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
// Copyright 2019 Google LLC // // Licensed under the Apache License, Version 2.0 (the "License"); // you may not use this file except in compliance with the License. // You may obtain a copy of the License at // // http://www.apache.org/licenses/LICENSE-2.0 // // Unless required by applicable law or agreed to in writing, software // distributed under the License is distributed on an "AS IS" BASIS, // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. // See the License for the specific language governing permissions and // limitations under the License. //! # Commonly used functionality adapters. //! //! At the moment, this crate contains the declaration of various errors use { rust_icu_sys as sys, std::{ffi, os}, thiserror::Error, }; /// Represents a Unicode error, resulting from operations of low-level ICU libraries. /// /// This is modeled after absl::Status in the Abseil library, which provides ways /// for users to avoid dealing with all the numerous error codes directly. #[derive(Error, Debug)] pub enum Error { /// The error originating in the underlying sys library. /// /// At the moment it is possible to produce an Error which has a zero error code (i.e. no /// error), because it makes it unnecessary for users to deal with error codes directly. It /// does make for a bit weird API, so we may turn it around a bit. Ideally, it should not be /// possible to have an Error that isn't really an error. #[error("ICU error code: {}", _0)] Sys(sys::UErrorCode), /// Errors originating from the wrapper code. For example when pre-converting input into /// UTF8 for input that happens to be malformed. #[error(transparent)] Wrapper(anyhow::Error), } impl Error { /// The error code denoting no error has happened. pub const OK_CODE: sys::UErrorCode = sys::UErrorCode::U_ZERO_ERROR; /// Returns true if this error code corresponds to no error. pub fn is_ok(code: sys::UErrorCode) -> bool { code == Self::OK_CODE } /// Creates a new error from the supplied status. Ok is returned if the error code does not /// correspond to an error code (as opposed to OK or a warning code). pub fn ok_or_warning(status: sys::UErrorCode) -> Result<(), Self> { if Self::is_ok(status) || status < Self::OK_CODE { Ok(()) } else { Err(Error::Sys(status)) } } /// Creates a new error from the supplied status. Ok is returned if the /// error code does not constitute an error in preflight mode. /// /// This error check explicitly ignores the buffer overflow error when reporting whether it /// contains an error condition. /// /// Preflight calls to ICU libraries do a dummy scan of the input to determine the buffer sizes /// required on the output in case of conversion calls such as `ucal_strFromUTF8`. The way /// this call is made is to offer a zero-capacity buffer (which could be pointed to by a `NULL` /// pointer), and then call the respective function. The function will compute the buffer /// size, but will also return a bogus buffer overflow error. pub fn ok_preflight(status: sys::UErrorCode) -> Result<(), Self> { if status > Self::OK_CODE && status != sys::UErrorCode::U_BUFFER_OVERFLOW_ERROR { Err(Error::Sys(status)) } else { Ok(()) } } /// Returns true if this error has the supplied `code`. pub fn is_code(&self, code: sys::UErrorCode) -> bool { if let Error::Sys(c) = self { return *c == code; } false } /// Returns true if the error is an error, not a warning. /// /// The ICU4C library has error codes for errors and warnings. pub fn is_err(&self) -> bool { match self { Error::Sys(code) => *code > sys::UErrorCode::U_ZERO_ERROR, Error::Wrapper(_) => true, } } /// Return true if there was an error in a preflight call. /// /// This error check explicitly ignores the buffer overflow error when reporting whether it /// contains an error condition. /// /// Preflight calls to ICU libraries do a dummy scan of the input to determine the buffer sizes /// required on the output in case of conversion calls such as `ucal_strFromUTF8`. The way /// this call is made is to offer a zero-capacity buffer (which could be pointed to by a `NULL` /// pointer), and then call the respective function. The function will compute the buffer /// size, but will also return a bogus buffer overflow error. pub fn is_preflight_err(&self) -> bool { // We may expand the set of error codes that are exempt from error checks in preflight. self.is_err() && !self.is_code(sys::UErrorCode::U_BUFFER_OVERFLOW_ERROR) } /// Returns true if the error is, in fact, a warning (nonfatal). pub fn is_warn(&self) -> bool { match self { Error::Sys(c) => *c < sys::UErrorCode::U_ZERO_ERROR, _ => false, } } pub fn wrapper(source: impl Into<anyhow::Error>) -> Self { Self::Wrapper(source.into()) } } impl From<ffi::NulError> for Error { fn from(e: ffi::NulError) -> Self { Self::wrapper(e) } } impl From<std::str::Utf8Error> for Error { fn from(e: std::str::Utf8Error) -> Self { Self::wrapper(e) } } impl From<std::string::FromUtf8Error> for Error { fn from(e: std::string::FromUtf8Error) -> Self { Self::wrapper(e) } } /// Used to simulate an array of C-style strings. #[derive(Debug)] pub struct CStringVec { // The internal representation of the vector of C strings. rep: Vec<ffi::CString>, // Same as rep, but converted into C pointers. c_rep: Vec<*const os::raw::c_char>, } impl CStringVec { /// Creates a new C string vector from the provided rust strings. /// /// C strings are continuous byte regions that end in `\0` and do not /// contain `\0` anywhere else. /// /// Use `as_c_array` to get an unowned raw pointer to the array, to pass /// into FFI C code. pub fn new(strings: &[&str]) -> Result<Self, Error> { let mut rep = Vec::with_capacity(strings.len()); // Convert all to asciiz strings and insert into the vector. for elem in strings { let asciiz = ffi::CString::new(*elem)?; rep.push(asciiz); } let c_rep = rep.iter().map(|e| e.as_ptr()).collect(); Ok(CStringVec { rep, c_rep }) } /// Returns the underlying array of C strings as a C array pointer. The /// array must not change after construction to ensure that this pointer /// remains valid. pub fn as_c_array(&self) -> *const *const os::raw::c_char { self.c_rep.as_ptr() as *const *const os::raw::c_char } /// Returns the number of elements in the vector. pub fn len(&self) -> usize { self.rep.len() } } #[cfg(test)] mod tests { use super::*; #[test] fn test_error_code() { let error = Error::ok_or_warning(sys::UErrorCode::U_BUFFER_OVERFLOW_ERROR) .err() .unwrap(); assert!(error.is_code(sys::UErrorCode::U_BUFFER_OVERFLOW_ERROR)); assert!(!error.is_preflight_err()); assert!(!error.is_code(sys::UErrorCode::U_ZERO_ERROR)); } #[test] fn test_into_char_array() { let values = vec!["eenie", "meenie", "minie", "moe"]; let c_array = CStringVec::new(&values).expect("success"); assert_eq!(c_array.len(), 4); } #[test] fn test_with_embedded_nul_byte() { let values = vec!["hell\0x00o"]; let _c_array = CStringVec::new(&values).expect_err("should fail"); } }