5 Commits

6 changed files with 76 additions and 44 deletions

View File

@@ -14,7 +14,7 @@ jobs:
- uses: actions/checkout@v4
- uses: erlef/setup-beam@v1
with:
otp-version: "28.0.4"
otp-version: "28.1"
gleam-version: "1.12.0"
rebar3-version: "3.25.1"
elixir-version: "1.18.4"

View File

@@ -12,3 +12,7 @@
- Improved parsing performance significantly and reduced memory usage up to 50%
- Significantly improved IPV4 parsing performance
## v2.0.2
- Minor performance improvement for uris with userinfo

View File

@@ -16,7 +16,7 @@ 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 = "0.63.1", build_tools = ["gleam"], requirements = [], otp_app = "gleam_stdlib", source = "hex", outer_checksum = "E1D5EC07638F606E48F0EA1556044DD805F2ACE9092A6F6AFBE4A0CC4DA21C2F" },
{ name = "gleam_stdlib", version = "0.63.2", build_tools = ["gleam"], requirements = [], otp_app = "gleam_stdlib", source = "hex", outer_checksum = "962B25C667DA07F4CAB32001F44D3C41C1A89E58E3BBA54F183B482CF6122150" },
{ 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" },

View File

@@ -168,7 +168,7 @@ fn parse_authority(str: String) -> Result(#(Uri, String), Nil) {
}
fn parse_authority_part(str: String) -> Result(#(Uri, String), Nil) {
let #(userinfo, rest) = parse_userinfo(str, "")
let #(userinfo, rest) = parse_userinfo(str)
use #(host, rest) <- result.try(parse_host(rest))
@@ -180,8 +180,12 @@ fn parse_authority_part(str: String) -> Result(#(Uri, String), Nil) {
}
// userinfo = *( unreserved / pct-encoded / sub-delims / ":" )
fn parse_userinfo(str: String, userinfo: String) -> #(Option(String), String) {
fn parse_userinfo(str: String) -> #(Option(String), String) {
use <- bool.guard(when: !string.contains(str, "@"), return: #(None, str))
do_parse_userinfo(str, "")
}
fn do_parse_userinfo(str: String, userinfo: String) -> #(Option(String), String) {
case str {
"@" <> rest -> #(Some(userinfo), rest)
"" -> #(None, userinfo <> str)
@@ -202,7 +206,7 @@ fn parse_userinfo(str: String, userinfo: String) -> #(Option(String), String) {
str,
)
{
Ok(#(part, rest)) -> parse_userinfo(rest, userinfo <> part)
Ok(#(part, rest)) -> do_parse_userinfo(rest, userinfo <> part)
Error(_) -> #(None, userinfo <> str)
}
}
@@ -421,7 +425,7 @@ fn parse_ipv4address(str: String) {
// / "1" 2DIGIT ; 100-199
// / "2" %x30-34 DIGIT ; 200-249
// / "25" %x30-35 ; 250-255
pub fn parse_dec_octet(str: String) -> Result(#(String, String), Nil) {
fn parse_dec_octet(str: String) -> Result(#(String, String), Nil) {
try_parsers(
[
parse_this_then(_, [
@@ -486,7 +490,7 @@ pub fn parse_dec_octet(str: String) -> Result(#(String, String), Nil) {
}
// reg-name = *( unreserved / pct-encoded / sub-delims )
pub fn parse_reg_name(str: String) {
fn parse_reg_name(str: String) {
// can't error
case do_parse_reg_name(str, "") {

View File

@@ -1,6 +1,5 @@
import gleam/uri as uri2
import gluri as uri
import gluri/internal/parser
import glychee/benchmark
import glychee/configuration
@@ -15,43 +14,43 @@ pub fn main() {
// 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 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 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() {
@@ -75,6 +74,10 @@ pub fn parse_benchmark() {
"long",
"https://github.com/gleam-lang/stdlib/issues/523#issuecomment-3288230480",
),
benchmark.Data(
"with user",
"https://test_name:user%20$$$@github.com/gleam-lang/stdlib/issues/523#issuecomment-3288230480",
),
benchmark.Data("ipv4", "https://192.255.36.4/"),
],
)

View File

@@ -9,6 +9,27 @@ pub fn main() {
startest.run(startest.default_config())
}
pub fn parse_general_tests() {
describe("general parsing", [
it("mailto parsing", fn() {
uri.parse("mailto:Joe@example.com")
|> should.equal(Ok(
Uri(..empty, scheme: Some("mailto"), path: "Joe@example.com"),
))
uri.parse("mailto:Joe@example.com?hello#bye")
|> should.equal(Ok(
Uri(
..empty,
scheme: Some("mailto"),
path: "Joe@example.com",
query: Some("hello"),
fragment: Some("bye"),
),
))
}),
])
}
pub fn parse_scheme_tests() {
describe("scheme parsing", [
it("simple parse", fn() {