From 83136d715803d5fb8d8d5de1aa04505d2b5746b2 Mon Sep 17 00:00:00 2001
From: mykola2312 <nikolay_tihonov@i.ua>
Date: Sat, 24 Feb 2024 15:20:38 +0200
Subject: [PATCH] write function to parse url via url crate

---
 Cargo.lock          | 17 +++++++++--------
 Cargo.toml          |  1 +
 src/bot/sanitize.rs | 18 +++++++++++++++---
 3 files changed, 25 insertions(+), 11 deletions(-)

diff --git a/Cargo.lock b/Cargo.lock
index 1e11d37..d07e0f3 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -294,9 +294,9 @@ checksum = "00b0228411908ca8685dba7fc2cdd70ec9990a6e753e89b6ac91a84c40fbaf4b"
 
 [[package]]
 name = "form_urlencoded"
-version = "1.2.0"
+version = "1.2.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a62bc1cf6f830c2ec14a513a9fb124d0a213a629668a4186f329db21fe045652"
+checksum = "e13624c2627564efccf4934284bdd98cbaa14e79b0b5a141218e507b3a823456"
 dependencies = [
  "percent-encoding",
 ]
@@ -523,9 +523,9 @@ checksum = "b9e0384b61958566e926dc50660321d12159025e767c18e043daf26b70104c39"
 
 [[package]]
 name = "idna"
-version = "0.4.0"
+version = "0.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7d20d6b07bfbc108882d88ed8e37d39636dcc260e15e30c45e6ba089610b917c"
+checksum = "634d9b1461af396cad843f47fdba5597a4f9e6ddd4bfb6ff5d85028c25cb12f6"
 dependencies = [
  "unicode-bidi",
  "unicode-normalization",
@@ -668,6 +668,7 @@ dependencies = [
  "serde_json",
  "teloxide",
  "tokio",
+ "url",
 ]
 
 [[package]]
@@ -777,9 +778,9 @@ dependencies = [
 
 [[package]]
 name = "percent-encoding"
-version = "2.3.0"
+version = "2.3.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9b2a4787296e9989611394c33f193f676704af1686e70b8f8033ab5ba9a35a94"
+checksum = "e3148f5046208a5d56bcfc03053e3ca6334e51da8dfb19b6cdc8b306fae3283e"
 
 [[package]]
 name = "pin-project"
@@ -1423,9 +1424,9 @@ dependencies = [
 
 [[package]]
 name = "url"
-version = "2.4.1"
+version = "2.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "143b538f18257fac9cad154828a57c6bf5157e1aa604d4816b5995bf6de87ae5"
+checksum = "31e6302e3bb753d46e83516cae55ae196fc0c309407cf11ab35cc51a4c2a4633"
 dependencies = [
  "form_urlencoded",
  "idna",
diff --git a/Cargo.toml b/Cargo.toml
index 34f5b2f..a072f31 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -14,3 +14,4 @@ serde = { version = "1.0.196", features = ["derive"] }
 serde_json = "1.0.113"
 ordered-float = "4.2.0"
 regex = "1.10.3"
+url = "2.5.0"
diff --git a/src/bot/sanitize.rs b/src/bot/sanitize.rs
index ca47400..7954af7 100644
--- a/src/bot/sanitize.rs
+++ b/src/bot/sanitize.rs
@@ -1,4 +1,5 @@
 use regex::Regex;
+use url::Url;
 
 // https://stackoverflow.com/questions/6038061/regular-expression-to-find-urls-within-a-string
 const RE_URL: &str =
@@ -12,16 +13,27 @@ pub fn extract_url(text: &str) -> Option<&str> {
     }
 }
 
+pub fn parse_url(url: &str) -> Option<Url> {
+    Url::parse(url).ok()
+}
+
 #[cfg(test)]
 mod tests {
-    use crate::bot::sanitize::extract_url;
+    use crate::bot::sanitize::{extract_url, parse_url};
 
     #[test]
     fn test_extract_url() {
-        // https://www.youtube.com/watch?v=00000000000
-
         assert_eq!(extract_url("test http://www.test.com/id/1"), Some("http://www.test.com/id/1"));
         assert_eq!(extract_url("https://www.test.com 3"), Some("https://www.test.com"));
         assert_eq!(extract_url("there is no any url"), None);
     }
+
+    #[test]
+    fn test_parse_url() {
+        let url = parse_url("https://www.youtube.com/watch?v=00000000000").unwrap();
+        assert_eq!(url.host_str().unwrap(), "www.youtube.com");
+
+        let url = parse_url("https://youtu.be/00000000000").unwrap();
+        assert_eq!(url.host_str().unwrap(), "youtu.be");
+    }
 }
\ No newline at end of file