Add getter methods

Author: fqaiser94

crates/iceberg/src/puffin/blob.rs

@@ -23,16 +23,48 @@ pub const APACHE_DATASKETCHES_THETA_V1: &str = "apache-datasketches-theta-v1";
 /// The blob
 #[derive(Debug, PartialEq, Clone)]
 pub struct Blob {
+    pub(crate) r#type: String,
+    pub(crate) fields: Vec<i32>,
+    pub(crate) snapshot_id: i64,
+    pub(crate) sequence_number: i64,
+    pub(crate) data: Vec<u8>,
+    pub(crate) properties: HashMap<String, String>,
+}
+
+impl Blob {
+    #[inline]
     /// See blob types: https://iceberg.apache.org/puffin-spec/#blob-types
-    pub r#type: String,
+    pub fn blob_type(&self) -> &String {
+        &self.r#type
+    }
+
+    #[inline]
     /// List of field IDs the blob was computed for; the order of items is used to compute sketches stored in the blob.
-    pub fields: Vec<i32>,
+    pub fn fields(&self) -> &Vec<i32> {
+        &self.fields
+    }
+
+    #[inline]
     /// ID of the Iceberg table's snapshot the blob was computed from
-    pub snapshot_id: i64,
+    pub fn snapshot_id(&self) -> &i64 {
+        &self.snapshot_id
+    }
+
+    #[inline]
     /// Sequence number of the Iceberg table's snapshot the blob was computed from
-    pub sequence_number: i64,
+    pub fn sequence_number(&self) -> &i64 {
+        &self.r#sequence_number
+    }
+
+    #[inline]
     /// The uncompressed blob data
-    pub data: Vec<u8>,
+    pub fn data(&self) -> &Vec<u8> {
+        &self.data
+    }
+
+    #[inline]
     /// Arbitrary meta-information about the blob
-    pub properties: HashMap<String, String>,
+    pub fn properties(&self) -> &HashMap<String, String> {
+        &self.properties
+    }
 }

crates/iceberg/src/puffin/metadata.rs

@@ -33,26 +33,68 @@ pub const CREATED_BY_PROPERTY: &str = "created-by";
 #[derive(Debug, PartialEq, Eq, Serialize, Deserialize, Clone)]
 #[serde(rename_all = "kebab-case")]
 pub struct BlobMetadata {
+    pub(crate) r#type: String,
+    pub(crate) fields: Vec<i32>,
+    pub(crate) snapshot_id: i64,
+    pub(crate) sequence_number: i64,
+    pub(crate) offset: u64,
+    pub(crate) length: u64,
+    #[serde(skip_serializing_if = "CompressionCodec::is_none")]
+    #[serde(default)]
+    pub(crate) compression_codec: CompressionCodec,
+    #[serde(skip_serializing_if = "HashMap::is_empty")]
+    #[serde(default)]
+    pub(crate) properties: HashMap<String, String>,
+}
+
+impl BlobMetadata {
+    #[inline]
     /// See blob types: https://iceberg.apache.org/puffin-spec/#blob-types
-    pub r#type: String,
+    pub fn blob_type(&self) -> &String {
+        &self.r#type
+    }
+
+    #[inline]
     /// List of field IDs the blob was computed for; the order of items is used to compute sketches stored in the blob.
-    pub fields: Vec<i32>,
+    pub fn fields(&self) -> &Vec<i32> {
+        &self.fields
+    }
+
+    #[inline]
     /// ID of the Iceberg table's snapshot the blob was computed from
-    pub snapshot_id: i64,
+    pub fn snapshot_id(&self) -> &i64 {
+        &self.snapshot_id
+    }
+
+    #[inline]
     /// Sequence number of the Iceberg table's snapshot the blob was computed from
-    pub sequence_number: i64,
+    pub fn sequence_number(&self) -> &i64 {
+        &self.r#sequence_number
+    }
+
+    #[inline]
     /// The offset in the file where the blob contents start
-    pub offset: u64,
+    pub fn offset(&self) -> &u64 {
+        &self.offset
+    }
+
+    #[inline]
     /// The length of the blob stored in the file (after compression, if compressed)
-    pub length: u64,
+    pub fn length(&self) -> &u64 {
+        &self.length
+    }
+
+    #[inline]
     /// The compression codec used to compress the data
-    #[serde(skip_serializing_if = "CompressionCodec::is_none")]
-    #[serde(default)]
-    pub compression_codec: CompressionCodec,
+    pub fn compression_codec(&self) -> &CompressionCodec {
+        &self.compression_codec
+    }
+
+    #[inline]
     /// Arbitrary meta-information about the blob
-    #[serde(skip_serializing_if = "HashMap::is_empty")]
-    #[serde(default)]
-    pub properties: HashMap<String, String>,
+    pub fn properties(&self) -> &HashMap<String, String> {
+        &self.properties
+    }
 }
 
 #[derive(Clone, Copy, PartialEq, Eq, Hash, Debug)]
@@ -92,12 +134,10 @@ impl Flag {
 /// For more information, see: https://iceberg.apache.org/puffin-spec/#filemetadata
 #[derive(Debug, PartialEq, Eq, Serialize, Deserialize, Clone)]
 pub struct FileMetadata {
-    /// Metadata about blobs in file
-    pub blobs: Vec<BlobMetadata>,
-    /// Arbitrary meta-information, like writer identification/version.
+    pub(crate) blobs: Vec<BlobMetadata>,
     #[serde(skip_serializing_if = "HashMap::is_empty")]
     #[serde(default)]
-    pub properties: HashMap<String, String>,
+    pub(crate) properties: HashMap<String, String>,
 }
 
 impl FileMetadata {
@@ -247,6 +287,18 @@ impl FileMetadata {
             FileMetadata::extract_footer_payload_as_str(&footer_bytes, footer_payload_length)?;
         FileMetadata::from_json_str(&footer_payload_str)
     }
+
+    #[inline]
+    /// Metadata about blobs in file
+    pub fn blobs(&self) -> &Vec<BlobMetadata> {
+        &self.blobs
+    }
+
+    #[inline]
+    /// Arbitrary meta-information, like writer identification/version.
+    pub fn properties(&self) -> &HashMap<String, String> {
+        &self.properties
+    }
 }
 
 #[cfg(test)]