Request tags, with refusing and limits per tag (closes #175)

Each request gets sorted tags from the lookup, new address lists, the
reputation sources and the rule matches, logged as `tags`. New settings:
`SWW_TAG_LISTS`, `SWW_TAG_DENY`, `SWW_TAG_LIMIT_PERCENT`. They replace
`SWW_ASN_LIMIT_PERCENT`, `SWW_COUNTRY_LIMIT_PERCENT` and
`SWW_DENIED_COUNTRIES`. New heuristics `tag_lists` and `tag_deny`; the
latter runs before the rate limits and again after the Core Rule Set.

Model: opus-5-5

internal/discriminator/biased.go

@@ -17,30 +17,31 @@ type percentage struct {
 }
 
 // biasedThresholdsSet reports whether a biased threshold can lower a
-// client's limits: one of its lists is not empty,
-// SWW_UNKNOWN_LIMIT_PERCENT is below 100, or SWW_ASN_LIMIT_PERCENT_URL
-// is set. The client's lookup is then needed before its request goes on.
+// client's limits by its lookup: SWW_ASN_BYTES_PERCENT or
+// SWW_COUNTRY_BYTES_PERCENT is not empty, SWW_UNKNOWN_LIMIT_PERCENT is
+// below 100, or SWW_ASN_LIMIT_PERCENT_URL is set. The client's lookup is
+// then needed before its request goes on.
 func biasedThresholdsSet(cfg *config.Config) bool {
-	return len(cfg.ASNLimitPercent) > 0 || len(cfg.CountryLimitPercent) > 0 ||
-		len(cfg.ASNBytesPercent) > 0 || len(cfg.CountryBytesPercent) > 0 ||
+	return len(cfg.ASNBytesPercent) > 0 || len(cfg.CountryBytesPercent) > 0 ||
 		cfg.UnknownLimitPercent < whole || cfg.ASNLimitPercentURL != ""
 }
 
 // limitPercentages returns the client's limit percentages, for the rate
-// limits and for the byte limits, by its AS number and country as looked
-// up, each "" when unknown, and the blocklists, DNSBL zones and AbuseIPDB
-// that list it. Each is the lowest of those the settings give it, the
-// first of them in the order below when several are lowest: the
-// percentage SWW_ASN_LIMIT_PERCENT gives its AS number, the one the file
-// SWW_ASN_LIMIT_PERCENT_URL names gives it, the one
-// SWW_COUNTRY_LIMIT_PERCENT gives its country, for a client without a
-// country, unless its request skipped the wait for GeoJS,
-// SWW_UNKNOWN_LIMIT_PERCENT, for a client a blocklist lists, the
-// percentage of SWW_BLOCKLIST_ACTION while it is limit, and for a client
-// a DNSBL zone's verdict lists, or whose AbuseIPDB score is a hit, the
-// percentage of SWW_REPUTATION_ACTION while it is limit. For the byte
-// limits, SWW_ASN_BYTES_PERCENT and SWW_COUNTRY_BYTES_PERCENT take the
-// place of the first three for an AS number or a country they list.
+// limits and for the byte limits, by its tags, its AS number and country
+// as looked up, each "" when unknown, and the blocklists, DNSBL zones and
+// AbuseIPDB that list it. Each is the lowest of those the settings give
+// it, the first of them in the order below when several are lowest: the
+// percentage SWW_TAG_LIMIT_PERCENT gives its asn= tag, the one the file
+// SWW_ASN_LIMIT_PERCENT_URL names gives its AS number, the one
+// SWW_TAG_LIMIT_PERCENT gives its country= tag, the lowest it gives its
+// other tags, for a client without a country, unless its request skipped
+// the wait for GeoJS, SWW_UNKNOWN_LIMIT_PERCENT, for a client a blocklist
+// lists, the percentage of SWW_BLOCKLIST_ACTION while it is limit, and
+// for a client a DNSBL zone's verdict lists, or whose AbuseIPDB score is
+// a hit, the percentage of SWW_REPUTATION_ACTION while it is limit. For
+// the byte limits, SWW_ASN_BYTES_PERCENT and SWW_COUNTRY_BYTES_PERCENT
+// take the place of the first three for an AS number or a country they
+// list.
 func (rq *request) limitPercentages() (percentage, percentage) {
 	cfg := rq.h.config
 	asn, country := rq.line.ASN, rq.line.Country
@@ -65,10 +66,18 @@ func (rq *request) limitPercentages() (percentage, percentage) {
 		reputationListed = percentage{cfg.ReputationLimitPercent, "SWW_REPUTATION_ACTION"}
 	}
 
-	asnRequests := lowest(given(cfg.ASNLimitPercent, asn, "SWW_ASN_LIMIT_PERCENT"),
-		fetched)
-	countryRequests := given(cfg.CountryLimitPercent, country,
-		"SWW_COUNTRY_LIMIT_PERCENT")
+	const tagSetting = "SWW_TAG_LIMIT_PERCENT"
+
+	asnRequests := lowest(given(cfg.TagLimitPercent, asnTag(asn), tagSetting), fetched)
+	countryRequests := given(cfg.TagLimitPercent, countryTag(country), tagSetting)
+
+	otherTags := percentage{percent: whole}
+
+	for _, tag := range rq.line.Tags {
+		if tag != asnTag(asn) && tag != countryTag(country) {
+			otherTags = lowest(otherTags, given(cfg.TagLimitPercent, tag, tagSetting))
+		}
+	}
 
 	asnBytes, countryBytes := asnRequests, countryRequests
 	if _, listed := cfg.ASNBytesPercent[asn]; listed {
@@ -79,12 +88,13 @@ func (rq *request) limitPercentages() (percentage, percentage) {
 		countryBytes = given(cfg.CountryBytesPercent, country, "SWW_COUNTRY_BYTES_PERCENT")
 	}
 
-	return lowest(asnRequests, countryRequests, unknown, blocklisted, reputationListed),
-		lowest(asnBytes, countryBytes, unknown, blocklisted, reputationListed)
+	return lowest(asnRequests, countryRequests, otherTags, unknown, blocklisted,
+			reputationListed),
+		lowest(asnBytes, countryBytes, otherTags, unknown, blocklisted, reputationListed)
 }
 
 // given returns the percentage percents, the setting named setting, gives
-// code, an AS number or a country, or whole when it does not list code.
+// code, an AS number, a country or a tag, or whole when it does not list code.
 func given(percents map[string]int64, code, setting string) percentage {
 	percent, listed := percents[code]
 	if !listed {