improve port parsing to be more resilient against trailing characters #31232
Conversation
|
@dotnet-bot test Outerloop Windows x64 Debug Build |
| { | ||
| host = value.Substring(0, separatorIndex); | ||
| if (!ushort.TryParse(value.AsSpan(separatorIndex + 1), out port)) | ||
| int endIndex = separatorIndex + 1; |
There was a problem hiding this comment.
Why does this string parsing code exist at all? Why not just convert the string to a URI and then extract the host and port portions of it?
There was a problem hiding this comment.
It was like that at some point. It did not handle well cases when the value is for example just IP (or ip + port) The other issue was handling special characters in password. So we end up breaking it to fragments on our own.
| host = value.Substring(0, separatorIndex); | ||
| if (!ushort.TryParse(value.AsSpan(separatorIndex + 1), out port)) | ||
| int endIndex = separatorIndex + 1; | ||
| // Strip any trailing characters after port num,ber. |
| break; | ||
| } | ||
| endIndex += 1; | ||
| } |
There was a problem hiding this comment.
add newline after this brace.
| // Strip any trailing characters after port num,ber. | ||
| while (endIndex < value.Length) | ||
| { | ||
| if (!char.IsDigit(value[endIndex])) |
There was a problem hiding this comment.
Note that char.IsDigit returns true for things other than '0'-'9'. It may not matter here other than for very esoteric cases.
…dotnet#31232) improve port parsing to be more resilient against trailing characters
…dotnet/corefx#31232) improve port parsing to be more resilient against trailing characters Commit migrated from dotnet/corefx@56d177d
Primarily covers 'http://10.30.62.64:7890/' accepted by curl.
Fixes #31326