Improve URL regex

(cherry picked from commit d2bbdd2)
ref #18066

base/libgit2/callbacks.jl

@@ -209,12 +209,10 @@ function credentials_callback(libgit2credptr::Ptr{Ptr{Void}}, url_ptr::Cstring,
     url = unsafe_string(url_ptr)
 
     # parse url for schema and host
-    urlparts = match(urlmatcher, url)
-    schema = urlparts.captures[1]
-    urlusername = urlparts.captures[4]
-    urlusername = urlusername === nothing ? "" : String(urlusername)
-    host = urlparts.captures[5]
-    schema = schema === nothing ? "" : schema*"://"
+    urlparts = match(URL_REGEX, url)
+    schema = urlparts[:scheme] === nothing ? "" : urlparts[:scheme] * "://"
+    urlusername = urlparts[:user] === nothing ? "" : urlparts[:user]
+    host = urlparts[:host]
 
     # get credentials object from payload pointer
     @assert payload_ptr != C_NULL

base/libgit2/utils.jl

@@ -1,6 +1,12 @@
 # This file is a part of Julia. License is MIT: http://julialang.org/license
 
-const urlmatcher = r"^(http[s]?|git|ssh)?(:\/\/)?((\w+)@)?([A-Za-z0-9\-\.]+)(:[0-9]+)?(.*)$"
+const URL_REGEX = r"""
+^(?:(?<scheme>https?|git|ssh)\:\/\/)?
+(?:(?<user>.*?)(?:\:(?<password>.*?))?@)?
+(?<host>[A-Za-z0-9\-\.]+)
+(?:\:(?<port>\d+)?)?
+(?<path>.*?)$
+"""x
 
 function version()
     major = Ref{Cint}(0)

test/libgit2.jl

@@ -76,6 +76,39 @@ const LIBGIT2_MIN_VER = v"0.23.0"
     @test sig3.email == sig.email
 #end
 
+#@testset "URL parsing" begin
+    # Use all named group
+    m = match(LibGit2.URL_REGEX, "https://user:pass@hostname.com:80/path/repo.git")
+    @test m[:scheme] == "https"
+    @test m[:user] == "user"
+    @test m[:password] == "pass"
+    @test m[:host] == "hostname.com"
+    @test m[:port] == "80"
+    @test m[:path] == "/path/repo.git"
+
+    # Realistic example HTTP example
+    m = match(LibGit2.URL_REGEX, "https://github.com/JuliaLang/Example.jl.git")
+    @test m[:scheme] == "https"
+    @test m[:user] == nothing
+    @test m[:password] == nothing
+    @test m[:host] == "github.com"
+    @test m[:port] == nothing
+    @test m[:path] == "/JuliaLang/Example.jl.git"
+
+    # Realistic example SSH example
+    m = match(LibGit2.URL_REGEX, "git@github.com:JuliaLang/Example.jl.git")
+    @test m[:scheme] == nothing
+    @test m[:user] == "git"
+    @test m[:password] == nothing
+    @test m[:host] == "github.com"
+    @test m[:port] == nothing
+    @test m[:path] == "JuliaLang/Example.jl.git"
+
+    # Make sure that a username can contain special characters
+    m = match(LibGit2.URL_REGEX, "user-name@hostname.com")
+    @test m[:user] == "user-name"
+#end
+
 mktempdir() do dir
     # test parameters
     repo_url = "https://github.com/JuliaLang/Example.jl"