1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
/****************************************************************
*                                                               *
* Copyright (c) 2019-2021 YottaDB LLC and/or its subsidiaries.  *
* All rights reserved.                                          *
*                                                               *
*       This source code contains the intellectual property     *
*       of its copyright holder(s), and is made available       *
*       under a license.  If you do not know the terms of       *
*       the license, please stop and do not read further.       *
*                                                               *
****************************************************************/

//! [YottaDB] is a NoSQL Database suitable for high-performance systems.
//!
//! YottaDB runs in-process, like SQLite, with no need for a daemon.
//! This crate is a Rust wrapper around the C implementation of YottaDB.
//!
//! There are two major APIs:
//! - [`craw`], the FFI bindings generated directly by bindgen.
//!     These are not recommended for normal use,
//!     but are available in case the Context API is missing functionality.
//! - The main Context API, which is a safe wrapper around the C API which
//!     stores the current tptoken and an error buffer so you don't have to keep track of them yourself.
//!     The reason this metadata is necessary is because this crate binds to the threaded version of
//!     YottaDB, which requires a `tptoken` and `err_buffer`. See [transaction processing] for more
//!     details on transactions and `tptoken`s.
//!
//! Most operations are encapsulated in methods in the [`KeyContext`] struct.
//! Iteration helpers are available to iterate over values in the database in a variety of ways.
//!
//! # Examples
//!
//! A basic database operation (set a value, retrieve it, and delete it)
//!
//! ```
//! use yottadb::{Context, KeyContext, DeleteType, YDBResult};
//!
//! fn main() -> YDBResult<()> {
//!     let ctx = Context::new();
//!     let mut key = KeyContext::new(&ctx, "^MyGlobal", &["SubscriptA", "42"]);
//!     key.set("This is a persistent message")?;
//!     let buffer = key.get()?;
//!     assert_eq!(&buffer, b"This is a persistent message");
//!     key.delete(DeleteType::DelNode)?;
//!     Ok(())
//! }
//! ```
//!
//! # Intrinsic Variables
//!
//! YottaDB has several intrinsic variables which are documented [online][intrinsics].
//! To get the value of these variables, call `get_st` on a `Key` with the name of the variable.
//!
//! ## Example
//!
//! Get the instrinsic variable [`$tlevel`][tlevel], which gives the current transaction level.
//!
//! ```
//! use yottadb::{Context, KeyContext, YDB_NOTTP, YDBResult};
//!
//! fn main() -> YDBResult<()> {
//!     let ctx = Context::new();
//!     let mut key = KeyContext::variable(&ctx, "$tlevel");
//!     let tlevel: usize = String::from_utf8_lossy(&key.get()?).parse()
//!         .expect("$tlevel should be an integer");
//!     assert_eq!(tlevel, 0_usize);
//!     Ok(())
//! }
//! ```
//!
//! # Features
//!
//! Since `yottadb` is a set of bindings to a C library, it uses `bindgen` to generate the bindings.
//! There are two ways to do this:
//! - `features = ["vendor"]`, the default. This compiles `bindgen` from source.
//! - `default-features = false`. This requires you to have bindgen already installed locally.
//!
//! Using vendoring means you can use `yottadb` in any user environment,
//! even when you don't have admin priviledges to install programs.
//! Using a pre-installed version means compile times are much lower.
//!
//! # Signal handling
//!
//! YottaDB performs its own signal handling in addition to any signal handlers you may have set up.
//! Since many functions in C are not async-safe, it defers any action until the next time `ydb_eintr_handler` is called.
//! All YDB functions will automatically call `ydb_eintr_handler` if necessary,
//! so in most cases this should not affect your application. However, there are some rare cases
//! when the handler will not be called:
//! - If you have a tight loop inside a [`Context::tp`] that does not call a YDB function
//!
//! For example, the following loop will run forever even if sent SIGINT:
//! ```no_run
//! # fn main() -> yottadb::YDBResult<()> {
//! use yottadb::{Context, KeyContext};
//! let ctx = Context::new();
//! let loop_callback = |_| loop {
//!     std::thread::sleep(std::time::Duration::from_secs(1));
//!     println!("finished sleep");
//! };
//! ctx.tp(loop_callback, "BATCH", &[]).unwrap();
//! # Ok(())
//! # }
//! ```
//!
//! To avoid this, call [`Context::eintr_handler`] in the loop:
//!
//! ```no_run
//! # fn main() -> yottadb::YDBResult<()> {
//! # use yottadb::{Context, KeyContext};
//! # let ctx = Context::new();
//! # let release = ctx.release()?;
//! loop {
//!     std::thread::sleep(std::time::Duration::from_secs(1));
//!     ctx.eintr_handler()?;
//!     println!("finished sleep");
//! }
//! # }
//! ```
//!
//! However, you should endeavor to keep transactions as short as possible -
//! both for performance, since YottaDB uses optimistic concurrency control,
//! and for reliability, since operations will not be committed until the transaction concludes.
//!
//! As a last-ditch method of error-recovery, YottaDB will exit immediately
//! if sent three interrupt signals in a row, even if `eintr_handler` is not called.
//!
//! YottaDB does not register any signal handlers until the first time `ydb_init` is called,
//! and deregisters its handlers after `ydb_exit`.
//!
//! ## See also
//!
//! - The [C documentation on signals](https://docs.yottadb.com/MultiLangProgGuide/programmingnotes.html#signals)
//! - [`Context::eintr_handler`]
//! - [`Context::tp`](crate::Context::tp)
//!
//! [YottaDB]: https://yottadb.com/
//! [transaction processing]: https://docs.yottadb.com/MultiLangProgGuide/MultiLangProgGuide.html#transaction-processing
//! [intrinsics]: https://docs.yottadb.com/MultiLangProgGuide/MultiLangProgGuide.html#intrinsic-special-variables
//! [tlevel]: https://docs.yottadb.com/MultiLangProgGuide/MultiLangProgGuide.html#tlevel
#![deny(missing_docs)]
#![allow(clippy::upper_case_acronyms)]

/// This is the entry-point of the `yottadb` crate. See
/// <https://doc.rust-lang.org/book/ch07-02-defining-modules-to-control-scope-and-privacy.html>
/// for more information; it calls this module the "crate root".
///
/// The root contains functions and data shared between both the simple and context API.
#[allow(unused)]
const INTERNAL_DOCS: () = ();

/// Public to reduce churn when upgrading versions, but it's recommended to use the top-level re-exports instead.
#[doc(hidden)]
pub mod context_api;
#[allow(missing_docs)]
pub mod craw;
mod simple_api;

pub use craw::{YDB_ERR_GVUNDEF, YDB_ERR_LVUNDEF};
#[doc(inline)] // needed because of rust-lang/rust#81890
pub use context_api::*; // glob import so we catch all the iterators
pub use simple_api::{
    call_in::{CallInDescriptor, CallInTableDescriptor},
    DataReturn, DeleteType, Key, TransactionStatus, TpToken, YDBError, YDBResult,
};
#[doc(hidden)]
/// This has to be public so that it can be used by `ci_t!`.
/// However, it is not a supported part of the API.
pub use simple_api::resize_call;

// This is not just a convenience for users; there is no way to construct a `TpToken` outside of
// YDBRust because the fields are private.
/// The default transaction processing token if no transaction is in progress.
pub const YDB_NOTTP: TpToken = TpToken(
    // This can't use `TpToken::default` because constants cannot use trait functions.
    craw::YDB_NOTTP,
);

/// Cleans up the process connection/access to all databases and all yottadb data structures.
///
/// If you have already made a call to YottaDB, any future calls to YottaDB after calling `yottadb::ydb_exit()`
/// will return `YDBError { status: YDB_ERR_CALLINAFTERXIT }`.
/// If you have never before made a call to YottaDB, `exit()` has no effect.
///
/// A typical application should not need to call `yottadb::ydb_exit()`
/// since YottaDB will automatically clean up on process termination.
///
/// This has no effect on any [`Key`]s, which will be automatically dropped when they go out of scope.
///
/// # Errors
/// - `YDB_ERR_INVYDBEXIT` if `ydb_exit()` is called through M FFI (e.g. through [`ci_t`])
///
/// Possible errors for this function include:
/// - [error return codes](https://docs.yottadb.com/MultiLangProgGuide/cprogram.html#error-return-code)
///
/// # Example
// this is no_run because otherwise all other tests will get YDB_ERR_CALLINAFTERXIT
/// ```no_run
/// yottadb::ydb_exit();
/// ```
pub fn ydb_exit() -> std::os::raw::c_int {
    unsafe { craw::ydb_exit() }
}

/// The locks in this module prevents data races in the test suite.
///
/// In particular, `delete_excl` and some of the call-in tests interfere with all other concurrent tests.
#[cfg(test)]
mod test_lock {
    use std::sync::{RwLock, RwLockReadGuard, RwLockWriteGuard};
    use once_cell::sync::Lazy;

    #[derive(Debug)]
    #[must_use = "if unused the RwLock will immediately unlock"]
    pub(crate) enum LockGuard<'a> {
        Read(RwLockReadGuard<'a, ()>),
        Write(RwLockWriteGuard<'a, ()>),
    }

    impl LockGuard<'_> {
        /// Use a `Read` lock if the test can be run concurrently with others.
        ///
        /// You may have to use unique variable names for this property to hold.
        pub(crate) fn read() -> Self {
            // If one tests panics, don't cause all others to panic as well.
            let guard = match TEST_LOCK.read() {
                Ok(g) => g,
                Err(poison) => poison.into_inner(),
            };
            LockGuard::Read(guard)
        }

        /// Use a `Write` lock if the test interferes with all other concurrent tests.
        pub(crate) fn write() -> Self {
            // If one tests panics, don't cause all others to panic as well.
            let guard = match TEST_LOCK.write() {
                Ok(g) => g,
                Err(poison) => poison.into_inner(),
            };
            LockGuard::Write(guard)
        }
    }

    /// This lock holds no data, it is used only for synchronization.
    pub(crate) static TEST_LOCK: Lazy<RwLock<()>> = Lazy::new(|| RwLock::new(()));
}