diff options
Diffstat (limited to 'vendor/icu_list')
| -rw-r--r-- | vendor/icu_list/.cargo-checksum.json | 2 | ||||
| -rw-r--r-- | vendor/icu_list/Cargo.lock | 761 | ||||
| -rw-r--r-- | vendor/icu_list/Cargo.toml | 43 | ||||
| -rw-r--r-- | vendor/icu_list/examples/and_list.rs | 4 | ||||
| -rw-r--r-- | vendor/icu_list/src/lazy_automaton.rs | 79 | ||||
| -rw-r--r-- | vendor/icu_list/src/lib.rs | 3 | ||||
| -rw-r--r-- | vendor/icu_list/src/list_formatter.rs | 43 | ||||
| -rw-r--r-- | vendor/icu_list/src/patterns.rs | 283 | ||||
| -rw-r--r-- | vendor/icu_list/src/provider.rs | 465 | ||||
| -rw-r--r-- | vendor/icu_list/src/provider/mod.rs | 261 | ||||
| -rw-r--r-- | vendor/icu_list/src/provider/serde_dfa.rs | 244 | ||||
| -rw-r--r-- | vendor/icu_list/src/string_matcher.rs | 213 |
12 files changed, 985 insertions, 1416 deletions
diff --git a/vendor/icu_list/.cargo-checksum.json b/vendor/icu_list/.cargo-checksum.json index 31667c731..0a33455b4 100644 --- a/vendor/icu_list/.cargo-checksum.json +++ b/vendor/icu_list/.cargo-checksum.json @@ -1 +1 @@ -{"files":{"Cargo.lock":"f650478383a2f1a7aed551b2d4650d83a2e8cc3d44c749a0e4fcfc649130b9da","Cargo.toml":"ac843eda0523eadb2d7c3bb39ea34ca5c45648f2e00c1aca87607a427f583e73","LICENSE":"4ad7541d66a407234e2c84902124cef325c29f3e966353efdb800bedb8b8da21","README.md":"33b424bdceea5edc4d3638592b007bf825e25a96e7fe9101a3ad04ea63637150","examples/and_list.rs":"2cf046f8af9892949db8c8b259ac060b9de5781a5941bec5d346cf82e9371774","src/error.rs":"3cf8a7962af6b43a91d9227e4d97824e4b2852d3d50fc00fb29c779768c1b915","src/lib.rs":"c0bd00a3d83d4285d33cef2af12bd9cae380ad9c81702ba7552ce14283e9651d","src/list_formatter.rs":"a5f6f9c8bf35c6e1d4cb712a73b31454c0d081572bbc0efd9997bf6013346825","src/provider.rs":"1ca4026dbc00c901763a41a4b4ad0083c32888b95d11f464a9e3389c7d976fec","src/string_matcher.rs":"6ce2a72cd61e3d87715dafb396d8e6ea4189a3e7dcb84d4188657a477b67b1e6"},"package":"c40218275f081c4493f190357c5395647b06734c2dc3dcb41cc099a0f60168b1"}
\ No newline at end of file +{"files":{"Cargo.lock":"11c80ad227499af3696bc71b715a91d202e3b178d24fd37776ca4221438d36ce","Cargo.toml":"2a69f60d480d23b4cfcba46039d83efe1eba24bd7ecf9cb6e7b08e7448a57b65","LICENSE":"4ad7541d66a407234e2c84902124cef325c29f3e966353efdb800bedb8b8da21","README.md":"33b424bdceea5edc4d3638592b007bf825e25a96e7fe9101a3ad04ea63637150","examples/and_list.rs":"76af2fbe31c6641a360726b201bfb043a35ee8dcf37b626640221fe889b194e4","src/error.rs":"3cf8a7962af6b43a91d9227e4d97824e4b2852d3d50fc00fb29c779768c1b915","src/lazy_automaton.rs":"eff2e95ae5c889908ba71480e73fc31d9cb2e793e9dc6dee8683c686e8ed8019","src/lib.rs":"24090cea0e42067e06fc9d5248c1ae6a03f12bd62ec0e9e2c237d2fc1251dcdb","src/list_formatter.rs":"a21de479a5ec86ed46ce0ad508e8059efb9ecd95ce869523be30118ed6a86660","src/patterns.rs":"3013aca1dd51d1b9a98e8573abac9116650635b2931157e9fb16ee10c8d3ab9c","src/provider/mod.rs":"0dbd37b8b83d40a2f5d7945cc55b4aa8477cec98442472c4b7a6875b226c8195","src/provider/serde_dfa.rs":"4f0489d176e87ad463efd36b4ead93b0b6bb0a39d26b1f96413bfb00576f7bac"},"package":"01a65ff0cab77c33c7e165c858eaa6e84a09f1e485dd495d9d0ae61083c6f786"}
\ No newline at end of file diff --git a/vendor/icu_list/Cargo.lock b/vendor/icu_list/Cargo.lock index 8d049ce78..759405aa8 100644 --- a/vendor/icu_list/Cargo.lock +++ b/vendor/icu_list/Cargo.lock @@ -3,258 +3,43 @@ version = 3 [[package]] -name = "aho-corasick" -version = "0.7.19" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "b4f55bd91a0978cbfd91c457a164bab8b4001c833b7f323132c0a4e1922dd44e" -dependencies = [ - "memchr", -] - -[[package]] name = "atomic-polyfill" -version = "0.1.10" +version = "0.1.11" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "9c041a8d9751a520ee19656232a18971f18946a7900f1520ee4400002244dd89" +checksum = "e3ff7eb3f316534d83a8a2c3d1674ace8a5a71198eba31e2e2b597833f699b28" dependencies = [ "critical-section", ] [[package]] -name = "atty" -version = "0.2.14" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d9b39be18770d11421cdb1b9947a45dd3f37e93092cbf377614828a319d5fee8" -dependencies = [ - "hermit-abi", - "libc", - "winapi", -] - -[[package]] name = "autocfg" version = "1.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "d468802bab17cbc0cc575e9b053f41e72aa36bfa6b7f55e3529ffa43161b97fa" [[package]] -name = "bare-metal" -version = "0.2.5" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5deb64efa5bd81e31fcd1938615a6d98c82eafcbcd787162b6f63b91d6bac5b3" -dependencies = [ - "rustc_version 0.2.3", -] - -[[package]] -name = "bare-metal" -version = "1.0.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f8fe8f5a8a398345e52358e18ff07cc17a568fbca5c6f73873d3a62056309603" - -[[package]] -name = "bit_field" -version = "0.10.1" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "dcb6dd1c2376d2e096796e234a70e17e94cc2d5d54ff8ce42b28cef1d0d359a4" - -[[package]] -name = "bitfield" -version = "0.13.2" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "46afbd2983a5d5a7bd740ccb198caf5b82f45c40c09c0eed36052d91cb92e719" - -[[package]] -name = "bitflags" -version = "1.3.2" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "bef38d45163c2f1dde094a7dfd33ccf595c92905c8f8f4fdc18d06fb1037718a" - -[[package]] -name = "bstr" -version = "0.2.17" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "ba3569f383e8f1598449f1a423e72e99569137b47740b1da11ef19af3d5c3223" -dependencies = [ - "lazy_static", - "memchr", - "regex-automata 0.1.10", - "serde", -] - -[[package]] -name = "bumpalo" -version = "3.11.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c1ad822118d20d2c234f427000d5acc36eabe1e29a348c89b63dd60b13f28e5d" - -[[package]] name = "byteorder" version = "1.4.3" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "14c189c53d098945499cdfa7ecc63567cf3886b3332b312a5b4585d8d3a6a610" [[package]] -name = "cast" -version = "0.3.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "37b2a672a2cb129a2e41c10b1224bb368f9f37a2b16b612598138befd7b37eb5" - -[[package]] -name = "cfg-if" -version = "1.0.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd" - -[[package]] -name = "clap" -version = "2.34.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a0610544180c38b88101fecf2dd634b174a62eef6946f84dfc6a7127512b381c" -dependencies = [ - "bitflags", - "textwrap", - "unicode-width", -] - -[[package]] name = "cobs" version = "0.2.3" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "67ba02a97a2bd10f4b59b25c7973101c79642302776489e030cd13cdab09ed15" [[package]] -name = "cortex-m" -version = "0.7.6" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "70858629a458fdfd39f9675c4dc309411f2a3f83bede76988d81bf1a0ecee9e0" -dependencies = [ - "bare-metal 0.2.5", - "bitfield", - "embedded-hal", - "volatile-register", -] - -[[package]] -name = "criterion" -version = "0.3.6" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "b01d6de93b2b6c65e17c634a26653a29d107b3c98c607c765bf38d041531cd8f" -dependencies = [ - "atty", - "cast", - "clap", - "criterion-plot", - "csv", - "itertools", - "lazy_static", - "num-traits", - "oorandom", - "plotters", - "rayon", - "regex", - "serde", - "serde_cbor", - "serde_derive", - "serde_json", - "tinytemplate", - "walkdir", -] - -[[package]] -name = "criterion-plot" -version = "0.4.5" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "2673cc8207403546f45f5fd319a974b1e6983ad1a3ee7e6041650013be041876" -dependencies = [ - "cast", - "itertools", -] - -[[package]] name = "critical-section" -version = "0.2.7" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "95da181745b56d4bd339530ec393508910c909c784e8962d15d722bacf0bcbcd" -dependencies = [ - "bare-metal 1.0.0", - "cfg-if", - "cortex-m", - "riscv", -] - -[[package]] -name = "crossbeam-channel" -version = "0.5.6" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c2dd04ddaf88237dc3b8d8f9a3c1004b506b54b3313403944054d23c0870c521" -dependencies = [ - "cfg-if", - "crossbeam-utils", -] - -[[package]] -name = "crossbeam-deque" -version = "0.8.2" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "715e8152b692bba2d374b53d4875445368fdf21a94751410af607a5ac677d1fc" -dependencies = [ - "cfg-if", - "crossbeam-epoch", - "crossbeam-utils", -] - -[[package]] -name = "crossbeam-epoch" -version = "0.9.10" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "045ebe27666471bb549370b4b0b3e51b07f56325befa4284db65fc89c02511b1" -dependencies = [ - "autocfg", - "cfg-if", - "crossbeam-utils", - "memoffset", - "once_cell", - "scopeguard", -] - -[[package]] -name = "crossbeam-utils" -version = "0.8.11" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "51887d4adc7b564537b15adcfb307936f8075dfcd5f00dde9a9f1d29383682bc" -dependencies = [ - "cfg-if", - "once_cell", -] - -[[package]] -name = "csv" -version = "1.1.6" +version = "1.1.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "22813a6dc45b335f9bade10bf7271dc477e81113e89eb251a0bc2a8a81c536e1" -dependencies = [ - "bstr", - "csv-core", - "itoa 0.4.8", - "ryu", - "serde", -] - -[[package]] -name = "csv-core" -version = "0.1.10" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "2b2466559f260f48ad25fe6317b3c8dac77b5bdb5763ac7d9d6103530663bc90" -dependencies = [ - "memchr", -] +checksum = "6548a0ad5d2549e111e1f6a11a6c2e2d00ce6a3dafe22948d67c2b443f775e52" [[package]] name = "databake" -version = "0.1.2" +version = "0.1.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c87777d6d7bde863ba217aa87521dc857239de1f36d66aac46fd173fb0495858" +checksum = "df626c4717e455cd7a70a82c4358630554a07e4341f86dd095c625f1474a2857" dependencies = [ "databake-derive", "proc-macro2", @@ -264,9 +49,9 @@ dependencies = [ [[package]] name = "databake-derive" -version = "0.1.1" +version = "0.1.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "905c7a060fc0c84c0452d97473b1177dd7a5cbc7670cfbae4a7fe22e42f6432e" +checksum = "be51a53c468489ae1ef0efa9f6b10706f426c0dde06d66122ffef1f0c51e87dc" dependencies = [ "proc-macro2", "quote", @@ -276,9 +61,9 @@ dependencies = [ [[package]] name = "deduplicating_array" -version = "0.1.2" +version = "0.1.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1e7f0807b2feeeda87369e8b4cf467f250f39841c8f9427bf3a972b878588937" +checksum = "135a278b07263e55438c15a3021b4947288f981ae387666f5015add8fbc76f5b" dependencies = [ "serde", ] @@ -295,28 +80,6 @@ dependencies = [ ] [[package]] -name = "either" -version = "1.8.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "90e5c1c8368803113bf0c9584fc495a58b86dc8a29edbf8fe877d21d9507e797" - -[[package]] -name = "embedded-hal" -version = "0.2.7" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "35949884794ad573cf46071e41c9b60efb0cb311e3ca01f7af807af1debc66ff" -dependencies = [ - "nb 0.1.3", - "void", -] - -[[package]] -name = "half" -version = "1.8.2" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "eabb4a44450da02c90444cf74558da904edde8fb4e9035a9a6a4e15445af0bd7" - -[[package]] name = "hash32" version = "0.2.1" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -333,51 +96,32 @@ checksum = "db04bc24a18b9ea980628ecf00e6c0264f3c1426dac36c00cb49b6fbad8b0743" dependencies = [ "atomic-polyfill", "hash32", - "rustc_version 0.4.0", + "rustc_version", "serde", "spin", "stable_deref_trait", ] [[package]] -name = "hermit-abi" -version = "0.1.19" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "62b467343b94ba476dcb2500d242dadbb39557df889310ac77c5d99100aaac33" -dependencies = [ - "libc", -] - -[[package]] -name = "icu_benchmark_macros" -version = "0.7.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6c867656f2d9c90b13709ac88e710a9d6afe33998c1dfa22384bab8804e8b3d4" - -[[package]] name = "icu_list" -version = "1.0.0" +version = "1.1.0" dependencies = [ - "criterion", "databake", "deduplicating_array", "displaydoc", - "icu_benchmark_macros", - "icu_locid", "icu_provider", "postcard", - "regex-automata 0.2.0", + "regex-automata", "serde", "serde_json", "writeable", - "zerovec", ] [[package]] name = "icu_locid" -version = "1.0.0" +version = "1.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "34b3de5d99a0e275fe6193b9586dbf37364daebc0d39c89b5cf8376a53b789e8" +checksum = "71d7a98ecb812760b5f077e55a4763edeefa7ccc30d6eb5680a70841ede81928" dependencies = [ "displaydoc", "litemap", @@ -387,9 +131,9 @@ dependencies = [ [[package]] name = "icu_provider" -version = "1.0.0" +version = "1.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e629bc2b6591ed9e4467d8a0fa2a597b70cff64ff8170e54a3f0f3257b99873f" +checksum = "a86816c97bc4e613086497f9479f63e120315e056763e8c4435604f98d21d82d" dependencies = [ "displaydoc", "icu_locid", @@ -404,9 +148,9 @@ dependencies = [ [[package]] name = "icu_provider_macros" -version = "1.0.0" +version = "1.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "38cf6f5b65cf81f0b4298da647101acbfe6ae0e25263f92bd7a22597e9d6d606" +checksum = "9ddb07844c2ffc4c28840e799e9e54ff054393cf090740decf25624e9d94b93a" dependencies = [ "proc-macro2", "quote", @@ -414,52 +158,16 @@ dependencies = [ ] [[package]] -name = "itertools" -version = "0.10.5" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "b0fd2260e829bddf4cb6ea802289de2f86d6a7a690192fbe91b3f46e0f2c8473" -dependencies = [ - "either", -] - -[[package]] -name = "itoa" -version = "0.4.8" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "b71991ff56294aa922b450139ee08b3bfc70982c6b2c7562771375cf73542dd4" - -[[package]] name = "itoa" -version = "1.0.3" +version = "1.0.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6c8af84674fe1f223a982c933a0ee1086ac4d4052aa0fb8060c12c6ad838e754" - -[[package]] -name = "js-sys" -version = "0.3.60" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "49409df3e3bf0856b916e2ceaca09ee28e6871cf7d9ce97a692cacfdb2a25a47" -dependencies = [ - "wasm-bindgen", -] - -[[package]] -name = "lazy_static" -version = "1.4.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e2abad23fbc42b3700f2f279844dc832adb2b2eb069b2df918f455c4e18cc646" - -[[package]] -name = "libc" -version = "0.2.133" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c0f80d65747a3e43d1596c7c5492d95d5edddaabd45a7fcdb02b95f644164966" +checksum = "fad582f4b9e86b6caa621cabeb0963332d92eea04729ab12892c2533951e6440" [[package]] name = "litemap" -version = "0.6.0" +version = "0.6.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f34a3f4798fac63fb48cf277eefa38f94d3443baff555bb98e4f56bc9092368e" +checksum = "575d8a551c59104b4df91269921e5eab561aa1b77c618dac0414b5d44a4617de" [[package]] name = "lock_api" @@ -472,104 +180,12 @@ dependencies = [ ] [[package]] -name = "log" -version = "0.4.17" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "abb12e687cfb44aa40f41fc3978ef76448f9b6038cad6aef4259d3c095a2382e" -dependencies = [ - "cfg-if", -] - -[[package]] name = "memchr" version = "2.5.0" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "2dffe52ecf27772e601905b7522cb4ef790d2cc203488bbd0e2fe85fcb74566d" [[package]] -name = "memoffset" -version = "0.6.5" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5aa361d4faea93603064a027415f07bd8e1d5c88c9fbf68bf56a285428fd79ce" -dependencies = [ - "autocfg", -] - -[[package]] -name = "nb" -version = "0.1.3" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "801d31da0513b6ec5214e9bf433a77966320625a37860f910be265be6e18d06f" -dependencies = [ - "nb 1.0.0", -] - -[[package]] -name = "nb" -version = "1.0.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "546c37ac5d9e56f55e73b677106873d9d9f5190605e41a856503623648488cae" - -[[package]] -name = "num-traits" -version = "0.2.15" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "578ede34cf02f8924ab9447f50c28075b4d3e5b269972345e7e0372b38c6cdcd" -dependencies = [ - "autocfg", -] - -[[package]] -name = "num_cpus" -version = "1.13.1" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "19e64526ebdee182341572e50e9ad03965aa510cd94427a4549448f285e957a1" -dependencies = [ - "hermit-abi", - "libc", -] - -[[package]] -name = "once_cell" -version = "1.15.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e82dad04139b71a90c080c8463fe0dc7902db5192d939bd0950f074d014339e1" - -[[package]] -name = "oorandom" -version = "11.1.3" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "0ab1bc2a289d34bd04a330323ac98a1b4bc82c9d9fcb1e66b63caa84da26b575" - -[[package]] -name = "plotters" -version = "0.3.4" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "2538b639e642295546c50fcd545198c9d64ee2a38620a628724a3b266d5fbf97" -dependencies = [ - "num-traits", - "plotters-backend", - "plotters-svg", - "wasm-bindgen", - "web-sys", -] - -[[package]] -name = "plotters-backend" -version = "0.3.4" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "193228616381fecdc1224c62e96946dfbc73ff4384fba576e052ff8c1bea8142" - -[[package]] -name = "plotters-svg" -version = "0.3.3" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f9a81d2759aae1dae668f783c308bc5c8ebd191ff4184aaa1b37f65a6ae5a56f" -dependencies = [ - "plotters-backend", -] - -[[package]] name = "postcard" version = "1.0.2" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -582,64 +198,23 @@ dependencies = [ [[package]] name = "proc-macro2" -version = "1.0.44" +version = "1.0.50" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "7bd7356a8122b6c4a24a82b278680c73357984ca2fc79a0f9fa6dea7dced7c58" +checksum = "6ef7d57beacfaf2d8aee5937dab7b7f28de3cb8b1828479bb5de2a7106f2bae2" dependencies = [ "unicode-ident", ] [[package]] name = "quote" -version = "1.0.21" +version = "1.0.23" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "bbe448f377a7d6961e30f5955f9b8d106c3f5e449d493ee1b125c1d43c2b5179" +checksum = "8856d8364d252a14d474036ea1358d63c9e6965c8e5c1885c18f73d70bff9c7b" dependencies = [ "proc-macro2", ] [[package]] -name = "rayon" -version = "1.5.3" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "bd99e5772ead8baa5215278c9b15bf92087709e9c1b2d1f97cdb5a183c933a7d" -dependencies = [ - "autocfg", - "crossbeam-deque", - "either", - "rayon-core", -] - -[[package]] -name = "rayon-core" -version = "1.9.3" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "258bcdb5ac6dad48491bb2992db6b7cf74878b0384908af124823d118c99683f" -dependencies = [ - "crossbeam-channel", - "crossbeam-deque", - "crossbeam-utils", - "num_cpus", -] - -[[package]] -name = "regex" -version = "1.6.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "4c4eb3267174b8c6c2f654116623910a0fef09c4753f8dd83db29c48a0df988b" -dependencies = [ - "aho-corasick", - "memchr", - "regex-syntax", -] - -[[package]] -name = "regex-automata" -version = "0.1.10" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6c230d73fb8d8c1b9c0b3135c5142a8acee3a0558fb8db5cf1cb65f8d7862132" - -[[package]] name = "regex-automata" version = "0.2.0" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -651,39 +226,9 @@ dependencies = [ [[package]] name = "regex-syntax" -version = "0.6.27" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a3f87b73ce11b1619a3c6332f45341e0047173771e8b8b73f87bfeefb7b56244" - -[[package]] -name = "riscv" -version = "0.7.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6907ccdd7a31012b70faf2af85cd9e5ba97657cc3987c4f13f8e4d2c2a088aba" -dependencies = [ - "bare-metal 1.0.0", - "bit_field", - "riscv-target", -] - -[[package]] -name = "riscv-target" -version = "0.1.2" +version = "0.6.28" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "88aa938cda42a0cf62a20cfe8d139ff1af20c2e681212b5b34adb5a58333f222" -dependencies = [ - "lazy_static", - "regex", -] - -[[package]] -name = "rustc_version" -version = "0.2.3" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "138e3e0acb6c9fb258b19b67cb8abd63c00679d2851805ea151465464fe9030a" -dependencies = [ - "semver 0.9.0", -] +checksum = "456c603be3e8d448b072f410900c09faf164fbce2d480456f50eea6e25f9c848" [[package]] name = "rustc_version" @@ -691,23 +236,14 @@ version = "0.4.0" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "bfa0f585226d2e68097d4f95d113b15b83a82e819ab25717ec0590d9584ef366" dependencies = [ - "semver 1.0.14", + "semver", ] [[package]] name = "ryu" -version = "1.0.11" +version = "1.0.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "4501abdff3ae82a1c1b477a17252eb69cee9e66eb915c1abaa4f44d873df9f09" - -[[package]] -name = "same-file" -version = "1.0.6" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "93fc1dc3aaa9bfed95e02e6eadabb4baf7e3078b0bd1b4d7b6b0b68378900502" -dependencies = [ - "winapi-util", -] +checksum = "7b4b9743ed687d4b4bcedf9ff5eaa7398495ae14e61cba0a295704edbc7decde" [[package]] name = "scopeguard" @@ -717,49 +253,24 @@ checksum = "d29ab0c6d3fc0ee92fe66e2d99f700eab17a8d57d1c1d3b748380fb20baa78cd" [[package]] name = "semver" -version = "0.9.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1d7eb9ef2c18661902cc47e535f9bc51b78acd254da71d375c2f6720d9a40403" -dependencies = [ - "semver-parser", -] - -[[package]] -name = "semver" -version = "1.0.14" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e25dfac463d778e353db5be2449d1cce89bd6fd23c9f1ea21310ce6e5a1b29c4" - -[[package]] -name = "semver-parser" -version = "0.7.0" +version = "1.0.16" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "388a1df253eca08550bef6c72392cfe7c30914bf41df5269b68cbd6ff8f570a3" +checksum = "58bc9567378fc7690d6b2addae4e60ac2eeea07becb2c64b9f218b53865cba2a" [[package]] name = "serde" -version = "1.0.145" +version = "1.0.152" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "728eb6351430bccb993660dfffc5a72f91ccc1295abaa8ce19b27ebe4f75568b" +checksum = "bb7d1f0d3021d347a83e556fc4683dea2ea09d87bccdf88ff5c12545d89d5efb" dependencies = [ "serde_derive", ] [[package]] -name = "serde_cbor" -version = "0.11.2" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "2bef2ebfde456fb76bbcf9f59315333decc4fda0b2b44b420243c11e0f5ec1f5" -dependencies = [ - "half", - "serde", -] - -[[package]] name = "serde_derive" -version = "1.0.145" +version = "1.0.152" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "81fa1584d3d1bcacd84c277a0dfe21f5b0f6accf4a23d04d4c6d61f1af522b4c" +checksum = "af487d118eecd09402d70a5d72551860e788df87b464af30e5ea6a38c75c541e" dependencies = [ "proc-macro2", "quote", @@ -768,11 +279,11 @@ dependencies = [ [[package]] name = "serde_json" -version = "1.0.85" +version = "1.0.91" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e55a28e3aaef9d5ce0506d0a14dbba8054ddc7e499ef522dd8b26859ec9d4a44" +checksum = "877c235533714907a8c2464236f5c4b2a17262ef1bd71f38f35ea592c8da6883" dependencies = [ - "itoa 1.0.3", + "itoa", "ryu", "serde", ] @@ -794,9 +305,9 @@ checksum = "a8f112729512f8e442d81f95a8a7ddf2b7c6b8a1a6f509a95864142b30cab2d3" [[package]] name = "syn" -version = "1.0.101" +version = "1.0.107" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e90cde112c4b9690b8cbe810cba9ddd8bc1d7472e2cae317b69e9438c1cba7d2" +checksum = "1f4064b5b16e03ae50984a5a8ed5d4f8803e6bc1fd170a3cda91a1be4b18e3f5" dependencies = [ "proc-macro2", "quote", @@ -816,44 +327,19 @@ dependencies = [ ] [[package]] -name = "textwrap" -version = "0.11.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d326610f408c7a4eb6f51c37c330e496b08506c9457c9d34287ecc38809fb060" -dependencies = [ - "unicode-width", -] - -[[package]] name = "tinystr" -version = "0.7.0" +version = "0.7.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f8aeafdfd935e4a7fe16a91ab711fa52d54df84f9c8f7ca5837a9d1d902ef4c2" +checksum = "7ac3f5b6856e931e15e07b478e98c8045239829a65f9156d4fa7e7788197a5ef" dependencies = [ "displaydoc", ] [[package]] -name = "tinytemplate" -version = "1.2.1" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "be4d6b5f19ff7664e8c98d03e2139cb510db9b0a60b55f8e8709b689d939b6bc" -dependencies = [ - "serde", - "serde_json", -] - -[[package]] name = "unicode-ident" -version = "1.0.4" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "dcc811dc4066ac62f84f11307873c4850cb653bfa9b1719cee2bd2204a4bc5dd" - -[[package]] -name = "unicode-width" -version = "0.1.10" +version = "1.0.6" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c0edd1e5b14653f783770bce4a4dabb4a5108a5370a5f5d8cfe8710c361f6c8b" +checksum = "84a22b9f218b40614adcb3f4ff08b703773ad44fa9423e4e0d346d5db86e4ebc" [[package]] name = "unicode-xid" @@ -862,143 +348,16 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "f962df74c8c05a667b5ee8bcf162993134c104e96440b663c8daa176dc772d8c" [[package]] -name = "vcell" -version = "0.1.3" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "77439c1b53d2303b20d9459b1ade71a83c716e3f9c34f3228c00e6f185d6c002" - -[[package]] -name = "void" -version = "1.0.2" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6a02e4885ed3bc0f2de90ea6dd45ebcbb66dacffe03547fadbb0eeae2770887d" - -[[package]] -name = "volatile-register" -version = "0.2.1" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "9ee8f19f9d74293faf70901bc20ad067dc1ad390d2cbf1e3f75f721ffee908b6" -dependencies = [ - "vcell", -] - -[[package]] -name = "walkdir" -version = "2.3.2" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "808cf2735cd4b6866113f648b791c6adc5714537bc222d9347bb203386ffda56" -dependencies = [ - "same-file", - "winapi", - "winapi-util", -] - -[[package]] -name = "wasm-bindgen" -version = "0.2.83" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "eaf9f5aceeec8be17c128b2e93e031fb8a4d469bb9c4ae2d7dc1888b26887268" -dependencies = [ - "cfg-if", - "wasm-bindgen-macro", -] - -[[package]] -name = "wasm-bindgen-backend" -version = "0.2.83" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "4c8ffb332579b0557b52d268b91feab8df3615f265d5270fec2a8c95b17c1142" -dependencies = [ - "bumpalo", - "log", - "once_cell", - "proc-macro2", - "quote", - "syn", - "wasm-bindgen-shared", -] - -[[package]] -name = "wasm-bindgen-macro" -version = "0.2.83" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "052be0f94026e6cbc75cdefc9bae13fd6052cdcaf532fa6c45e7ae33a1e6c810" -dependencies = [ - "quote", - "wasm-bindgen-macro-support", -] - -[[package]] -name = "wasm-bindgen-macro-support" -version = "0.2.83" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "07bc0c051dc5f23e307b13285f9d75df86bfdf816c5721e573dec1f9b8aa193c" -dependencies = [ - "proc-macro2", - "quote", - "syn", - "wasm-bindgen-backend", - "wasm-bindgen-shared", -] - -[[package]] -name = "wasm-bindgen-shared" -version = "0.2.83" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1c38c045535d93ec4f0b4defec448e4291638ee608530863b1e2ba115d4fff7f" - -[[package]] -name = "web-sys" -version = "0.3.60" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "bcda906d8be16e728fd5adc5b729afad4e444e106ab28cd1c7256e54fa61510f" -dependencies = [ - "js-sys", - "wasm-bindgen", -] - -[[package]] -name = "winapi" -version = "0.3.9" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5c839a674fcd7a98952e593242ea400abe93992746761e38641405d28b00f419" -dependencies = [ - "winapi-i686-pc-windows-gnu", - "winapi-x86_64-pc-windows-gnu", -] - -[[package]] -name = "winapi-i686-pc-windows-gnu" -version = "0.4.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6" - -[[package]] -name = "winapi-util" -version = "0.1.5" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "70ec6ce85bb158151cae5e5c87f95a8e97d2c0c4b001223f33a334e3ce5de178" -dependencies = [ - "winapi", -] - -[[package]] -name = "winapi-x86_64-pc-windows-gnu" -version = "0.4.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f" - -[[package]] name = "writeable" -version = "0.5.0" +version = "0.5.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f8e6ab4f5da1b24daf2c590cfac801bacb27b15b4f050e84eb60149ea726f06b" +checksum = "92d74a687e3b9a7a129db0a8c82b4d464eb9c36f5a66ca68572a7e5f1cfdb5bc" [[package]] name = "yoke" -version = "0.6.2" +version = "0.7.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1fe1d55ca72c32d573bfbd5cb2f0ca65a497854c44762957a6d3da96041a5184" +checksum = "222180af14a6b54ef2c33493c1eff77ae95a3687a21b243e752624006fb8f26e" dependencies = [ "serde", "stable_deref_trait", @@ -1008,9 +367,9 @@ dependencies = [ [[package]] name = "yoke-derive" -version = "0.6.0" +version = "0.7.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "58c2c5bb7c929b85c1b9ec69091b0d835f0878b4fd9eb67973b25936e06c4374" +checksum = "ca800d73d6b7a7ee54f2608205c98b549fca71c9500c1abcb3abdc7708b4a8cb" dependencies = [ "proc-macro2", "quote", @@ -1029,9 +388,9 @@ dependencies = [ [[package]] name = "zerofrom-derive" -version = "0.1.0" +version = "0.1.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "8785f47d6062c1932866147f91297286a9f350b3070e9d9f0b6078e37d623c1a" +checksum = "2e8aa86add9ddbd2409c1ed01e033cd457d79b1b1229b64922c25095c595e829" dependencies = [ "proc-macro2", "quote", @@ -1041,21 +400,19 @@ dependencies = [ [[package]] name = "zerovec" -version = "0.9.0" +version = "0.9.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "b9d919a74c17749ccb17beaf6405562e413cd94e98ba52ca1e64bbe7eefbd8b8" +checksum = "154df60c74c4a844bc04a53cef4fc18a909d3ea07e19f5225eaba86209da3aa6" dependencies = [ - "serde", - "yoke", "zerofrom", "zerovec-derive", ] [[package]] name = "zerovec-derive" -version = "0.9.0" +version = "0.9.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "490e5f878c2856225e884c35927e7ea6db3c24cdb7229b72542c7526ad7ed49e" +checksum = "c630983d26a5f0c061dad3bf22df69a7329b4939a9752bc5f19f1cbd8e2263db" dependencies = [ "proc-macro2", "quote", diff --git a/vendor/icu_list/Cargo.toml b/vendor/icu_list/Cargo.toml index 805f2aea4..2d3f3feb1 100644 --- a/vendor/icu_list/Cargo.toml +++ b/vendor/icu_list/Cargo.toml @@ -12,7 +12,7 @@ [package] edition = "2018" name = "icu_list" -version = "1.0.0" +version = "1.1.0" authors = ["The ICU4X Project Developers"] include = [ "src/**/*", @@ -30,30 +30,30 @@ license = "Unicode-DFS-2016" repository = "https://github.com/unicode-org/icu4x" resolver = "2" -[lib] -path = "src/lib.rs" +[package.metadata.docs.rs] +all-features = true + +[package.metadata.cargo-all-features] +denylist = ["bench"] [[example]] name = "and_list" [dependencies.databake] -version = "0.1.0" +version = "0.1.3" features = ["derive"] optional = true [dependencies.deduplicating_array] -version = "0.1" +version = "0.1.3" optional = true [dependencies.displaydoc] version = "0.2.3" default-features = false -[dependencies.icu_locid] -version = "1.0.0" - [dependencies.icu_provider] -version = "1.0.0" +version = "1.1.0" features = ["macros"] [dependencies.regex-automata] @@ -70,17 +70,7 @@ optional = true default-features = false [dependencies.writeable] -version = "0.5" - -[dependencies.zerovec] -version = "0.9" -features = ["yoke"] - -[dev-dependencies.criterion] -version = "0.3.3" - -[dev-dependencies.icu_benchmark_macros] -version = "0.7" +version = "0.5.1" [dev-dependencies.postcard] version = "1.0.0" @@ -94,21 +84,16 @@ bench = [] datagen = [ "serde", "std", - "databake", + "dep:databake", + "regex-automata/alloc", ] serde = [ + "dep:deduplicating_array", "dep:serde", "icu_provider/serde", - "zerovec/serde", - "deduplicating_array", ] serde_human = [ "serde", "regex-automata/alloc", ] -std = [ - "icu_provider/std", - "icu_locid/std", - "regex-automata/std", - "regex-automata/alloc", -] +std = ["icu_provider/std"] diff --git a/vendor/icu_list/examples/and_list.rs b/vendor/icu_list/examples/and_list.rs index 9d869d9fb..08cfa3bda 100644 --- a/vendor/icu_list/examples/and_list.rs +++ b/vendor/icu_list/examples/and_list.rs @@ -6,8 +6,8 @@ icu_benchmark_macros::static_setup!(); -use icu_list::{ListFormatter, ListLength}; -use icu_locid::locale; +use icu::list::{ListFormatter, ListLength}; +use icu::locid::locale; #[no_mangle] fn main(_argc: isize, _argv: *const *const u8) -> isize { diff --git a/vendor/icu_list/src/lazy_automaton.rs b/vendor/icu_list/src/lazy_automaton.rs new file mode 100644 index 000000000..3431b3c9d --- /dev/null +++ b/vendor/icu_list/src/lazy_automaton.rs @@ -0,0 +1,79 @@ +// This file is part of ICU4X. For terms of use, please see the file +// called LICENSE at the top level of the ICU4X source tree +// (online at: https://github.com/unicode-org/icu4x/blob/main/LICENSE ). + +use regex_automata::dfa::sparse::DFA; +use regex_automata::dfa::Automaton; +use regex_automata::util::id::StateID; +use writeable::Writeable; + +pub trait LazyAutomaton: Automaton { + // Like Automaton::find_earliest_fwd, but doesn't require a materialized string. + fn matches_earliest_fwd_lazy<S: Writeable + ?Sized>(&self, haystack: &S) -> bool; +} + +impl<T: AsRef<[u8]>> LazyAutomaton for DFA<T> { + fn matches_earliest_fwd_lazy<S: Writeable + ?Sized>(&self, haystack: &S) -> bool { + struct DFAStepper<'a> { + dfa: &'a DFA<&'a [u8]>, + state: StateID, + } + + impl core::fmt::Write for DFAStepper<'_> { + fn write_str(&mut self, s: &str) -> core::fmt::Result { + for &byte in s.as_bytes() { + self.state = self.dfa.next_state(self.state, byte); + if self.dfa.is_match_state(self.state) || self.dfa.is_dead_state(self.state) { + // We matched or are in a no-match-cycle, return early + return Err(core::fmt::Error); + } + } + Ok(()) + } + } + + let mut stepper = DFAStepper { + // If start == 0 the start state does not depend on the actual string, so + // we can just pass an empty slice. + state: self.start_state_forward(None, &[], 0, 0), + dfa: &self.as_ref(), + }; + + if haystack.write_to(&mut stepper).is_ok() { + stepper.state = self.next_eoi_state(stepper.state); + } + + self.is_match_state(stepper.state) + } +} + +#[cfg(test)] +#[test] +fn test() { + use crate::provider::SerdeDFA; + use alloc::borrow::Cow; + + let matcher = SerdeDFA::new(Cow::Borrowed("11(000)*$")).unwrap(); + + for writeable in [1i32, 11, 110, 11000, 211000] { + assert_eq!( + matcher + .deref() + .find_earliest_fwd(writeable.write_to_string().as_bytes()) + .unwrap() + .is_some(), + matcher.deref().matches_earliest_fwd_lazy(&writeable) + ); + } + + struct ExitEarlyTest; + + impl writeable::Writeable for ExitEarlyTest { + fn write_to<W: core::fmt::Write + ?Sized>(&self, sink: &mut W) -> core::fmt::Result { + sink.write_str("12")?; + unreachable!() + } + } + + assert!(!matcher.deref().matches_earliest_fwd_lazy(&ExitEarlyTest)); +} diff --git a/vendor/icu_list/src/lib.rs b/vendor/icu_list/src/lib.rs index 18f2156a6..61aec0fa3 100644 --- a/vendor/icu_list/src/lib.rs +++ b/vendor/icu_list/src/lib.rs @@ -93,8 +93,9 @@ extern crate alloc; mod error; +mod lazy_automaton; mod list_formatter; -mod string_matcher; +mod patterns; pub mod provider; diff --git a/vendor/icu_list/src/list_formatter.rs b/vendor/icu_list/src/list_formatter.rs index 36f5fbb7b..93f035eab 100644 --- a/vendor/icu_list/src/list_formatter.rs +++ b/vendor/icu_list/src/list_formatter.rs @@ -72,8 +72,39 @@ impl ListFormatter { ); /// Returns a [`Writeable`] composed of the input [`Writeable`]s and the language-dependent - /// formatting. The first layer of parts contains [`parts::ELEMENT`] for input - /// elements, and [`parts::LITERAL`] for list literals. + /// formatting. + /// + /// The [`Writeable`] is annotated with [`parts::ELEMENT`] for input elements, + /// and [`parts::LITERAL`] for list literals. + /// + /// # Example + /// + /// ``` + /// use icu::list::*; + /// # use icu::locid::locale; + /// # use writeable::*; + /// let formatteur = ListFormatter::try_new_and_with_length_unstable( + /// &icu_testdata::unstable(), + /// &locale!("fr").into(), + /// ListLength::Wide, + /// ) + /// .unwrap(); + /// let pays = ["Italie", "France", "Espagne", "Allemagne"]; + /// + /// assert_writeable_parts_eq!( + /// formatteur.format(pays.iter()), + /// "Italie, France, Espagne et Allemagne", + /// [ + /// (0, 6, parts::ELEMENT), + /// (6, 8, parts::LITERAL), + /// (8, 14, parts::ELEMENT), + /// (14, 16, parts::LITERAL), + /// (16, 23, parts::ELEMENT), + /// (23, 27, parts::LITERAL), + /// (27, 36, parts::ELEMENT), + /// ] + /// ); + /// ``` pub fn format<'a, W: Writeable + 'a, I: Iterator<Item = W> + Clone + 'a>( &'a self, values: I, @@ -99,6 +130,9 @@ pub mod parts { use writeable::Part; /// The [`Part`] used by [`FormattedList`](super::FormattedList) to mark the part of the string that is an element. + /// + /// * `category`: `"list"` + /// * `value`: `"element"` pub const ELEMENT: Part = Part { category: "list", value: "element", @@ -106,6 +140,9 @@ pub mod parts { /// The [`Part`] used by [`FormattedList`](super::FormattedList) to mark the part of the string that is a list literal, /// such as ", " or " and ". + /// + /// * `category`: `"list"` + /// * `value`: `"literal"` pub const LITERAL: Part = Part { category: "list", value: "literal", @@ -234,7 +271,7 @@ mod tests { fn formatter(length: ListLength) -> ListFormatter { ListFormatter { - data: DataPayload::from_owned(crate::provider::test::test_patterns()), + data: DataPayload::from_owned(crate::patterns::test::test_patterns()), length, } } diff --git a/vendor/icu_list/src/patterns.rs b/vendor/icu_list/src/patterns.rs new file mode 100644 index 000000000..8cfcb98c1 --- /dev/null +++ b/vendor/icu_list/src/patterns.rs @@ -0,0 +1,283 @@ +// This file is part of ICU4X. For terms of use, please see the file +// called LICENSE at the top level of the ICU4X source tree +// (online at: https://github.com/unicode-org/icu4x/blob/main/LICENSE ). + +use crate::lazy_automaton::LazyAutomaton; +use crate::provider::*; +use crate::ListLength; +#[cfg(feature = "datagen")] +use alloc::borrow::Cow; +#[cfg(feature = "datagen")] +use icu_provider::DataError; +use writeable::{LengthHint, Writeable}; + +impl<'data> ListFormatterPatternsV1<'data> { + /// Creates a new [`ListFormatterPatternsV1`] from the given patterns. Fails if any pattern is invalid. + /// + /// See [`ListJoinerPattern::from_str`]. `allow_prefix` will be true for `pair` and `end` patterns, + /// `allow_suffix` for `start` and `pair` patterns. + #[cfg(feature = "datagen")] + pub fn try_new( + [start, middle, end, pair, short_start, short_middle, short_end, short_pair, narrow_start, narrow_middle, narrow_end, narrow_pair]: [&str; 12], + ) -> Result<Self, DataError> { + Ok(Self([ + ListJoinerPattern::from_str(start, true, false)?.into(), + ListJoinerPattern::from_str(middle, false, false)?.into(), + ListJoinerPattern::from_str(end, false, true)?.into(), + ListJoinerPattern::from_str(pair, true, true)?.into(), + ListJoinerPattern::from_str(short_start, true, false)?.into(), + ListJoinerPattern::from_str(short_middle, false, false)?.into(), + ListJoinerPattern::from_str(short_end, false, true)?.into(), + ListJoinerPattern::from_str(short_pair, true, true)?.into(), + ListJoinerPattern::from_str(narrow_start, true, false)?.into(), + ListJoinerPattern::from_str(narrow_middle, false, false)?.into(), + ListJoinerPattern::from_str(narrow_end, false, true)?.into(), + ListJoinerPattern::from_str(narrow_pair, true, true)?.into(), + ])) + } + + /// Adds a special case to all `pattern`s that will evaluate to + /// `alternative_pattern` when `regex` matches the following element. + /// The regex is interpreted case-insensitive and anchored to the beginning, but + /// to improve efficiency does not search for full matches. If a full match is + /// required, use `$`. + #[cfg(feature = "datagen")] + pub fn make_conditional( + &mut self, + pattern: &str, + regex: &SerdeDFA<'static>, + alternative_pattern: &str, + ) -> Result<(), DataError> { + let old = ListJoinerPattern::from_str(pattern, true, true)?; + for i in 0..12 { + #[allow(clippy::indexing_slicing)] // self.0 is &[_; 12] + if self.0[i].default == old { + self.0[i].special_case = Some(SpecialCasePattern { + condition: regex.clone(), + pattern: ListJoinerPattern::from_str( + alternative_pattern, + i % 4 == 0 || i % 4 == 3, // allow_prefix = start or pair + i % 4 == 2 || i % 4 == 3, // allow_suffix = end or pair + )?, + }); + } + } + Ok(()) + } + + /// The range of the number of bytes required by the list literals to join a + /// list of length `len`. If none of the patterns are conditional, this is exact. + pub(crate) fn size_hint(&self, style: ListLength, len: usize) -> LengthHint { + match len { + 0 | 1 => LengthHint::exact(0), + 2 => self.pair(style).size_hint(), + n => { + self.start(style).size_hint() + + self.middle(style).size_hint() * (n - 3) + + self.end(style).size_hint() + } + } + } +} + +type PatternParts<'a> = (&'a str, &'a str, &'a str); + +impl<'a> ConditionalListJoinerPattern<'a> { + pub(crate) fn parts<'b, W: Writeable + ?Sized>( + &'a self, + following_value: &'b W, + ) -> PatternParts<'a> { + match &self.special_case { + Some(SpecialCasePattern { condition, pattern }) + if condition.deref().matches_earliest_fwd_lazy(following_value) => + { + pattern.borrow_tuple() + } + _ => self.default.borrow_tuple(), + } + } + + /// The expected length of this pattern + fn size_hint(&'a self) -> LengthHint { + let mut hint = self.default.size_hint(); + if let Some(special_case) = &self.special_case { + hint |= special_case.pattern.size_hint() + } + hint + } +} + +impl<'data> ListJoinerPattern<'data> { + /// Construct the pattern from a CLDR pattern string + #[cfg(feature = "datagen")] + pub fn from_str( + pattern: &str, + allow_prefix: bool, + allow_suffix: bool, + ) -> Result<Self, DataError> { + match (pattern.find("{0}"), pattern.find("{1}")) { + (Some(index_0), Some(index_1)) + if index_0 < index_1 + && (allow_prefix || index_0 == 0) + && (allow_suffix || index_1 == pattern.len() - 3) => + { + if (index_0 > 0 && !cfg!(test)) || index_1 - 3 >= 256 { + return Err(DataError::custom( + "Found valid pattern that cannot be stored in ListFormatterPatternsV1", + ) + .with_debug_context(pattern)); + } + #[allow(clippy::indexing_slicing)] // find + Ok(ListJoinerPattern { + string: Cow::Owned(alloc::format!( + "{}{}{}", + &pattern[0..index_0], + &pattern[index_0 + 3..index_1], + &pattern[index_1 + 3..] + )), + index_0: index_0 as u8, + index_1: (index_1 - 3) as u8, + }) + } + _ => Err(DataError::custom("Invalid list pattern").with_debug_context(pattern)), + } + } + + fn borrow_tuple(&'data self) -> PatternParts<'data> { + #![allow(clippy::indexing_slicing)] // by invariant + let index_0 = self.index_0 as usize; + let index_1 = self.index_1 as usize; + ( + &self.string[0..index_0], + &self.string[index_0..index_1], + &self.string[index_1..], + ) + } + + fn size_hint(&self) -> LengthHint { + LengthHint::exact(self.string.len()) + } +} + +#[cfg(feature = "datagen")] +impl<'data> From<ListJoinerPattern<'data>> for ConditionalListJoinerPattern<'data> { + fn from(default: ListJoinerPattern<'data>) -> Self { + Self { + default, + special_case: None, + } + } +} + +#[cfg(all(test, feature = "datagen"))] +pub mod test { + use super::*; + + pub fn test_patterns() -> ListFormatterPatternsV1<'static> { + let mut patterns = ListFormatterPatternsV1::try_new([ + // Wide: general + "@{0}:{1}", + "{0},{1}", + "{0}.{1}!", + "${0};{1}+", + // Short: different pattern lengths + "{0}1{1}", + "{0}12{1}", + "{0}12{1}34", + "{0}123{1}456", + // Narrow: conditionals + "{0}: {1}", + "{0}, {1}", + "{0}. {1}", + "{0}. {1}", + ]) + .unwrap(); + patterns + .make_conditional( + "{0}. {1}", + &SerdeDFA::new(Cow::Borrowed("A")).unwrap(), + "{0} :o {1}", + ) + .unwrap(); + patterns + } + + #[test] + fn rejects_bad_patterns() { + assert!(ListJoinerPattern::from_str("{0} and", true, true).is_err()); + assert!(ListJoinerPattern::from_str("and {1}", true, true).is_err()); + assert!(ListJoinerPattern::from_str("{1} and {0}", true, true).is_err()); + assert!(ListJoinerPattern::from_str("{1{0}}", true, true).is_err()); + assert!(ListJoinerPattern::from_str("{0\u{202e}} and {1}", true, true).is_err()); + assert!(ListJoinerPattern::from_str("{{0}} {{1}}", true, true).is_ok()); + + assert!(ListJoinerPattern::from_str("{0} and {1} ", true, true).is_ok()); + assert!(ListJoinerPattern::from_str("{0} and {1} ", true, false).is_err()); + assert!(ListJoinerPattern::from_str(" {0} and {1}", true, true).is_ok()); + assert!(ListJoinerPattern::from_str(" {0} and {1}", false, true).is_err()); + } + + #[test] + fn produces_correct_parts() { + assert_eq!( + test_patterns().pair(ListLength::Wide).parts(""), + ("$", ";", "+") + ); + } + + #[test] + fn produces_correct_parts_conditionally() { + assert_eq!( + test_patterns().end(ListLength::Narrow).parts("A"), + ("", " :o ", "") + ); + assert_eq!( + test_patterns().end(ListLength::Narrow).parts("a"), + ("", " :o ", "") + ); + assert_eq!( + test_patterns().end(ListLength::Narrow).parts("ab"), + ("", " :o ", "") + ); + assert_eq!( + test_patterns().end(ListLength::Narrow).parts("B"), + ("", ". ", "") + ); + assert_eq!( + test_patterns().end(ListLength::Narrow).parts("BA"), + ("", ". ", "") + ); + } + + #[test] + fn size_hint_works() { + let pattern = test_patterns(); + + assert_eq!( + pattern.size_hint(ListLength::Short, 0), + LengthHint::exact(0) + ); + assert_eq!( + pattern.size_hint(ListLength::Short, 1), + LengthHint::exact(0) + ); + + // pair pattern "{0}123{1}456" + assert_eq!( + pattern.size_hint(ListLength::Short, 2), + LengthHint::exact(6) + ); + + // patterns "{0}1{1}", "{0}12{1}" (x197), and "{0}12{1}34" + assert_eq!( + pattern.size_hint(ListLength::Short, 200), + LengthHint::exact(1 + 2 * 197 + 4) + ); + + // patterns "{0}: {1}", "{0}, {1}" (x197), and "{0} :o {1}" or "{0}. {1}" + assert_eq!( + pattern.size_hint(ListLength::Narrow, 200), + LengthHint::exact(2 + 197 * 2) + LengthHint::between(2, 4) + ); + } +} diff --git a/vendor/icu_list/src/provider.rs b/vendor/icu_list/src/provider.rs deleted file mode 100644 index 27f3e4fec..000000000 --- a/vendor/icu_list/src/provider.rs +++ /dev/null @@ -1,465 +0,0 @@ -// This file is part of ICU4X. For terms of use, please see the file -// called LICENSE at the top level of the ICU4X source tree -// (online at: https://github.com/unicode-org/icu4x/blob/main/LICENSE ). - -// Provider structs must be stable -#![allow(clippy::exhaustive_structs, clippy::exhaustive_enums)] - -//! Data provider struct definitions for this ICU4X component. -//! -//! Read more about data providers: [`icu_provider`] - -use crate::ListLength; -use alloc::borrow::Cow; -use icu_provider::DataMarker; -use icu_provider::{yoke, zerofrom}; -use writeable::{LengthHint, Writeable}; - -pub use crate::string_matcher::StringMatcher; - -/// Symbols and metadata required for [`ListFormatter`](crate::ListFormatter). -#[icu_provider::data_struct( - AndListV1Marker = "list/and@1", - OrListV1Marker = "list/or@1", - UnitListV1Marker = "list/unit@1" -)] -#[derive(Clone, Debug)] -#[cfg_attr( - feature = "datagen", - derive(serde::Serialize, databake::Bake), - databake(path = icu_list::provider), -)] -pub struct ListFormatterPatternsV1<'data>( - #[cfg_attr(feature = "datagen", serde(with = "deduplicating_array"))] - /// The patterns in the order start, middle, end, pair, short_start, short_middle, - /// short_end, short_pair, narrow_start, narrow_middle, narrow_end, narrow_pair, - pub [ConditionalListJoinerPattern<'data>; 12], -); - -#[cfg(feature = "serde")] -impl<'de> serde::Deserialize<'de> for ListFormatterPatternsV1<'de> { - fn deserialize<D>(deserializer: D) -> Result<Self, D::Error> - where - D: serde::de::Deserializer<'de>, - { - #[cfg(not(feature = "serde_human"))] - if deserializer.is_human_readable() { - use serde::de::Error; - return Err(D::Error::custom( - "Deserializing human-readable ListFormatter data requires the 'serde_human' feature", - )); - } - - Ok(ListFormatterPatternsV1(deduplicating_array::deserialize( - deserializer, - )?)) - } -} - -pub(crate) struct ErasedListV1Marker; - -impl DataMarker for ErasedListV1Marker { - type Yokeable = ListFormatterPatternsV1<'static>; -} - -impl<'data> ListFormatterPatternsV1<'data> { - pub(crate) fn start(&self, style: ListLength) -> &ConditionalListJoinerPattern<'data> { - #![allow(clippy::indexing_slicing)] // style as usize < 3 - &self.0[4 * (style as usize)] - } - - pub(crate) fn middle(&self, style: ListLength) -> &ConditionalListJoinerPattern<'data> { - #![allow(clippy::indexing_slicing)] // style as usize < 3 - &self.0[4 * (style as usize) + 1] - } - - pub(crate) fn end(&self, style: ListLength) -> &ConditionalListJoinerPattern<'data> { - #![allow(clippy::indexing_slicing)] // style as usize < 3 - &self.0[4 * (style as usize) + 2] - } - - pub(crate) fn pair(&self, style: ListLength) -> &ConditionalListJoinerPattern<'data> { - #![allow(clippy::indexing_slicing)] // style as usize < 3 - &self.0[4 * (style as usize) + 3] - } - - /// The range of the number of bytes required by the list literals to join a - /// list of length `len`. If none of the patterns are conditional, this is exact. - pub(crate) fn size_hint(&self, style: ListLength, len: usize) -> LengthHint { - match len { - 0 | 1 => LengthHint::exact(0), - 2 => self.pair(style).size_hint(), - n => { - self.start(style).size_hint() - + self.middle(style).size_hint() * (n - 3) - + self.end(style).size_hint() - } - } - } -} - -/// A pattern that can behave conditionally on the next element. -#[derive(Clone, Debug, PartialEq, yoke::Yokeable, zerofrom::ZeroFrom)] -#[cfg_attr( - feature = "datagen", - derive(serde::Serialize, databake::Bake), - databake(path = icu_list::provider), -)] -#[cfg_attr(feature = "serde", derive(serde::Deserialize))] -pub struct ConditionalListJoinerPattern<'data> { - /// The default pattern - #[cfg_attr(feature = "serde", serde(borrow))] - pub default: ListJoinerPattern<'data>, - /// And optional special case - #[cfg_attr(feature = "serde", serde(borrow))] - pub special_case: Option<SpecialCasePattern<'data>>, -} - -/// The special case of a [`ConditionalListJoinerPattern`] -#[derive(Clone, Debug, PartialEq, yoke::Yokeable, zerofrom::ZeroFrom)] -#[cfg_attr( - feature = "datagen", - derive(serde::Serialize, databake::Bake), - databake(path = icu_list::provider), -)] -#[cfg_attr(feature = "serde", derive(serde::Deserialize))] -pub struct SpecialCasePattern<'data> { - /// The condition on the following element - #[cfg_attr(feature = "serde", serde(borrow))] - pub condition: StringMatcher<'data>, - /// The pattern if the condition matches - #[cfg_attr(feature = "serde", serde(borrow))] - pub pattern: ListJoinerPattern<'data>, -} - -/// A pattern containing two numeric placeholders ("{0}, and {1}.") -#[derive(Clone, Debug, PartialEq, yoke::Yokeable, zerofrom::ZeroFrom)] -#[cfg_attr(feature = "datagen", derive(serde::Serialize))] -pub struct ListJoinerPattern<'data> { - /// The pattern string without the placeholders - string: Cow<'data, str>, - /// The index of the first placeholder. Always <= index_1. - // Always 0 for CLDR data, so we don't need to serialize it. - // In-memory we have free space for it as index_1 doesn't - // fill a word. - #[cfg_attr(feature = "datagen", serde(skip))] - index_0: u8, - /// The index of the second placeholder. Always < string.len(). - index_1: u8, -} - -#[cfg(feature = "serde")] -impl<'de: 'data, 'data> serde::Deserialize<'de> for ListJoinerPattern<'data> { - fn deserialize<D>(deserializer: D) -> Result<Self, D::Error> - where - D: serde::Deserializer<'de>, - { - #[derive(serde::Deserialize)] - struct Dummy<'data> { - #[cfg_attr(feature = "serde", serde(borrow))] - string: Cow<'data, str>, - index_1: u8, - } - let Dummy { string, index_1 } = Dummy::deserialize(deserializer)?; - - if index_1 as usize > string.len() { - use serde::de::Error; - Err(D::Error::custom("invalid index_1")) - } else { - Ok(ListJoinerPattern { - string, - index_0: 0, - index_1, - }) - } - } -} - -impl<'a> ListJoinerPattern<'a> { - /// Constructs a [`ListJoinerPattern`] from raw parts. Used by databake. - /// - /// # Safety - /// index_1 may be at most string.len() - pub const unsafe fn from_parts_unchecked(string: &'a str, index_1: u8) -> Self { - Self { - string: Cow::Borrowed(string), - index_0: 0, - index_1, - } - } -} - -pub(crate) type PatternParts<'a> = (&'a str, &'a str, &'a str); - -impl<'a> ConditionalListJoinerPattern<'a> { - pub(crate) fn parts<'b, W: Writeable + ?Sized>( - &'a self, - following_value: &'b W, - ) -> PatternParts<'a> { - match &self.special_case { - Some(SpecialCasePattern { condition, pattern }) - // TODO: Implement lookahead instead of materializing here. - if condition.test(&*following_value.write_to_string()) => - { - pattern.borrow_tuple() - } - _ => self.default.borrow_tuple(), - } - } - - /// The expected length of this pattern - pub fn size_hint(&'a self) -> LengthHint { - let mut hint = self.default.size_hint(); - if let Some(special_case) = &self.special_case { - hint |= special_case.pattern.size_hint() - } - hint - } -} - -impl<'data> ListJoinerPattern<'data> { - fn borrow_tuple(&'data self) -> PatternParts<'data> { - #![allow(clippy::indexing_slicing)] // by invariant - let index_0 = self.index_0 as usize; - let index_1 = self.index_1 as usize; - ( - &self.string[0..index_0], - &self.string[index_0..index_1], - &self.string[index_1..], - ) - } - - fn size_hint(&self) -> LengthHint { - LengthHint::exact(self.string.len()) - } -} - -#[cfg(feature = "datagen")] -mod datagen { - #![allow(clippy::indexing_slicing)] // datagen - - use super::*; - use icu_provider::DataError; - - impl<'data> ListFormatterPatternsV1<'data> { - /// The patterns in the order start, middle, end, pair, short_start, short_middle, - /// short_end, short_pair, narrow_start, narrow_middle, narrow_end, narrow_pair, - pub fn try_new(patterns: [&str; 12]) -> Result<Self, DataError> { - Ok(Self([ - ListJoinerPattern::from_str(patterns[0], true, false)?.into(), - ListJoinerPattern::from_str(patterns[1], false, false)?.into(), - ListJoinerPattern::from_str(patterns[2], false, true)?.into(), - ListJoinerPattern::from_str(patterns[3], true, true)?.into(), - ListJoinerPattern::from_str(patterns[4], true, false)?.into(), - ListJoinerPattern::from_str(patterns[5], false, false)?.into(), - ListJoinerPattern::from_str(patterns[6], false, true)?.into(), - ListJoinerPattern::from_str(patterns[7], true, true)?.into(), - ListJoinerPattern::from_str(patterns[8], true, false)?.into(), - ListJoinerPattern::from_str(patterns[9], false, false)?.into(), - ListJoinerPattern::from_str(patterns[10], false, true)?.into(), - ListJoinerPattern::from_str(patterns[11], true, true)?.into(), - ])) - } - - /// Adds a special case to all `pattern`s that will evaluate to - /// `alternative_pattern` when `regex` matches the following element. - /// The regex is interpreted case-insensitive and anchored to the beginning, but - /// to improve efficiency does not search for full matches. If a full match is - /// required, use `$`. - pub fn make_conditional( - &mut self, - pattern: &str, - regex: &StringMatcher<'static>, - alternative_pattern: &str, - ) -> Result<(), DataError> { - let old = ListJoinerPattern::from_str(pattern, true, true)?; - for i in 0..12 { - if self.0[i].default == old { - self.0[i].special_case = Some(SpecialCasePattern { - condition: regex.clone(), - pattern: ListJoinerPattern::from_str( - alternative_pattern, - i % 4 == 0 || i % 4 == 3, // allow_prefix = start or pair - i % 4 == 2 || i % 4 == 3, // allow_suffix = end or pair - )?, - }); - } - } - Ok(()) - } - } - - impl<'data> ListJoinerPattern<'data> { - /// Construct the pattern from a CLDR pattern string - pub fn from_str( - pattern: &str, - allow_prefix: bool, - allow_suffix: bool, - ) -> Result<Self, DataError> { - match (pattern.find("{0}"), pattern.find("{1}")) { - (Some(index_0), Some(index_1)) - if index_0 < index_1 - && (allow_prefix || index_0 == 0) - && (allow_suffix || index_1 == pattern.len() - 3) => - { - if (index_0 > 0 && !cfg!(test)) || index_1 - 3 >= 256 { - return Err(DataError::custom( - "Found valid pattern that cannot be stored in ListFormatterPatternsV1", - ) - .with_debug_context(pattern)); - } - Ok(ListJoinerPattern { - string: Cow::Owned(alloc::format!( - "{}{}{}", - &pattern[0..index_0], - &pattern[index_0 + 3..index_1], - &pattern[index_1 + 3..] - )), - index_0: index_0 as u8, - index_1: (index_1 - 3) as u8, - }) - } - _ => Err(DataError::custom("Invalid list pattern").with_debug_context(pattern)), - } - } - } - - impl<'data> From<ListJoinerPattern<'data>> for ConditionalListJoinerPattern<'data> { - fn from(default: ListJoinerPattern<'data>) -> Self { - Self { - default, - special_case: None, - } - } - } - - impl databake::Bake for ListJoinerPattern<'_> { - fn bake(&self, env: &databake::CrateEnv) -> databake::TokenStream { - env.insert("icu_list"); - let string = (&*self.string).bake(env); - let index_1 = self.index_1.bake(env); - // Safe because our own data is safe - databake::quote! { unsafe { - ::icu_list::provider::ListJoinerPattern::from_parts_unchecked(#string, #index_1) - }} - } - } -} - -#[cfg(all(test, feature = "datagen"))] -pub(crate) mod test { - use super::*; - - pub fn test_patterns() -> ListFormatterPatternsV1<'static> { - let mut patterns = ListFormatterPatternsV1::try_new([ - // Wide: general - "@{0}:{1}", - "{0},{1}", - "{0}.{1}!", - "${0};{1}+", - // Short: different pattern lengths - "{0}1{1}", - "{0}12{1}", - "{0}12{1}34", - "{0}123{1}456", - // Narrow: conditionals - "{0}: {1}", - "{0}, {1}", - "{0}. {1}", - "{0}. {1}", - ]) - .unwrap(); - patterns - .make_conditional("{0}. {1}", &StringMatcher::new("A").unwrap(), "{0} :o {1}") - .unwrap(); - patterns - } - - #[test] - fn rejects_bad_patterns() { - assert!(ListJoinerPattern::from_str("{0} and", true, true).is_err()); - assert!(ListJoinerPattern::from_str("and {1}", true, true).is_err()); - assert!(ListJoinerPattern::from_str("{1} and {0}", true, true).is_err()); - assert!(ListJoinerPattern::from_str("{1{0}}", true, true).is_err()); - assert!(ListJoinerPattern::from_str("{0\u{202e}} and {1}", true, true).is_err()); - assert!(ListJoinerPattern::from_str("{{0}} {{1}}", true, true).is_ok()); - - assert!(ListJoinerPattern::from_str("{0} and {1} ", true, true).is_ok()); - assert!(ListJoinerPattern::from_str("{0} and {1} ", true, false).is_err()); - assert!(ListJoinerPattern::from_str(" {0} and {1}", true, true).is_ok()); - assert!(ListJoinerPattern::from_str(" {0} and {1}", false, true).is_err()); - } - - #[test] - fn produces_correct_parts() { - assert_eq!( - test_patterns().pair(ListLength::Wide).parts(""), - ("$", ";", "+") - ); - } - - #[test] - fn produces_correct_parts_conditionally() { - assert_eq!( - test_patterns().end(ListLength::Narrow).parts("A"), - ("", " :o ", "") - ); - assert_eq!( - test_patterns().end(ListLength::Narrow).parts("a"), - ("", " :o ", "") - ); - assert_eq!( - test_patterns().end(ListLength::Narrow).parts("ab"), - ("", " :o ", "") - ); - assert_eq!( - test_patterns().end(ListLength::Narrow).parts("B"), - ("", ". ", "") - ); - assert_eq!( - test_patterns().end(ListLength::Narrow).parts("BA"), - ("", ". ", "") - ); - } - - #[test] - fn size_hint_works() { - let pattern = test_patterns(); - - assert_eq!( - pattern.size_hint(ListLength::Short, 0), - LengthHint::exact(0) - ); - assert_eq!( - pattern.size_hint(ListLength::Short, 1), - LengthHint::exact(0) - ); - - // pair pattern "{0}123{1}456" - assert_eq!( - pattern.size_hint(ListLength::Short, 2), - LengthHint::exact(6) - ); - - // patterns "{0}1{1}", "{0}12{1}" (x197), and "{0}12{1}34" - assert_eq!( - pattern.size_hint(ListLength::Short, 200), - LengthHint::exact(1 + 2 * 197 + 4) - ); - - // patterns "{0}: {1}", "{0}, {1}" (x197), and "{0} :o {1}" or "{0}. {1}" - assert_eq!( - pattern.size_hint(ListLength::Narrow, 200), - LengthHint::exact(2 + 197 * 2) + LengthHint::between(2, 4) - ); - } - - #[test] - fn databake() { - databake::test_bake!( - ListJoinerPattern, - const: unsafe { crate::provider::ListJoinerPattern::from_parts_unchecked(", ", 2u8) }, - icu_list - ); - } -} diff --git a/vendor/icu_list/src/provider/mod.rs b/vendor/icu_list/src/provider/mod.rs new file mode 100644 index 000000000..efab7c8bc --- /dev/null +++ b/vendor/icu_list/src/provider/mod.rs @@ -0,0 +1,261 @@ +// This file is part of ICU4X. For terms of use, please see the file +// called LICENSE at the top level of the ICU4X source tree +// (online at: https://github.com/unicode-org/icu4x/blob/main/LICENSE ). + +// Provider structs must be stable +#![allow(clippy::exhaustive_structs, clippy::exhaustive_enums)] + +//! 🚧 \[Unstable\] Data provider struct definitions for this ICU4X component. +//! +//! <div class="stab unstable"> +//! 🚧 This code is considered unstable; it may change at any time, in breaking or non-breaking ways, +//! including in SemVer minor releases. While the serde representation of data structs is guaranteed +//! to be stable, their Rust representation might not be. Use with caution. +//! </div> +//! +//! Read more about data providers: [`icu_provider`] + +use crate::ListLength; +use alloc::borrow::Cow; +use icu_provider::DataMarker; +use icu_provider::{yoke, zerofrom}; + +mod serde_dfa; +pub use serde_dfa::SerdeDFA; + +/// Symbols and metadata required for [`ListFormatter`](crate::ListFormatter). +/// +/// <div class="stab unstable"> +/// 🚧 This code is considered unstable; it may change at any time, in breaking or non-breaking ways, +/// including in SemVer minor releases. While the serde representation of data structs is guaranteed +/// to be stable, their Rust representation might not be. Use with caution. +/// </div> +#[icu_provider::data_struct( + AndListV1Marker = "list/and@1", + OrListV1Marker = "list/or@1", + UnitListV1Marker = "list/unit@1" +)] +#[derive(Clone, Debug)] +#[cfg_attr( + feature = "datagen", + derive(serde::Serialize, databake::Bake), + databake(path = icu_list::provider), +)] +pub struct ListFormatterPatternsV1<'data>( + #[cfg_attr(feature = "datagen", serde(with = "deduplicating_array"))] + /// The patterns in the order start, middle, end, pair, short_start, short_middle, + /// short_end, short_pair, narrow_start, narrow_middle, narrow_end, narrow_pair, + pub [ConditionalListJoinerPattern<'data>; 12], +); + +#[cfg(feature = "serde")] +impl<'de> serde::Deserialize<'de> for ListFormatterPatternsV1<'de> { + fn deserialize<D>(deserializer: D) -> Result<Self, D::Error> + where + D: serde::de::Deserializer<'de>, + { + #[cfg(not(feature = "serde_human"))] + if deserializer.is_human_readable() { + use serde::de::Error; + return Err(D::Error::custom( + "Deserializing human-readable ListFormatter data requires the 'serde_human' feature", + )); + } + + Ok(ListFormatterPatternsV1(deduplicating_array::deserialize( + deserializer, + )?)) + } +} + +pub(crate) struct ErasedListV1Marker; + +impl DataMarker for ErasedListV1Marker { + type Yokeable = ListFormatterPatternsV1<'static>; +} + +impl<'data> ListFormatterPatternsV1<'data> { + pub(crate) fn start(&self, style: ListLength) -> &ConditionalListJoinerPattern<'data> { + #![allow(clippy::indexing_slicing)] // style as usize < 3 + &self.0[4 * (style as usize)] + } + + pub(crate) fn middle(&self, style: ListLength) -> &ConditionalListJoinerPattern<'data> { + #![allow(clippy::indexing_slicing)] // style as usize < 3 + &self.0[4 * (style as usize) + 1] + } + + pub(crate) fn end(&self, style: ListLength) -> &ConditionalListJoinerPattern<'data> { + #![allow(clippy::indexing_slicing)] // style as usize < 3 + &self.0[4 * (style as usize) + 2] + } + + pub(crate) fn pair(&self, style: ListLength) -> &ConditionalListJoinerPattern<'data> { + #![allow(clippy::indexing_slicing)] // style as usize < 3 + &self.0[4 * (style as usize) + 3] + } +} + +/// A pattern that can behave conditionally on the next element. +/// +/// <div class="stab unstable"> +/// 🚧 This code is considered unstable; it may change at any time, in breaking or non-breaking ways, +/// including in SemVer minor releases. While the serde representation of data structs is guaranteed +/// to be stable, their Rust representation might not be. Use with caution. +/// </div> +#[derive(Clone, Debug, yoke::Yokeable, zerofrom::ZeroFrom)] +#[cfg_attr( + feature = "datagen", + derive(PartialEq, serde::Serialize, databake::Bake), + databake(path = icu_list::provider), +)] +#[cfg_attr(feature = "serde", derive(serde::Deserialize))] +pub struct ConditionalListJoinerPattern<'data> { + /// The default pattern + #[cfg_attr(feature = "serde", serde(borrow))] + pub default: ListJoinerPattern<'data>, + /// And optional special case + #[cfg_attr( + feature = "serde", + serde(borrow, deserialize_with = "SpecialCasePattern::deserialize_option") + )] + pub special_case: Option<SpecialCasePattern<'data>>, +} + +/// The special case of a [`ConditionalListJoinerPattern`] +/// +/// <div class="stab unstable"> +/// 🚧 This code is considered unstable; it may change at any time, in breaking or non-breaking ways, +/// including in SemVer minor releases. While the serde representation of data structs is guaranteed +/// to be stable, their Rust representation might not be. Use with caution. +/// </div> +#[derive(Clone, Debug, yoke::Yokeable, zerofrom::ZeroFrom)] +#[cfg_attr( + feature = "datagen", + derive(PartialEq, serde::Serialize, databake::Bake), + databake(path = icu_list::provider), +)] +pub struct SpecialCasePattern<'data> { + /// The condition on the following element + pub condition: SerdeDFA<'data>, + /// The pattern if the condition matches + pub pattern: ListJoinerPattern<'data>, +} + +#[cfg(feature = "serde")] +impl<'data> SpecialCasePattern<'data> { + // If the condition doesn't deserialize, the whole special case becomes `None` + fn deserialize_option<'de: 'data, D>(deserializer: D) -> Result<Option<Self>, D::Error> + where + D: serde::de::Deserializer<'de>, + { + use serde::Deserialize; + + #[derive(Deserialize)] + struct SpecialCasePatternOptionalDfa<'data> { + #[cfg_attr( + feature = "serde", + serde(borrow, deserialize_with = "SerdeDFA::maybe_deserialize") + )] + pub condition: Option<SerdeDFA<'data>>, + #[cfg_attr(feature = "serde", serde(borrow))] + pub pattern: ListJoinerPattern<'data>, + } + + Ok( + match Option::<SpecialCasePatternOptionalDfa<'data>>::deserialize(deserializer)? { + Some(SpecialCasePatternOptionalDfa { + condition: Some(condition), + pattern, + }) => Some(SpecialCasePattern { condition, pattern }), + _ => None, + }, + ) + } +} + +/// A pattern containing two numeric placeholders ("{0}, and {1}.") +/// +/// <div class="stab unstable"> +/// 🚧 This code is considered unstable; it may change at any time, in breaking or non-breaking ways, +/// including in SemVer minor releases. While the serde representation of data structs is guaranteed +/// to be stable, their Rust representation might not be. Use with caution. +/// </div> +#[derive(Clone, Debug, PartialEq, yoke::Yokeable, zerofrom::ZeroFrom)] +#[cfg_attr(feature = "datagen", derive(serde::Serialize))] +pub struct ListJoinerPattern<'data> { + /// The pattern string without the placeholders + pub(crate) string: Cow<'data, str>, + /// The index of the first placeholder. Always <= index_1. + // Always 0 for CLDR data, so we don't need to serialize it. + // In-memory we have free space for it as index_1 doesn't + // fill a word. + #[cfg_attr(feature = "datagen", serde(skip))] + pub(crate) index_0: u8, + /// The index of the second placeholder. Always < string.len(). + pub(crate) index_1: u8, +} + +#[cfg(feature = "serde")] +impl<'de: 'data, 'data> serde::Deserialize<'de> for ListJoinerPattern<'data> { + fn deserialize<D>(deserializer: D) -> Result<Self, D::Error> + where + D: serde::Deserializer<'de>, + { + #[derive(serde::Deserialize)] + struct Dummy<'data> { + #[cfg_attr(feature = "serde", serde(borrow))] + string: Cow<'data, str>, + index_1: u8, + } + let Dummy { string, index_1 } = Dummy::deserialize(deserializer)?; + + if index_1 as usize > string.len() { + use serde::de::Error; + Err(D::Error::custom("invalid index_1")) + } else { + Ok(ListJoinerPattern { + string, + index_0: 0, + index_1, + }) + } + } +} + +impl<'a> ListJoinerPattern<'a> { + /// Constructs a [`ListJoinerPattern`] from raw parts. Used by databake. + /// + /// # Safety + /// index_1 may be at most string.len() + pub const unsafe fn from_parts_unchecked(string: &'a str, index_1: u8) -> Self { + Self { + string: Cow::Borrowed(string), + index_0: 0, + index_1, + } + } +} + +#[cfg(feature = "datagen")] +impl databake::Bake for ListJoinerPattern<'_> { + fn bake(&self, env: &databake::CrateEnv) -> databake::TokenStream { + env.insert("icu_list"); + let string = (&*self.string).bake(env); + let index_1 = self.index_1.bake(env); + // Safe because our own data is safe + databake::quote! { unsafe { + ::icu_list::provider::ListJoinerPattern::from_parts_unchecked(#string, #index_1) + }} + } +} + +#[cfg(all(test, feature = "datagen"))] +#[test] +fn databake() { + databake::test_bake!( + ListJoinerPattern, + const: unsafe { crate::provider::ListJoinerPattern::from_parts_unchecked(", ", 2u8) }, + icu_list + ); +} diff --git a/vendor/icu_list/src/provider/serde_dfa.rs b/vendor/icu_list/src/provider/serde_dfa.rs new file mode 100644 index 000000000..e2424e1e9 --- /dev/null +++ b/vendor/icu_list/src/provider/serde_dfa.rs @@ -0,0 +1,244 @@ +// This file is part of ICU4X. For terms of use, please see the file +// called LICENSE at the top level of the ICU4X source tree +// (online at: https://github.com/unicode-org/icu4x/blob/main/LICENSE ). + +use alloc::borrow::Cow; +use icu_provider::{yoke, zerofrom}; +use regex_automata::dfa::sparse::DFA; + +/// A serde-compatible version of [regex_automata::dfa::sparse::DFA]. This does not implement +/// [`serde::Deserialize`] directly, as binary deserialization is not supported in big-endian +/// platforms. `Self::maybe_deserialize` can be used to deserialize to `Option<SerdeDFA>`. +/// +/// <div class="stab unstable"> +/// 🚧 This code is considered unstable; it may change at any time, in breaking or non-breaking ways, +/// including in SemVer minor releases. While the serde representation of data structs is guaranteed +/// to be stable, their Rust representation might not be. Use with caution. +/// </div> +#[derive(Clone, Debug, yoke::Yokeable, zerofrom::ZeroFrom)] +pub struct SerdeDFA<'data> { + // Safety: These always represent a valid DFA (DFA::from_bytes(dfa_bytes).is_ok()) + dfa_bytes: Cow<'data, [u8]>, + pattern: Option<Cow<'data, str>>, +} + +#[cfg(feature = "datagen")] +impl PartialEq for SerdeDFA<'_> { + fn eq(&self, other: &Self) -> bool { + self.dfa_bytes == other.dfa_bytes + } +} + +#[cfg(feature = "datagen")] +impl databake::Bake for SerdeDFA<'_> { + fn bake(&self, env: &databake::CrateEnv) -> databake::TokenStream { + env.insert("icu_list"); + let le_bytes = self.deref().to_bytes_little_endian().as_slice().bake(env); + let be_bytes = self.deref().to_bytes_big_endian().as_slice().bake(env); + // Safe because of `to_bytes_little_endian`/`to_bytes_big_endian`'s invariant. + databake::quote! { + unsafe { + ::icu_list::provider::SerdeDFA::from_dfa_bytes_unchecked( + if cfg!(target_endian = "little") { + &#le_bytes + } else { + &#be_bytes + } + ) + } + } + } +} + +#[cfg(feature = "datagen")] +impl serde::Serialize for SerdeDFA<'_> { + fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error> + where + S: serde::ser::Serializer, + { + if serializer.is_human_readable() { + self.pattern + .as_ref() + .map(|pattern| pattern.serialize(serializer)) + .unwrap_or_else(|| { + use serde::ser::Error; + Err(S::Error::custom( + "cannot serialize a deserialized bincode SerdeDFA to JSON", + )) + }) + } else { + self.deref().to_bytes_little_endian().serialize(serializer) + } + } +} + +#[cfg(feature = "serde")] +impl<'data> SerdeDFA<'data> { + /// Deserializes to `Option<Self>`. Will return `None` for non-human-readable serialization + /// formats on big-endian systems, as `regex_automata` serialization is endian-sensitive. + pub fn maybe_deserialize<'de: 'data, D>(deserializer: D) -> Result<Option<Self>, D::Error> + where + D: serde::de::Deserializer<'de>, + { + use icu_provider::serde::borrow_de_utils::CowBytesWrap; + use serde::Deserialize; + + #[cfg(feature = "serde_human")] + if deserializer.is_human_readable() { + #[cfg(not(feature = "std"))] + use alloc::string::ToString; + use serde::de::Error; + return SerdeDFA::new(Cow::<str>::deserialize(deserializer)?) + .map(Some) + .map_err(|e| D::Error::custom(e.to_string())); + } + + let dfa_bytes = <CowBytesWrap<'de>>::deserialize(deserializer)?.0; + + if cfg!(target_endian = "big") { + return Ok(None); + } + + // Verify safety invariant + DFA::from_bytes(&dfa_bytes).map_err(|e| { + use serde::de::Error; + D::Error::custom(alloc::format!("Invalid DFA bytes: {}", e)) + })?; + + Ok(Some(SerdeDFA { + dfa_bytes, + pattern: None, + })) + } +} + +impl<'data> SerdeDFA<'data> { + /// Creates a `SerdeDFA` from raw bytes. Used internally by databake. + /// + /// # Safety + /// + /// `dfa_bytes` has to be a valid DFA (regex_automata::dfa::sparse::DFA::from_bytes(dfa_bytes).is_ok()) + pub const unsafe fn from_dfa_bytes_unchecked(dfa_bytes: &'data [u8]) -> Self { + Self { + dfa_bytes: Cow::Borrowed(dfa_bytes), + pattern: None, + } + } + + /// Creates a `SerdeDFA` from a regex. + #[cfg(any(feature = "datagen", feature = "serde_human",))] + pub fn new(pattern: Cow<'data, str>) -> Result<Self, icu_provider::DataError> { + use regex_automata::{ + dfa::dense::{Builder, Config}, + SyntaxConfig, + }; + + let mut builder = Builder::new(); + let dfa = builder + .syntax(SyntaxConfig::new().case_insensitive(true)) + .configure(Config::new().anchored(true).minimize(true)) + .build(&pattern) + .map_err(|_| { + icu_provider::DataError::custom("Cannot build DFA").with_display_context(&pattern) + })? + .to_sparse() + .map_err(|_| { + icu_provider::DataError::custom("Cannot sparsify DFA") + .with_display_context(&pattern) + })?; + + Ok(Self { + dfa_bytes: dfa.to_bytes_native_endian().into(), + pattern: Some(pattern), + }) + } + + /// Returns the represented [`DFA`] + #[allow(clippy::unwrap_used)] // by invariant + pub fn deref(&'data self) -> DFA<&'data [u8]> { + // Safe due to struct invariant. + unsafe { DFA::from_bytes_unchecked(&self.dfa_bytes).unwrap().0 } + } +} + +#[cfg(all(test, feature = "datagen"))] +mod test { + use super::*; + + #[test] + fn test_serde_dfa() { + use regex_automata::dfa::Automaton; + + let matcher = SerdeDFA::new(Cow::Borrowed("abc")).unwrap(); + + assert!(matcher.deref().find_earliest_fwd(b"ab").unwrap().is_none()); + assert!(matcher.deref().find_earliest_fwd(b"abc").unwrap().is_some()); + assert!(matcher + .deref() + .find_earliest_fwd(b"abcde") + .unwrap() + .is_some()); + assert!(matcher + .deref() + .find_earliest_fwd(b" abcde") + .unwrap() + .is_none()); + } + + #[derive(serde::Deserialize)] + struct OptionSerdeDFA<'data>( + #[serde(borrow, deserialize_with = "SerdeDFA::maybe_deserialize")] Option<SerdeDFA<'data>>, + ); + + #[test] + #[cfg(target_endian = "little")] + fn test_postcard_serialization() { + let matcher = SerdeDFA::new(Cow::Borrowed("abc*")).unwrap(); + + let mut bytes = postcard::to_stdvec(&matcher).unwrap(); + assert_eq!( + postcard::from_bytes::<OptionSerdeDFA>(&bytes).unwrap().0, + Some(matcher) + ); + + // A corrupted byte leads to an error + bytes[17] ^= 255; + assert!(postcard::from_bytes::<OptionSerdeDFA>(&bytes).is_err()); + bytes[17] ^= 255; + + // An extra byte leads to an error + bytes.insert(123, 40); + assert!(postcard::from_bytes::<OptionSerdeDFA>(&bytes).is_err()); + bytes.remove(123); + + // Missing bytes lead to an error + assert!(postcard::from_bytes::<OptionSerdeDFA>(&bytes[0..bytes.len() - 5]).is_err()); + } + + #[test] + #[cfg(feature = "serde_human")] + fn test_json_serialization() { + let matcher = SerdeDFA::new(Cow::Borrowed("abc*")).unwrap(); + + let json = serde_json::to_string(&matcher).unwrap(); + assert_eq!( + serde_json::from_str::<OptionSerdeDFA>(&json).unwrap().0, + Some(matcher) + ); + assert!(serde_json::from_str::<OptionSerdeDFA>(".*[").is_err()); + } + + #[test] + #[ignore] // https://github.com/rust-lang/rust/issues/98906 + fn databake() { + databake::test_bake!( + SerdeDFA, + const: unsafe { crate::provider::SerdeDFA::from_dfa_bytes_unchecked(if cfg!(target_endian = "little") { + &[1] // TODO: set this when activating the test + } else { + &[2] // TODO: set this when activating the test + })}, + icu_list + ); + } +} diff --git a/vendor/icu_list/src/string_matcher.rs b/vendor/icu_list/src/string_matcher.rs deleted file mode 100644 index ba4833605..000000000 --- a/vendor/icu_list/src/string_matcher.rs +++ /dev/null @@ -1,213 +0,0 @@ -// This file is part of ICU4X. For terms of use, please see the file -// called LICENSE at the top level of the ICU4X source tree -// (online at: https://github.com/unicode-org/icu4x/blob/main/LICENSE ). - -use alloc::borrow::Cow; -#[cfg(any(feature = "serde_human", feature = "datagen"))] -use alloc::string::ToString; -use icu_provider::{yoke, zerofrom}; -use regex_automata::dfa::sparse::DFA; -use regex_automata::dfa::Automaton; - -/// A precompiled regex -#[derive(Clone, Debug, yoke::Yokeable, zerofrom::ZeroFrom)] -#[allow(clippy::exhaustive_structs)] // not a public API -pub struct StringMatcher<'data> { - // Safety: These always represent a valid DFA (DFA::from_bytes(dfa_bytes).is_ok()) - dfa_bytes: Cow<'data, [u8]>, - pattern: Option<Cow<'data, str>>, -} - -impl PartialEq for StringMatcher<'_> { - fn eq(&self, other: &Self) -> bool { - self.dfa_bytes == other.dfa_bytes - } -} - -#[cfg(feature = "datagen")] -impl databake::Bake for StringMatcher<'_> { - fn bake(&self, env: &databake::CrateEnv) -> databake::TokenStream { - env.insert("icu_list"); - let bytes = (&&*self.dfa_bytes).bake(env); - // Safe because our own data is safe - databake::quote! { - unsafe { ::icu_list::provider::StringMatcher::from_dfa_bytes_unchecked(#bytes) } - } - } -} - -#[cfg(feature = "datagen")] -impl serde::Serialize for StringMatcher<'_> { - fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error> - where - S: serde::ser::Serializer, - { - if serializer.is_human_readable() { - self.pattern - .as_ref() - .map(|pattern| pattern.serialize(serializer)) - .unwrap_or_else(|| { - use serde::ser::Error; - Err(S::Error::custom( - "cannot serialize a deserialized bincode StringMatcher to JSON", - )) - }) - } else { - self.dfa_bytes.serialize(serializer) - } - } -} - -#[cfg(feature = "serde")] -impl<'de: 'data, 'data> serde::Deserialize<'de> for StringMatcher<'data> { - fn deserialize<D>(deserializer: D) -> Result<Self, D::Error> - where - D: serde::de::Deserializer<'de>, - { - use icu_provider::serde::borrow_de_utils::CowBytesWrap; - - #[cfg(feature = "serde_human")] - if deserializer.is_human_readable() { - use serde::de::Error; - return StringMatcher::new(<&str>::deserialize(deserializer)?) - .map_err(|e| D::Error::custom(e.to_string())); - } - - if cfg!(target_endian = "big") { - // TODO: Convert LE to BE. For now we just behave like the - // accept-nothing DFA on BE systems. - return Ok(StringMatcher { - dfa_bytes: Cow::Borrowed(&[]), - pattern: None, - }); - } - - let dfa_bytes = <CowBytesWrap<'de>>::deserialize(deserializer)?.0; - - // Verify safety invariant - DFA::from_bytes(&dfa_bytes).map_err(|e| { - use serde::de::Error; - D::Error::custom(alloc::format!("Invalid DFA bytes: {}", e)) - })?; - - Ok(StringMatcher { - dfa_bytes, - pattern: None, - }) - } -} - -impl<'data> StringMatcher<'data> { - /// Creates a `StringMatcher` from a serialized DFA. Used internally by databake. - /// - /// # Safety - /// - /// `dfa_bytes` has to be a valid DFA (regex_automata::dfa::sparse::DFA::from_bytes(dfa_bytes).is_ok()) - pub const unsafe fn from_dfa_bytes_unchecked(dfa_bytes: &'data [u8]) -> Self { - Self { - dfa_bytes: Cow::Borrowed(dfa_bytes), - pattern: None, - } - } - - /// Creates a `StringMatcher` from regex. - #[cfg(any(feature = "datagen", feature = "serde_human",))] - pub fn new(pattern: &str) -> Result<Self, icu_provider::DataError> { - use regex_automata::{ - dfa::dense::{Builder, Config}, - SyntaxConfig, - }; - - let mut builder = Builder::new(); - let dfa = builder - .syntax(SyntaxConfig::new().case_insensitive(true)) - .configure(Config::new().anchored(true).minimize(true)) - .build(pattern) - .map_err(|_| { - icu_provider::DataError::custom("Cannot build DFA").with_display_context(&pattern) - })? - .to_sparse() - .map_err(|_| { - icu_provider::DataError::custom("Cannot sparsify DFA") - .with_display_context(&pattern) - })?; - - Ok(Self { - dfa_bytes: dfa.to_bytes_little_endian().into(), - pattern: Some(pattern.to_string().into()), - }) - } - - #[allow(clippy::unwrap_used)] // by invariant - pub(crate) fn test(&self, string: &str) -> bool { - cfg!(target_endian = "little") - && matches!( - // Safe due to struct invariant. - unsafe { DFA::from_bytes_unchecked(&self.dfa_bytes).unwrap().0 } - .find_earliest_fwd(string.as_bytes()), - Ok(Some(_)) - ) - } -} - -#[cfg(all(test, feature = "datagen"))] -mod test { - use super::*; - - #[test] - fn test_string_matcher() { - let matcher = StringMatcher::new("abc.*").unwrap(); - assert!(!matcher.test("ab")); - assert!(matcher.test("abc")); - assert!(matcher.test("abcde")); - } - - #[test] - fn test_postcard_serialization() { - let matcher = StringMatcher::new("abc*").unwrap(); - - let mut bytes = postcard::to_stdvec(&matcher).unwrap(); - assert_eq!( - postcard::from_bytes::<StringMatcher>(&bytes).unwrap(), - matcher - ); - - // A corrupted byte leads to an error - bytes[17] ^= 255; - assert!(postcard::from_bytes::<StringMatcher>(&bytes).is_err()); - bytes[17] ^= 255; - - // An extra byte leads to an error - bytes.insert(123, 40); - assert!(postcard::from_bytes::<StringMatcher>(&bytes).is_err()); - bytes.remove(123); - - // Missing bytes lead to an error - assert!(postcard::from_bytes::<StringMatcher>(&bytes[0..bytes.len() - 5]).is_err()); - } - - #[test] - #[cfg(feature = "serde_human")] - fn test_json_serialization() { - let matcher = StringMatcher::new("abc*").unwrap(); - - let json = serde_json::to_string(&matcher).unwrap(); - assert_eq!( - serde_json::from_str::<StringMatcher>(&json).unwrap(), - matcher - ); - assert!(serde_json::from_str::<StringMatcher>(".*[").is_err()); - } - - #[test] - #[ignore] // https://github.com/rust-lang/rust/issues/98906 - fn databake() { - databake::test_bake!( - StringMatcher, - const: unsafe { - crate::provider::StringMatcher::from_dfa_bytes_unchecked(&[49u8, 50u8, 51u8, ]) - }, - icu_list - ); - } -} |