == "rev") {
-                if (rev)
-                    throw BadURL("URL '%s' contains multiple commit hashes", url.url);
-                rev = Hash::parseAny(value, htSHA1);
-            }
-            else if (name == "ref") {
-                if (!std::regex_match(value, refRegex))
-                    throw BadURL("URL '%s' contains an invalid branch/tag name", url.url);
-                if (ref)
-                    throw BadURL("URL '%s' contains multiple branch/tag names", url.url);
-                ref = value;
-            }
-            else if (name == "host") {
-                if (!std::regex_match(value, hostRegex))
-                    throw BadURL("URL '%s' contains an invalid instance host", url.url);
-                host_url = value;
+            if (name == "rev" || name == "ref") {
+                if (refOrRev) {
+                    throw BadURL("URL '%s' already contains a ref or rev", url.url);
+                } else {
+                    refOrRev = value;
+                }
+            } else if (name == "lastModified") {
+                if (auto n = string2Int<uint64_t>(value)) {
+                    attrs.emplace(name, *n);
+                } else {
+                    throw Error(
+                        "Attribute 'lastModified' in URL '%s' must be an integer",
+                        url.to_string()
+                    );
+                }
+            } else {
+                attrs.emplace(name, value);
             }
-            // FIXME: barf on unsupported attributes
         }
 
-        if (ref && rev)
-            throw BadURL("URL '%s' contains both a commit hash and a branch/tag name %s %s", url.url, *ref, rev->gitRev());
+        if (refOrRev) attrs.emplace("refOrRev", *refOrRev);
 
-        Input input;
-        input.attrs.insert_or_assign("type", type());
-        input.attrs.insert_or_assign("owner", path[0]);
-        input.attrs.insert_or_assign("repo", path[1]);
-        if (rev) input.attrs.insert_or_assign("rev", rev->gitRev());
-        if (ref) input.attrs.insert_or_assign("ref", *ref);
-        if (host_url) input.attrs.insert_or_assign("host", *host_url);
-
-        return input;
+        return inputFromAttrs(attrs);
     }
 
     std::optional<Input> inputFromAttrs(const Attrs & attrs) const override
     {
-        if (maybeGetStrAttr(attrs, "type") != type()) return {};
+        // Attributes can contain refOrRev and it needs to be figured out
+        // which one it is (see inputFromURL for when that may happen).
+        // The correct one (ref or rev) will be written into finalAttrs and
+        // it needs to be mutable for that.
+        Attrs finalAttrs(attrs);
+        auto type_ = maybeGetStrAttr(finalAttrs, "type");
+        if (type_ != type()) return {};
+
+        auto owner = getStrAttr(finalAttrs, "owner");
+        auto repo = getStrAttr(finalAttrs, "repo");
+
+        auto url = fmt("%s:%s/%s", *type_, owner, repo);
+        if (auto host = maybeGetStrAttr(finalAttrs, "host")) {
+            if (!std::regex_match(*host, hostRegex)) {
+                throw BadURL("URL '%s' contains an invalid instance host", url);
+            }
+        }
 
-        for (auto & [name, value] : attrs)
-            if (name != "type" && name != "owner" && name != "repo" && name != "ref" && name != "rev" && name != "narHash" && name != "lastModified" && name != "host")
-                throw Error("unsupported input attribute '%s'", name);
+        if (auto refOrRev = maybeGetStrAttr(finalAttrs, "refOrRev")) {
+            finalAttrs.erase("refOrRev");
+            if (std::regex_match(*refOrRev, revRegex)) {
+                finalAttrs.emplace("rev", *refOrRev);
+            } else if (std::regex_match(*refOrRev, refRegex)) {
+                finalAttrs.emplace("ref", *refOrRev);
+            } else {
+                throw Error(
+                    "in URL '%s', '%s' is not a commit hash or a branch/tag name",
+                    url,
+                    *refOrRev
+                );
+            }
+        } else if (auto ref = maybeGetStrAttr(finalAttrs, "ref")) {
+            if (!std::regex_match(*ref, refRegex)) {
+                throw BadURL("URL '%s' contains an invalid branch/tag name", url);
+            }
+        }
 
-        getStrAttr(attrs, "owner");
-        getStrAttr(attrs, "repo");
+        for (auto & [name, value] : finalAttrs) {
+            if (name != "type" && name != "owner" && name != "repo" && name != "ref" && name != "rev" && name != "narHash" && name != "lastModified" && name != "host") {
+                throw Error("unsupported input attribute '%s'", name);
+            }
+        }
 
         Input input;
-        input.attrs = attrs;
+        input.attrs = finalAttrs;
         return input;
     }
 
diff --git a/src/libfetchers/indirect.cc b/src/libfetchers/indirect.cc
index c73505b31..8c0176e84 100644
--- a/src/libfetchers/indirect.cc
+++ b/src/libfetchers/indirect.cc
@@ -17,6 +17,8 @@ struct IndirectInputScheme : InputScheme
         std::optional<Hash> rev;
         std::optional<std::string> ref;
 
+        Attrs attrs;
+
         if (path.size() == 1) {
         } else if (path.size() == 2) {
             if (std::regex_match(path[1], revRegex))
@@ -26,29 +28,21 @@ struct IndirectInputScheme : InputScheme
             else
                 throw BadURL("in flake URL '%s', '%s' is not a commit hash or branch/tag name", url.url, path[1]);
         } else if (path.size() == 3) {
-            if (!std::regex_match(path[1], refRegex))
-                throw BadURL("in flake URL '%s', '%s' is not a branch/tag name", url.url, path[1]);
             ref = path[1];
-            if (!std::regex_match(path[2], revRegex))
-                throw BadURL("in flake URL '%s', '%s' is not a commit hash", url.url, path[2]);
             rev = Hash::parseAny(path[2], htSHA1);
         } else
             throw BadURL("GitHub URL '%s' is invalid", url.url);
 
         std::string id = path[0];
-        if (!std::regex_match(id, flakeRegex))
-            throw BadURL("'%s' is not a valid flake ID", id);
 
-        // FIXME: forbid query params?
+        attrs.emplace("type", "indirect");
+        attrs.emplace("id", id);
+        if (rev) attrs.emplace("rev", rev->gitRev());
+        if (ref) attrs.emplace("ref", *ref);
 
-        Input input;
-        input.direct = false;
-        input.attrs.insert_or_assign("type", "indirect");
-        input.attrs.insert_or_assign("id", id);
-        if (rev) input.attrs.insert_or_assign("rev", rev->gitRev());
-        if (ref) input.attrs.insert_or_assign("ref", *ref);
+        emplaceURLQueryIntoAttrs(url, attrs, {}, {});
 
-        return input;
+        return inputFromAttrs(attrs);
     }
 
     std::optional<Input> inputFromAttrs(const Attrs & attrs) const override
@@ -63,6 +57,18 @@ struct IndirectInputScheme : InputScheme
         if (!std::regex_match(id, flakeRegex))
             throw BadURL("'%s' is not a valid flake ID", id);
 
+        // TODO come up with a nicer error message for those two.
+        if (auto rev = maybeGetStrAttr(attrs, "rev")) {
+            if (!std::regex_match(*rev, revRegex)) {
+                throw BadURL("in flake '%s', '%s' is not a commit hash", id, *rev);
+            }
+        }
+        if (auto ref = maybeGetStrAttr(attrs, "ref")) {
+            if (!std::regex_match(*ref, refRegex)) {
+                throw BadURL("in flake '%s', '%s' is not a valid branch/tag name", id, *ref);
+            }
+        }
+
         Input input;
         input.direct = false;
         input.attrs = attrs;
diff --git a/src/libfetchers/mercurial.cc b/src/libfetchers/mercurial.cc
index 4fffa71d3..6015f8ec7 100644
--- a/src/libfetchers/mercurial.cc
+++ b/src/libfetchers/mercurial.cc
@@ -57,12 +57,7 @@ struct MercurialInputScheme : InputScheme
         Attrs attrs;
         attrs.emplace("type", "hg");
 
-        for (auto &[name, value] : url.query) {
-            if (name == "rev" || name == "ref")
-                attrs.emplace(name, value);
-            else
-                url2.query.emplace(name, value);
-        }
+        emplaceURLQueryIntoAttrs(url, attrs, {"revCount"}, {});
 
         attrs.emplace("url", url2.to_string());
 
diff --git a/src/libfetchers/tarball.cc b/src/libfetchers/tarball.cc
index c903895e2..8dfdecda8 100644
--- a/src/libfetchers/tarball.cc
+++ b/src/libfetchers/tarball.cc
@@ -201,29 +201,17 @@ struct CurlInputScheme : InputScheme
         if (!isValidURL(_url, requireTree))
             return std::nullopt;
 
-        Input input;
-
         auto url = _url;
 
-        url.scheme = parseUrlScheme(url.scheme).transport;
-
-        auto narHash = url.query.find("narHash");
-        if (narHash != url.query.end())
-            input.attrs.insert_or_assign("narHash", narHash->second);
+        Attrs attrs;
+        attrs.emplace("type", inputType());
 
-        if (auto i = get(url.query, "rev"))
-            input.attrs.insert_or_assign("rev", *i);
-
-        if (auto i = get(url.query, "revCount"))
-            if (auto n = string2Int<uint64_t>(*i))
-                input.attrs.insert_or_assign("revCount", *n);
+        url.scheme = parseUrlScheme(url.scheme).transport;
 
-        url.query.erase("rev");
-        url.query.erase("revCount");
+        emplaceURLQueryIntoAttrs(url, attrs, {"revCount"}, {});
 
-        input.attrs.insert_or_assign("type", inputType());
-        input.attrs.insert_or_assign("url", url.to_string());
-        return input;
+        attrs.emplace("url", url.to_string());
+        return inputFromAttrs(attrs);
     }
 
     std::optional<Input> inputFromAttrs(const Attrs & attrs) const override
@@ -235,7 +223,7 @@ struct CurlInputScheme : InputScheme
         std::set<std::string> allowedNames = {"type", "url", "narHash", "name", "unpack", "rev", "revCount", "lastModified"};
         for (auto & [name, value] : attrs)
             if (!allowedNames.count(name))
-                throw Error("unsupported %s input attribute '%s'", *type, name);
+                throw Error("unsupported %s input attribute '%s'. If you wanted to fetch a tarball with a query parameter, please use '{ type = \"tarball\"; url = \"...\"; }'", *type, name);
 
         Input input;
         input.attrs = attrs;