ecr/src/extract.rs

use crate::veprintln;
use anyhow::{Context, Result};
use std::fs::File;
use std::io::{BufReader, Read};
use std::path::Path;

/// Extract a tarball to the specified directory
pub fn extract_tarball(tarball: &Path, dest: &Path) -> Result<()> {
    let filename = tarball.file_name().and_then(|n| n.to_str()).unwrap_or("");

    // OCI bundle files are always named "oci-…" by generate_cache_filename.
    // Use the filename prefix as a zero-I/O dispatch signal instead of
    // scanning the archive for a layers.manifest entry (which required
    // reading the entire compressed archive twice for large OCI images).
    if filename.starts_with("oci-") {
        veprintln!("Detected multi-layer OCI image, extracting layers...");
        return extract_multi_layer_oci(tarball, dest);
    }

    let file = File::open(tarball)
        .with_context(|| format!("Failed to open tarball: {}", tarball.display()))?;

    let reader = BufReader::new(file);

    // Detect compression format from filename
    if filename.ends_with(".tar.gz") || filename.ends_with(".tgz") {
        extract_gz(reader, dest)?;
    } else if filename.ends_with(".tar.xz") || filename.ends_with(".txz") {
        extract_xz(reader, dest)?;
    } else if filename.ends_with(".tar.zst") || filename.ends_with(".tar.zstd") {
        extract_zst(reader, dest)?;
    } else if filename.ends_with(".tar") {
        extract_tar(reader, dest)?;
    } else {
        // Try to detect from magic bytes
        let mut magic = [0u8; 6];
        let mut peek_reader = BufReader::new(File::open(tarball)?);
        peek_reader.read_exact(&mut magic)?;

        match magic {
            [0x1f, 0x8b, ..] => {
                // gzip magic
                drop(peek_reader);
                let file = File::open(tarball)?;
                extract_gz(BufReader::new(file), dest)?;
            }
            [0xfd, b'7', b'z', b'X', b'Z', 0x00] => {
                // xz magic
                drop(peek_reader);
                let file = File::open(tarball)?;
                extract_xz(BufReader::new(file), dest)?;
            }
            [0x28, 0xb5, 0x2f, 0xfd, ..] => {
                // zstd magic
                drop(peek_reader);
                let file = File::open(tarball)?;
                extract_zst(BufReader::new(file), dest)?;
            }
            _ => {
                // Assume uncompressed tar
                drop(peek_reader);
                let file = File::open(tarball)?;
                extract_tar(BufReader::new(file), dest)?;
            }
        }
    }

    Ok(())
}

/// Extract a multi-layer OCI image
fn extract_multi_layer_oci(tarball: &Path, dest: &Path) -> Result<()> {
    let filename = tarball.file_name().and_then(|n| n.to_str()).unwrap_or("");

    // Extract the outer OCI bundle (layers.manifest + layer_N.tar.gz files)
    // into a temp directory.  This is our own format, not an OCI layer, so
    // plain unpack is fine here.
    let temp_dir = tempfile::tempdir().context("Failed to create temp directory for OCI layers")?;

    let file = File::open(tarball)
        .with_context(|| format!("Failed to open OCI tarball: {}", tarball.display()))?;
    let reader = BufReader::new(file);

    if filename.ends_with(".tar.gz") || filename.ends_with(".tgz") {
        tar::Archive::new(flate2::read::GzDecoder::new(reader))
            .unpack(temp_dir.path())
            .context("Failed to unpack OCI bundle")?;
    } else if filename.ends_with(".tar.xz") || filename.ends_with(".txz") {
        tar::Archive::new(xz2::read::XzDecoder::new(reader))
            .unpack(temp_dir.path())
            .context("Failed to unpack OCI bundle")?;
    } else if filename.ends_with(".tar.zst") || filename.ends_with(".tar.zstd") {
        tar::Archive::new(zstd::stream::read::Decoder::new(reader)?)
            .unpack(temp_dir.path())
            .context("Failed to unpack OCI bundle")?;
    } else {
        tar::Archive::new(reader)
            .unpack(temp_dir.path())
            .context("Failed to unpack OCI bundle")?;
    }

    // Read the layers manifest
    let manifest = std::fs::read_to_string(temp_dir.path().join("layers.manifest"))
        .context("Failed to read layers.manifest")?;

    // Apply each layer in order with full whiteout handling.
    for layer_name in manifest.lines() {
        let layer_path = temp_dir.path().join(layer_name);
        if !layer_path.exists() {
            continue;
        }
        veprintln!("Extracting layer: {}", layer_name);
        extract_oci_layer(&layer_path, dest)?;
    }

    Ok(())
}

/// Decompress and extract one OCI layer tarball into `dest`, honouring
/// whiteout markers.  Compression is inferred from the filename then from
/// magic bytes.
fn extract_oci_layer(layer_path: &Path, dest: &Path) -> Result<()> {
    use std::io::Read;
    let layer_name = layer_path
        .file_name()
        .and_then(|n| n.to_str())
        .unwrap_or("");
    let file = File::open(layer_path)
        .with_context(|| format!("Failed to open layer: {}", layer_path.display()))?;
    let reader = BufReader::new(file);

    if layer_name.ends_with(".tar.gz") || layer_name.ends_with(".tgz") {
        extract_archive_with_whiteouts(
            tar::Archive::new(flate2::read::GzDecoder::new(reader)),
            dest,
        )
    } else if layer_name.ends_with(".tar.xz") || layer_name.ends_with(".txz") {
        extract_archive_with_whiteouts(tar::Archive::new(xz2::read::XzDecoder::new(reader)), dest)
    } else if layer_name.ends_with(".tar.zst") || layer_name.ends_with(".tar.zstd") {
        extract_archive_with_whiteouts(
            tar::Archive::new(zstd::stream::read::Decoder::new(reader)?),
            dest,
        )
    } else {
        // Fall back to magic-byte detection
        let mut magic = [0u8; 6];
        let mut peek = BufReader::new(File::open(layer_path)?);
        let _ = peek.read_exact(&mut magic); // short reads are fine for detection
        drop(peek);
        match magic {
            [0x1f, 0x8b, ..] => extract_archive_with_whiteouts(
                tar::Archive::new(flate2::read::GzDecoder::new(BufReader::new(File::open(
                    layer_path,
                )?))),
                dest,
            ),
            [0xfd, b'7', b'z', b'X', b'Z', 0x00] => extract_archive_with_whiteouts(
                tar::Archive::new(xz2::read::XzDecoder::new(BufReader::new(File::open(
                    layer_path,
                )?))),
                dest,
            ),
            [0x28, 0xb5, 0x2f, 0xfd, ..] => extract_archive_with_whiteouts(
                tar::Archive::new(zstd::stream::read::Decoder::new(BufReader::new(
                    File::open(layer_path)?,
                ))?),
                dest,
            ),
            _ => extract_archive_with_whiteouts(
                tar::Archive::new(BufReader::new(File::open(layer_path)?)),
                dest,
            ),
        }
    }
}

/// Apply one OCI layer archive to `dest`, interpreting Docker whiteout markers:
///
/// - `.wh.<name>` — Delete `<name>` from a lower layer that was already
///   extracted into `dest`.
/// - `.wh..wh..opq` — Opaque whiteout: the directory that contains this entry
///   is new in this layer; delete everything already in that
///   directory from lower layers before applying new content.
///
/// All other entries are extracted normally via `Entry::unpack_in`.
fn extract_archive_with_whiteouts<R: std::io::Read>(
    mut archive: tar::Archive<R>,
    dest: &Path,
) -> Result<()> {
    archive.set_preserve_permissions(true);
    archive.set_preserve_ownerships(false);
    archive.set_unpack_xattrs(false);

    for entry in archive.entries().context("Failed to iterate tar entries")? {
        let mut entry = entry.context("Failed to read tar entry")?;

        // Clone the path before any mutable borrow of entry (needed for unpack_in)
        let path = entry.path().context("Invalid tar entry path")?.into_owned();

        let filename = path
            .file_name()
            .map(|n| n.to_string_lossy().into_owned())
            .unwrap_or_default();

        if filename == ".wh..wh..opq" {
            // Opaque whiteout: clear all previously-extracted content in the
            // parent directory so only this layer's content is visible.
            let parent = path.parent().unwrap_or(Path::new(""));
            let dest_dir = dest.join(parent);
            if dest_dir.symlink_metadata().is_ok() {
                for child in std::fs::read_dir(&dest_dir)
                    .with_context(|| format!("Failed to read {}", dest_dir.display()))?
                {
                    let child = child?;
                    let child_path = child.path();
                    remove_path(&child_path).with_context(|| {
                        format!("Opaque whiteout: failed to remove {}", child_path.display())
                    })?;
                }
            }
            // Do not extract the .wh..wh..opq marker itself.
        } else if let Some(real_name) = filename.strip_prefix(".wh.") {
            // Regular whiteout: delete the named path from lower layers.
            let parent = path.parent().unwrap_or(Path::new(""));
            let target = dest.join(parent).join(real_name);
            // symlink_metadata (lstat) does not follow symlinks, so a dangling
            // symlink is correctly detected and removed rather than silently skipped.
            if target.symlink_metadata().is_ok() {
                remove_path(&target)
                    .with_context(|| format!("Whiteout: failed to remove {}", target.display()))?;
            }
            // Do not extract the .wh.* marker itself.
        } else {
            entry
                .unpack_in(dest)
                .with_context(|| format!("Failed to extract {}", path.display()))?;
        }
    }
    Ok(())
}

/// Remove a path: uses remove_dir_all for real directories, remove_file for
/// everything else (regular files, symlinks — including symlinks-to-dirs).
fn remove_path(path: &Path) -> std::io::Result<()> {
    // symlink_metadata does not follow symlinks, so a symlink-to-dir correctly
    // reports file_type().is_symlink() rather than is_dir().
    let meta = std::fs::symlink_metadata(path)?;
    if meta.is_dir() {
        std::fs::remove_dir_all(path)
    } else {
        std::fs::remove_file(path)
    }
}

fn extract_gz<R: std::io::Read>(reader: R, dest: &Path) -> Result<()> {
    let gz_decoder = flate2::read::GzDecoder::new(reader);
    let mut archive = tar::Archive::new(gz_decoder);

    archive.set_preserve_permissions(true);
    archive.set_preserve_ownerships(false);
    archive.set_unpack_xattrs(false);

    archive
        .unpack(dest)
        .context("Failed to extract gzip archive")?;

    Ok(())
}

fn extract_xz<R: std::io::Read>(reader: R, dest: &Path) -> Result<()> {
    let xz_decoder = xz2::read::XzDecoder::new(reader);
    let mut archive = tar::Archive::new(xz_decoder);

    archive.set_preserve_permissions(true);
    archive.set_preserve_ownerships(false);
    archive.set_unpack_xattrs(false);

    archive
        .unpack(dest)
        .context("Failed to extract xz archive")?;

    Ok(())
}

fn extract_zst<R: std::io::Read>(reader: R, dest: &Path) -> Result<()> {
    let zst_decoder = zstd::Decoder::new(reader)?;
    let mut archive = tar::Archive::new(zst_decoder);

    archive.set_preserve_permissions(true);
    archive.set_preserve_ownerships(false);
    archive.set_unpack_xattrs(false);

    archive
        .unpack(dest)
        .context("Failed to extract zstd archive")?;

    Ok(())
}

fn extract_tar<R: std::io::Read>(reader: R, dest: &Path) -> Result<()> {
    let mut archive = tar::Archive::new(reader);

    archive.set_preserve_permissions(true);
    archive.set_preserve_ownerships(false);
    archive.set_unpack_xattrs(false);

    archive
        .unpack(dest)
        .context("Failed to extract tar archive")?;

    Ok(())
}