Add some phonenumberutil functions

Cargo.lock

@@ -41,6 +41,27 @@ version = "0.8.21"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "d0a5c400df2834b80a4c3327b3aad3a4c4cd4de0629063962b03235697506a28"
 
+[[package]]
+name = "csv"
+version = "1.3.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "acdc4883a9c96732e4733212c01447ebd805833b7275a73ca3ee080fd77afdaf"
+dependencies = [
+ "csv-core",
+ "itoa",
+ "ryu",
+ "serde",
+]
+
+[[package]]
+name = "csv-core"
+version = "0.1.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7d02f3b0da4c6504f86e9cd789d8dbafab48c2321be74e9987593de5a894d93d"
+dependencies = [
+ "memchr",
+]
+
 [[package]]
 name = "dashmap"
 version = "6.1.0"
@@ -55,6 +76,37 @@ dependencies = [
  "parking_lot_core",
 ]
 
+[[package]]
+name = "dec_from_char"
+version = "0.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5467a30cfe9ac2281f28ab2fc7e82b3036544bcb340f648425769fe9eb737708"
+dependencies = [
+ "dec_from_char_gen",
+]
+
+[[package]]
+name = "dec_from_char_gen"
+version = "0.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "22ec31e992a6f011b864a5eda50b01cf01fa8d9d53241f4902f93d0f8024faa6"
+dependencies = [
+ "csv",
+ "quote",
+ "serde",
+]
+
+[[package]]
+name = "displaydoc"
+version = "0.2.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "97369cbbc041bc366949bc74d34658d6cda5621039731c6310521892a3a20ae0"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
 [[package]]
 name = "either"
 version = "1.15.0"
@@ -128,6 +180,95 @@ dependencies = [
  "windows-sys 0.59.0",
 ]
 
+[[package]]
+name = "icu_collections"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "200072f5d0e3614556f94a9930d5dc3e0662a652823904c3a75dc3b0af7fee47"
+dependencies = [
+ "displaydoc",
+ "potential_utf",
+ "yoke",
+ "zerofrom",
+ "zerovec",
+]
+
+[[package]]
+name = "icu_locale_core"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0cde2700ccaed3872079a65fb1a78f6c0a36c91570f28755dda67bc8f7d9f00a"
+dependencies = [
+ "displaydoc",
+ "litemap",
+ "tinystr",
+ "writeable",
+ "zerovec",
+]
+
+[[package]]
+name = "icu_normalizer"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "436880e8e18df4d7bbc06d58432329d6458cc84531f7ac5f024e93deadb37979"
+dependencies = [
+ "displaydoc",
+ "icu_collections",
+ "icu_normalizer_data",
+ "icu_properties",
+ "icu_provider",
+ "smallvec",
+ "utf16_iter",
+ "utf8_iter",
+ "write16",
+ "zerovec",
+]
+
+[[package]]
+name = "icu_normalizer_data"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "00210d6893afc98edb752b664b8890f0ef174c8adbb8d0be9710fa66fbbf72d3"
+
+[[package]]
+name = "icu_properties"
+version = "2.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "016c619c1eeb94efb86809b015c58f479963de65bdb6253345c1a1276f22e32b"
+dependencies = [
+ "displaydoc",
+ "icu_collections",
+ "icu_locale_core",
+ "icu_properties_data",
+ "icu_provider",
+ "potential_utf",
+ "zerotrie",
+ "zerovec",
+]
+
+[[package]]
+name = "icu_properties_data"
+version = "2.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "298459143998310acd25ffe6810ed544932242d3f07083eee1084d83a71bd632"
+
+[[package]]
+name = "icu_provider"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "03c80da27b5f4187909049ee2d72f276f0d9f99a42c306bd0131ecfe04d8e5af"
+dependencies = [
+ "displaydoc",
+ "icu_locale_core",
+ "stable_deref_trait",
+ "tinystr",
+ "writeable",
+ "yoke",
+ "zerofrom",
+ "zerotrie",
+ "zerovec",
+]
+
 [[package]]
 name = "indexmap"
 version = "2.10.0"
@@ -162,6 +303,12 @@ version = "0.9.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "cd945864f07fe9f5371a27ad7b52a172b4b499999f1d97574c9fa68373937e12"
 
+[[package]]
+name = "litemap"
+version = "0.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "241eaef5fd12c88705a01fc1066c48c4b36e0dd4377dcdc7ec3942cea7a69956"
+
 [[package]]
 name = "lock_api"
 version = "0.4.13"
@@ -203,6 +350,15 @@ dependencies = [
  "windows-targets 0.52.6",
 ]
 
+[[package]]
+name = "potential_utf"
+version = "0.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e5a7c30837279ca13e7c867e9e40053bc68740f988cb07f7ca6df43cc734b585"
+dependencies = [
+ "zerovec",
+]
+
 [[package]]
 name = "proc-macro2"
 version = "1.0.95"
@@ -321,7 +477,9 @@ name = "rlibphonenumbers"
 version = "0.1.0"
 dependencies = [
  "dashmap",
+ "dec_from_char",
  "fast-cat",
+ "icu_normalizer",
  "itoa",
  "log",
  "protobuf",
@@ -329,6 +487,7 @@ dependencies = [
  "regex",
  "strum",
  "thiserror 2.0.12",
+ "tinystr",
 ]
 
 [[package]]
@@ -363,18 +522,50 @@ version = "1.0.21"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8a0d197bd2c9dc6e53b84da9556a69ba4cdfab8619eb41a8bd1cc2027a0f6b1d"
 
+[[package]]
+name = "ryu"
+version = "1.0.20"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "28d3b2b1366ec20994f1fd18c3c594f05c5dd4bc44d8bb0c1c632c8d6829481f"
+
 [[package]]
 name = "scopeguard"
 version = "1.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "94143f37725109f92c262ed2cf5e59bce7498c01bcc1502d7b9afe439a4e9f49"
 
+[[package]]
+name = "serde"
+version = "1.0.219"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5f0e2c6ed6606019b4e29e69dbaba95b11854410e5347d525002456dbbb786b6"
+dependencies = [
+ "serde_derive",
+]
+
+[[package]]
+name = "serde_derive"
+version = "1.0.219"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5b0276cf7f2c73365f7157c8123c21cd9a50fbbd844757af28ca1f5925fc2a00"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
 [[package]]
 name = "smallvec"
 version = "1.15.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "67b1b7a3b5fe4f1376887184045fcf45c69e92af734b7aaddc05fb777b6fbd03"
 
+[[package]]
+name = "stable_deref_trait"
+version = "1.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a8f112729512f8e442d81f95a8a7ddf2b7c6b8a1a6f509a95864142b30cab2d3"
+
 [[package]]
 name = "strum"
 version = "0.27.1"
@@ -408,6 +599,17 @@ dependencies = [
  "unicode-ident",
 ]
 
+[[package]]
+name = "synstructure"
+version = "0.13.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "728a70f3dbaf5bab7f0c4b1ac8d7ae5ea60a4b5549c8a5914361c99147a709d2"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
 [[package]]
 name = "tempfile"
 version = "3.20.0"
@@ -461,12 +663,34 @@ dependencies = [
  "syn",
 ]
 
+[[package]]
+name = "tinystr"
+version = "0.8.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5d4f6d1145dcb577acf783d4e601bc1d76a13337bb54e6233add580b07344c8b"
+dependencies = [
+ "displaydoc",
+ "zerovec",
+]
+
 [[package]]
 name = "unicode-ident"
 version = "1.0.18"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "5a5f39404a5da50712a4c1eecf25e90dd62b613502b7e925fd4e4d19b5c96512"
 
+[[package]]
+name = "utf16_iter"
+version = "1.0.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c8232dd3cdaed5356e0f716d285e4b40b932ac434100fe9b7e0e8e935b9e6246"
+
+[[package]]
+name = "utf8_iter"
+version = "1.0.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b6c140620e7ffbb22c2dee59cafe6084a59b5ffc27a8859a5f0d494b5d52b6be"
+
 [[package]]
 name = "wasi"
 version = "0.14.2+wasi-0.2.4"
@@ -642,3 +866,93 @@ checksum = "6f42320e61fe2cfd34354ecb597f86f413484a798ba44a8ca1165c58d42da6c1"
 dependencies = [
  "bitflags",
 ]
+
+[[package]]
+name = "write16"
+version = "1.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d1890f4022759daae28ed4fe62859b1236caebfc61ede2f63ed4e695f3f6d936"
+
+[[package]]
+name = "writeable"
+version = "0.6.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ea2f10b9bb0928dfb1b42b65e1f9e36f7f54dbdf08457afefb38afcdec4fa2bb"
+
+[[package]]
+name = "yoke"
+version = "0.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5f41bb01b8226ef4bfd589436a297c53d118f65921786300e427be8d487695cc"
+dependencies = [
+ "serde",
+ "stable_deref_trait",
+ "yoke-derive",
+ "zerofrom",
+]
+
+[[package]]
+name = "yoke-derive"
+version = "0.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "38da3c9736e16c5d3c8c597a9aaa5d1fa565d0532ae05e27c24aa62fb32c0ab6"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+ "synstructure",
+]
+
+[[package]]
+name = "zerofrom"
+version = "0.1.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "50cc42e0333e05660c3587f3bf9d0478688e15d870fab3346451ce7f8c9fbea5"
+dependencies = [
+ "zerofrom-derive",
+]
+
+[[package]]
+name = "zerofrom-derive"
+version = "0.1.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d71e5d6e06ab090c67b5e44993ec16b72dcbaabc526db883a360057678b48502"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+ "synstructure",
+]
+
+[[package]]
+name = "zerotrie"
+version = "0.2.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "36f0bbd478583f79edad978b407914f61b2972f5af6fa089686016be8f9af595"
+dependencies = [
+ "displaydoc",
+ "yoke",
+ "zerofrom",
+]
+
+[[package]]
+name = "zerovec"
+version = "0.11.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4a05eb080e015ba39cc9e23bbe5e7fb04d5fb040350f99f34e338d5fdd294428"
+dependencies = [
+ "yoke",
+ "zerofrom",
+ "zerovec-derive",
+]
+
+[[package]]
+name = "zerovec-derive"
+version = "0.11.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5b96237efa0c878c64bd89c436f661be4e46b2f3eff1ebb976f7ef2321d2f58f"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]

Cargo.toml

@@ -22,6 +22,9 @@ itoa = "1.0.15"
 # concatenation of strings
 fast-cat = "0.1.1"
 strum = { version = "0.27.1", features = ["derive"] }
+icu_normalizer = "2.0.0"
+tinystr = "0.8.1"
+dec_from_char = "0.1.1"
 
 [build-dependencies]
 thiserror = "2.0.12"

build/rust_build.rs

@@ -72,6 +72,7 @@ fn parse_prefixes(path: &str, prefixes: &mut BTreeMap<i32, String>) -> Result<()
 
 fn main() -> Result<(), BuildError> {
     protobuf_codegen::Codegen::new()
+        .pure()
         .includes(["resources"])
         .input("resources/phonemetadata.proto")
         .input("resources/phonenumber.proto")

src/lib.rs

@@ -6,7 +6,8 @@ mod phonenumberutil;
 mod regexp_cache;
 mod regex_based_matcher;
 pub mod i18n;
-pub mod regex_util;
+pub(crate) mod regex_util;
+pub(crate) mod string_util;
 
 /// I decided to create this module because there are many 
 /// boilerplate places in the code that can be replaced with macros, 

src/phonenumberutil/errors.rs

@@ -1,3 +1,6 @@
+use core::error;
+use std::num::ParseIntError;
+
 use thiserror::Error;
 
 use crate::regexp_cache::ErrorInvalidRegex;
@@ -5,5 +8,38 @@ use crate::regexp_cache::ErrorInvalidRegex;
 #[derive(Debug, PartialEq, Error)]
 pub enum PhoneNumberUtilError {
     #[error("{0}")]
-    InvalidRegexError(#[from] ErrorInvalidRegex)
+    InvalidRegexError(#[from] ErrorInvalidRegex),
-}
+    #[error("Parse error: {0}")]
+    ParseError(#[from] ParseError),
+    #[error("Extract number error: {0}")]
+    ExtractNumberError(#[from] ExtractNumberError)
+}   
+
+#[derive(Debug, PartialEq, Error)]
+pub enum ParseError {
+    // Removed as OK variant
+    // NoParsingError,
+    #[error("Invalid country code")]
+    InvalidCountryCodeError, // INVALID_COUNTRY_CODE in the java version.
+    #[error("Not a number")]
+    NotANumber,
+    #[error("Too short after idd")]
+    TooShortAfterIdd,
+    #[error("Too short Nsn")]
+    TooShortNsn,
+    #[error("Too long nsn")]
+    TooLongNsn, // TOO_LONG in the java version.
+    #[error("{0}")]
+    InvalidRegexError(#[from] ErrorInvalidRegex),
+    #[error("{0}")]
+    ParseNumberAsIntError(#[from] ParseIntError)
+}
+
+#[derive(Debug, PartialEq, Error)]
+pub enum ExtractNumberError {
+    #[error("No valid start character found")]
+    NoValidStartCharacter,
+    #[error("Invalid number")]
+    NotANumber,
+}
+

src/phonenumberutil/helper_types.rs

@@ -0,0 +1,31 @@
+use std::borrow::Cow;
+
+use crate::proto_gen::phonenumber::phone_number::CountryCodeSource;
+
+#[derive(Debug)]
+pub struct PhoneNumberWithCountryCodeSource<'a> {
+    pub phone_number: Cow<'a, str>,
+    pub country_code_source: CountryCodeSource
+}
+
+impl<'a> PhoneNumberWithCountryCodeSource<'a> {
+    pub fn new(phone_number: Cow<'a, str>, country_code_source: CountryCodeSource) -> Self {
+        Self { phone_number, country_code_source }
+    }
+}
+
+#[derive(Debug)]
+pub struct PhoneNumberAndCarrierCode<'a> {
+    pub carrier_code: Option<&'a str>,
+    pub phone_number: Cow<'a, str>
+}
+
+impl<'a> PhoneNumberAndCarrierCode<'a> {
+    pub fn new<B: Into<Cow<'a, str>>>(carrier_code: Option<&'a str>, phone_number: B) -> Self {
+        Self { carrier_code, phone_number: phone_number.into() }
+    }
+    
+    pub fn new_phone<B: Into<Cow<'a, str>>>(phone_number: B) -> Self {
+        Self { carrier_code: None, phone_number: phone_number.into() }
+    }
+}

src/phonenumberutil/mod.rs

@@ -4,30 +4,15 @@ mod errors;
 mod enums;
 mod phonenumberutil;
 mod phone_number_regexps_and_mappings;
+pub(self) mod helper_types;
 
 use std::sync::LazyLock;
 
 pub use enums::{MatchType, PhoneNumberFormat, PhoneNumberType, ValidationResultErr, ValidNumberLenType};
 use thiserror::Error;
 
-use crate::phonenumberutil::phonenumberutil::PhoneNumberUtil;
+// use crate::phonenumberutil::phonenumberutil::PhoneNumberUtil;
 
-#[derive(Debug, Error)]
+// static PHONE_NUMBER_UTIL: LazyLock<PhoneNumberUtil> = LazyLock::new(|| {
-pub enum ErrorType {
+//     PhoneNumberUtil::new()
-    #[error("No parsing")]
+// });
-    NoParsingError,
-    #[error("Invalid country code")]
-    InvalidCountryCodeError, // INVALID_COUNTRY_CODE in the java version.
-    #[error("Not a number")]
-    NotANumber,
-    #[error("Too short after idd")]
-    TooShortAfterIdd,
-    #[error("Too short Nsn")]
-    TooShortNsn,
-    #[error("Too long nsn")]
-    TooLongNsn, // TOO_LONG in the java version.
-}
-
-static PHONE_NUMBER_UTIL: LazyLock<PhoneNumberUtil> = LazyLock::new(|| {
-    PhoneNumberUtil::new()
-});

src/regex_based_matcher.rs

@@ -21,7 +21,7 @@ impl RegexBasedMatcher {
 
         // find first occurrence
         if allow_prefix_match {
-            Ok(regexp.consume_start(phone_number).is_some())
+            Ok(regexp.matches_start(phone_number))
         } else {
             Ok(regexp.full_match(phone_number))
         }

src/regex_util.rs

@@ -1,6 +1,4 @@
-use std::borrow::Cow;
+use regex::{Captures, Match, Regex};
-
-use regex::{Captures, Regex};
 
 pub trait RegexFullMatch {
     /// Eq of C fullMatch
@@ -8,18 +6,12 @@ pub trait RegexFullMatch {
 }
 
 pub trait RegexConsume {
-    /// Eq of C Consume
+    fn matches_start<'a>(&self, s: &'a str) -> bool {
-    fn consume_start<'a>(&self, s: &'a str) -> Option<Cow<'a, str>> {
+        self.find_start(s).is_some()
-        self.consume_start_capturing(s).map(| res| res.0)
     }
 
-    fn consume_start_capturing<'a>(&self, s: &'a str) -> Option<(Cow<'a, str>, Captures<'a>)>;
+    fn captures_start<'a>(&self, s: &'a str) -> Option<Captures<'a>>;
-    
+    fn find_start<'a>(&self, s: &'a str) -> Option<Match<'a>>;
-    fn find_and_consume<'a>(&self, s: &'a str) -> Option<Cow<'a, str>> {
-        self.find_and_consume_capturing(s).map(| res| res.0)
-    }
-    
-    fn find_and_consume_capturing<'a>(&self, s: &'a str) -> Option<(Cow<'a, str>, Captures<'a>)>;
 }
 
 trait RegexMatchStart {
@@ -48,24 +40,21 @@ impl RegexMatchStart for Regex {
 }
 
 impl RegexConsume for Regex {
-    fn consume_start_capturing<'a>(&self, s: &'a str) -> Option<(Cow<'a, str>, Captures<'a>)> {
+    fn captures_start<'a>(&self, s: &'a str) -> Option<Captures<'a>> {
-        _consume(self, s, true)
+        let captures = self.captures(s)?;
+        let full_capture = captures.get(0)?;
+        if full_capture.start() != 0 {
+            return None
+        }
+
+        Some(captures)
     }
 
-    fn find_and_consume_capturing<'a>(&self, s: &'a str) -> Option<(Cow<'a, str>, Captures<'a>)> {
+    fn find_start<'a>(&self, s: &'a str) -> Option<Match<'a>> {
-        _consume(self, s, false)
+        let found = self.find(s)?;
+        if found.start() != 0 {
+            return None
+        }
+        Some(found)
     }
 }
-
-fn _consume<'a>(
-    r: &Regex, input: &'a str, 
-    anchor_at_start: bool
-) -> Option<(Cow<'a, str>, Captures<'a>)> {
-    let captures = r.captures(input)?;
-    let full_capture = captures.get(0)?;
-    if anchor_at_start && full_capture.start() != 0 {
-        return None
-    }
-
-    Some((Cow::Borrowed(&input[full_capture.end()..]), captures))
-}

src/string_util.rs

@@ -0,0 +1,36 @@
+use std::borrow::Cow;
+
+/// Strips prefix of given string Cow. Returns option with `Some` if 
+/// prefix found and stripped.
+/// 
+/// Calls `drain` if string is owned and returns slice if string is borrowed
+pub fn strip_cow_prefix<'a>(cow: Cow<'a, str>, prefix: &str) -> Option<Cow<'a, str>> {
+    match cow {
+        Cow::Borrowed(s) => s.strip_prefix(prefix).map(| s | Cow::Borrowed(s)),
+        Cow::Owned(mut s) => {
+            if s.starts_with(prefix) {
+                s.drain(0..prefix.len());                               
+                return Some(Cow::Owned(s));
+            }
+            None
+        }
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use std::borrow::Cow;
+
+    use crate::string_util::strip_cow_prefix;
+
+    #[test]
+    fn test_usage() {
+        let str_to_strip = Cow::Owned("test0:test".to_owned());
+        let stripped = strip_cow_prefix(str_to_strip, "test0");
+        assert_eq!(stripped, Some(Cow::Owned(":test".to_owned())));
+
+        let str_to_strip = Cow::Owned("test:test0".to_owned());
+        let stripped = strip_cow_prefix(str_to_strip, "test0");
+        assert_eq!(stripped, None)
+    }
+}