iceberg/scan/task.rs
1// Licensed to the Apache Software Foundation (ASF) under one
2// or more contributor license agreements. See the NOTICE file
3// distributed with this work for additional information
4// regarding copyright ownership. The ASF licenses this file
5// to you under the Apache License, Version 2.0 (the
6// "License"); you may not use this file except in compliance
7// with the License. You may obtain a copy of the License at
8//
9// http://www.apache.org/licenses/LICENSE-2.0
10//
11// Unless required by applicable law or agreed to in writing,
12// software distributed under the License is distributed on an
13// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
14// KIND, either express or implied. See the License for the
15// specific language governing permissions and limitations
16// under the License.
17
18use futures::stream::BoxStream;
19use serde::{Deserialize, Serialize};
20
21use crate::Result;
22use crate::expr::BoundPredicate;
23use crate::spec::{DataContentType, DataFileFormat, ManifestEntryRef, Schema, SchemaRef};
24
25/// A stream of [`FileScanTask`].
26pub type FileScanTaskStream = BoxStream<'static, Result<FileScanTask>>;
27
28/// A task to scan part of file.
29#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
30pub struct FileScanTask {
31 /// The start offset of the file to scan.
32 pub start: u64,
33 /// The length of the file to scan.
34 pub length: u64,
35 /// The number of records in the file to scan.
36 ///
37 /// This is an optional field, and only available if we are
38 /// reading the entire data file.
39 pub record_count: Option<u64>,
40
41 /// The data file path corresponding to the task.
42 pub data_file_path: String,
43
44 /// The format of the file to scan.
45 pub data_file_format: DataFileFormat,
46
47 /// The schema of the file to scan.
48 pub schema: SchemaRef,
49 /// The field ids to project.
50 pub project_field_ids: Vec<i32>,
51 /// The predicate to filter.
52 #[serde(skip_serializing_if = "Option::is_none")]
53 pub predicate: Option<BoundPredicate>,
54
55 /// The list of delete files that may need to be applied to this data file
56 pub deletes: Vec<FileScanTaskDeleteFile>,
57}
58
59impl FileScanTask {
60 /// Returns the data file path of this file scan task.
61 pub fn data_file_path(&self) -> &str {
62 &self.data_file_path
63 }
64
65 /// Returns the project field id of this file scan task.
66 pub fn project_field_ids(&self) -> &[i32] {
67 &self.project_field_ids
68 }
69
70 /// Returns the predicate of this file scan task.
71 pub fn predicate(&self) -> Option<&BoundPredicate> {
72 self.predicate.as_ref()
73 }
74
75 /// Returns the schema of this file scan task as a reference
76 pub fn schema(&self) -> &Schema {
77 &self.schema
78 }
79
80 /// Returns the schema of this file scan task as a SchemaRef
81 pub fn schema_ref(&self) -> SchemaRef {
82 self.schema.clone()
83 }
84}
85
86#[derive(Debug)]
87pub(crate) struct DeleteFileContext {
88 pub(crate) manifest_entry: ManifestEntryRef,
89 pub(crate) partition_spec_id: i32,
90}
91
92impl From<&DeleteFileContext> for FileScanTaskDeleteFile {
93 fn from(ctx: &DeleteFileContext) -> Self {
94 FileScanTaskDeleteFile {
95 file_path: ctx.manifest_entry.file_path().to_string(),
96 file_type: ctx.manifest_entry.content_type(),
97 partition_spec_id: ctx.partition_spec_id,
98 equality_ids: ctx.manifest_entry.data_file.equality_ids.clone(),
99 }
100 }
101}
102
103/// A task to scan part of file.
104#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
105pub struct FileScanTaskDeleteFile {
106 /// The delete file path
107 pub file_path: String,
108
109 /// delete file type
110 pub file_type: DataContentType,
111
112 /// partition id
113 pub partition_spec_id: i32,
114
115 /// equality ids for equality deletes (null for anything other than equality-deletes)
116 pub equality_ids: Option<Vec<i32>>,
117}