[processor/geoip] Add maxmind geoprovider

.chloggen/add_maxmind_geoprovider.yaml

@@ -0,0 +1,27 @@
+# Use this changelog template to create an entry for release notes.
+
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: enhancement
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: geoipprocessor
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Add MaxMind geoip provider for GeoIP2-City and GeoLite2-City databases.
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [32663]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext:
+
+# If your change doesn't affect end users or the exported elements of any package,
+# you should instead start your pull request title with [chore] or use the "Skip Changelog" label.
+# Optional: The change log or logs in which this entry should be included.
+# e.g. '[user]' or '[user, api]'
+# Include 'user' if the change is relevant to end users.
+# Include 'api' if there is a change to a library API.
+# Default: '[user]'
+change_logs: []

processor/geoipprocessor/geoip_processor.go

@@ -6,6 +6,7 @@ package geoipprocessor // import "github.com/open-telemetry/opentelemetry-collec
 import (
 	"context"
 	"errors"
+	"fmt"
 	"net"
 
 	"go.opentelemetry.io/collector/pdata/pcommon"
@@ -17,7 +18,11 @@ import (
 	"github.com/open-telemetry/opentelemetry-collector-contrib/processor/geoipprocessor/internal/provider"
 )
 
-var errIPNotFound = errors.New("no IP address found in the resource attributes")
+var (
+	errIPNotFound    = errors.New("no IP address found in the resource attributes")
+	errParseIP       = errors.New("could not parse IP address")
+	errUnspecifiedIP = errors.New("unspecified address")
+)
 
 // newGeoIPProcessor creates a new instance of geoIPProcessor with the specified fields.
 type geoIPProcessor struct {
@@ -31,15 +36,26 @@ func newGeoIPProcessor(resourceAttributes []attribute.Key) *geoIPProcessor {
 	}
 }
 
+// parseIP parses a string to a net.IP type and returns an error if the IP is invalid or unspecified.
+func parseIP(strIP string) (net.IP, error) {
+	ip := net.ParseIP(strIP)
+	if ip == nil {
+		return nil, fmt.Errorf("%w address: %s", errParseIP, strIP)
+	} else if ip.IsUnspecified() {
+		return nil, fmt.Errorf("%w address: %s", errUnspecifiedIP, strIP)
+	}
+	return ip, nil
+}
+
 // ipFromResourceAttributes extracts an IP address from the given resource's attributes based on the specified fields.
 // It returns the first IP address if found, or an error if no valid IP address is found.
 func ipFromResourceAttributes(attributes []attribute.Key, resource pcommon.Resource) (net.IP, error) {
 	for _, attr := range attributes {
 		if ipField, found := resource.Attributes().Get(string(attr)); found {
-			ipAttribute := net.ParseIP(ipField.AsString())
 			// The attribute might contain a domain name. Skip any net.ParseIP error until we have a fine-grained error propagation strategy.
 			// TODO: propagate an error once error_mode configuration option is available (e.g. transformprocessor)
-			if ipAttribute != nil {
+			ipAttribute, err := parseIP(ipField.AsString())
+			if err == nil && ipAttribute != nil {
 				return ipAttribute, nil
 			}
 		}

processor/geoipprocessor/geoip_processor_test.go

@@ -131,6 +131,23 @@ func TestProcessPdata(t *testing.T) {
 				}),
 			},
 		},
+		{
+			name:               "default source.ip attribute with an unspecified IP address should be skipped",
+			resourceAttributes: defaultResourceAttributes,
+			initResourceAttributes: []generateResourceFunc{
+				withAttributes([]attribute.KeyValue{
+					attribute.String(string(semconv.SourceAddressKey), "0.0.0.0"),
+				}),
+			},
+			geoLocationMock: func(context.Context, net.IP) (attribute.Set, error) {
+				return attribute.NewSet([]attribute.KeyValue{attribute.String("geo.city_name", "barcelona")}...), nil
+			},
+			expectedResourceAttributes: []generateResourceFunc{
+				withAttributes([]attribute.KeyValue{
+					attribute.String(string(semconv.SourceAddressKey), "0.0.0.0"),
+				}),
+			},
+		},
 		{
 			name:               "custom resource attribute",
 			resourceAttributes: []attribute.Key{"ip"},

processor/geoipprocessor/go.mod

@@ -28,11 +28,15 @@ require (
 	github.com/knadh/koanf/maps v0.1.1 // indirect
 	github.com/knadh/koanf/providers/confmap v0.1.0 // indirect
 	github.com/knadh/koanf/v2 v2.1.1 // indirect
+	github.com/maxmind/MaxMind-DB v0.0.0-20240605211347-880f6b4b5eb6
+	github.com/maxmind/mmdbwriter v1.0.0 // indirect
 	github.com/mitchellh/copystructure v1.2.0 // indirect
 	github.com/mitchellh/reflectwalk v1.0.2 // indirect
 	github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd // indirect
 	github.com/modern-go/reflect2 v1.0.2 // indirect
 	github.com/open-telemetry/opentelemetry-collector-contrib/pkg/pdatautil v0.103.0 // indirect
+	github.com/oschwald/geoip2-golang v1.11.0
+	github.com/oschwald/maxminddb-golang v1.13.0 // indirect
 	github.com/pmezard/go-difflib v1.0.0 // indirect
 	github.com/prometheus/client_golang v1.19.1 // indirect
 	github.com/prometheus/client_model v0.6.1 // indirect
@@ -49,6 +53,7 @@ require (
 	go.opentelemetry.io/otel/trace v1.27.0 // indirect
 	go.uber.org/multierr v1.11.0 // indirect
 	go.uber.org/zap v1.27.0 // indirect
+	go4.org/netipx v0.0.0-20230824141953-6213f710f925 // indirect
 	golang.org/x/net v0.25.0 // indirect
 	golang.org/x/sys v0.20.0 // indirect
 	golang.org/x/text v0.15.0 // indirect

processor/geoipprocessor/internal/convention/attributes.go

@@ -0,0 +1,40 @@
+// Copyright The OpenTelemetry Authors
+// SPDX-License-Identifier: Apache-2.0
+
+package conventions // import "github.com/open-telemetry/opentelemetry-collector-contrib/processor/geoipprocessor/internal/convention"
+
+// TODO: replace for semconv once https://github.com/open-telemetry/semantic-conventions/issues/1033 is closed.
+const (
+	// AttributeGeoCityName represents the attribute name for the city name in geographical data.
+	AttributeGeoCityName = "geo.city_name"
+
+	// AttributeGeoPostalCode represents the attribute name for the city postal code.
+	AttributeGeoPostalCode = "geo.postal_code"
+
+	// AttributeGeoCountryName represents the attribute name for the country name in geographical data.
+	AttributeGeoCountryName = "geo.country_name"
+
+	// AttributeGeoCountryIsoCode represents the attribute name for the Two-letter ISO Country Code.
+	AttributeGeoCountryIsoCode = "geo.country_iso_code"
+
+	// AttributeGeoContinentName represents the attribute name for the continent name in geographical data.
+	AttributeGeoContinentName = "geo.continent_name"
+
+	// AttributeGeoContinentIsoCode represents the attribute name for the Two-letter Continent Code.
+	AttributeGeoContinentCode = "geo.continent_code"
+
+	// AttributeGeoRegionName represents the attribute name for the region name in geographical data.
+	AttributeGeoRegionName = "geo.region_name"
+
+	// AttributeGeoRegionIsoCode represents the attribute name for the Two-letter ISO Region Code.
+	AttributeGeoRegionIsoCode = "geo.region_iso_code"
+
+	// AttributeGeoTimezone represents the attribute name for the timezone.
+	AttributeGeoTimezone = "geo.timezone"
+
+	// AttributeGeoLocationLat represents the attribute name for the latitude.
+	AttributeGeoLocationLat = "geo.location.lat"
+
+	// AttributeGeoLocationLon represents the attribute name for the longitude.
+	AttributeGeoLocationLon = "geo.location.lon"
+)

processor/geoipprocessor/internal/provider/geoipprovider.go

@@ -7,9 +7,15 @@ import (
 	"context"
 	"net"
 
+	"go.opentelemetry.io/collector/component"
 	"go.opentelemetry.io/otel/attribute"
 )
 
+// Config is the configuration of a GeoIPProvider.
+type Config interface {
+	component.ConfigValidator
+}
+
 // GeoIPProvider defines methods for obtaining the geographical location based on the provided IP address.
 type GeoIPProvider interface {
 	// Location returns a set of attributes representing the geographical location for the given IP address. It requires a context for managing request lifetime.

processor/geoipprocessor/internal/provider/maxmindprovider/README.md

@@ -0,0 +1,14 @@
+# MaxMind GeoIP Provider
+
+This package provides a MaxMind GeoIP provider for use with the OpenTelemetry GeoIP processor. It leverages the [geoip2-golang package](https://github.com/oschwald/geoip2-golang) to query geographical information associated with IP addresses from MaxMind databases. See recommended clients: https://dev.maxmind.com/geoip/docs/databases#api-clients
+
+# Features
+
+- Supports GeoIP2-City and GeoLite2-City database types.
+- Retrieves and returns geographical metadata for a given IP address. The generated attributes follow the internal [Geo conventions](../../convention/attributes.go).
+
+## Configuration
+
+The following configuration must be provided:
+
+- `database_path`: local file path to a GeoIP2-City or GeoLite2-City database.

processor/geoipprocessor/internal/provider/maxmindprovider/config.go

@@ -0,0 +1,27 @@
+// Copyright The OpenTelemetry Authors
+// SPDX-License-Identifier: Apache-2.0
+
+package maxmind // import "github.com/open-telemetry/opentelemetry-collector-contrib/processor/geoipprocessor/internal/provider/maxmindprovider"
+
+import (
+	"errors"
+
+	"github.com/open-telemetry/opentelemetry-collector-contrib/processor/geoipprocessor/internal/provider"
+)
+
+// Config defines configuration for MaxMind provider.
+type Config struct {
+	// DatabasePath section allows specifying a local GeoIP database
+	// file to retrieve the geographical metadata from.
+	DatabasePath string `mapstructure:"database_path"`
+}
+
+var _ provider.Config = (*Config)(nil)
+
+// Validate implements provider.Config.
+func (c *Config) Validate() error {
+	if c.DatabasePath == "" {
+		return errors.New("a local geoIP database path must be provided")
+	}
+	return nil
+}

processor/geoipprocessor/internal/provider/maxmindprovider/provider.go

@@ -0,0 +1,103 @@
+// Copyright The OpenTelemetry Authors
+// SPDX-License-Identifier: Apache-2.0
+
+package maxmind // import "github.com/open-telemetry/opentelemetry-collector-contrib/processor/geoipprocessor/internal/provider/maxmindprovider"
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"net"
+
+	"github.com/oschwald/geoip2-golang"
+	"go.opentelemetry.io/otel/attribute"
+
+	conventions "github.com/open-telemetry/opentelemetry-collector-contrib/processor/geoipprocessor/internal/convention"
+	"github.com/open-telemetry/opentelemetry-collector-contrib/processor/geoipprocessor/internal/provider"
+)
+
+var (
+	// defaultLanguageCode specifies English as the default Geolocation language code, see https://dev.maxmind.com/geoip/docs/web-services/responses#languages
+	defaultLanguageCode = "en"
+	geoIP2CityDBType    = "GeoIP2-City"
+	geoLite2CityDBType  = "GeoLite2-City"
+
+	errUnsupportedDB   = errors.New("unsupported geo IP database type")
+	errNoMetadataFound = errors.New("no geo IP metadata found")
+)
+
+type maxMindProvider struct {
+	geoReader *geoip2.Reader
+	// language code to be used in name retrieval, e.g. "en" or "pt-BR"
+	langCode string
+}
+
+var _ provider.GeoIPProvider = (*maxMindProvider)(nil)
+
+func newMaxMindProvider(cfg *Config) (*maxMindProvider, error) {
+	geoReader, err := geoip2.Open(cfg.DatabasePath)
+	if err != nil {
+		return nil, fmt.Errorf("could not open geoip database: %w", err)
+	}
+
+	return &maxMindProvider{geoReader: geoReader, langCode: defaultLanguageCode}, nil
+}
+
+// Location implements provider.GeoIPProvider for MaxMind. If a non City database type is used or no metadata is found in the database, an error will be returned.
+func (g *maxMindProvider) Location(_ context.Context, ipAddress net.IP) (attribute.Set, error) {
+	switch g.geoReader.Metadata().DatabaseType {
+	case geoIP2CityDBType, geoLite2CityDBType:
+		attrs, err := g.cityAttributes(ipAddress)
+		if err != nil {
+			return attribute.Set{}, err
+		} else if len(*attrs) == 0 {
+			return attribute.Set{}, errNoMetadataFound
+		}
+		return attribute.NewSet(*attrs...), nil
+	default:
+		return attribute.Set{}, fmt.Errorf("%w type: %s", errUnsupportedDB, g.geoReader.Metadata().DatabaseType)
+	}
+}
+
+// cityAttributes returns a list of key-values containing geographical metadata associated to the provided IP. The key names are populated using the internal geo IP conventions package. If the an invalid or nil IP is provided, an error is returned.
+func (g *maxMindProvider) cityAttributes(ipAddress net.IP) (*[]attribute.KeyValue, error) {
+	attributes := make([]attribute.KeyValue, 0, 11)
+
+	city, err := g.geoReader.City(ipAddress)
+	if err != nil {
+		return nil, err
+	}
+
+	// The exact set of top-level keys varies based on the particular GeoIP2 web service you are using. If a key maps to an undefined or empty value, it is not included in the JSON object. The following anonymous function appends the given key-value only if the value is not empty.
+	appendIfNotEmpty := func(keyName, value string) {
+		if value != "" {
+			attributes = append(attributes, attribute.String(keyName, value))
+		}
+	}
+
+	// city
+	appendIfNotEmpty(conventions.AttributeGeoCityName, city.City.Names[g.langCode])
+	// country
+	appendIfNotEmpty(conventions.AttributeGeoCountryName, city.Country.Names[g.langCode])
+	appendIfNotEmpty(conventions.AttributeGeoCountryIsoCode, city.Country.IsoCode)
+	// continent
+	appendIfNotEmpty(conventions.AttributeGeoContinentName, city.Continent.Names[g.langCode])
+	appendIfNotEmpty(conventions.AttributeGeoContinentCode, city.Continent.Code)
+	// postal code
+	appendIfNotEmpty(conventions.AttributeGeoPostalCode, city.Postal.Code)
+	// region
+	if len(city.Subdivisions) > 0 {
+		// The most specific subdivision is located at the last array position, see https://github.com/maxmind/GeoIP2-java/blob/2fe4c65424fed2c3c2449e5530381b6452b0560f/src/main/java/com/maxmind/geoip2/model/AbstractCityResponse.java#L112
+		mostSpecificSubdivision := city.Subdivisions[len(city.Subdivisions)-1]
+		appendIfNotEmpty(conventions.AttributeGeoRegionName, mostSpecificSubdivision.Names[g.langCode])
+		appendIfNotEmpty(conventions.AttributeGeoRegionIsoCode, mostSpecificSubdivision.IsoCode)
+	}
+
+	// location
+	appendIfNotEmpty(conventions.AttributeGeoTimezone, city.Location.TimeZone)
+	if city.Location.Latitude != 0 && city.Location.Longitude != 0 {
+		attributes = append(attributes, attribute.Float64(conventions.AttributeGeoLocationLat, city.Location.Latitude), attribute.Float64(conventions.AttributeGeoLocationLon, city.Location.Longitude))
+	}
+
+	return &attributes, err
+}