add ZipEntry API in src/unstable/read.rs

- also use git dep for zstd to pull in removal of R: BufRead bound
- move new methods into unstable/read.rs
- support getting aes verification info from a raw ZipEntry
- flesh out other methods from ZipFile -> ZipEntry
- now create a streaming abstraction
- move some methods around
- make AesModeInfo pub(crate) to make its intention clear
- make ArchiveEntry more publicly accessible
- make ZipFile impl EntryData

Author: cosmicexplorer

src/aes.rs

@@ -66,7 +66,7 @@ pub struct AesReader<R> {
     data_length: u64,
 }
 
-impl<R: Read> AesReader<R> {
+impl<R> AesReader<R> {
     pub const fn new(reader: R, aes_mode: AesMode, compressed_size: u64) -> AesReader<R> {
         let data_length = compressed_size
             - (PWD_VERIFY_LENGTH + AUTH_CODE_LENGTH + aes_mode.salt_length()) as u64;
@@ -77,7 +77,9 @@ impl<R: Read> AesReader<R> {
             data_length,
         }
     }
+}
 
+impl<R: Read> AesReader<R> {
     /// Read the AES header bytes and validate the password.
     ///
     /// Even if the validation succeeds, there is still a 1 in 65536 chance that an incorrect
@@ -150,7 +152,7 @@ impl<R: Read> AesReader<R> {
 /// There is a 1 in 65536 chance that an invalid password passes that check.
 /// After the data has been read and decrypted an HMAC will be checked and provide a final means
 /// to check if either the password is invalid or if the data has been changed.
-pub struct AesReaderValid<R: Read> {
+pub struct AesReaderValid<R> {
     reader: R,
     data_remaining: u64,
     cipher: Cipher,
@@ -214,7 +216,7 @@ impl<R: Read> Read for AesReaderValid<R> {
     }
 }
 
-impl<R: Read> AesReaderValid<R> {
+impl<R> AesReaderValid<R> {
     /// Consumes this decoder, returning the underlying reader.
     pub fn into_inner(self) -> R {
         self.reader

src/crc32.rs

@@ -83,6 +83,110 @@ impl<R: Read> Read for Crc32Reader<R> {
     }
 }
 
+pub(crate) mod non_crypto {
+    use std::io;
+    use std::io::prelude::*;
+
+    use crc32fast::Hasher;
+
+    /// Reader that validates the CRC32 when it reaches the EOF.
+    pub struct Crc32Reader<R> {
+        inner: R,
+        hasher: Hasher,
+        check: u32,
+    }
+
+    impl<R> Crc32Reader<R> {
+        /// Get a new Crc32Reader which checks the inner reader against checksum.
+        pub(crate) fn new(inner: R, checksum: u32) -> Self {
+            Crc32Reader {
+                inner,
+                hasher: Hasher::new(),
+                check: checksum,
+            }
+        }
+
+        fn check_matches(&self) -> Result<(), &'static str> {
+            let res = self.hasher.clone().finalize();
+            if self.check == res {
+                Ok(())
+            } else {
+                /* TODO: make this into our own Crc32Error error type! */
+                Err("Invalid checksum")
+            }
+        }
+    }
+
+    impl<R: Read> Read for Crc32Reader<R> {
+        fn read(&mut self, buf: &mut [u8]) -> io::Result<usize> {
+            /* We want to make sure we only check the hash when the input stream is exhausted. */
+            if buf.is_empty() {
+                /* If the input buf is empty (this shouldn't happen, but isn't guaranteed), we
+                 * still want to "pull" from the source in case it surfaces an i/o error. This will
+                 * always return a count of Ok(0) if successful. */
+                return self.inner.read(buf);
+            }
+
+            let count = self.inner.read(buf)?;
+            if count == 0 {
+                return self
+                    .check_matches()
+                    .map(|()| 0)
+                    /* TODO: use io::Error::other for MSRV >=1.74 */
+                    .map_err(|e| io::Error::new(io::ErrorKind::Other, e));
+            }
+            self.hasher.update(&buf[..count]);
+            Ok(count)
+        }
+    }
+
+    #[cfg(test)]
+    mod test {
+        use super::*;
+
+        #[test]
+        fn test_empty_reader() {
+            let data: &[u8] = b"";
+            let mut buf = [0; 1];
+
+            let mut reader = Crc32Reader::new(data, 0);
+            assert_eq!(reader.read(&mut buf).unwrap(), 0);
+
+            let mut reader = Crc32Reader::new(data, 1);
+            assert!(reader
+                .read(&mut buf)
+                .unwrap_err()
+                .to_string()
+                .contains("Invalid checksum"));
+        }
+
+        #[test]
+        fn test_byte_by_byte() {
+            let data: &[u8] = b"1234";
+            let mut buf = [0; 1];
+
+            let mut reader = Crc32Reader::new(data, 0x9be3e0a3);
+            assert_eq!(reader.read(&mut buf).unwrap(), 1);
+            assert_eq!(reader.read(&mut buf).unwrap(), 1);
+            assert_eq!(reader.read(&mut buf).unwrap(), 1);
+            assert_eq!(reader.read(&mut buf).unwrap(), 1);
+            assert_eq!(reader.read(&mut buf).unwrap(), 0);
+            // Can keep reading 0 bytes after the end
+            assert_eq!(reader.read(&mut buf).unwrap(), 0);
+        }
+
+        #[test]
+        fn test_zero_read() {
+            let data: &[u8] = b"1234";
+            let mut buf = [0; 5];
+
+            let mut reader = Crc32Reader::new(data, 0x9be3e0a3);
+            assert_eq!(reader.read(&mut buf[..0]).unwrap(), 0);
+            assert_eq!(reader.read(&mut buf).unwrap(), 4);
+        }
+    }
+}
+
 #[cfg(test)]
 mod test {
     use super::*;

src/lib.rs

@@ -32,8 +32,8 @@
 #![warn(missing_docs)]
 #![allow(unexpected_cfgs)] // Needed for cfg(fuzzing) on nightly as of 2024-05-06
 pub use crate::compression::{CompressionMethod, SUPPORTED_COMPRESSION_METHODS};
-pub use crate::read::HasZipMetadata;
 pub use crate::read::ZipArchive;
+pub use crate::read::{EntryData, HasZipMetadata};
 pub use crate::spec::{ZIP64_BYTES_THR, ZIP64_ENTRY_THR};
 pub use crate::types::{AesMode, DateTime};
 pub use crate::write::ZipWriter;