tools/internal/parser: remove workarounds for fixed PSL blocks (publi…

…csuffix#2010)

* tools/internal/parser: update exceptions for cleaned legacy blocks

* tools/internal/parser: remove one-off hacks for fixed blocks

tools/internal/parser/exceptions.go

@@ -341,7 +341,7 @@ var missingEmail = []string{
 		"hashbang.sh",
 	),
 	lines(
-		"// HostyHosting (hostyhosting.com)",
+		"// HostyHosting (https://hostyhosting.com)",
 		"hostyhosting.io",
 	),
 	lines(
@@ -363,7 +363,7 @@ var missingEmail = []string{
 		"torun.pl",
 	),
 	lines(
-		"// TASK geographical domains (www.task.gda.pl/uslugi/dns)",
+		"// TASK geographical domains (https://www.task.gda.pl/uslugi/dns)",
 		"gda.pl",
 		"gdansk.pl",
 		"gdynia.pl",

tools/internal/parser/parser.go

@@ -334,15 +334,6 @@ func splitNameish(line string) (name string, url *url.URL, submitter *mail.Addre
 		}
 	}
 
-	// A single entry uses the unicode fullwidth colon codepoint
-	// (U+FF1A) instead of an ascii colon. Correct that before
-	// attempting a parse.
-	//
-	// TODO: fix the source and delete this hack.
-	if strings.Contains(line, "Future Versatile Group") {
-		line = strings.Replace(line, "\uff1a", ":", -1)
-	}
-
 	name, rest, ok := strings.Cut(line, ":")
 	if !ok {
 		return "", nil, nil
@@ -374,14 +365,6 @@ func splitNameAndURLInParens(line string) (name string, url *url.URL, ok bool) {
 	name = strings.TrimSpace(line[:idx])
 	urlStr := strings.TrimSpace(line[idx+1 : len(line)-1])
 
-	// Two PSL entries omit the scheme at the front of the URL, which
-	// makes them invalid by getURL's standards.
-	//
-	// TODO: fix the source and delete this hack.
-	if urlStr == "www.task.gda.pl/uslugi/dns" || urlStr == "hostyhosting.com" {
-		urlStr = "https://" + urlStr
-	}
-
 	if u := getURL(urlStr); u != nil {
 		return name, u, true
 	}
@@ -440,16 +423,6 @@ func getSubmitter(line string) *mail.Address {
 		return addr
 	}
 
-	// One current entry is missing the closing chevron on the email,
-	// which makes it an invalid address.
-	//
-	// TODO: fix the source and delete this hack.
-	if strings.HasSuffix(line, "torproject.org") {
-		if addr, err := mail.ParseAddress(line + ">"); err == nil {
-			return addr
-		}
-	}
-
 	// One current entry uses old school email obfuscation to foil
 	// spam bots, which makes it an invalid address.
 	//