Merge pull request #1686 from cruessler/move-lookup-entry-to-gix-object

Add lookup_entry and lookup_entry_by_path to TreeRef

Author: Byron

Cargo.lock

@@ -50,6 +50,7 @@ gix-hashtable = { version = "^0.6.0", path = "../gix-hashtable" }
 gix-validate = { version = "^0.9.2", path = "../gix-validate" }
 gix-actor = { version = "^0.33.1", path = "../gix-actor" }
 gix-date = { version = "^0.9.2", path = "../gix-date" }
+gix-path = { version = "^0.10.12", path = "../gix-path" }
 gix-utils = { version = "^0.1.13", path = "../gix-utils" }
 
 itoa = "1.0.1"

gix-object/Cargo.toml

@@ -8,6 +8,77 @@ impl<'a> TreeRefIter<'a> {
     pub fn from_bytes(data: &'a [u8]) -> TreeRefIter<'a> {
         TreeRefIter { data }
     }
+
+    /// Follow a sequence of `path` components starting from this instance, and look them up in `odb` one by one using `buffer`
+    /// until the last component is looked up and its tree entry is returned.
+    ///
+    /// # Performance Notes
+    ///
+    /// Searching tree entries is currently done in sequence, which allows the search to be allocation free. It would be possible
+    /// to reuse a vector and use a binary search instead, which might be able to improve performance over all.
+    /// However, a benchmark should be created first to have some data and see which trade-off to choose here.
+    pub fn lookup_entry<I, P>(
+        &self,
+        odb: impl crate::Find,
+        buffer: &'a mut Vec<u8>,
+        path: I,
+    ) -> Result<Option<tree::Entry>, crate::find::Error>
+    where
+        I: IntoIterator<Item = P>,
+        P: PartialEq<BStr>,
+    {
+        buffer.clear();
+
+        let mut path = path.into_iter().peekable();
+        buffer.extend_from_slice(self.data);
+        while let Some(component) = path.next() {
+            match TreeRefIter::from_bytes(buffer)
+                .filter_map(Result::ok)
+                .find(|entry| component.eq(entry.filename))
+            {
+                Some(entry) => {
+                    if path.peek().is_none() {
+                        return Ok(Some(entry.into()));
+                    } else {
+                        let next_id = entry.oid.to_owned();
+                        let obj = odb.try_find(&next_id, buffer)?;
+                        let Some(obj) = obj else { return Ok(None) };
+                        if !obj.kind.is_tree() {
+                            return Ok(None);
+                        }
+                    }
+                }
+                None => return Ok(None),
+            }
+        }
+        Ok(None)
+    }
+
+    /// Like [`Self::lookup_entry()`], but takes any [`AsRef<Path>`](`std::path::Path`) directly via `relative_path`,
+    /// a path relative to this tree.
+    /// `odb` and `buffer` are used to lookup intermediate trees.
+    ///
+    /// # Note
+    ///
+    /// If any path component contains illformed UTF-8 and thus can't be converted to bytes on platforms which can't do so natively,
+    /// the returned component will be empty which makes the lookup fail.
+    pub fn lookup_entry_by_path(
+        &self,
+        odb: impl crate::Find,
+        buffer: &'a mut Vec<u8>,
+        relative_path: impl AsRef<std::path::Path>,
+    ) -> Result<Option<tree::Entry>, crate::find::Error> {
+        use crate::bstr::ByteSlice;
+        self.lookup_entry(
+            odb,
+            buffer,
+            relative_path.as_ref().components().map(|c: std::path::Component<'_>| {
+                gix_path::os_str_into_bstr(c.as_os_str())
+                    .unwrap_or_else(|_| "".into())
+                    .as_bytes()
+            }),
+        )
+    }
 }
 
 impl<'a> TreeRef<'a> {

gix-object/src/tree/ref_iter.rs

@@ -1,4 +1,9 @@
-use gix_object::{bstr::ByteSlice, tree, tree::EntryRef, TreeRefIter};
+use gix_object::{
+    bstr::ByteSlice,
+    tree::{self, EntryRef},
+    TreeRefIter,
+};
+use pretty_assertions::assert_eq;
 
 use crate::{fixture_name, hex_to_id};
 
@@ -52,3 +57,74 @@ fn everything() -> crate::Result {
     );
     Ok(())
 }
+
+mod lookup_entry {
+    use crate::hex_to_id;
+    use gix_object::tree::EntryKind;
+    use utils::entry;
+
+    #[test]
+    fn top_level_directory() -> crate::Result {
+        assert_eq!(
+            utils::lookup_entry_by_path("bin")?,
+            entry(
+                "bin",
+                EntryKind::Blob,
+                hex_to_id("e69de29bb2d1d6434b8b29ae775ad8c2e48c5391")
+            )
+        );
+        Ok(())
+    }
+
+    #[test]
+    fn nested_file() -> crate::Result {
+        assert_eq!(
+            utils::lookup_entry_by_path("file/a")?,
+            entry(
+                "a",
+                EntryKind::Blob,
+                hex_to_id("e69de29bb2d1d6434b8b29ae775ad8c2e48c5391")
+            )
+        );
+        Ok(())
+    }
+
+    #[test]
+    fn non_existing_nested_file() -> crate::Result {
+        for path in ["file/does-not-exist", "non-existing", "file/a/through-file"] {
+            let actual = utils::lookup_entry_by_path(path)?;
+            assert_eq!(actual, None);
+        }
+        Ok(())
+    }
+
+    mod utils {
+        use crate::hex_to_id;
+
+        use gix_object::{tree, FindExt};
+
+        pub(super) fn entry(filename: &str, mode: tree::EntryKind, oid: gix_hash::ObjectId) -> Option<tree::Entry> {
+            Some(tree::Entry {
+                mode: mode.into(),
+                filename: filename.into(),
+                oid,
+            })
+        }
+
+        pub(super) fn tree_odb() -> gix_testtools::Result<gix_odb::Handle> {
+            let root = gix_testtools::scripted_fixture_read_only("make_trees.sh")?;
+            Ok(gix_odb::at(root.join(".git/objects"))?)
+        }
+
+        pub(super) fn lookup_entry_by_path(path: &str) -> gix_testtools::Result<Option<gix_object::tree::Entry>> {
+            let odb = tree_odb()?;
+            let root_tree_id = hex_to_id("ff7e7d2aecae1c3fb15054b289a4c58aa65b8646");
+
+            let mut buf = Vec::new();
+            let root_tree = odb.find_tree_iter(&root_tree_id, &mut buf)?;
+
+            let mut buf = Vec::new();
+            root_tree.lookup_entry_by_path(&odb, &mut buf, path)
+        }
+    }
+}