build: Update dependencies and increase version

If the URI doesn't contain @ then there can't be a userinfo so
completely ignore this part of parsing to improve performance

.github/workflows/test.yml

@@ -14,10 +14,10 @@ jobs:
       - uses: actions/checkout@v4
       - uses: erlef/setup-beam@v1
         with:
-          otp-version: "27.1.2"
+          otp-version: "28.0.4"
           gleam-version: "1.12.0"
-          rebar3-version: "3"
-          # elixir-version: "1"
+          rebar3-version: "3.25.1"
+          elixir-version: "1.18.4"
       - run: gleam deps download
       - run: gleam test
       - run: gleam format --check src test

CHANGELOG.md

@@ -7,3 +7,8 @@
 ## v2.0.0
 
 - Removed types.Uri. Now gluri uses the stdlib Uri type (and empty)
+
+## v2.0.1
+
+- Improved parsing performance significantly and reduced memory usage up to 50%
+- Significantly improved IPV4 parsing performance

gleam.toml

@@ -1,5 +1,5 @@
 name = "gluri"
-version = "2.0.0"
+version = "2.0.1"
 
 # Fill out these fields if you intend to generate HTML documentation or publish
 # your project to the Hex package manager.
@@ -20,3 +20,4 @@ splitter = ">= 1.1.0 and < 2.0.0"
 [dev-dependencies]
 gleeunit = ">= 1.0.0 and < 2.0.0"
 startest = ">= 0.7.0 and < 1.0.0"
+glychee = ">= 1.1.2 and < 2.0.0"

manifest.toml

@@ -3,8 +3,10 @@
 
 packages = [
   { name = "argv", version = "1.0.2", build_tools = ["gleam"], requirements = [], otp_app = "argv", source = "hex", outer_checksum = "BA1FF0929525DEBA1CE67256E5ADF77A7CDDFE729E3E3F57A5BDCAA031DED09D" },
+  { name = "benchee", version = "1.4.0", build_tools = ["mix"], requirements = ["deep_merge", "statistex", "table"], otp_app = "benchee", source = "hex", outer_checksum = "299CD10DD8CE51C9EA3DDB74BB150F93D25E968F93E4C1FA31698A8E4FA5D715" },
   { name = "bigben", version = "1.0.1", build_tools = ["gleam"], requirements = ["birl", "gleam_erlang", "gleam_otp", "gleam_stdlib"], otp_app = "bigben", source = "hex", outer_checksum = "190E489610A80D76C48BACC75EB8314BD184FF0220AB0F251ABE760B993B91BB" },
   { name = "birl", version = "1.8.0", build_tools = ["gleam"], requirements = ["gleam_regexp", "gleam_stdlib", "ranger"], otp_app = "birl", source = "hex", outer_checksum = "2AC7BA26F998E3DFADDB657148BD5DDFE966958AD4D6D6957DD0D22E5B56C400" },
+  { name = "deep_merge", version = "1.0.0", build_tools = ["mix"], requirements = [], otp_app = "deep_merge", source = "hex", outer_checksum = "CE708E5F094B9CD4E8F2BE4F00D2F4250C4095BE93F8CD6D018C753894885430" },
   { name = "exception", version = "2.1.0", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "exception", source = "hex", outer_checksum = "329D269D5C2A314F7364BD2711372B6F2C58FA6F39981572E5CA68624D291F8C" },
   { name = "filepath", version = "1.1.2", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "filepath", source = "hex", outer_checksum = "B06A9AF0BF10E51401D64B98E4B627F1D2E48C154967DA7AF4D0914780A6D40A" },
   { name = "gleam_community_ansi", version = "1.4.3", build_tools = ["gleam"], requirements = ["gleam_community_colour", "gleam_regexp", "gleam_stdlib"], otp_app = "gleam_community_ansi", source = "hex", outer_checksum = "8A62AE9CC6EA65BEA630D95016D6C07E4F9973565FA3D0DE68DC4200D8E0DD27" },
@@ -14,21 +16,24 @@ packages = [
   { name = "gleam_json", version = "3.0.2", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "gleam_json", source = "hex", outer_checksum = "874FA3C3BB6E22DD2BB111966BD40B3759E9094E05257899A7C08F5DE77EC049" },
   { name = "gleam_otp", version = "1.1.0", build_tools = ["gleam"], requirements = ["gleam_erlang", "gleam_stdlib"], otp_app = "gleam_otp", source = "hex", outer_checksum = "7987CBEBC8060B88F14575DEF546253F3116EBE2A5DA6FD82F38243FCE97C54B" },
   { name = "gleam_regexp", version = "1.1.1", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "gleam_regexp", source = "hex", outer_checksum = "9C215C6CA84A5B35BB934A9B61A9A306EC743153BE2B0425A0D032E477B062A9" },
-  { name = "gleam_stdlib", version = "0.63.0", build_tools = ["gleam"], requirements = [], otp_app = "gleam_stdlib", source = "hex", outer_checksum = "5E216C7D5E8BE22359C9D7DAA2CFBD66039BC12565542F34CD033C5BB57071ED" },
+  { name = "gleam_stdlib", version = "0.63.1", build_tools = ["gleam"], requirements = [], otp_app = "gleam_stdlib", source = "hex", outer_checksum = "E1D5EC07638F606E48F0EA1556044DD805F2ACE9092A6F6AFBE4A0CC4DA21C2F" },
   { name = "gleam_time", version = "1.4.0", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "gleam_time", source = "hex", outer_checksum = "DCDDC040CE97DA3D2A925CDBBA08D8A78681139745754A83998641C8A3F6587E" },
   { name = "gleam_yielder", version = "1.1.0", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "gleam_yielder", source = "hex", outer_checksum = "8E4E4ECFA7982859F430C57F549200C7749823C106759F4A19A78AEA6687717A" },
   { name = "gleeunit", version = "1.6.1", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "gleeunit", source = "hex", outer_checksum = "FDC68A8C492B1E9B429249062CD9BAC9B5538C6FBF584817205D0998C42E1DAC" },
   { name = "glint", version = "1.2.1", build_tools = ["gleam"], requirements = ["gleam_community_ansi", "gleam_community_colour", "gleam_stdlib", "snag"], otp_app = "glint", source = "hex", outer_checksum = "2214C7CEFDE457CEE62140C3D4899B964E05236DA74E4243DFADF4AF29C382BB" },
+  { name = "glychee", version = "1.1.2", build_tools = ["gleam"], requirements = ["benchee"], otp_app = "glychee", source = "hex", outer_checksum = "41784216C213F223095BB3FC3EDDB60CC537835B2340A868EA3931193F7F3824" },
   { name = "ranger", version = "1.4.0", build_tools = ["gleam"], requirements = ["gleam_stdlib", "gleam_yielder"], otp_app = "ranger", source = "hex", outer_checksum = "C8988E8F8CDBD3E7F4D8F2E663EF76490390899C2B2885A6432E942495B3E854" },
   { name = "simplifile", version = "2.3.0", build_tools = ["gleam"], requirements = ["filepath", "gleam_stdlib"], otp_app = "simplifile", source = "hex", outer_checksum = "0A868DAC6063D9E983477981839810DC2E553285AB4588B87E3E9C96A7FB4CB4" },
   { name = "snag", version = "1.1.0", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "snag", source = "hex", outer_checksum = "7E9F06390040EB5FAB392CE642771484136F2EC103A92AE11BA898C8167E6E17" },
   { name = "splitter", version = "1.1.0", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "splitter", source = "hex", outer_checksum = "05564A381580395DCDEFF4F88A64B021E8DAFA6540AE99B4623962F52976AA9D" },
   { name = "startest", version = "0.7.0", build_tools = ["gleam"], requirements = ["argv", "bigben", "birl", "exception", "gleam_community_ansi", "gleam_erlang", "gleam_javascript", "gleam_regexp", "gleam_stdlib", "glint", "simplifile", "tom"], otp_app = "startest", source = "hex", outer_checksum = "71B9CB82C4B8779A4BD54C7151DF7D0B0F778D0DDE805B782B44EFA7BA8F50DA" },
+  { name = "statistex", version = "1.1.0", build_tools = ["mix"], requirements = [], otp_app = "statistex", source = "hex", outer_checksum = "F5950EA26AD43246BA2CCE54324AC394A4E7408FDCF98B8E230F503A0CBA9CF5" },
   { name = "tom", version = "2.0.0", build_tools = ["gleam"], requirements = ["gleam_stdlib", "gleam_time"], otp_app = "tom", source = "hex", outer_checksum = "74D0C5A3761F7A7D06994755D4D5AD854122EF8E9F9F76A3E7547606D8C77091" },
 ]
 
 [requirements]
 gleam_stdlib = { version = ">= 0.44.0 and < 2.0.0" }
 gleeunit = { version = ">= 1.0.0 and < 2.0.0" }
+glychee = { version = ">= 1.1.2 and < 2.0.0" }
 splitter = { version = ">= 1.1.0 and < 2.0.0" }
 startest = { version = ">= 0.7.0 and < 1.0.0" }

src/gluri.gleam

@@ -3,7 +3,7 @@ import gleam/int
 import gleam/list
 import gleam/option.{Some}
 import gleam/string
-import gleam/uri.{type Uri, Uri}
+import gleam/uri.{type Uri}
 import gluri/internal/parser
 import gluri/internal/utils
 

src/gluri/internal/utils.gleam

@@ -7,17 +7,21 @@ import gleam/string
 import gleam/uri.{type Uri, Uri}
 import splitter.{type Splitter}
 
-pub const scheme_port = [
-  #("http", 80),
-  #("https", 443),
-  #("ftp", 21),
-  #("ws", 80),
-  #("wss", 443),
+type Scheme {
+  Scheme(name: String, port: Int)
+}
+
+const scheme_port = [
+  Scheme("http", 80),
+  Scheme("https", 443),
+  Scheme("ftp", 21),
+  Scheme("ws", 80),
+  Scheme("wss", 443),
 ]
 
 pub fn get_port_for_scheme(scheme: String) -> Option(Int) {
-  list.find(scheme_port, fn(sp) { sp.0 == scheme })
-  |> result.map(fn(sp) { sp.1 })
+  list.find(scheme_port, fn(sp) { sp.name == scheme })
+  |> result.map(fn(sp) { sp.port })
   |> option.from_result
 }
 
@@ -82,6 +86,152 @@ fn merge_paths(base: Uri, relative: Uri) -> String {
   }
 }
 
+pub fn try_parsers(
+  over list: List(fn(String) -> Result(#(a, String), Nil)),
+  against static_data: String,
+) -> Result(#(a, String), Nil) {
+  case list {
+    [] -> Error(Nil)
+    [first, ..rest] ->
+      case first(static_data) {
+        Error(_) -> try_parsers(rest, static_data)
+        Ok(r) -> Ok(r)
+      }
+  }
+}
+
+pub fn parse_min_max(
+  str: f,
+  min: Int,
+  max: Int,
+  parse_fn: fn(f) -> Result(#(String, f), g),
+) -> Result(#(String, f), Nil) {
+  do_parse_min_max(str, "", min, max, parse_fn)
+}
+
+pub fn do_parse_min_max(
+  str: d,
+  acc: String,
+  min: Int,
+  max: Int,
+  parse_fn: fn(d) -> Result(#(String, d), e),
+) -> Result(#(String, d), Nil) {
+  case parse_fn(str) {
+    Error(_) -> {
+      case min > 0 {
+        True -> Error(Nil)
+        False -> Ok(#(acc, str))
+      }
+    }
+    Ok(#(l, rest)) -> {
+      case max {
+        1 -> Ok(#(acc <> l, rest))
+        _ -> do_parse_min_max(rest, acc <> l, min - 1, max - 1, parse_fn)
+      }
+    }
+  }
+}
+
+pub fn parse_optional(
+  to_parse str: String,
+  with opt_fn: fn(String) -> Result(#(String, String), Nil),
+) -> #(String, String) {
+  case opt_fn(str) {
+    Error(Nil) -> #("", str)
+    Ok(r) -> r
+  }
+}
+
+pub fn parse_optional_result(
+  to_parse str: String,
+  with opt_fn: fn(String) -> Result(#(String, String), Nil),
+) -> Result(#(String, String), Nil) {
+  parse_optional(str, opt_fn) |> Ok
+}
+
+pub fn parse_this_then(
+  to_parse str: String,
+  with parsers: List(fn(String) -> Result(#(String, String), Nil)),
+) -> Result(#(String, String), Nil) {
+  do_parse_this_then(str, "", parsers)
+}
+
+fn do_parse_this_then(
+  to_parse str: String,
+  from initial: String,
+  with parsers: List(fn(String) -> Result(#(String, String), Nil)),
+) -> Result(#(String, String), Nil) {
+  case parsers {
+    [] -> Ok(#(initial, str))
+    [head, ..tail] -> {
+      case head(str) {
+        Ok(#(res, rest)) -> do_parse_this_then(rest, initial <> res, tail)
+        Error(_) -> Error(Nil)
+      }
+    }
+  }
+}
+
+pub fn parse_multiple(
+  to_parse str: String,
+  with to_run: fn(String) -> Result(#(String, String), Nil),
+) -> Result(#(String, String), Nil) {
+  case do_parse_multiple(str, to_run, "") {
+    Ok(#("", _)) | Error(Nil) -> Error(Nil)
+    Ok(#(r, rest)) -> Ok(#(r, rest))
+  }
+}
+
+fn do_parse_multiple(
+  to_parse str: String,
+  with to_run: fn(String) -> Result(#(String, String), Nil),
+  acc ret: String,
+) -> Result(#(String, String), Nil) {
+  case str {
+    "" -> Ok(#(ret, str))
+    _ ->
+      case to_run(str) {
+        Ok(#(r, rest)) -> do_parse_multiple(rest, to_run, ret <> r)
+        Error(_) -> Ok(#(ret, str))
+      }
+  }
+}
+
+pub fn combine_uris(uris: List(Uri)) -> Uri {
+  list.fold(uris, Uri(None, None, None, None, "", None, None), fn(acc, uri) {
+    let acc = case uri {
+      Uri(Some(scheme), _, _, _, _, _, _) -> Uri(..acc, scheme: Some(scheme))
+      _ -> acc
+    }
+    let acc = case uri {
+      Uri(_, Some(userinfo), _, _, _, _, _) ->
+        Uri(..acc, userinfo: Some(userinfo))
+      _ -> acc
+    }
+    let acc = case uri {
+      Uri(_, _, Some(host), _, _, _, _) -> Uri(..acc, host: Some(host))
+      _ -> acc
+    }
+    let acc = case uri {
+      Uri(_, _, _, Some(port), _, _, _) -> Uri(..acc, port: Some(port))
+      _ -> acc
+    }
+    let acc = case uri {
+      Uri(_, _, _, _, path, _, _) if path != "" -> Uri(..acc, path: path)
+      _ -> acc
+    }
+    let acc = case uri {
+      Uri(_, _, _, _, _, Some(query), _) -> Uri(..acc, query: Some(query))
+      _ -> acc
+    }
+    case uri {
+      Uri(_, _, _, _, _, _, Some(fragment)) ->
+        Uri(..acc, fragment: Some(fragment))
+      _ -> acc
+    }
+  })
+}
+
 pub fn normalise(uri: Uri) -> Uri {
   let percent_splitter = splitter.new(["%"])
   let percent_normaliser = normalise_percent(percent_splitter, _)
@@ -248,6 +398,10 @@ pub fn parse_hex_digit(str) {
   }
 }
 
+pub fn parse_hex_digits(str, min, max) {
+  parse_min_max(str, min, max, parse_hex_digit)
+}
+
 fn encoding_not_needed(i: Int) -> Bool {
   // $-_.+!*'()
   case i {

test/benchmark.gleam

@@ -0,0 +1,81 @@
+import gleam/uri as uri2
+import gluri as uri
+import gluri/internal/parser
+import glychee/benchmark
+import glychee/configuration
+
+@target(erlang)
+pub fn main() {
+  configuration.initialize()
+  configuration.set_pair(configuration.Warmup, 2)
+  configuration.set_pair(configuration.Parallel, 2)
+
+  parse_benchmark()
+  // reg_name_benchmark()
+  // ip_benchmark()
+}
+
+@target(erlang)
+pub fn ip_benchmark() {
+  benchmark.run(
+    [
+      benchmark.Function("ip_benchmark", fn(data) {
+        fn() {
+          let _ = parser.parse_dec_octet(data)
+          Nil
+        }
+      }),
+    ],
+    [
+      benchmark.Data("173", "173"),
+      benchmark.Data("5", "5"),
+      benchmark.Data("200", "200"),
+      benchmark.Data("255", "255"),
+      benchmark.Data("fail", "2b"),
+    ],
+  )
+}
+
+@target(erlang)
+pub fn reg_name_benchmark() {
+  benchmark.run(
+    [
+      benchmark.Function("reg_name_benchmark", fn(data) {
+        fn() {
+          let _ = parser.parse_reg_name(data)
+          Nil
+        }
+      }),
+    ],
+    [
+      benchmark.Data("long", "github.com"),
+    ],
+  )
+}
+
+@target(erlang)
+pub fn parse_benchmark() {
+  benchmark.run(
+    [
+      benchmark.Function("parse_benchmark", fn(data) {
+        fn() {
+          let _ = uri.parse(data)
+          Nil
+        }
+      }),
+      benchmark.Function("stdlib_parse_benchmark", fn(data) {
+        fn() {
+          let _ = uri2.parse(data)
+          Nil
+        }
+      }),
+    ],
+    [
+      benchmark.Data(
+        "long",
+        "https://github.com/gleam-lang/stdlib/issues/523#issuecomment-3288230480",
+      ),
+      benchmark.Data("ipv4", "https://192.255.36.4/"),
+    ],
+  )
+}

test/scratch.gleam

@@ -1,12 +1,20 @@
 // import gleam/result
 
-// import gleam/uri as uri2
+import gleam/uri as uri2
+
 // import splitter
 // import types.{Uri}
 
 import gluri as uri
 
 pub fn main() {
-  uri.parse("http://my_host.com") |> echo
+  // uri.parse("https://192.255.36.4/") |> echo
+  // uri.parse(
+  //   "https://github.com/gleam-lang/stdlib/issues/523#issuecomment-3288230480",
+  // )
+  // |> echo
+  let _ = uri.parse("/abc/def") |> echo
+  let _ = uri2.parse("/abc/def") |> echo
+  let _ = uri.parse("/abc/") |> echo
   Nil
 }

uri.abnf

@@ -90,5 +90,3 @@ RFC 3986                   URI Generic Syntax               January 2005
    gen-delims    = ":" / "/" / "?" / "#" / "[" / "]" / "@"
    sub-delims    = "!" / "$" / "&" / "'" / "(" / ")"
                  / "*" / "+" / "," / ";" / "="
-                 
-                 " # % < > [ \ ] ^ ` { | }