Take the Tor exit list from Onionoo, which has the IPv6 exits (closes #234)

check.torproject.org lists IPv4 exits only, so is_tor was false for an IPv6
exit. tor_exits now comes from Onionoo's details document, read by a new
format that takes every address in exit_addresses and or_addresses.

Model: opus-5-5

internal/iplist/parse.go

@@ -6,6 +6,7 @@ import (
 	"errors"
 	"fmt"
 	"io"
+	"net"
 	"net/netip"
 	"strings"
 
@@ -32,12 +33,32 @@ var (
 	errUnknownFormat = errors.New("unknown list format")
 	errNoNetwork     = errors.New("no network on the line")
 	errBadRange      = errors.New("first and last address make no range")
+	errNoAddresses   = errors.New("no relay addresses in the document")
 )
 
 // parse reads a list file written in format and returns the addresses it
-// holds. A line it cannot read fails the whole file, so a file that is not
+// holds. Anything it cannot read fails the whole file, so a file that is not
 // the list, such as an error page, never replaces a good copy.
 func parse(r io.Reader, format Format) (*netipx.IPSet, error) {
+	switch format {
+	case FormatLines:
+		return parseLines(r, addAddressOrNetwork)
+	case FormatSpamhausJSON:
+		return parseLines(r, addSpamhausNetwork)
+	case FormatDShield:
+		return parseLines(r, addDShieldNetwork)
+	case FormatOnionooJSON:
+		return parseOnionoo(r)
+	default:
+		return nil, errUnknownFormat
+	}
+}
+
+// parseLines reads a list file line by line, skipping blank lines and
+// comments, and calls addLine with each other line.
+func parseLines(
+	r io.Reader, addLine func(*netipx.IPSetBuilder, string) error,
+) (*netipx.IPSet, error) {
 	var builder netipx.IPSetBuilder
 
 	scanner := bufio.NewScanner(r)
@@ -48,7 +69,7 @@ func parse(r io.Reader, format Format) (*netipx.IPSet, error) {
 			continue
 		}
 
-		err := addLine(&builder, line, format)
+		err := addLine(&builder, line)
 		if err != nil {
 			return nil, fmt.Errorf("line %d: %w", lineNumber, err)
 		}
@@ -62,20 +83,6 @@ func parse(r io.Reader, format Format) (*netipx.IPSet, error) {
 	return builder.IPSet()
 }
 
-// addLine adds the address or network on one line of a list file to builder.
-func addLine(builder *netipx.IPSetBuilder, line string, format Format) error {
-	switch format {
-	case FormatLines:
-		return addAddressOrNetwork(builder, line)
-	case FormatSpamhausJSON:
-		return addSpamhausNetwork(builder, line)
-	case FormatDShield:
-		return addDShieldNetwork(builder, line)
-	default:
-		return errUnknownFormat
-	}
-}
-
 // addAddressOrNetwork adds an address, such as 192.0.2.1, or a network, such
 // as 192.0.2.0/24, to builder.
 func addAddressOrNetwork(builder *netipx.IPSetBuilder, text string) error {
@@ -156,3 +163,79 @@ func addDShieldNetwork(builder *netipx.IPSetBuilder, line string) error {
 
 	return nil
 }
+
+// parseOnionoo returns every address in the exit_addresses and or_addresses
+// fields of the relays in an Onionoo details document. A document with no
+// addresses fails, because the Tor network always has relays.
+func parseOnionoo(r io.Reader) (*netipx.IPSet, error) {
+	data, err := io.ReadAll(r)
+	if err != nil {
+		return nil, err
+	}
+
+	//nolint:tagliatelle // Onionoo's own field names
+	var details struct {
+		Relays []struct {
+			ExitAddresses []string `json:"exit_addresses"`
+			ORAddresses   []string `json:"or_addresses"`
+		} `json:"relays"`
+	}
+
+	err = json.Unmarshal(data, &details)
+	if err != nil {
+		return nil, err
+	}
+
+	var builder netipx.IPSetBuilder
+
+	for _, relay := range details.Relays {
+		for _, text := range relay.ExitAddresses {
+			err = addAddress(&builder, text)
+			if err != nil {
+				return nil, err
+			}
+		}
+
+		for _, text := range relay.ORAddresses {
+			err = addAddressWithPort(&builder, text)
+			if err != nil {
+				return nil, err
+			}
+		}
+	}
+
+	addresses, err := builder.IPSet()
+	if err != nil {
+		return nil, err
+	}
+
+	if len(addresses.Ranges()) == 0 {
+		return nil, errNoAddresses
+	}
+
+	return addresses, nil
+}
+
+// addAddress adds an address, such as 192.0.2.1, to builder.
+func addAddress(builder *netipx.IPSetBuilder, text string) error {
+	addr, err := netip.ParseAddr(text)
+	if err != nil {
+		return err
+	}
+
+	builder.Add(addr)
+
+	return nil
+}
+
+// addAddressWithPort adds the address of text, such as 192.0.2.1:443 or
+// [2001:db8::1]:443, to builder. Onionoo allows a list of ports after the
+// address, so the part after the address is not read.
+func addAddressWithPort(builder *netipx.IPSetBuilder, text string) error {
+	host, _, err := net.SplitHostPort(text)
+	if err != nil {
+		return err
+	}
+
+	return addAddress(builder, host)
+}