commonware_storage/journal/contiguous/mod.rs
1//! Contiguous journals with position-based access.
2//!
3//! This module provides position-based journal implementations where items are stored
4//! contiguously and can be accessed by their position (0-indexed). Both [fixed]-size and
5//! [variable]-size item journals are supported.
6
7use super::Error;
8use futures::Stream;
9use std::{future::Future, num::NonZeroUsize, ops::Range};
10use tracing::warn;
11
12pub mod fixed;
13pub mod variable;
14
15#[cfg(test)]
16mod tests;
17
18/// Core trait for contiguous journals supporting sequential append operations.
19///
20/// A contiguous journal maintains a consecutively increasing position counter where each
21/// appended item receives a unique position starting from 0.
22pub trait Contiguous: Send + Sync {
23 /// The type of items stored in the journal.
24 type Item;
25
26 /// Returns [start, end) where `start` and `end - 1` are the indices of the oldest and newest
27 /// retained operations respectively.
28 fn bounds(&self) -> Range<u64>;
29
30 /// Return the total number of items that have been appended to the journal.
31 ///
32 /// This count is NOT affected by pruning. The next appended item will receive this
33 /// position as its value.
34 ///
35 /// Equivalent to `bounds().end`.
36 fn size(&self) -> u64 {
37 self.bounds().end
38 }
39
40 /// Return a stream of all items in the journal starting from `start_pos`.
41 ///
42 /// Each item is yielded as a tuple `(position, item)` where position is the item's
43 /// stable position in the journal.
44 ///
45 /// # Errors
46 ///
47 /// Returns an error if `start_pos` exceeds the journal size or if any storage/decoding
48 /// errors occur during replay.
49 fn replay(
50 &self,
51 start_pos: u64,
52 buffer: NonZeroUsize,
53 ) -> impl std::future::Future<
54 Output = Result<impl Stream<Item = Result<(u64, Self::Item), Error>> + Send + '_, Error>,
55 > + Send;
56
57 /// Read the item at the given position.
58 ///
59 /// # Errors
60 ///
61 /// - Returns [Error::ItemPruned] if the item at `position` has been pruned.
62 /// - Returns [Error::ItemOutOfRange] if the item at `position` does not exist.
63 fn read(&self, position: u64) -> impl Future<Output = Result<Self::Item, Error>> + Send;
64}
65
66/// A [Contiguous] journal that supports appending, rewinding, and pruning.
67pub trait MutableContiguous: Contiguous + Send + Sync {
68 /// Append a new item to the journal, returning its position.
69 ///
70 /// Positions are consecutively increasing starting from 0. The position of each item
71 /// is stable across pruning (i.e., if item X has position 5, it will always have
72 /// position 5 even if earlier items are pruned).
73 ///
74 /// # Errors
75 ///
76 /// Returns an error if the underlying storage operation fails or if the item cannot
77 /// be encoded.
78 fn append(
79 &mut self,
80 item: Self::Item,
81 ) -> impl std::future::Future<Output = Result<u64, Error>> + Send;
82
83 /// Prune items at positions strictly less than `min_position`.
84 ///
85 /// Returns `true` if any data was pruned, `false` otherwise.
86 ///
87 /// # Behavior
88 ///
89 /// - If `min_position > bounds.end`, the prune is capped to `bounds.end` (no error is returned)
90 /// - Some items with positions less than `min_position` may be retained due to
91 /// section/blob alignment
92 /// - This operation is not atomic, but implementations guarantee the journal is left in a
93 /// recoverable state if a crash occurs during pruning
94 ///
95 /// # Errors
96 ///
97 /// Returns an error if the underlying storage operation fails.
98 fn prune(
99 &mut self,
100 min_position: u64,
101 ) -> impl std::future::Future<Output = Result<bool, Error>> + Send;
102
103 /// Rewind the journal to the given size, discarding items from the end.
104 ///
105 /// After rewinding to size N, the journal will contain exactly N items (positions 0 to N-1),
106 /// and the next append will receive position N.
107 ///
108 /// # Behavior
109 ///
110 /// - If `size > bounds.end`, returns [Error::InvalidRewind]
111 /// - If `size == bounds.end`, this is a no-op
112 /// - If `size < bounds.start`, returns [Error::ItemPruned] (can't rewind to pruned data)
113 /// - This operation is not atomic, but implementations guarantee the journal is left in a
114 /// recoverable state if a crash occurs during rewinding
115 ///
116 /// # Warnings
117 ///
118 /// - This operation is not guaranteed to survive restarts until `commit` or `sync` is called.
119 ///
120 /// # Errors
121 ///
122 /// Returns [Error::InvalidRewind] if size is invalid (too large or points to pruned data).
123 /// Returns an error if the underlying storage operation fails.
124 fn rewind(&mut self, size: u64) -> impl std::future::Future<Output = Result<(), Error>> + Send;
125
126 /// Rewinds the journal to the last item matching `predicate`. If no item matches, the journal
127 /// is rewound to the pruning boundary, discarding all unpruned items.
128 ///
129 /// # Warnings
130 ///
131 /// - This operation is not guaranteed to survive restarts until `commit` or `sync` is called.
132 fn rewind_to<'a, P>(
133 &'a mut self,
134 mut predicate: P,
135 ) -> impl std::future::Future<Output = Result<u64, Error>> + Send + 'a
136 where
137 P: FnMut(&Self::Item) -> bool + Send + 'a,
138 {
139 async move {
140 let bounds = self.bounds();
141 let mut rewind_size = bounds.end;
142
143 while rewind_size > bounds.start {
144 let item = self.read(rewind_size - 1).await?;
145 if predicate(&item) {
146 break;
147 }
148 rewind_size -= 1;
149 }
150
151 if rewind_size != bounds.end {
152 let rewound_items = bounds.end - rewind_size;
153 warn!(
154 journal_size = bounds.end,
155 rewound_items, "rewinding journal items"
156 );
157 self.rewind(rewind_size).await?;
158 }
159
160 Ok(rewind_size)
161 }
162 }
163}