first commit

This commit is contained in:
Darko Draskovic
2023-03-14 15:12:23 +01:00
commit 22c743c4dd
1140 changed files with 423491 additions and 0 deletions
+191
View File
@@ -0,0 +1,191 @@
Apache License
Version 2.0, January 2004
https://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
Copyright 2015-2020 Mainflux
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
https://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
+5
View File
@@ -0,0 +1,5 @@
docker_mfxkit:
docker build --no-cache --tag=mainflux/mfxkit -f docker/Dockerfile .
run:
docker-compose -f docker/docker-compose.yml up
+48
View File
@@ -0,0 +1,48 @@
# Mfxkit - Mainflux Starter Kit
Mfxkit service provides a barebones HTTP API and Service interface implementation for development of a core Mainflux service.
## How-to
Copy `mfxkit` directory to the `mainflux` root directory, e.g. `~/go/src/github.com/mainflux/mainflux/`. Copy `cmd/mfxkit` directory to `mainflux/cmd` directory.
In `mainflux` root directory run
```
MF_MFXKIT_LOG_LEVEL=info go run cmd/mfxkit/main.go
```
You should get a message similar to this one
```
{"level":"info","message":"Mfxkit service started using http on port 9021","ts":"2021-03-03T11:16:27.027381203Z"}
```
In the other terminal window run
```
curl -i -X POST -H "Content-Type: application/json" localhost:9021/mfxkit -d '{"secret":"secret"}'
```
If everything goes well, you should get
```
HTTP/1.1 200 OK
Content-Type: application/json
Date: Wed, 03 Mar 2021 11:17:10 GMT
Content-Length: 30
{"greeting":"Hello World :)"}
```
To change the secret or the port, prefix the `go run` command with environment variable assignments, e.g.
```
MF_MFXKIT_LOG_LEVEL=info MF_MFXKIT_SECRET=secret2 MF_MFXKIT_HTTP_PORT=9022 go run cmd/mfxkit/main.go
```
To see the change in action, run
```
curl -i -X POST -H "Content-Type: application/json" localhost:9022/mfxkit -d '{"secret":"secret2"}'
```
+61
View File
@@ -0,0 +1,61 @@
# Mfxkit
Mfxkit service provides a barebones HTTP API and Service interface implementation for development of a core Mainflux service.
## Configuration
The service is configured using the environment variables from the following table. Note that any unset variables will be replaced with their default values.
| Variable | Description | Default |
|-----------------------|---------------------------------------------------------|---------|
| MF_MFXKIT_LOG_LEVEL | Log level for mfxkit service (debug, info, warn, error) | error |
| MF_MFXKIT_HTTP_PORT | Mfxkit service HTTP port | 9021 |
| MF_MFXKIT_SERVER_CERT | Path to server certificate in pem format | |
| MF_MFXKIT_SERVER_KEY | Path to server key in pem format | |
| MF_JAEGER_URL | Jaeger server URL | |
| MF_MFXKIT_SECRET | Mfxkit service secret | secret |
## Deployment
The service is distributed as a Docker container. The following snippet provides a compose file template that can be used to deploy the service container locally:
```yaml
version: "3"
services:
mfxkit:
image: mainflux/mfxkit:[version]
container_name: [instance name]
ports:
- [host machine port]:[configured HTTP port]
environment:
MF_MFXKIT_LOG_LEVEL: [Kit log level]
MF_MFXKIT_HTTP_PORT: [Service HTTP port]
MF_MFXKIT_SERVER_CERT: [String path to server cert in pem format]
MF_MFXKIT_SERVER_KEY: [String path to server key in pem format]
MF_MFXKIT_SECRET: [Mfxkit service secret]
MF_JAEGER_URL: [Jaeger server URL]
```
To start the service outside of the container, execute the following shell script:
```bash
# download the latest version of the service
go get github.com/mainflux/mainflux
cd $GOPATH/src/github.com/mainflux/mainflux
# compile the mfxkit
make mfxkit
# copy binary to bin
make install
# set the environment variables and run the service
MF_MFXKIT_LOG_LEVEL=[Kit log level] MF_MFXKIT_HTTP_PORT=[Service HTTP port] MF_MFXKIT_SERVER_CERT: [String path to server cert in pem format] MF_MFXKIT_SERVER_KEY: [String path to server key in pem format] MF_JAEGER_URL=[Jaeger server URL] MF_MFXKIT_SECRET: [Mfxkit service secret] $GOBIN/mainflux-kit
```
## Usage
For more information about service capabilities and its usage, please check out the [API documentation](swagger.yaml).
[doc]: http://mainflux.readthedocs.io
+5
View File
@@ -0,0 +1,5 @@
// Copyright (c) Mainflux
// SPDX-License-Identifier: Apache-2.0
// Package http contains implementation of kit service HTTP API.
package http
+31
View File
@@ -0,0 +1,31 @@
// Copyright (c) Mainflux
// SPDX-License-Identifier: Apache-2.0
package http
import (
"context"
"github.com/go-kit/kit/endpoint"
"github.com/ultravioletrs/cocosvm/agent"
)
func pingEndpoint(svc agent.Service) endpoint.Endpoint {
return func(ctx context.Context, request interface{}) (interface{}, error) {
req := request.(pingReq)
if err := req.validate(); err != nil {
return nil, err
}
greeting, err := svc.Ping(req.Secret)
if err != nil {
return nil, err
}
res := pingRes{
Greeting: greeting,
}
return res, nil
}
}
+22
View File
@@ -0,0 +1,22 @@
// Copyright (c) Mainflux
// SPDX-License-Identifier: Apache-2.0
package http
import "github.com/ultravioletrs/cocosvm/agent"
type apiReq interface {
validate() error
}
type pingReq struct {
Secret string `json:"secret"`
}
func (req pingReq) validate() error {
if req.Secret == "" {
return agent.ErrMalformedEntity
}
return nil
}
+28
View File
@@ -0,0 +1,28 @@
// Copyright (c) Mainflux
// SPDX-License-Identifier: Apache-2.0
package http
import (
"net/http"
"github.com/mainflux/mainflux"
)
var _ mainflux.Response = (*pingRes)(nil)
type pingRes struct {
Greeting string `json:"greeting"`
}
func (res pingRes) Code() int {
return http.StatusOK
}
func (res pingRes) Headers() map[string]string {
return map[string]string{}
}
func (res pingRes) Empty() bool {
return false
}
+143
View File
@@ -0,0 +1,143 @@
// Copyright (c) Mainflux
// SPDX-License-Identifier: Apache-2.0
package http
import (
"context"
"encoding/json"
"errors"
"io"
"net/http"
"strconv"
"strings"
kitot "github.com/go-kit/kit/tracing/opentracing"
kithttp "github.com/go-kit/kit/transport/http"
"github.com/go-zoo/bone"
"github.com/mainflux/mainflux"
opentracing "github.com/opentracing/opentracing-go"
"github.com/prometheus/client_golang/prometheus/promhttp"
cckit "github.com/ultravioletrs/cocosvm/agent"
)
const (
contentType = "application/json"
)
var (
errUnsupportedContentType = errors.New("unsupported content type")
errInvalidQueryParams = errors.New("invalid query params")
)
// MakeHandler returns a HTTP handler for API endpoints.
func MakeHandler(tracer opentracing.Tracer, svc cckit.Service) http.Handler {
opts := []kithttp.ServerOption{
kithttp.ServerErrorEncoder(encodeError),
}
r := bone.New()
r.Post("/mfxkit", kithttp.NewServer(
kitot.TraceServer(tracer, "ping")(pingEndpoint(svc)),
decodePing,
encodeResponse,
opts...,
))
r.GetFunc("/version", mainflux.Version("things"))
r.Handle("/metrics", promhttp.Handler())
return r
}
func decodePing(_ context.Context, r *http.Request) (interface{}, error) {
if !strings.Contains(r.Header.Get("Content-Type"), contentType) {
return nil, errUnsupportedContentType
}
req := pingReq{}
if err := json.NewDecoder(r.Body).Decode(&req); err != nil {
return nil, err
}
return req, nil
}
func encodeResponse(_ context.Context, w http.ResponseWriter, response interface{}) error {
w.Header().Set("Content-Type", contentType)
if ar, ok := response.(mainflux.Response); ok {
for k, v := range ar.Headers() {
w.Header().Set(k, v)
}
w.WriteHeader(ar.Code())
if ar.Empty() {
return nil
}
}
return json.NewEncoder(w).Encode(response)
}
func encodeError(_ context.Context, err error, w http.ResponseWriter) {
w.Header().Set("Content-Type", contentType)
switch err {
case cckit.ErrMalformedEntity:
w.WriteHeader(http.StatusBadRequest)
case cckit.ErrUnauthorizedAccess:
w.WriteHeader(http.StatusForbidden)
case errUnsupportedContentType:
w.WriteHeader(http.StatusUnsupportedMediaType)
case errInvalidQueryParams:
w.WriteHeader(http.StatusBadRequest)
case io.ErrUnexpectedEOF:
w.WriteHeader(http.StatusBadRequest)
case io.EOF:
w.WriteHeader(http.StatusBadRequest)
default:
switch err.(type) {
case *json.SyntaxError:
w.WriteHeader(http.StatusBadRequest)
case *json.UnmarshalTypeError:
w.WriteHeader(http.StatusBadRequest)
default:
w.WriteHeader(http.StatusInternalServerError)
}
}
}
func readUintQuery(r *http.Request, key string, def uint64) (uint64, error) {
vals := bone.GetQuery(r, key)
if len(vals) > 1 {
return 0, errInvalidQueryParams
}
if len(vals) == 0 {
return def, nil
}
strval := vals[0]
val, err := strconv.ParseUint(strval, 10, 64)
if err != nil {
return 0, errInvalidQueryParams
}
return val, nil
}
func readStringQuery(r *http.Request, key string) (string, error) {
vals := bone.GetQuery(r, key)
if len(vals) > 1 {
return "", errInvalidQueryParams
}
if len(vals) == 0 {
return "", nil
}
return vals[0], nil
}
+6
View File
@@ -0,0 +1,6 @@
// Copyright (c) Mainflux
// SPDX-License-Identifier: Apache-2.0
// Package api contains API-related concerns: endpoint definitions, middlewares
// and all resource representations.
package api
+40
View File
@@ -0,0 +1,40 @@
// Copyright (c) Mainflux
// SPDX-License-Identifier: Apache-2.0
//go:build !test
// +build !test
package api
import (
"fmt"
"time"
log "github.com/mainflux/mainflux/logger"
"github.com/ultravioletrs/cocosvm/agent"
)
var _ agent.Service = (*loggingMiddleware)(nil)
type loggingMiddleware struct {
logger log.Logger
svc agent.Service
}
// LoggingMiddleware adds logging facilities to the core service.
func LoggingMiddleware(svc agent.Service, logger log.Logger) agent.Service {
return &loggingMiddleware{logger, svc}
}
func (lm *loggingMiddleware) Ping(secret string) (response string, err error) {
defer func(begin time.Time) {
message := fmt.Sprintf("Method ping for secret %s took %s to complete", secret, time.Since(begin))
if err != nil {
lm.logger.Warn(fmt.Sprintf("%s with error: %s.", message, err))
return
}
lm.logger.Info(fmt.Sprintf("%s without errors.", message))
}(time.Now())
return lm.svc.Ping(secret)
}
+41
View File
@@ -0,0 +1,41 @@
// Copyright (c) Mainflux
// SPDX-License-Identifier: Apache-2.0
//go:build !test
// +build !test
package api
import (
"time"
"github.com/go-kit/kit/metrics"
"github.com/ultravioletrs/cocosvm/agent"
)
var _ agent.Service = (*metricsMiddleware)(nil)
type metricsMiddleware struct {
counter metrics.Counter
latency metrics.Histogram
svc agent.Service
}
// MetricsMiddleware instruments core service by tracking request count and
// latency.
func MetricsMiddleware(svc agent.Service, counter metrics.Counter, latency metrics.Histogram) agent.Service {
return &metricsMiddleware{
counter: counter,
latency: latency,
svc: svc,
}
}
func (ms *metricsMiddleware) Ping(secret string) (response string, err error) {
defer func(begin time.Time) {
ms.counter.With("method", "ping").Add(1)
ms.latency.With("method", "ping").Observe(time.Since(begin).Seconds())
}(time.Now())
return ms.svc.Ping(secret)
}
+7
View File
@@ -0,0 +1,7 @@
// Copyright (c) Mainflux
// SPDX-License-Identifier: Apache-2.0
// Package things contains the domain concept definitions needed to support
// Mainflux mfxkit service functionality.
package agent
+45
View File
@@ -0,0 +1,45 @@
// Copyright (c) Mainflux
// SPDX-License-Identifier: Apache-2.0
package agent
import (
"errors"
)
var (
// ErrMalformedEntity indicates malformed entity specification (e.g.
// invalid username or password).
ErrMalformedEntity = errors.New("malformed entity specification")
// ErrUnauthorizedAccess indicates missing or invalid credentials provided
// when accessing a protected resource.
ErrUnauthorizedAccess = errors.New("missing or invalid credentials provided")
)
// Service specifies an API that must be fullfiled by the domain service
// implementation, and all of its decorators (e.g. logging & metrics).
type Service interface {
// Ping compares a given string with secret
Ping(string) (string, error)
}
type mfxkitService struct {
secret string
}
var _ Service = (*mfxkitService)(nil)
// New instantiates the mfxkit service implementation.
func New(secret string) Service {
return &mfxkitService{
secret: secret,
}
}
func (ks *mfxkitService) Ping(secret string) (string, error) {
if ks.secret != secret {
return "", ErrUnauthorizedAccess
}
return "Hello World :)", nil
}
+153
View File
@@ -0,0 +1,153 @@
//
// Copyright (c) 2019
// Mainflux
//
// SPDX-License-Identifier: Apache-2.0
//
package main
import (
"fmt"
"io"
"io/ioutil"
"log"
"net/http"
"os"
"os/signal"
"syscall"
"github.com/mainflux/mainflux"
"github.com/mainflux/mainflux/logger"
"github.com/ultravioletrs/cocosvm/agent"
"github.com/ultravioletrs/cocosvm/agent/api"
agenthttpapi "github.com/ultravioletrs/cocosvm/agent/api/agent/http"
kitprometheus "github.com/go-kit/kit/metrics/prometheus"
opentracing "github.com/opentracing/opentracing-go"
stdprometheus "github.com/prometheus/client_golang/prometheus"
jconfig "github.com/uber/jaeger-client-go/config"
)
const (
defLogLevel = "error"
defHTTPPort = "9021"
defJaegerURL = ""
defServerCert = ""
defServerKey = ""
defSecret = "secret"
envLogLevel = "MF_MFXKIT_LOG_LEVEL"
envHTTPPort = "MF_MFXKIT_HTTP_PORT"
envServerCert = "MF_MFXKIT_SERVER_CERT"
envServerKey = "MF_MFXKIT_SERVER_KEY"
envSecret = "MF_MFXKIT_SECRET"
envJaegerURL = "MF_JAEGER_URL"
)
type config struct {
logLevel string
httpPort string
authHTTPPort string
authGRPCPort string
serverCert string
serverKey string
secret string
jaegerURL string
}
func main() {
cfg := loadConfig()
logger, err := logger.New(os.Stdout, cfg.logLevel)
if err != nil {
log.Fatalf(err.Error())
}
mfxkitTracer, mfxkitCloser := initJaeger("mfxkit", cfg.jaegerURL, logger)
defer mfxkitCloser.Close()
svc := newService(cfg.secret, logger)
errs := make(chan error, 2)
go startHTTPServer(agenthttpapi.MakeHandler(mfxkitTracer, svc), cfg.httpPort, cfg, logger, errs)
go func() {
c := make(chan os.Signal)
signal.Notify(c, syscall.SIGINT)
errs <- fmt.Errorf("%s", <-c)
}()
err = <-errs
logger.Error(fmt.Sprintf("Mfxkit service terminated: %s", err))
}
func loadConfig() config {
return config{
logLevel: mainflux.Env(envLogLevel, defLogLevel),
httpPort: mainflux.Env(envHTTPPort, defHTTPPort),
serverCert: mainflux.Env(envServerCert, defServerCert),
serverKey: mainflux.Env(envServerKey, defServerKey),
jaegerURL: mainflux.Env(envJaegerURL, defJaegerURL),
secret: mainflux.Env(envSecret, defSecret),
}
}
func initJaeger(svcName, url string, logger logger.Logger) (opentracing.Tracer, io.Closer) {
if url == "" {
return opentracing.NoopTracer{}, ioutil.NopCloser(nil)
}
tracer, closer, err := jconfig.Configuration{
ServiceName: svcName,
Sampler: &jconfig.SamplerConfig{
Type: "const",
Param: 1,
},
Reporter: &jconfig.ReporterConfig{
LocalAgentHostPort: url,
LogSpans: true,
},
}.NewTracer()
if err != nil {
logger.Error(fmt.Sprintf("Failed to init Jaeger client: %s", err))
os.Exit(1)
}
return tracer, closer
}
func newService(secret string, logger logger.Logger) agent.Service {
svc := agent.New(secret)
svc = api.LoggingMiddleware(svc, logger)
svc = api.MetricsMiddleware(
svc,
kitprometheus.NewCounterFrom(stdprometheus.CounterOpts{
Namespace: "mfxkit",
Subsystem: "api",
Name: "request_count",
Help: "Number of requests received.",
}, []string{"method"}),
kitprometheus.NewSummaryFrom(stdprometheus.SummaryOpts{
Namespace: "mfxkit",
Subsystem: "api",
Name: "request_latency_microseconds",
Help: "Total duration of requests in microseconds.",
}, []string{"method"}),
)
return svc
}
func startHTTPServer(handler http.Handler, port string, cfg config, logger logger.Logger, errs chan error) {
p := fmt.Sprintf(":%s", port)
if cfg.serverCert != "" || cfg.serverKey != "" {
logger.Info(fmt.Sprintf("Mfxkit service started using https on port %s with cert %s key %s",
port, cfg.serverCert, cfg.serverKey))
errs <- http.ListenAndServeTLS(p, cfg.serverCert, cfg.serverKey, handler)
return
}
logger.Info(fmt.Sprintf("Mfxkit service started using http on port %s", cfg.httpPort))
errs <- http.ListenAndServe(p, handler)
}
+9
View File
@@ -0,0 +1,9 @@
# Docker: Environment variables in Compose
## Mfxkit
MF_MFXKIT_LOG_LEVEL=debug
MF_MFXKIT_HTTP_PORT=9021
MF_MFXKIT_SECRET=secret
MF_MFXKIT_SERVER_CERT=""
MF_MFXKIT_SERVER_KEY=""
MF_JAEGER_URL="jaeger:6831"
+11
View File
@@ -0,0 +1,11 @@
FROM golang:1.16-alpine AS builder
WORKDIR /go/src/github.com/mainflux/mfxkit
COPY . .
RUN CGO_ENABLED=0 GOARCH=amd64 \
go build -mod=vendor -ldflags "-s -w" -o build/mainflux-mfxkit cmd/mfxkit/main.go \
&& mv build/mainflux-mfxkit /exe
FROM scratch
COPY --from=builder /exe /
ENTRYPOINT ["/exe"]
+30
View File
@@ -0,0 +1,30 @@
###
# Copyright (c) 2015-2017 Mainflux
#
# Mainflux is licensed under an Apache license, version 2.0 license.
# All rights not explicitly granted in the Apache license, version 2.0 are reserved.
# See the included LICENSE file for more details.
###
version: "3.7"
networks:
docker_mainflux-base-net:
external: true
services:
mfxkit:
image: mainflux/mfxkit:latest
container_name: mainflux-mfxkit
restart: on-failure
environment:
MF_MFXKIT_LOG_LEVEL: ${MF_MFXKIT_LOG_LEVEL}
MF_MFXKIT_HTTP_PORT: ${MF_MFXKIT_HTTP_PORT}
MF_MFXKIT_SERVER_CERT: ${MF_MFXKIT_SERVER_CERT}
MF_MFXKIT_SERVER_KEY: ${MF_MFXKIT_SERVER_KEY}
MF_JAEGER_URL: ${MF_JAEGER_URL}
MF_MFXKIT_SECRET: ${MF_MFXKIT_SECRET}
ports:
- ${MF_MFXKIT_HTTP_PORT}:${MF_MFXKIT_HTTP_PORT}
networks:
- docker_mainflux-base-net
+34
View File
@@ -0,0 +1,34 @@
module github.com/ultravioletrs/cocosvm
go 1.19
require (
github.com/go-kit/kit v0.12.0
github.com/go-zoo/bone v1.3.0
github.com/mainflux/mainflux v0.12.0
github.com/opentracing/opentracing-go v1.2.0
github.com/prometheus/client_golang v1.14.0
github.com/uber/jaeger-client-go v2.30.0+incompatible
)
require (
github.com/beorn7/perks v1.0.1 // indirect
github.com/cespare/xxhash/v2 v2.1.2 // indirect
github.com/go-kit/log v0.2.0 // indirect
github.com/go-logfmt/logfmt v0.5.1 // indirect
github.com/golang/protobuf v1.5.2 // indirect
github.com/matttproud/golang_protobuf_extensions v1.0.1 // indirect
github.com/pkg/errors v0.9.1 // indirect
github.com/prometheus/client_model v0.3.0 // indirect
github.com/prometheus/common v0.37.0 // indirect
github.com/prometheus/procfs v0.8.0 // indirect
github.com/subosito/gotenv v1.2.0 // indirect
github.com/uber/jaeger-lib v2.0.0+incompatible // indirect
go.uber.org/atomic v1.9.0 // indirect
golang.org/x/net v0.0.0-20220225172249-27dd8689420f // indirect
golang.org/x/sys v0.0.0-20220520151302-bc2c85ada10a // indirect
golang.org/x/text v0.3.7 // indirect
google.golang.org/genproto v0.0.0-20210917145530-b395a37504d4 // indirect
google.golang.org/grpc v1.40.0 // indirect
google.golang.org/protobuf v1.28.1 // indirect
)
+1166
View File
File diff suppressed because it is too large Load Diff
+20
View File
@@ -0,0 +1,20 @@
Copyright (C) 2013 Blake Mizerany
Permission is hereby granted, free of charge, to any person obtaining
a copy of this software and associated documentation files (the
"Software"), to deal in the Software without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to
the following conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
File diff suppressed because it is too large Load Diff
+316
View File
@@ -0,0 +1,316 @@
// Package quantile computes approximate quantiles over an unbounded data
// stream within low memory and CPU bounds.
//
// A small amount of accuracy is traded to achieve the above properties.
//
// Multiple streams can be merged before calling Query to generate a single set
// of results. This is meaningful when the streams represent the same type of
// data. See Merge and Samples.
//
// For more detailed information about the algorithm used, see:
//
// Effective Computation of Biased Quantiles over Data Streams
//
// http://www.cs.rutgers.edu/~muthu/bquant.pdf
package quantile
import (
"math"
"sort"
)
// Sample holds an observed value and meta information for compression. JSON
// tags have been added for convenience.
type Sample struct {
Value float64 `json:",string"`
Width float64 `json:",string"`
Delta float64 `json:",string"`
}
// Samples represents a slice of samples. It implements sort.Interface.
type Samples []Sample
func (a Samples) Len() int { return len(a) }
func (a Samples) Less(i, j int) bool { return a[i].Value < a[j].Value }
func (a Samples) Swap(i, j int) { a[i], a[j] = a[j], a[i] }
type invariant func(s *stream, r float64) float64
// NewLowBiased returns an initialized Stream for low-biased quantiles
// (e.g. 0.01, 0.1, 0.5) where the needed quantiles are not known a priori, but
// error guarantees can still be given even for the lower ranks of the data
// distribution.
//
// The provided epsilon is a relative error, i.e. the true quantile of a value
// returned by a query is guaranteed to be within (1±Epsilon)*Quantile.
//
// See http://www.cs.rutgers.edu/~muthu/bquant.pdf for time, space, and error
// properties.
func NewLowBiased(epsilon float64) *Stream {
ƒ := func(s *stream, r float64) float64 {
return 2 * epsilon * r
}
return newStream(ƒ)
}
// NewHighBiased returns an initialized Stream for high-biased quantiles
// (e.g. 0.01, 0.1, 0.5) where the needed quantiles are not known a priori, but
// error guarantees can still be given even for the higher ranks of the data
// distribution.
//
// The provided epsilon is a relative error, i.e. the true quantile of a value
// returned by a query is guaranteed to be within 1-(1±Epsilon)*(1-Quantile).
//
// See http://www.cs.rutgers.edu/~muthu/bquant.pdf for time, space, and error
// properties.
func NewHighBiased(epsilon float64) *Stream {
ƒ := func(s *stream, r float64) float64 {
return 2 * epsilon * (s.n - r)
}
return newStream(ƒ)
}
// NewTargeted returns an initialized Stream concerned with a particular set of
// quantile values that are supplied a priori. Knowing these a priori reduces
// space and computation time. The targets map maps the desired quantiles to
// their absolute errors, i.e. the true quantile of a value returned by a query
// is guaranteed to be within (Quantile±Epsilon).
//
// See http://www.cs.rutgers.edu/~muthu/bquant.pdf for time, space, and error properties.
func NewTargeted(targetMap map[float64]float64) *Stream {
// Convert map to slice to avoid slow iterations on a map.
// ƒ is called on the hot path, so converting the map to a slice
// beforehand results in significant CPU savings.
targets := targetMapToSlice(targetMap)
ƒ := func(s *stream, r float64) float64 {
var m = math.MaxFloat64
var f float64
for _, t := range targets {
if t.quantile*s.n <= r {
f = (2 * t.epsilon * r) / t.quantile
} else {
f = (2 * t.epsilon * (s.n - r)) / (1 - t.quantile)
}
if f < m {
m = f
}
}
return m
}
return newStream(ƒ)
}
type target struct {
quantile float64
epsilon float64
}
func targetMapToSlice(targetMap map[float64]float64) []target {
targets := make([]target, 0, len(targetMap))
for quantile, epsilon := range targetMap {
t := target{
quantile: quantile,
epsilon: epsilon,
}
targets = append(targets, t)
}
return targets
}
// Stream computes quantiles for a stream of float64s. It is not thread-safe by
// design. Take care when using across multiple goroutines.
type Stream struct {
*stream
b Samples
sorted bool
}
func newStream(ƒ invariant) *Stream {
x := &stream{ƒ: ƒ}
return &Stream{x, make(Samples, 0, 500), true}
}
// Insert inserts v into the stream.
func (s *Stream) Insert(v float64) {
s.insert(Sample{Value: v, Width: 1})
}
func (s *Stream) insert(sample Sample) {
s.b = append(s.b, sample)
s.sorted = false
if len(s.b) == cap(s.b) {
s.flush()
}
}
// Query returns the computed qth percentiles value. If s was created with
// NewTargeted, and q is not in the set of quantiles provided a priori, Query
// will return an unspecified result.
func (s *Stream) Query(q float64) float64 {
if !s.flushed() {
// Fast path when there hasn't been enough data for a flush;
// this also yields better accuracy for small sets of data.
l := len(s.b)
if l == 0 {
return 0
}
i := int(math.Ceil(float64(l) * q))
if i > 0 {
i -= 1
}
s.maybeSort()
return s.b[i].Value
}
s.flush()
return s.stream.query(q)
}
// Merge merges samples into the underlying streams samples. This is handy when
// merging multiple streams from separate threads, database shards, etc.
//
// ATTENTION: This method is broken and does not yield correct results. The
// underlying algorithm is not capable of merging streams correctly.
func (s *Stream) Merge(samples Samples) {
sort.Sort(samples)
s.stream.merge(samples)
}
// Reset reinitializes and clears the list reusing the samples buffer memory.
func (s *Stream) Reset() {
s.stream.reset()
s.b = s.b[:0]
}
// Samples returns stream samples held by s.
func (s *Stream) Samples() Samples {
if !s.flushed() {
return s.b
}
s.flush()
return s.stream.samples()
}
// Count returns the total number of samples observed in the stream
// since initialization.
func (s *Stream) Count() int {
return len(s.b) + s.stream.count()
}
func (s *Stream) flush() {
s.maybeSort()
s.stream.merge(s.b)
s.b = s.b[:0]
}
func (s *Stream) maybeSort() {
if !s.sorted {
s.sorted = true
sort.Sort(s.b)
}
}
func (s *Stream) flushed() bool {
return len(s.stream.l) > 0
}
type stream struct {
n float64
l []Sample
ƒ invariant
}
func (s *stream) reset() {
s.l = s.l[:0]
s.n = 0
}
func (s *stream) insert(v float64) {
s.merge(Samples{{v, 1, 0}})
}
func (s *stream) merge(samples Samples) {
// TODO(beorn7): This tries to merge not only individual samples, but
// whole summaries. The paper doesn't mention merging summaries at
// all. Unittests show that the merging is inaccurate. Find out how to
// do merges properly.
var r float64
i := 0
for _, sample := range samples {
for ; i < len(s.l); i++ {
c := s.l[i]
if c.Value > sample.Value {
// Insert at position i.
s.l = append(s.l, Sample{})
copy(s.l[i+1:], s.l[i:])
s.l[i] = Sample{
sample.Value,
sample.Width,
math.Max(sample.Delta, math.Floor(s.ƒ(s, r))-1),
// TODO(beorn7): How to calculate delta correctly?
}
i++
goto inserted
}
r += c.Width
}
s.l = append(s.l, Sample{sample.Value, sample.Width, 0})
i++
inserted:
s.n += sample.Width
r += sample.Width
}
s.compress()
}
func (s *stream) count() int {
return int(s.n)
}
func (s *stream) query(q float64) float64 {
t := math.Ceil(q * s.n)
t += math.Ceil(s.ƒ(s, t) / 2)
p := s.l[0]
var r float64
for _, c := range s.l[1:] {
r += p.Width
if r+c.Width+c.Delta > t {
return p.Value
}
p = c
}
return p.Value
}
func (s *stream) compress() {
if len(s.l) < 2 {
return
}
x := s.l[len(s.l)-1]
xi := len(s.l) - 1
r := s.n - 1 - x.Width
for i := len(s.l) - 2; i >= 0; i-- {
c := s.l[i]
if c.Width+x.Width+x.Delta <= s.ƒ(s, r) {
x.Width += c.Width
s.l[xi] = x
// Remove element at i.
copy(s.l[i:], s.l[i+1:])
s.l = s.l[:len(s.l)-1]
xi -= 1
} else {
x = c
xi = i
}
r -= c.Width
}
}
func (s *stream) samples() Samples {
samples := make(Samples, len(s.l))
copy(samples, s.l)
return samples
}
+22
View File
@@ -0,0 +1,22 @@
Copyright (c) 2016 Caleb Spare
MIT License
Permission is hereby granted, free of charge, to any person obtaining
a copy of this software and associated documentation files (the
"Software"), to deal in the Software without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to
the following conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+69
View File
@@ -0,0 +1,69 @@
# xxhash
[![Go Reference](https://pkg.go.dev/badge/github.com/cespare/xxhash/v2.svg)](https://pkg.go.dev/github.com/cespare/xxhash/v2)
[![Test](https://github.com/cespare/xxhash/actions/workflows/test.yml/badge.svg)](https://github.com/cespare/xxhash/actions/workflows/test.yml)
xxhash is a Go implementation of the 64-bit
[xxHash](http://cyan4973.github.io/xxHash/) algorithm, XXH64. This is a
high-quality hashing algorithm that is much faster than anything in the Go
standard library.
This package provides a straightforward API:
```
func Sum64(b []byte) uint64
func Sum64String(s string) uint64
type Digest struct{ ... }
func New() *Digest
```
The `Digest` type implements hash.Hash64. Its key methods are:
```
func (*Digest) Write([]byte) (int, error)
func (*Digest) WriteString(string) (int, error)
func (*Digest) Sum64() uint64
```
This implementation provides a fast pure-Go implementation and an even faster
assembly implementation for amd64.
## Compatibility
This package is in a module and the latest code is in version 2 of the module.
You need a version of Go with at least "minimal module compatibility" to use
github.com/cespare/xxhash/v2:
* 1.9.7+ for Go 1.9
* 1.10.3+ for Go 1.10
* Go 1.11 or later
I recommend using the latest release of Go.
## Benchmarks
Here are some quick benchmarks comparing the pure-Go and assembly
implementations of Sum64.
| input size | purego | asm |
| --- | --- | --- |
| 5 B | 979.66 MB/s | 1291.17 MB/s |
| 100 B | 7475.26 MB/s | 7973.40 MB/s |
| 4 KB | 17573.46 MB/s | 17602.65 MB/s |
| 10 MB | 17131.46 MB/s | 17142.16 MB/s |
These numbers were generated on Ubuntu 18.04 with an Intel i7-8700K CPU using
the following commands under Go 1.11.2:
```
$ go test -tags purego -benchtime 10s -bench '/xxhash,direct,bytes'
$ go test -benchtime 10s -bench '/xxhash,direct,bytes'
```
## Projects using this package
- [InfluxDB](https://github.com/influxdata/influxdb)
- [Prometheus](https://github.com/prometheus/prometheus)
- [VictoriaMetrics](https://github.com/VictoriaMetrics/VictoriaMetrics)
- [FreeCache](https://github.com/coocood/freecache)
- [FastCache](https://github.com/VictoriaMetrics/fastcache)
+235
View File
@@ -0,0 +1,235 @@
// Package xxhash implements the 64-bit variant of xxHash (XXH64) as described
// at http://cyan4973.github.io/xxHash/.
package xxhash
import (
"encoding/binary"
"errors"
"math/bits"
)
const (
prime1 uint64 = 11400714785074694791
prime2 uint64 = 14029467366897019727
prime3 uint64 = 1609587929392839161
prime4 uint64 = 9650029242287828579
prime5 uint64 = 2870177450012600261
)
// NOTE(caleb): I'm using both consts and vars of the primes. Using consts where
// possible in the Go code is worth a small (but measurable) performance boost
// by avoiding some MOVQs. Vars are needed for the asm and also are useful for
// convenience in the Go code in a few places where we need to intentionally
// avoid constant arithmetic (e.g., v1 := prime1 + prime2 fails because the
// result overflows a uint64).
var (
prime1v = prime1
prime2v = prime2
prime3v = prime3
prime4v = prime4
prime5v = prime5
)
// Digest implements hash.Hash64.
type Digest struct {
v1 uint64
v2 uint64
v3 uint64
v4 uint64
total uint64
mem [32]byte
n int // how much of mem is used
}
// New creates a new Digest that computes the 64-bit xxHash algorithm.
func New() *Digest {
var d Digest
d.Reset()
return &d
}
// Reset clears the Digest's state so that it can be reused.
func (d *Digest) Reset() {
d.v1 = prime1v + prime2
d.v2 = prime2
d.v3 = 0
d.v4 = -prime1v
d.total = 0
d.n = 0
}
// Size always returns 8 bytes.
func (d *Digest) Size() int { return 8 }
// BlockSize always returns 32 bytes.
func (d *Digest) BlockSize() int { return 32 }
// Write adds more data to d. It always returns len(b), nil.
func (d *Digest) Write(b []byte) (n int, err error) {
n = len(b)
d.total += uint64(n)
if d.n+n < 32 {
// This new data doesn't even fill the current block.
copy(d.mem[d.n:], b)
d.n += n
return
}
if d.n > 0 {
// Finish off the partial block.
copy(d.mem[d.n:], b)
d.v1 = round(d.v1, u64(d.mem[0:8]))
d.v2 = round(d.v2, u64(d.mem[8:16]))
d.v3 = round(d.v3, u64(d.mem[16:24]))
d.v4 = round(d.v4, u64(d.mem[24:32]))
b = b[32-d.n:]
d.n = 0
}
if len(b) >= 32 {
// One or more full blocks left.
nw := writeBlocks(d, b)
b = b[nw:]
}
// Store any remaining partial block.
copy(d.mem[:], b)
d.n = len(b)
return
}
// Sum appends the current hash to b and returns the resulting slice.
func (d *Digest) Sum(b []byte) []byte {
s := d.Sum64()
return append(
b,
byte(s>>56),
byte(s>>48),
byte(s>>40),
byte(s>>32),
byte(s>>24),
byte(s>>16),
byte(s>>8),
byte(s),
)
}
// Sum64 returns the current hash.
func (d *Digest) Sum64() uint64 {
var h uint64
if d.total >= 32 {
v1, v2, v3, v4 := d.v1, d.v2, d.v3, d.v4
h = rol1(v1) + rol7(v2) + rol12(v3) + rol18(v4)
h = mergeRound(h, v1)
h = mergeRound(h, v2)
h = mergeRound(h, v3)
h = mergeRound(h, v4)
} else {
h = d.v3 + prime5
}
h += d.total
i, end := 0, d.n
for ; i+8 <= end; i += 8 {
k1 := round(0, u64(d.mem[i:i+8]))
h ^= k1
h = rol27(h)*prime1 + prime4
}
if i+4 <= end {
h ^= uint64(u32(d.mem[i:i+4])) * prime1
h = rol23(h)*prime2 + prime3
i += 4
}
for i < end {
h ^= uint64(d.mem[i]) * prime5
h = rol11(h) * prime1
i++
}
h ^= h >> 33
h *= prime2
h ^= h >> 29
h *= prime3
h ^= h >> 32
return h
}
const (
magic = "xxh\x06"
marshaledSize = len(magic) + 8*5 + 32
)
// MarshalBinary implements the encoding.BinaryMarshaler interface.
func (d *Digest) MarshalBinary() ([]byte, error) {
b := make([]byte, 0, marshaledSize)
b = append(b, magic...)
b = appendUint64(b, d.v1)
b = appendUint64(b, d.v2)
b = appendUint64(b, d.v3)
b = appendUint64(b, d.v4)
b = appendUint64(b, d.total)
b = append(b, d.mem[:d.n]...)
b = b[:len(b)+len(d.mem)-d.n]
return b, nil
}
// UnmarshalBinary implements the encoding.BinaryUnmarshaler interface.
func (d *Digest) UnmarshalBinary(b []byte) error {
if len(b) < len(magic) || string(b[:len(magic)]) != magic {
return errors.New("xxhash: invalid hash state identifier")
}
if len(b) != marshaledSize {
return errors.New("xxhash: invalid hash state size")
}
b = b[len(magic):]
b, d.v1 = consumeUint64(b)
b, d.v2 = consumeUint64(b)
b, d.v3 = consumeUint64(b)
b, d.v4 = consumeUint64(b)
b, d.total = consumeUint64(b)
copy(d.mem[:], b)
d.n = int(d.total % uint64(len(d.mem)))
return nil
}
func appendUint64(b []byte, x uint64) []byte {
var a [8]byte
binary.LittleEndian.PutUint64(a[:], x)
return append(b, a[:]...)
}
func consumeUint64(b []byte) ([]byte, uint64) {
x := u64(b)
return b[8:], x
}
func u64(b []byte) uint64 { return binary.LittleEndian.Uint64(b) }
func u32(b []byte) uint32 { return binary.LittleEndian.Uint32(b) }
func round(acc, input uint64) uint64 {
acc += input * prime2
acc = rol31(acc)
acc *= prime1
return acc
}
func mergeRound(acc, val uint64) uint64 {
val = round(0, val)
acc ^= val
acc = acc*prime1 + prime4
return acc
}
func rol1(x uint64) uint64 { return bits.RotateLeft64(x, 1) }
func rol7(x uint64) uint64 { return bits.RotateLeft64(x, 7) }
func rol11(x uint64) uint64 { return bits.RotateLeft64(x, 11) }
func rol12(x uint64) uint64 { return bits.RotateLeft64(x, 12) }
func rol18(x uint64) uint64 { return bits.RotateLeft64(x, 18) }
func rol23(x uint64) uint64 { return bits.RotateLeft64(x, 23) }
func rol27(x uint64) uint64 { return bits.RotateLeft64(x, 27) }
func rol31(x uint64) uint64 { return bits.RotateLeft64(x, 31) }
+13
View File
@@ -0,0 +1,13 @@
// +build !appengine
// +build gc
// +build !purego
package xxhash
// Sum64 computes the 64-bit xxHash digest of b.
//
//go:noescape
func Sum64(b []byte) uint64
//go:noescape
func writeBlocks(d *Digest, b []byte) int
+215
View File
@@ -0,0 +1,215 @@
// +build !appengine
// +build gc
// +build !purego
#include "textflag.h"
// Register allocation:
// AX h
// SI pointer to advance through b
// DX n
// BX loop end
// R8 v1, k1
// R9 v2
// R10 v3
// R11 v4
// R12 tmp
// R13 prime1v
// R14 prime2v
// DI prime4v
// round reads from and advances the buffer pointer in SI.
// It assumes that R13 has prime1v and R14 has prime2v.
#define round(r) \
MOVQ (SI), R12 \
ADDQ $8, SI \
IMULQ R14, R12 \
ADDQ R12, r \
ROLQ $31, r \
IMULQ R13, r
// mergeRound applies a merge round on the two registers acc and val.
// It assumes that R13 has prime1v, R14 has prime2v, and DI has prime4v.
#define mergeRound(acc, val) \
IMULQ R14, val \
ROLQ $31, val \
IMULQ R13, val \
XORQ val, acc \
IMULQ R13, acc \
ADDQ DI, acc
// func Sum64(b []byte) uint64
TEXT ·Sum64(SB), NOSPLIT, $0-32
// Load fixed primes.
MOVQ ·prime1v(SB), R13
MOVQ ·prime2v(SB), R14
MOVQ ·prime4v(SB), DI
// Load slice.
MOVQ b_base+0(FP), SI
MOVQ b_len+8(FP), DX
LEAQ (SI)(DX*1), BX
// The first loop limit will be len(b)-32.
SUBQ $32, BX
// Check whether we have at least one block.
CMPQ DX, $32
JLT noBlocks
// Set up initial state (v1, v2, v3, v4).
MOVQ R13, R8
ADDQ R14, R8
MOVQ R14, R9
XORQ R10, R10
XORQ R11, R11
SUBQ R13, R11
// Loop until SI > BX.
blockLoop:
round(R8)
round(R9)
round(R10)
round(R11)
CMPQ SI, BX
JLE blockLoop
MOVQ R8, AX
ROLQ $1, AX
MOVQ R9, R12
ROLQ $7, R12
ADDQ R12, AX
MOVQ R10, R12
ROLQ $12, R12
ADDQ R12, AX
MOVQ R11, R12
ROLQ $18, R12
ADDQ R12, AX
mergeRound(AX, R8)
mergeRound(AX, R9)
mergeRound(AX, R10)
mergeRound(AX, R11)
JMP afterBlocks
noBlocks:
MOVQ ·prime5v(SB), AX
afterBlocks:
ADDQ DX, AX
// Right now BX has len(b)-32, and we want to loop until SI > len(b)-8.
ADDQ $24, BX
CMPQ SI, BX
JG fourByte
wordLoop:
// Calculate k1.
MOVQ (SI), R8
ADDQ $8, SI
IMULQ R14, R8
ROLQ $31, R8
IMULQ R13, R8
XORQ R8, AX
ROLQ $27, AX
IMULQ R13, AX
ADDQ DI, AX
CMPQ SI, BX
JLE wordLoop
fourByte:
ADDQ $4, BX
CMPQ SI, BX
JG singles
MOVL (SI), R8
ADDQ $4, SI
IMULQ R13, R8
XORQ R8, AX
ROLQ $23, AX
IMULQ R14, AX
ADDQ ·prime3v(SB), AX
singles:
ADDQ $4, BX
CMPQ SI, BX
JGE finalize
singlesLoop:
MOVBQZX (SI), R12
ADDQ $1, SI
IMULQ ·prime5v(SB), R12
XORQ R12, AX
ROLQ $11, AX
IMULQ R13, AX
CMPQ SI, BX
JL singlesLoop
finalize:
MOVQ AX, R12
SHRQ $33, R12
XORQ R12, AX
IMULQ R14, AX
MOVQ AX, R12
SHRQ $29, R12
XORQ R12, AX
IMULQ ·prime3v(SB), AX
MOVQ AX, R12
SHRQ $32, R12
XORQ R12, AX
MOVQ AX, ret+24(FP)
RET
// writeBlocks uses the same registers as above except that it uses AX to store
// the d pointer.
// func writeBlocks(d *Digest, b []byte) int
TEXT ·writeBlocks(SB), NOSPLIT, $0-40
// Load fixed primes needed for round.
MOVQ ·prime1v(SB), R13
MOVQ ·prime2v(SB), R14
// Load slice.
MOVQ b_base+8(FP), SI
MOVQ b_len+16(FP), DX
LEAQ (SI)(DX*1), BX
SUBQ $32, BX
// Load vN from d.
MOVQ d+0(FP), AX
MOVQ 0(AX), R8 // v1
MOVQ 8(AX), R9 // v2
MOVQ 16(AX), R10 // v3
MOVQ 24(AX), R11 // v4
// We don't need to check the loop condition here; this function is
// always called with at least one block of data to process.
blockLoop:
round(R8)
round(R9)
round(R10)
round(R11)
CMPQ SI, BX
JLE blockLoop
// Copy vN back to d.
MOVQ R8, 0(AX)
MOVQ R9, 8(AX)
MOVQ R10, 16(AX)
MOVQ R11, 24(AX)
// The number of bytes written is SI minus the old base pointer.
SUBQ b_base+8(FP), SI
MOVQ SI, ret+32(FP)
RET
+76
View File
@@ -0,0 +1,76 @@
// +build !amd64 appengine !gc purego
package xxhash
// Sum64 computes the 64-bit xxHash digest of b.
func Sum64(b []byte) uint64 {
// A simpler version would be
// d := New()
// d.Write(b)
// return d.Sum64()
// but this is faster, particularly for small inputs.
n := len(b)
var h uint64
if n >= 32 {
v1 := prime1v + prime2
v2 := prime2
v3 := uint64(0)
v4 := -prime1v
for len(b) >= 32 {
v1 = round(v1, u64(b[0:8:len(b)]))
v2 = round(v2, u64(b[8:16:len(b)]))
v3 = round(v3, u64(b[16:24:len(b)]))
v4 = round(v4, u64(b[24:32:len(b)]))
b = b[32:len(b):len(b)]
}
h = rol1(v1) + rol7(v2) + rol12(v3) + rol18(v4)
h = mergeRound(h, v1)
h = mergeRound(h, v2)
h = mergeRound(h, v3)
h = mergeRound(h, v4)
} else {
h = prime5
}
h += uint64(n)
i, end := 0, len(b)
for ; i+8 <= end; i += 8 {
k1 := round(0, u64(b[i:i+8:len(b)]))
h ^= k1
h = rol27(h)*prime1 + prime4
}
if i+4 <= end {
h ^= uint64(u32(b[i:i+4:len(b)])) * prime1
h = rol23(h)*prime2 + prime3
i += 4
}
for ; i < end; i++ {
h ^= uint64(b[i]) * prime5
h = rol11(h) * prime1
}
h ^= h >> 33
h *= prime2
h ^= h >> 29
h *= prime3
h ^= h >> 32
return h
}
func writeBlocks(d *Digest, b []byte) int {
v1, v2, v3, v4 := d.v1, d.v2, d.v3, d.v4
n := len(b)
for len(b) >= 32 {
v1 = round(v1, u64(b[0:8:len(b)]))
v2 = round(v2, u64(b[8:16:len(b)]))
v3 = round(v3, u64(b[16:24:len(b)]))
v4 = round(v4, u64(b[24:32:len(b)]))
b = b[32:len(b):len(b)]
}
d.v1, d.v2, d.v3, d.v4 = v1, v2, v3, v4
return n - len(b)
}
+15
View File
@@ -0,0 +1,15 @@
// +build appengine
// This file contains the safe implementations of otherwise unsafe-using code.
package xxhash
// Sum64String computes the 64-bit xxHash digest of s.
func Sum64String(s string) uint64 {
return Sum64([]byte(s))
}
// WriteString adds more data to d. It always returns len(s), nil.
func (d *Digest) WriteString(s string) (n int, err error) {
return d.Write([]byte(s))
}
+57
View File
@@ -0,0 +1,57 @@
// +build !appengine
// This file encapsulates usage of unsafe.
// xxhash_safe.go contains the safe implementations.
package xxhash
import (
"unsafe"
)
// In the future it's possible that compiler optimizations will make these
// XxxString functions unnecessary by realizing that calls such as
// Sum64([]byte(s)) don't need to copy s. See https://golang.org/issue/2205.
// If that happens, even if we keep these functions they can be replaced with
// the trivial safe code.
// NOTE: The usual way of doing an unsafe string-to-[]byte conversion is:
//
// var b []byte
// bh := (*reflect.SliceHeader)(unsafe.Pointer(&b))
// bh.Data = (*reflect.StringHeader)(unsafe.Pointer(&s)).Data
// bh.Len = len(s)
// bh.Cap = len(s)
//
// Unfortunately, as of Go 1.15.3 the inliner's cost model assigns a high enough
// weight to this sequence of expressions that any function that uses it will
// not be inlined. Instead, the functions below use a different unsafe
// conversion designed to minimize the inliner weight and allow both to be
// inlined. There is also a test (TestInlining) which verifies that these are
// inlined.
//
// See https://github.com/golang/go/issues/42739 for discussion.
// Sum64String computes the 64-bit xxHash digest of s.
// It may be faster than Sum64([]byte(s)) by avoiding a copy.
func Sum64String(s string) uint64 {
b := *(*[]byte)(unsafe.Pointer(&sliceHeader{s, len(s)}))
return Sum64(b)
}
// WriteString adds more data to d. It always returns len(s), nil.
// It may be faster than Write([]byte(s)) by avoiding a copy.
func (d *Digest) WriteString(s string) (n int, err error) {
d.Write(*(*[]byte)(unsafe.Pointer(&sliceHeader{s, len(s)})))
// d.Write always returns len(s), nil.
// Ignoring the return output and returning these fixed values buys a
// savings of 6 in the inliner's cost model.
return len(s), nil
}
// sliceHeader is similar to reflect.SliceHeader, but it assumes that the layout
// of the first two words is the same as the layout of a string.
type sliceHeader struct {
s string
cap int
}
+22
View File
@@ -0,0 +1,22 @@
The MIT License (MIT)
Copyright (c) 2015 Peter Bourgon
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
+5
View File
@@ -0,0 +1,5 @@
// Package endpoint defines an abstraction for RPCs.
//
// Endpoints are a fundamental building block for many Go kit components.
// Endpoints are implemented by servers, and called by clients.
package endpoint
+40
View File
@@ -0,0 +1,40 @@
package endpoint
import (
"context"
)
// Endpoint is the fundamental building block of servers and clients.
// It represents a single RPC method.
type Endpoint func(ctx context.Context, request interface{}) (response interface{}, err error)
// Nop is an endpoint that does nothing and returns a nil error.
// Useful for tests.
func Nop(context.Context, interface{}) (interface{}, error) { return struct{}{}, nil }
// Middleware is a chainable behavior modifier for endpoints.
type Middleware func(Endpoint) Endpoint
// Chain is a helper function for composing middlewares. Requests will
// traverse them in the order they're declared. That is, the first middleware
// is treated as the outermost middleware.
func Chain(outer Middleware, others ...Middleware) Middleware {
return func(next Endpoint) Endpoint {
for i := len(others) - 1; i >= 0; i-- { // reverse
next = others[i](next)
}
return outer(next)
}
}
// Failer may be implemented by Go kit response types that contain business
// logic error details. If Failed returns a non-nil error, the Go kit transport
// layer may interpret this as a business logic error, and may encode it
// differently than a regular, successful response.
//
// It's not necessary for your response types to implement Failer, but it may
// help for more sophisticated use cases. The addsvc example shows how Failer
// should be used by a complete application.
type Failer interface {
Failed() error
}
+160
View File
@@ -0,0 +1,160 @@
# package log
**Deprecation notice:** The core Go kit log packages (log, log/level, log/term, and
log/syslog) have been moved to their own repository at github.com/go-kit/log.
The corresponding packages in this directory remain for backwards compatibility.
Their types alias the types and their functions call the functions provided by
the new repository. Using either import path should be equivalent. Prefer the
new import path when practical.
______
`package log` provides a minimal interface for structured logging in services.
It may be wrapped to encode conventions, enforce type-safety, provide leveled
logging, and so on. It can be used for both typical application log events,
and log-structured data streams.
## Structured logging
Structured logging is, basically, conceding to the reality that logs are
_data_, and warrant some level of schematic rigor. Using a stricter,
key/value-oriented message format for our logs, containing contextual and
semantic information, makes it much easier to get insight into the
operational activity of the systems we build. Consequently, `package log` is
of the strong belief that "[the benefits of structured logging outweigh the
minimal effort involved](https://www.thoughtworks.com/radar/techniques/structured-logging)".
Migrating from unstructured to structured logging is probably a lot easier
than you'd expect.
```go
// Unstructured
log.Printf("HTTP server listening on %s", addr)
// Structured
logger.Log("transport", "HTTP", "addr", addr, "msg", "listening")
```
## Usage
### Typical application logging
```go
w := log.NewSyncWriter(os.Stderr)
logger := log.NewLogfmtLogger(w)
logger.Log("question", "what is the meaning of life?", "answer", 42)
// Output:
// question="what is the meaning of life?" answer=42
```
### Contextual Loggers
```go
func main() {
var logger log.Logger
logger = log.NewLogfmtLogger(log.NewSyncWriter(os.Stderr))
logger = log.With(logger, "instance_id", 123)
logger.Log("msg", "starting")
NewWorker(log.With(logger, "component", "worker")).Run()
NewSlacker(log.With(logger, "component", "slacker")).Run()
}
// Output:
// instance_id=123 msg=starting
// instance_id=123 component=worker msg=running
// instance_id=123 component=slacker msg=running
```
### Interact with stdlib logger
Redirect stdlib logger to Go kit logger.
```go
import (
"os"
stdlog "log"
kitlog "github.com/go-kit/kit/log"
)
func main() {
logger := kitlog.NewJSONLogger(kitlog.NewSyncWriter(os.Stdout))
stdlog.SetOutput(kitlog.NewStdlibAdapter(logger))
stdlog.Print("I sure like pie")
}
// Output:
// {"msg":"I sure like pie","ts":"2016/01/01 12:34:56"}
```
Or, if, for legacy reasons, you need to pipe all of your logging through the
stdlib log package, you can redirect Go kit logger to the stdlib logger.
```go
logger := kitlog.NewLogfmtLogger(kitlog.StdlibWriter{})
logger.Log("legacy", true, "msg", "at least it's something")
// Output:
// 2016/01/01 12:34:56 legacy=true msg="at least it's something"
```
### Timestamps and callers
```go
var logger log.Logger
logger = log.NewLogfmtLogger(log.NewSyncWriter(os.Stderr))
logger = log.With(logger, "ts", log.DefaultTimestampUTC, "caller", log.DefaultCaller)
logger.Log("msg", "hello")
// Output:
// ts=2016-01-01T12:34:56Z caller=main.go:15 msg=hello
```
## Levels
Log levels are supported via the [level package](https://godoc.org/github.com/go-kit/kit/log/level).
## Supported output formats
- [Logfmt](https://brandur.org/logfmt) ([see also](https://blog.codeship.com/logfmt-a-log-format-thats-easy-to-read-and-write))
- JSON
## Enhancements
`package log` is centered on the one-method Logger interface.
```go
type Logger interface {
Log(keyvals ...interface{}) error
}
```
This interface, and its supporting code like is the product of much iteration
and evaluation. For more details on the evolution of the Logger interface,
see [The Hunt for a Logger Interface](http://go-talks.appspot.com/github.com/ChrisHines/talks/structured-logging/structured-logging.slide#1),
a talk by [Chris Hines](https://github.com/ChrisHines).
Also, please see
[#63](https://github.com/go-kit/kit/issues/63),
[#76](https://github.com/go-kit/kit/pull/76),
[#131](https://github.com/go-kit/kit/issues/131),
[#157](https://github.com/go-kit/kit/pull/157),
[#164](https://github.com/go-kit/kit/issues/164), and
[#252](https://github.com/go-kit/kit/pull/252)
to review historical conversations about package log and the Logger interface.
Value-add packages and suggestions,
like improvements to [the leveled logger](https://godoc.org/github.com/go-kit/kit/log/level),
are of course welcome. Good proposals should
- Be composable with [contextual loggers](https://godoc.org/github.com/go-kit/kit/log#With),
- Not break the behavior of [log.Caller](https://godoc.org/github.com/go-kit/kit/log#Caller) in any wrapped contextual loggers, and
- Be friendly to packages that accept only an unadorned log.Logger.
## Benchmarks & comparisons
There are a few Go logging benchmarks and comparisons that include Go kit's package log.
- [imkira/go-loggers-bench](https://github.com/imkira/go-loggers-bench) includes kit/log
- [uber-common/zap](https://github.com/uber-common/zap), a zero-alloc logging library, includes a comparison with kit/log
+118
View File
@@ -0,0 +1,118 @@
// Package log provides a structured logger.
//
// Deprecated: Use github.com/go-kit/log instead.
//
// Structured logging produces logs easily consumed later by humans or
// machines. Humans might be interested in debugging errors, or tracing
// specific requests. Machines might be interested in counting interesting
// events, or aggregating information for off-line processing. In both cases,
// it is important that the log messages are structured and actionable.
// Package log is designed to encourage both of these best practices.
//
// Basic Usage
//
// The fundamental interface is Logger. Loggers create log events from
// key/value data. The Logger interface has a single method, Log, which
// accepts a sequence of alternating key/value pairs, which this package names
// keyvals.
//
// type Logger interface {
// Log(keyvals ...interface{}) error
// }
//
// Here is an example of a function using a Logger to create log events.
//
// func RunTask(task Task, logger log.Logger) string {
// logger.Log("taskID", task.ID, "event", "starting task")
// ...
// logger.Log("taskID", task.ID, "event", "task complete")
// }
//
// The keys in the above example are "taskID" and "event". The values are
// task.ID, "starting task", and "task complete". Every key is followed
// immediately by its value.
//
// Keys are usually plain strings. Values may be any type that has a sensible
// encoding in the chosen log format. With structured logging it is a good
// idea to log simple values without formatting them. This practice allows
// the chosen logger to encode values in the most appropriate way.
//
// Contextual Loggers
//
// A contextual logger stores keyvals that it includes in all log events.
// Building appropriate contextual loggers reduces repetition and aids
// consistency in the resulting log output. With, WithPrefix, and WithSuffix
// add context to a logger. We can use With to improve the RunTask example.
//
// func RunTask(task Task, logger log.Logger) string {
// logger = log.With(logger, "taskID", task.ID)
// logger.Log("event", "starting task")
// ...
// taskHelper(task.Cmd, logger)
// ...
// logger.Log("event", "task complete")
// }
//
// The improved version emits the same log events as the original for the
// first and last calls to Log. Passing the contextual logger to taskHelper
// enables each log event created by taskHelper to include the task.ID even
// though taskHelper does not have access to that value. Using contextual
// loggers this way simplifies producing log output that enables tracing the
// life cycle of individual tasks. (See the Contextual example for the full
// code of the above snippet.)
//
// Dynamic Contextual Values
//
// A Valuer function stored in a contextual logger generates a new value each
// time an event is logged. The Valuer example demonstrates how this feature
// works.
//
// Valuers provide the basis for consistently logging timestamps and source
// code location. The log package defines several valuers for that purpose.
// See Timestamp, DefaultTimestamp, DefaultTimestampUTC, Caller, and
// DefaultCaller. A common logger initialization sequence that ensures all log
// entries contain a timestamp and source location looks like this:
//
// logger := log.NewLogfmtLogger(log.NewSyncWriter(os.Stdout))
// logger = log.With(logger, "ts", log.DefaultTimestampUTC, "caller", log.DefaultCaller)
//
// Concurrent Safety
//
// Applications with multiple goroutines want each log event written to the
// same logger to remain separate from other log events. Package log provides
// two simple solutions for concurrent safe logging.
//
// NewSyncWriter wraps an io.Writer and serializes each call to its Write
// method. Using a SyncWriter has the benefit that the smallest practical
// portion of the logging logic is performed within a mutex, but it requires
// the formatting Logger to make only one call to Write per log event.
//
// NewSyncLogger wraps any Logger and serializes each call to its Log method.
// Using a SyncLogger has the benefit that it guarantees each log event is
// handled atomically within the wrapped logger, but it typically serializes
// both the formatting and output logic. Use a SyncLogger if the formatting
// logger may perform multiple writes per log event.
//
// Error Handling
//
// This package relies on the practice of wrapping or decorating loggers with
// other loggers to provide composable pieces of functionality. It also means
// that Logger.Log must return an error because some
// implementations—especially those that output log data to an io.Writer—may
// encounter errors that cannot be handled locally. This in turn means that
// Loggers that wrap other loggers should return errors from the wrapped
// logger up the stack.
//
// Fortunately, the decorator pattern also provides a way to avoid the
// necessity to check for errors every time an application calls Logger.Log.
// An application required to panic whenever its Logger encounters
// an error could initialize its logger as follows.
//
// fmtlogger := log.NewLogfmtLogger(log.NewSyncWriter(os.Stdout))
// logger := log.LoggerFunc(func(keyvals ...interface{}) error {
// if err := fmtlogger.Log(keyvals...); err != nil {
// panic(err)
// }
// return nil
// })
package log
+15
View File
@@ -0,0 +1,15 @@
package log
import (
"io"
"github.com/go-kit/log"
)
// NewJSONLogger returns a Logger that encodes keyvals to the Writer as a
// single JSON object. Each log event produces no more than one call to
// w.Write. The passed Writer must be safe for concurrent use by multiple
// goroutines if the returned Logger will be used concurrently.
func NewJSONLogger(w io.Writer) Logger {
return log.NewJSONLogger(w)
}
+51
View File
@@ -0,0 +1,51 @@
package log
import (
"github.com/go-kit/log"
)
// Logger is the fundamental interface for all log operations. Log creates a
// log event from keyvals, a variadic sequence of alternating keys and values.
// Implementations must be safe for concurrent use by multiple goroutines. In
// particular, any implementation of Logger that appends to keyvals or
// modifies or retains any of its elements must make a copy first.
type Logger = log.Logger
// ErrMissingValue is appended to keyvals slices with odd length to substitute
// the missing value.
var ErrMissingValue = log.ErrMissingValue
// With returns a new contextual logger with keyvals prepended to those passed
// to calls to Log. If logger is also a contextual logger created by With,
// WithPrefix, or WithSuffix, keyvals is appended to the existing context.
//
// The returned Logger replaces all value elements (odd indexes) containing a
// Valuer with their generated value for each call to its Log method.
func With(logger Logger, keyvals ...interface{}) Logger {
return log.With(logger, keyvals...)
}
// WithPrefix returns a new contextual logger with keyvals prepended to those
// passed to calls to Log. If logger is also a contextual logger created by
// With, WithPrefix, or WithSuffix, keyvals is prepended to the existing context.
//
// The returned Logger replaces all value elements (odd indexes) containing a
// Valuer with their generated value for each call to its Log method.
func WithPrefix(logger Logger, keyvals ...interface{}) Logger {
return log.WithPrefix(logger, keyvals...)
}
// WithSuffix returns a new contextual logger with keyvals appended to those
// passed to calls to Log. If logger is also a contextual logger created by
// With, WithPrefix, or WithSuffix, keyvals is appended to the existing context.
//
// The returned Logger replaces all value elements (odd indexes) containing a
// Valuer with their generated value for each call to its Log method.
func WithSuffix(logger Logger, keyvals ...interface{}) Logger {
return log.WithSuffix(logger, keyvals...)
}
// LoggerFunc is an adapter to allow use of ordinary functions as Loggers. If
// f is a function with the appropriate signature, LoggerFunc(f) is a Logger
// object that calls f.
type LoggerFunc = log.LoggerFunc
+15
View File
@@ -0,0 +1,15 @@
package log
import (
"io"
"github.com/go-kit/log"
)
// NewLogfmtLogger returns a logger that encodes keyvals to the Writer in
// logfmt format. Each log event produces no more than one call to w.Write.
// The passed Writer must be safe for concurrent use by multiple goroutines if
// the returned Logger will be used concurrently.
func NewLogfmtLogger(w io.Writer) Logger {
return log.NewLogfmtLogger(w)
}
+8
View File
@@ -0,0 +1,8 @@
package log
import "github.com/go-kit/log"
// NewNopLogger returns a logger that doesn't do anything.
func NewNopLogger() Logger {
return log.NewNopLogger()
}
+54
View File
@@ -0,0 +1,54 @@
package log
import (
"io"
"github.com/go-kit/log"
)
// StdlibWriter implements io.Writer by invoking the stdlib log.Print. It's
// designed to be passed to a Go kit logger as the writer, for cases where
// it's necessary to redirect all Go kit log output to the stdlib logger.
//
// If you have any choice in the matter, you shouldn't use this. Prefer to
// redirect the stdlib log to the Go kit logger via NewStdlibAdapter.
type StdlibWriter = log.StdlibWriter
// StdlibAdapter wraps a Logger and allows it to be passed to the stdlib
// logger's SetOutput. It will extract date/timestamps, filenames, and
// messages, and place them under relevant keys.
type StdlibAdapter = log.StdlibAdapter
// StdlibAdapterOption sets a parameter for the StdlibAdapter.
type StdlibAdapterOption = log.StdlibAdapterOption
// TimestampKey sets the key for the timestamp field. By default, it's "ts".
func TimestampKey(key string) StdlibAdapterOption {
return log.TimestampKey(key)
}
// FileKey sets the key for the file and line field. By default, it's "caller".
func FileKey(key string) StdlibAdapterOption {
return log.FileKey(key)
}
// MessageKey sets the key for the actual log message. By default, it's "msg".
func MessageKey(key string) StdlibAdapterOption {
return log.MessageKey(key)
}
// Prefix configures the adapter to parse a prefix from stdlib log events. If
// you provide a non-empty prefix to the stdlib logger, then your should provide
// that same prefix to the adapter via this option.
//
// By default, the prefix isn't included in the msg key. Set joinPrefixToMsg to
// true if you want to include the parsed prefix in the msg.
func Prefix(prefix string, joinPrefixToMsg bool) StdlibAdapterOption {
return log.Prefix(prefix, joinPrefixToMsg)
}
// NewStdlibAdapter returns a new StdlibAdapter wrapper around the passed
// logger. It's designed to be passed to log.SetOutput.
func NewStdlibAdapter(logger Logger, options ...StdlibAdapterOption) io.Writer {
return log.NewStdlibAdapter(logger, options...)
}
+37
View File
@@ -0,0 +1,37 @@
package log
import (
"io"
"github.com/go-kit/log"
)
// SwapLogger wraps another logger that may be safely replaced while other
// goroutines use the SwapLogger concurrently. The zero value for a SwapLogger
// will discard all log events without error.
//
// SwapLogger serves well as a package global logger that can be changed by
// importers.
type SwapLogger = log.SwapLogger
// NewSyncWriter returns a new writer that is safe for concurrent use by
// multiple goroutines. Writes to the returned writer are passed on to w. If
// another write is already in progress, the calling goroutine blocks until
// the writer is available.
//
// If w implements the following interface, so does the returned writer.
//
// interface {
// Fd() uintptr
// }
func NewSyncWriter(w io.Writer) io.Writer {
return log.NewSyncWriter(w)
}
// NewSyncLogger returns a logger that synchronizes concurrent use of the
// wrapped logger. When multiple goroutines use the SyncLogger concurrently
// only one goroutine will be allowed to log to the wrapped logger at a time.
// The other goroutines will block until the logger is available.
func NewSyncLogger(logger Logger) Logger {
return log.NewSyncLogger(logger)
}
+52
View File
@@ -0,0 +1,52 @@
package log
import (
"time"
"github.com/go-kit/log"
)
// A Valuer generates a log value. When passed to With, WithPrefix, or
// WithSuffix in a value element (odd indexes), it represents a dynamic
// value which is re-evaluated with each log event.
type Valuer = log.Valuer
// Timestamp returns a timestamp Valuer. It invokes the t function to get the
// time; unless you are doing something tricky, pass time.Now.
//
// Most users will want to use DefaultTimestamp or DefaultTimestampUTC, which
// are TimestampFormats that use the RFC3339Nano format.
func Timestamp(t func() time.Time) Valuer {
return log.Timestamp(t)
}
// TimestampFormat returns a timestamp Valuer with a custom time format. It
// invokes the t function to get the time to format; unless you are doing
// something tricky, pass time.Now. The layout string is passed to
// Time.Format.
//
// Most users will want to use DefaultTimestamp or DefaultTimestampUTC, which
// are TimestampFormats that use the RFC3339Nano format.
func TimestampFormat(t func() time.Time, layout string) Valuer {
return log.TimestampFormat(t, layout)
}
// Caller returns a Valuer that returns a file and line from a specified depth
// in the callstack. Users will probably want to use DefaultCaller.
func Caller(depth int) Valuer {
return log.Caller(depth)
}
var (
// DefaultTimestamp is a Valuer that returns the current wallclock time,
// respecting time zones, when bound.
DefaultTimestamp = log.DefaultTimestamp
// DefaultTimestampUTC is a Valuer that returns the current time in UTC
// when bound.
DefaultTimestampUTC = log.DefaultTimestampUTC
// DefaultCaller is a Valuer that returns the file and line where the Log
// method was invoked. It can only be used with log.With.
DefaultCaller = log.DefaultCaller
)
+98
View File
@@ -0,0 +1,98 @@
# package metrics
`package metrics` provides a set of uniform interfaces for service instrumentation.
It has
[counters](http://prometheus.io/docs/concepts/metric_types/#counter),
[gauges](http://prometheus.io/docs/concepts/metric_types/#gauge), and
[histograms](http://prometheus.io/docs/concepts/metric_types/#histogram),
and provides adapters to popular metrics packages, like
[expvar](https://golang.org/pkg/expvar),
[StatsD](https://github.com/etsy/statsd), and
[Prometheus](https://prometheus.io).
## Rationale
Code instrumentation is absolutely essential to achieve
[observability](https://speakerdeck.com/mattheath/observability-in-micro-service-architectures)
into a distributed system.
Metrics and instrumentation tools have coalesced around a few well-defined idioms.
`package metrics` provides a common, minimal interface those idioms for service authors.
## Usage
A simple counter, exported via expvar.
```go
import (
"github.com/go-kit/kit/metrics"
"github.com/go-kit/kit/metrics/expvar"
)
func main() {
var myCount metrics.Counter
myCount = expvar.NewCounter("my_count")
myCount.Add(1)
}
```
A histogram for request duration,
exported via a Prometheus summary with dynamically-computed quantiles.
```go
import (
"time"
stdprometheus "github.com/prometheus/client_golang/prometheus"
"github.com/go-kit/kit/metrics"
"github.com/go-kit/kit/metrics/prometheus"
)
func main() {
var dur metrics.Histogram = prometheus.NewSummaryFrom(stdprometheus.SummaryOpts{
Namespace: "myservice",
Subsystem: "api",
Name: "request_duration_seconds",
Help: "Total time spent serving requests.",
}, []string{})
// ...
}
func handleRequest(dur metrics.Histogram) {
defer func(begin time.Time) { dur.Observe(time.Since(begin).Seconds()) }(time.Now())
// handle request
}
```
A gauge for the number of goroutines currently running, exported via StatsD.
```go
import (
"context"
"net"
"os"
"runtime"
"time"
"github.com/go-kit/kit/metrics"
"github.com/go-kit/kit/metrics/statsd"
)
func main() {
statsd := statsd.New("foo_svc.", log.NewNopLogger())
report := time.NewTicker(5 * time.Second)
defer report.Stop()
go statsd.SendLoop(context.Background(), report.C, "tcp", "statsd.internal:8125")
goroutines := statsd.NewGauge("goroutine_count")
go exportGoroutines(goroutines)
// ...
}
func exportGoroutines(g metrics.Gauge) {
for range time.Tick(time.Second) {
g.Set(float64(runtime.NumGoroutine()))
}
}
```
For more information, see [the package documentation](https://godoc.org/github.com/go-kit/kit/metrics).
+97
View File
@@ -0,0 +1,97 @@
// Package metrics provides a framework for application instrumentation. It's
// primarily designed to help you get started with good and robust
// instrumentation, and to help you migrate from a less-capable system like
// Graphite to a more-capable system like Prometheus. If your organization has
// already standardized on an instrumentation system like Prometheus, and has no
// plans to change, it may make sense to use that system's instrumentation
// library directly.
//
// This package provides three core metric abstractions (Counter, Gauge, and
// Histogram) and implementations for almost all common instrumentation
// backends. Each metric has an observation method (Add, Set, or Observe,
// respectively) used to record values, and a With method to "scope" the
// observation by various parameters. For example, you might have a Histogram to
// record request durations, parameterized by the method that's being called.
//
// var requestDuration metrics.Histogram
// // ...
// requestDuration.With("method", "MyMethod").Observe(time.Since(begin))
//
// This allows a single high-level metrics object (requestDuration) to work with
// many code paths somewhat dynamically. The concept of With is fully supported
// in some backends like Prometheus, and not supported in other backends like
// Graphite. So, With may be a no-op, depending on the concrete implementation
// you choose. Please check the implementation to know for sure. For
// implementations that don't provide With, it's necessary to fully parameterize
// each metric in the metric name, e.g.
//
// // Statsd
// c := statsd.NewCounter("request_duration_MyMethod_200")
// c.Add(1)
//
// // Prometheus
// c := prometheus.NewCounter(stdprometheus.CounterOpts{
// Name: "request_duration",
// ...
// }, []string{"method", "status_code"})
// c.With("method", "MyMethod", "status_code", strconv.Itoa(code)).Add(1)
//
// Usage
//
// Metrics are dependencies, and should be passed to the components that need
// them in the same way you'd construct and pass a database handle, or reference
// to another component. Metrics should *not* be created in the global scope.
// Instead, instantiate metrics in your func main, using whichever concrete
// implementation is appropriate for your organization.
//
// latency := prometheus.NewSummaryFrom(stdprometheus.SummaryOpts{
// Namespace: "myteam",
// Subsystem: "foosvc",
// Name: "request_latency_seconds",
// Help: "Incoming request latency in seconds.",
// }, []string{"method", "status_code"})
//
// Write your components to take the metrics they will use as parameters to
// their constructors. Use the interface types, not the concrete types. That is,
//
// // NewAPI takes metrics.Histogram, not *prometheus.Summary
// func NewAPI(s Store, logger log.Logger, latency metrics.Histogram) *API {
// // ...
// }
//
// func (a *API) ServeFoo(w http.ResponseWriter, r *http.Request) {
// begin := time.Now()
// // ...
// a.latency.Observe(time.Since(begin).Seconds())
// }
//
// Finally, pass the metrics as dependencies when building your object graph.
// This should happen in func main, not in the global scope.
//
// api := NewAPI(store, logger, latency)
// http.ListenAndServe("/", api)
//
// Note that metrics are "write-only" interfaces.
//
// Implementation details
//
// All metrics are safe for concurrent use. Considerable design influence has
// been taken from https://github.com/codahale/metrics and
// https://prometheus.io.
//
// Each telemetry system has different semantics for label values, push vs.
// pull, support for histograms, etc. These properties influence the design of
// their respective packages. This table attempts to summarize the key points of
// distinction.
//
// SYSTEM DIM COUNTERS GAUGES HISTOGRAMS
// dogstatsd n batch, push-aggregate batch, push-aggregate native, batch, push-each
// statsd 1 batch, push-aggregate batch, push-aggregate native, batch, push-each
// graphite 1 batch, push-aggregate batch, push-aggregate synthetic, batch, push-aggregate
// expvar 1 atomic atomic synthetic, batch, in-place expose
// influx n custom custom custom
// prometheus n native native native
// pcp 1 native native native
// cloudwatch n batch push-aggregate batch push-aggregate synthetic, batch, push-aggregate
//
package metrics
+14
View File
@@ -0,0 +1,14 @@
package lv
// LabelValues is a type alias that provides validation on its With method.
// Metrics may include it as a member to help them satisfy With semantics and
// save some code duplication.
type LabelValues []string
// With validates the input, and returns a new aggregate labelValues.
func (lvs LabelValues) With(labelValues ...string) LabelValues {
if len(labelValues)%2 != 0 {
labelValues = append(labelValues, "unknown")
}
return append(lvs, labelValues...)
}
+145
View File
@@ -0,0 +1,145 @@
package lv
import "sync"
// NewSpace returns an N-dimensional vector space.
func NewSpace() *Space {
return &Space{}
}
// Space represents an N-dimensional vector space. Each name and unique label
// value pair establishes a new dimension and point within that dimension. Order
// matters, i.e. [a=1 b=2] identifies a different timeseries than [b=2 a=1].
type Space struct {
mtx sync.RWMutex
nodes map[string]*node
}
// Observe locates the time series identified by the name and label values in
// the vector space, and appends the value to the list of observations.
func (s *Space) Observe(name string, lvs LabelValues, value float64) {
s.nodeFor(name).observe(lvs, value)
}
// Add locates the time series identified by the name and label values in
// the vector space, and appends the delta to the last value in the list of
// observations.
func (s *Space) Add(name string, lvs LabelValues, delta float64) {
s.nodeFor(name).add(lvs, delta)
}
// Walk traverses the vector space and invokes fn for each non-empty time series
// which is encountered. Return false to abort the traversal.
func (s *Space) Walk(fn func(name string, lvs LabelValues, observations []float64) bool) {
s.mtx.RLock()
defer s.mtx.RUnlock()
for name, node := range s.nodes {
f := func(lvs LabelValues, observations []float64) bool { return fn(name, lvs, observations) }
if !node.walk(LabelValues{}, f) {
return
}
}
}
// Reset empties the current space and returns a new Space with the old
// contents. Reset a Space to get an immutable copy suitable for walking.
func (s *Space) Reset() *Space {
s.mtx.Lock()
defer s.mtx.Unlock()
n := NewSpace()
n.nodes, s.nodes = s.nodes, n.nodes
return n
}
func (s *Space) nodeFor(name string) *node {
s.mtx.Lock()
defer s.mtx.Unlock()
if s.nodes == nil {
s.nodes = map[string]*node{}
}
n, ok := s.nodes[name]
if !ok {
n = &node{}
s.nodes[name] = n
}
return n
}
// node exists at a specific point in the N-dimensional vector space of all
// possible label values. The node collects observations and has child nodes
// with greater specificity.
type node struct {
mtx sync.RWMutex
observations []float64
children map[pair]*node
}
type pair struct{ label, value string }
func (n *node) observe(lvs LabelValues, value float64) {
n.mtx.Lock()
defer n.mtx.Unlock()
if len(lvs) <= 0 {
n.observations = append(n.observations, value)
return
}
if len(lvs) < 2 {
panic("too few LabelValues; programmer error!")
}
head, tail := pair{lvs[0], lvs[1]}, lvs[2:]
if n.children == nil {
n.children = map[pair]*node{}
}
child, ok := n.children[head]
if !ok {
child = &node{}
n.children[head] = child
}
child.observe(tail, value)
}
func (n *node) add(lvs LabelValues, delta float64) {
n.mtx.Lock()
defer n.mtx.Unlock()
if len(lvs) <= 0 {
var value float64
if len(n.observations) > 0 {
value = last(n.observations) + delta
} else {
value = delta
}
n.observations = append(n.observations, value)
return
}
if len(lvs) < 2 {
panic("too few LabelValues; programmer error!")
}
head, tail := pair{lvs[0], lvs[1]}, lvs[2:]
if n.children == nil {
n.children = map[pair]*node{}
}
child, ok := n.children[head]
if !ok {
child = &node{}
n.children[head] = child
}
child.add(tail, delta)
}
func (n *node) walk(lvs LabelValues, fn func(LabelValues, []float64) bool) bool {
n.mtx.RLock()
defer n.mtx.RUnlock()
if len(n.observations) > 0 && !fn(lvs, n.observations) {
return false
}
for p, child := range n.children {
if !child.walk(append(lvs, p.label, p.value), fn) {
return false
}
}
return true
}
func last(a []float64) float64 {
return a[len(a)-1]
}
+25
View File
@@ -0,0 +1,25 @@
package metrics
// Counter describes a metric that accumulates values monotonically.
// An example of a counter is the number of received HTTP requests.
type Counter interface {
With(labelValues ...string) Counter
Add(delta float64)
}
// Gauge describes a metric that takes specific values over time.
// An example of a gauge is the current depth of a job queue.
type Gauge interface {
With(labelValues ...string) Gauge
Set(value float64)
Add(delta float64)
}
// Histogram describes a metric that takes repeated observations of the same
// kind of thing, and produces a statistical summary of those observations,
// typically expressed as quantiles or buckets. An example of a histogram is
// HTTP request latencies.
type Histogram interface {
With(labelValues ...string) Histogram
Observe(value float64)
}
+165
View File
@@ -0,0 +1,165 @@
// Package prometheus provides Prometheus implementations for metrics.
// Individual metrics are mapped to their Prometheus counterparts, and
// (depending on the constructor used) may be automatically registered in the
// global Prometheus metrics registry.
package prometheus
import (
"github.com/prometheus/client_golang/prometheus"
"github.com/go-kit/kit/metrics"
"github.com/go-kit/kit/metrics/internal/lv"
)
// Counter implements Counter, via a Prometheus CounterVec.
type Counter struct {
cv *prometheus.CounterVec
lvs lv.LabelValues
}
// NewCounterFrom constructs and registers a Prometheus CounterVec,
// and returns a usable Counter object.
func NewCounterFrom(opts prometheus.CounterOpts, labelNames []string) *Counter {
cv := prometheus.NewCounterVec(opts, labelNames)
prometheus.MustRegister(cv)
return NewCounter(cv)
}
// NewCounter wraps the CounterVec and returns a usable Counter object.
func NewCounter(cv *prometheus.CounterVec) *Counter {
return &Counter{
cv: cv,
}
}
// With implements Counter.
func (c *Counter) With(labelValues ...string) metrics.Counter {
return &Counter{
cv: c.cv,
lvs: c.lvs.With(labelValues...),
}
}
// Add implements Counter.
func (c *Counter) Add(delta float64) {
c.cv.With(makeLabels(c.lvs...)).Add(delta)
}
// Gauge implements Gauge, via a Prometheus GaugeVec.
type Gauge struct {
gv *prometheus.GaugeVec
lvs lv.LabelValues
}
// NewGaugeFrom constructs and registers a Prometheus GaugeVec,
// and returns a usable Gauge object.
func NewGaugeFrom(opts prometheus.GaugeOpts, labelNames []string) *Gauge {
gv := prometheus.NewGaugeVec(opts, labelNames)
prometheus.MustRegister(gv)
return NewGauge(gv)
}
// NewGauge wraps the GaugeVec and returns a usable Gauge object.
func NewGauge(gv *prometheus.GaugeVec) *Gauge {
return &Gauge{
gv: gv,
}
}
// With implements Gauge.
func (g *Gauge) With(labelValues ...string) metrics.Gauge {
return &Gauge{
gv: g.gv,
lvs: g.lvs.With(labelValues...),
}
}
// Set implements Gauge.
func (g *Gauge) Set(value float64) {
g.gv.With(makeLabels(g.lvs...)).Set(value)
}
// Add is supported by Prometheus GaugeVecs.
func (g *Gauge) Add(delta float64) {
g.gv.With(makeLabels(g.lvs...)).Add(delta)
}
// Summary implements Histogram, via a Prometheus SummaryVec. The difference
// between a Summary and a Histogram is that Summaries don't require predefined
// quantile buckets, but cannot be statistically aggregated.
type Summary struct {
sv *prometheus.SummaryVec
lvs lv.LabelValues
}
// NewSummaryFrom constructs and registers a Prometheus SummaryVec,
// and returns a usable Summary object.
func NewSummaryFrom(opts prometheus.SummaryOpts, labelNames []string) *Summary {
sv := prometheus.NewSummaryVec(opts, labelNames)
prometheus.MustRegister(sv)
return NewSummary(sv)
}
// NewSummary wraps the SummaryVec and returns a usable Summary object.
func NewSummary(sv *prometheus.SummaryVec) *Summary {
return &Summary{
sv: sv,
}
}
// With implements Histogram.
func (s *Summary) With(labelValues ...string) metrics.Histogram {
return &Summary{
sv: s.sv,
lvs: s.lvs.With(labelValues...),
}
}
// Observe implements Histogram.
func (s *Summary) Observe(value float64) {
s.sv.With(makeLabels(s.lvs...)).Observe(value)
}
// Histogram implements Histogram via a Prometheus HistogramVec. The difference
// between a Histogram and a Summary is that Histograms require predefined
// quantile buckets, and can be statistically aggregated.
type Histogram struct {
hv *prometheus.HistogramVec
lvs lv.LabelValues
}
// NewHistogramFrom constructs and registers a Prometheus HistogramVec,
// and returns a usable Histogram object.
func NewHistogramFrom(opts prometheus.HistogramOpts, labelNames []string) *Histogram {
hv := prometheus.NewHistogramVec(opts, labelNames)
prometheus.MustRegister(hv)
return NewHistogram(hv)
}
// NewHistogram wraps the HistogramVec and returns a usable Histogram object.
func NewHistogram(hv *prometheus.HistogramVec) *Histogram {
return &Histogram{
hv: hv,
}
}
// With implements Histogram.
func (h *Histogram) With(labelValues ...string) metrics.Histogram {
return &Histogram{
hv: h.hv,
lvs: h.lvs.With(labelValues...),
}
}
// Observe implements Histogram.
func (h *Histogram) Observe(value float64) {
h.hv.With(makeLabels(h.lvs...)).Observe(value)
}
func makeLabels(labelValues ...string) prometheus.Labels {
labels := prometheus.Labels{}
for i := 0; i < len(labelValues); i += 2 {
labels[labelValues[i]] = labelValues[i+1]
}
return labels
}
+36
View File
@@ -0,0 +1,36 @@
package metrics
import "time"
// Timer acts as a stopwatch, sending observations to a wrapped histogram.
// It's a bit of helpful syntax sugar for h.Observe(time.Since(x)).
type Timer struct {
h Histogram
t time.Time
u time.Duration
}
// NewTimer wraps the given histogram and records the current time.
func NewTimer(h Histogram) *Timer {
return &Timer{
h: h,
t: time.Now(),
u: time.Second,
}
}
// ObserveDuration captures the number of seconds since the timer was
// constructed, and forwards that observation to the histogram.
func (t *Timer) ObserveDuration() {
d := float64(time.Since(t.t).Nanoseconds()) / float64(t.u)
if d < 0 {
d = 0
}
t.h.Observe(d)
}
// Unit sets the unit of the float64 emitted by the timer.
// By default, the timer emits seconds.
func (t *Timer) Unit(u time.Duration) {
t.u = u
}
+6
View File
@@ -0,0 +1,6 @@
// Package sd provides utilities related to service discovery. That includes the
// client-side loadbalancer pattern, where a microservice subscribes to a
// service discovery system in order to reach remote instances; as well as the
// registrator pattern, where a microservice registers itself in a service
// discovery system. Implementations are provided for most common systems.
package sd
+143
View File
@@ -0,0 +1,143 @@
package sd
import (
"io"
"sort"
"sync"
"time"
"github.com/go-kit/kit/endpoint"
"github.com/go-kit/log"
)
// endpointCache collects the most recent set of instances from a service discovery
// system, creates endpoints for them using a factory function, and makes
// them available to consumers.
type endpointCache struct {
options endpointerOptions
mtx sync.RWMutex
factory Factory
cache map[string]endpointCloser
err error
endpoints []endpoint.Endpoint
logger log.Logger
invalidateDeadline time.Time
timeNow func() time.Time
}
type endpointCloser struct {
endpoint.Endpoint
io.Closer
}
// newEndpointCache returns a new, empty endpointCache.
func newEndpointCache(factory Factory, logger log.Logger, options endpointerOptions) *endpointCache {
return &endpointCache{
options: options,
factory: factory,
cache: map[string]endpointCloser{},
logger: logger,
timeNow: time.Now,
}
}
// Update should be invoked by clients with a complete set of current instance
// strings whenever that set changes. The cache manufactures new endpoints via
// the factory, closes old endpoints when they disappear, and persists existing
// endpoints if they survive through an update.
func (c *endpointCache) Update(event Event) {
c.mtx.Lock()
defer c.mtx.Unlock()
// Happy path.
if event.Err == nil {
c.updateCache(event.Instances)
c.err = nil
return
}
// Sad path. Something's gone wrong in sd.
c.logger.Log("err", event.Err)
if !c.options.invalidateOnError {
return // keep returning the last known endpoints on error
}
if c.err != nil {
return // already in the error state, do nothing & keep original error
}
c.err = event.Err
// set new deadline to invalidate Endpoints unless non-error Event is received
c.invalidateDeadline = c.timeNow().Add(c.options.invalidateTimeout)
return
}
func (c *endpointCache) updateCache(instances []string) {
// Deterministic order (for later).
sort.Strings(instances)
// Produce the current set of services.
cache := make(map[string]endpointCloser, len(instances))
for _, instance := range instances {
// If it already exists, just copy it over.
if sc, ok := c.cache[instance]; ok {
cache[instance] = sc
delete(c.cache, instance)
continue
}
// If it doesn't exist, create it.
service, closer, err := c.factory(instance)
if err != nil {
c.logger.Log("instance", instance, "err", err)
continue
}
cache[instance] = endpointCloser{service, closer}
}
// Close any leftover endpoints.
for _, sc := range c.cache {
if sc.Closer != nil {
sc.Closer.Close()
}
}
// Populate the slice of endpoints.
endpoints := make([]endpoint.Endpoint, 0, len(cache))
for _, instance := range instances {
// A bad factory may mean an instance is not present.
if _, ok := cache[instance]; !ok {
continue
}
endpoints = append(endpoints, cache[instance].Endpoint)
}
// Swap and trigger GC for old copies.
c.endpoints = endpoints
c.cache = cache
}
// Endpoints yields the current set of (presumably identical) endpoints, ordered
// lexicographically by the corresponding instance string.
func (c *endpointCache) Endpoints() ([]endpoint.Endpoint, error) {
// in the steady state we're going to have many goroutines calling Endpoints()
// concurrently, so to minimize contention we use a shared R-lock.
c.mtx.RLock()
if c.err == nil || c.timeNow().Before(c.invalidateDeadline) {
defer c.mtx.RUnlock()
return c.endpoints, nil
}
c.mtx.RUnlock()
// in case of an error, switch to an exclusive lock.
c.mtx.Lock()
defer c.mtx.Unlock()
// re-check condition due to a race between RUnlock() and Lock().
if c.err == nil || c.timeNow().Before(c.invalidateDeadline) {
return c.endpoints, nil
}
c.updateCache(nil) // close any remaining active endpoints
return nil, c.err
}
+90
View File
@@ -0,0 +1,90 @@
package sd
import (
"time"
"github.com/go-kit/kit/endpoint"
"github.com/go-kit/log"
)
// Endpointer listens to a service discovery system and yields a set of
// identical endpoints on demand. An error indicates a problem with connectivity
// to the service discovery system, or within the system itself; an Endpointer
// may yield no endpoints without error.
type Endpointer interface {
Endpoints() ([]endpoint.Endpoint, error)
}
// FixedEndpointer yields a fixed set of endpoints.
type FixedEndpointer []endpoint.Endpoint
// Endpoints implements Endpointer.
func (s FixedEndpointer) Endpoints() ([]endpoint.Endpoint, error) { return s, nil }
// NewEndpointer creates an Endpointer that subscribes to updates from Instancer src
// and uses factory f to create Endpoints. If src notifies of an error, the Endpointer
// keeps returning previously created Endpoints assuming they are still good, unless
// this behavior is disabled via InvalidateOnError option.
func NewEndpointer(src Instancer, f Factory, logger log.Logger, options ...EndpointerOption) *DefaultEndpointer {
opts := endpointerOptions{}
for _, opt := range options {
opt(&opts)
}
se := &DefaultEndpointer{
cache: newEndpointCache(f, logger, opts),
instancer: src,
ch: make(chan Event),
}
go se.receive()
src.Register(se.ch)
return se
}
// EndpointerOption allows control of endpointCache behavior.
type EndpointerOption func(*endpointerOptions)
// InvalidateOnError returns EndpointerOption that controls how the Endpointer
// behaves when then Instancer publishes an Event containing an error.
// Without this option the Endpointer continues returning the last known
// endpoints. With this option, the Endpointer continues returning the last
// known endpoints until the timeout elapses, then closes all active endpoints
// and starts returning an error. Once the Instancer sends a new update with
// valid resource instances, the normal operation is resumed.
func InvalidateOnError(timeout time.Duration) EndpointerOption {
return func(opts *endpointerOptions) {
opts.invalidateOnError = true
opts.invalidateTimeout = timeout
}
}
type endpointerOptions struct {
invalidateOnError bool
invalidateTimeout time.Duration
}
// DefaultEndpointer implements an Endpointer interface.
// When created with NewEndpointer function, it automatically registers
// as a subscriber to events from the Instances and maintains a list
// of active Endpoints.
type DefaultEndpointer struct {
cache *endpointCache
instancer Instancer
ch chan Event
}
func (de *DefaultEndpointer) receive() {
for event := range de.ch {
de.cache.Update(event)
}
}
// Close deregisters DefaultEndpointer from the Instancer and stops the internal go-routine.
func (de *DefaultEndpointer) Close() {
de.instancer.Deregister(de.ch)
close(de.ch)
}
// Endpoints implements Endpointer.
func (de *DefaultEndpointer) Endpoints() ([]endpoint.Endpoint, error) {
return de.cache.Endpoints()
}
+17
View File
@@ -0,0 +1,17 @@
package sd
import (
"io"
"github.com/go-kit/kit/endpoint"
)
// Factory is a function that converts an instance string (e.g. host:port) to a
// specific endpoint. Instances that provide multiple endpoints require multiple
// factories. A factory also returns an io.Closer that's invoked when the
// instance goes away and needs to be cleaned up. Factories may return nil
// closers.
//
// Users are expected to provide their own factory functions that assume
// specific transports, or can deduce transports by parsing the instance string.
type Factory func(instance string) (endpoint.Endpoint, io.Closer, error)
+38
View File
@@ -0,0 +1,38 @@
package sd
// Event represents a push notification generated from the underlying service discovery
// implementation. It contains either a full set of available resource instances, or
// an error indicating some issue with obtaining information from discovery backend.
// Examples of errors may include loosing connection to the discovery backend, or
// trying to look up resource instances using an incorrectly formatted key.
// After receiving an Event with an error the listenter should treat previously discovered
// resource instances as stale (although it may choose to continue using them).
// If the Instancer is able to restore connection to the discovery backend it must push
// another Event with the current set of resource instances.
type Event struct {
Instances []string
Err error
}
// Instancer listens to a service discovery system and notifies registered
// observers of changes in the resource instances. Every event sent to the channels
// contains a complete set of instances known to the Instancer. That complete set is
// sent immediately upon registering the channel, and on any future updates from
// discovery system.
type Instancer interface {
Register(chan<- Event)
Deregister(chan<- Event)
Stop()
}
// FixedInstancer yields a fixed set of instances.
type FixedInstancer []string
// Register implements Instancer.
func (d FixedInstancer) Register(ch chan<- Event) { ch <- Event{Instances: d} }
// Deregister implements Instancer.
func (d FixedInstancer) Deregister(ch chan<- Event) {}
// Stop implements Instancer.
func (d FixedInstancer) Stop() {}
+15
View File
@@ -0,0 +1,15 @@
package lb
import (
"errors"
"github.com/go-kit/kit/endpoint"
)
// Balancer yields endpoints according to some heuristic.
type Balancer interface {
Endpoint() (endpoint.Endpoint, error)
}
// ErrNoEndpoints is returned when no qualifying endpoints are available.
var ErrNoEndpoints = errors.New("no endpoints available")
+4
View File
@@ -0,0 +1,4 @@
// Package lb implements the client-side load balancer pattern. When combined
// with a service discovery system of record, it enables a more decentralized
// architecture, removing the need for separate load balancers like HAProxy.
package lb
+32
View File
@@ -0,0 +1,32 @@
package lb
import (
"math/rand"
"github.com/go-kit/kit/endpoint"
"github.com/go-kit/kit/sd"
)
// NewRandom returns a load balancer that selects services randomly.
func NewRandom(s sd.Endpointer, seed int64) Balancer {
return &random{
s: s,
r: rand.New(rand.NewSource(seed)),
}
}
type random struct {
s sd.Endpointer
r *rand.Rand
}
func (r *random) Endpoint() (endpoint.Endpoint, error) {
endpoints, err := r.s.Endpoints()
if err != nil {
return nil, err
}
if len(endpoints) <= 0 {
return nil, ErrNoEndpoints
}
return endpoints[r.r.Intn(len(endpoints))], nil
}
+117
View File
@@ -0,0 +1,117 @@
package lb
import (
"context"
"fmt"
"strings"
"time"
"github.com/go-kit/kit/endpoint"
)
// RetryError is an error wrapper that is used by the retry mechanism. All
// errors returned by the retry mechanism via its endpoint will be RetryErrors.
type RetryError struct {
RawErrors []error // all errors encountered from endpoints directly
Final error // the final, terminating error
}
func (e RetryError) Error() string {
var suffix string
if len(e.RawErrors) > 1 {
a := make([]string, len(e.RawErrors)-1)
for i := 0; i < len(e.RawErrors)-1; i++ { // last one is Final
a[i] = e.RawErrors[i].Error()
}
suffix = fmt.Sprintf(" (previously: %s)", strings.Join(a, "; "))
}
return fmt.Sprintf("%v%s", e.Final, suffix)
}
// Callback is a function that is given the current attempt count and the error
// received from the underlying endpoint. It should return whether the Retry
// function should continue trying to get a working endpoint, and a custom error
// if desired. The error message may be nil, but a true/false is always
// expected. In all cases, if the replacement error is supplied, the received
// error will be replaced in the calling context.
type Callback func(n int, received error) (keepTrying bool, replacement error)
// Retry wraps a service load balancer and returns an endpoint oriented load
// balancer for the specified service method. Requests to the endpoint will be
// automatically load balanced via the load balancer. Requests that return
// errors will be retried until they succeed, up to max times, or until the
// timeout is elapsed, whichever comes first.
func Retry(max int, timeout time.Duration, b Balancer) endpoint.Endpoint {
return RetryWithCallback(timeout, b, maxRetries(max))
}
func maxRetries(max int) Callback {
return func(n int, err error) (keepTrying bool, replacement error) {
return n < max, nil
}
}
func alwaysRetry(int, error) (keepTrying bool, replacement error) {
return true, nil
}
// RetryWithCallback wraps a service load balancer and returns an endpoint
// oriented load balancer for the specified service method. Requests to the
// endpoint will be automatically load balanced via the load balancer. Requests
// that return errors will be retried until they succeed, up to max times, until
// the callback returns false, or until the timeout is elapsed, whichever comes
// first.
func RetryWithCallback(timeout time.Duration, b Balancer, cb Callback) endpoint.Endpoint {
if cb == nil {
cb = alwaysRetry
}
if b == nil {
panic("nil Balancer")
}
return func(ctx context.Context, request interface{}) (response interface{}, err error) {
var (
newctx, cancel = context.WithTimeout(ctx, timeout)
responses = make(chan interface{}, 1)
errs = make(chan error, 1)
final RetryError
)
defer cancel()
for i := 1; ; i++ {
go func() {
e, err := b.Endpoint()
if err != nil {
errs <- err
return
}
response, err := e(newctx, request)
if err != nil {
errs <- err
return
}
responses <- response
}()
select {
case <-newctx.Done():
return nil, newctx.Err()
case response := <-responses:
return response, nil
case err := <-errs:
final.RawErrors = append(final.RawErrors, err)
keepTrying, replacement := cb(i, err)
if replacement != nil {
err = replacement
}
if !keepTrying {
final.Final = err
return nil, final
}
continue
}
}
}
}
+34
View File
@@ -0,0 +1,34 @@
package lb
import (
"sync/atomic"
"github.com/go-kit/kit/endpoint"
"github.com/go-kit/kit/sd"
)
// NewRoundRobin returns a load balancer that returns services in sequence.
func NewRoundRobin(s sd.Endpointer) Balancer {
return &roundRobin{
s: s,
c: 0,
}
}
type roundRobin struct {
s sd.Endpointer
c uint64
}
func (rr *roundRobin) Endpoint() (endpoint.Endpoint, error) {
endpoints, err := rr.s.Endpoints()
if err != nil {
return nil, err
}
if len(endpoints) <= 0 {
return nil, ErrNoEndpoints
}
old := atomic.AddUint64(&rr.c, 1) - 1
idx := old % uint64(len(endpoints))
return endpoints[idx], nil
}
+13
View File
@@ -0,0 +1,13 @@
package sd
// Registrar registers instance information to a service discovery system when
// an instance becomes alive and healthy, and deregisters that information when
// the service becomes unhealthy or goes away.
//
// Registrar implementations exist for various service discovery systems. Note
// that identifying instance information (e.g. host:port) must be given via the
// concrete constructor; this interface merely signals lifecycle changes.
type Registrar interface {
Register()
Deregister()
}
+4
View File
@@ -0,0 +1,4 @@
// Package opentracing provides Go kit integration to the OpenTracing project.
// OpenTracing implements a general purpose interface that microservices can
// program against, and which adapts to all major distributed tracing systems.
package opentracing
+124
View File
@@ -0,0 +1,124 @@
package opentracing
import (
"context"
"strconv"
"github.com/opentracing/opentracing-go"
otext "github.com/opentracing/opentracing-go/ext"
otlog "github.com/opentracing/opentracing-go/log"
"github.com/go-kit/kit/endpoint"
"github.com/go-kit/kit/sd/lb"
)
// TraceEndpoint returns a Middleware that wraps the `next` Endpoint in an
// OpenTracing Span called `operationName`.
//
// If `ctx` already has a Span, child span is created from it.
// If `ctx` doesn't yet have a Span, the new one is created.
func TraceEndpoint(tracer opentracing.Tracer, operationName string, opts ...EndpointOption) endpoint.Middleware {
cfg := &EndpointOptions{
Tags: make(opentracing.Tags),
}
for _, opt := range opts {
opt(cfg)
}
return func(next endpoint.Endpoint) endpoint.Endpoint {
return func(ctx context.Context, request interface{}) (response interface{}, err error) {
if cfg.GetOperationName != nil {
if newOperationName := cfg.GetOperationName(ctx, operationName); newOperationName != "" {
operationName = newOperationName
}
}
var span opentracing.Span
if parentSpan := opentracing.SpanFromContext(ctx); parentSpan != nil {
span = tracer.StartSpan(
operationName,
opentracing.ChildOf(parentSpan.Context()),
)
} else {
span = tracer.StartSpan(operationName)
}
defer span.Finish()
applyTags(span, cfg.Tags)
if cfg.GetTags != nil {
extraTags := cfg.GetTags(ctx)
applyTags(span, extraTags)
}
ctx = opentracing.ContextWithSpan(ctx, span)
defer func() {
if err != nil {
if lbErr, ok := err.(lb.RetryError); ok {
// handle errors originating from lb.Retry
fields := make([]otlog.Field, 0, len(lbErr.RawErrors))
for idx, rawErr := range lbErr.RawErrors {
fields = append(fields, otlog.String(
"gokit.retry.error."+strconv.Itoa(idx+1), rawErr.Error(),
))
}
otext.LogError(span, lbErr, fields...)
return
}
// generic error
otext.LogError(span, err)
return
}
// test for business error
if res, ok := response.(endpoint.Failer); ok && res.Failed() != nil {
span.LogFields(
otlog.String("gokit.business.error", res.Failed().Error()),
)
if cfg.IgnoreBusinessError {
return
}
// treating business error as real error in span.
otext.LogError(span, res.Failed())
return
}
}()
return next(ctx, request)
}
}
}
// TraceServer returns a Middleware that wraps the `next` Endpoint in an
// OpenTracing Span called `operationName` with server span.kind tag..
func TraceServer(tracer opentracing.Tracer, operationName string, opts ...EndpointOption) endpoint.Middleware {
opts = append(opts, WithTags(map[string]interface{}{
otext.SpanKindRPCServer.Key: otext.SpanKindRPCServer.Value,
}))
return TraceEndpoint(tracer, operationName, opts...)
}
// TraceClient returns a Middleware that wraps the `next` Endpoint in an
// OpenTracing Span called `operationName` with client span.kind tag.
func TraceClient(tracer opentracing.Tracer, operationName string, opts ...EndpointOption) endpoint.Middleware {
opts = append(opts, WithTags(map[string]interface{}{
otext.SpanKindRPCClient.Key: otext.SpanKindRPCClient.Value,
}))
return TraceEndpoint(tracer, operationName, opts...)
}
func applyTags(span opentracing.Span, tags opentracing.Tags) {
for key, value := range tags {
span.SetTag(key, value)
}
}
+74
View File
@@ -0,0 +1,74 @@
package opentracing
import (
"context"
"github.com/opentracing/opentracing-go"
)
// EndpointOptions holds the options for tracing an endpoint
type EndpointOptions struct {
// IgnoreBusinessError if set to true will not treat a business error
// identified through the endpoint.Failer interface as a span error.
IgnoreBusinessError bool
// GetOperationName is an optional function that can set the span operation name based on the existing one
// for the endpoint and information in the context.
//
// If the function is nil, or the returned name is empty, the existing name for the endpoint is used.
GetOperationName func(ctx context.Context, name string) string
// Tags holds the default tags which will be set on span
// creation by our Endpoint middleware.
Tags opentracing.Tags
// GetTags is an optional function that can extract tags
// from the context and add them to the span.
GetTags func(ctx context.Context) opentracing.Tags
}
// EndpointOption allows for functional options to endpoint tracing middleware.
type EndpointOption func(*EndpointOptions)
// WithOptions sets all configuration options at once by use of the EndpointOptions struct.
func WithOptions(options EndpointOptions) EndpointOption {
return func(o *EndpointOptions) {
*o = options
}
}
// WithIgnoreBusinessError if set to true will not treat a business error
// identified through the endpoint.Failer interface as a span error.
func WithIgnoreBusinessError(ignoreBusinessError bool) EndpointOption {
return func(o *EndpointOptions) {
o.IgnoreBusinessError = ignoreBusinessError
}
}
// WithOperationNameFunc allows to set function that can set the span operation name based on the existing one
// for the endpoint and information in the context.
func WithOperationNameFunc(getOperationName func(ctx context.Context, name string) string) EndpointOption {
return func(o *EndpointOptions) {
o.GetOperationName = getOperationName
}
}
// WithTags adds default tags for the spans created by the Endpoint tracer.
func WithTags(tags opentracing.Tags) EndpointOption {
return func(o *EndpointOptions) {
if o.Tags == nil {
o.Tags = make(opentracing.Tags)
}
for key, value := range tags {
o.Tags[key] = value
}
}
}
// WithTagsFunc set the func to extracts additional tags from the context.
func WithTagsFunc(getTags func(ctx context.Context) opentracing.Tags) EndpointOption {
return func(o *EndpointOptions) {
o.GetTags = getTags
}
}
+70
View File
@@ -0,0 +1,70 @@
package opentracing
import (
"context"
"encoding/base64"
"strings"
"github.com/opentracing/opentracing-go"
"github.com/opentracing/opentracing-go/ext"
"google.golang.org/grpc/metadata"
"github.com/go-kit/log"
)
// ContextToGRPC returns a grpc RequestFunc that injects an OpenTracing Span
// found in `ctx` into the grpc Metadata. If no such Span can be found, the
// RequestFunc is a noop.
func ContextToGRPC(tracer opentracing.Tracer, logger log.Logger) func(ctx context.Context, md *metadata.MD) context.Context {
return func(ctx context.Context, md *metadata.MD) context.Context {
if span := opentracing.SpanFromContext(ctx); span != nil {
// There's nothing we can do with an error here.
if err := tracer.Inject(span.Context(), opentracing.TextMap, metadataReaderWriter{md}); err != nil {
logger.Log("err", err)
}
}
return ctx
}
}
// GRPCToContext returns a grpc RequestFunc that tries to join with an
// OpenTracing trace found in `req` and starts a new Span called
// `operationName` accordingly. If no trace could be found in `req`, the Span
// will be a trace root. The Span is incorporated in the returned Context and
// can be retrieved with opentracing.SpanFromContext(ctx).
func GRPCToContext(tracer opentracing.Tracer, operationName string, logger log.Logger) func(ctx context.Context, md metadata.MD) context.Context {
return func(ctx context.Context, md metadata.MD) context.Context {
var span opentracing.Span
wireContext, err := tracer.Extract(opentracing.TextMap, metadataReaderWriter{&md})
if err != nil && err != opentracing.ErrSpanContextNotFound {
logger.Log("err", err)
}
span = tracer.StartSpan(operationName, ext.RPCServerOption(wireContext))
return opentracing.ContextWithSpan(ctx, span)
}
}
// A type that conforms to opentracing.TextMapReader and
// opentracing.TextMapWriter.
type metadataReaderWriter struct {
*metadata.MD
}
func (w metadataReaderWriter) Set(key, val string) {
key = strings.ToLower(key)
if strings.HasSuffix(key, "-bin") {
val = base64.StdEncoding.EncodeToString([]byte(val))
}
(*w.MD)[key] = append((*w.MD)[key], val)
}
func (w metadataReaderWriter) ForeachKey(handler func(key, val string) error) error {
for k, vals := range *w.MD {
for _, v := range vals {
if err := handler(k, v); err != nil {
return err
}
}
}
return nil
}
+71
View File
@@ -0,0 +1,71 @@
package opentracing
import (
"context"
"net"
"net/http"
"strconv"
opentracing "github.com/opentracing/opentracing-go"
"github.com/opentracing/opentracing-go/ext"
kithttp "github.com/go-kit/kit/transport/http"
"github.com/go-kit/log"
)
// ContextToHTTP returns an http RequestFunc that injects an OpenTracing Span
// found in `ctx` into the http headers. If no such Span can be found, the
// RequestFunc is a noop.
func ContextToHTTP(tracer opentracing.Tracer, logger log.Logger) kithttp.RequestFunc {
return func(ctx context.Context, req *http.Request) context.Context {
// Try to find a Span in the Context.
if span := opentracing.SpanFromContext(ctx); span != nil {
// Add standard OpenTracing tags.
ext.HTTPMethod.Set(span, req.Method)
ext.HTTPUrl.Set(span, req.URL.String())
host, portString, err := net.SplitHostPort(req.URL.Host)
if err == nil {
ext.PeerHostname.Set(span, host)
if port, err := strconv.Atoi(portString); err == nil {
ext.PeerPort.Set(span, uint16(port))
}
} else {
ext.PeerHostname.Set(span, req.URL.Host)
}
// There's nothing we can do with any errors here.
if err = tracer.Inject(
span.Context(),
opentracing.HTTPHeaders,
opentracing.HTTPHeadersCarrier(req.Header),
); err != nil {
logger.Log("err", err)
}
}
return ctx
}
}
// HTTPToContext returns an http RequestFunc that tries to join with an
// OpenTracing trace found in `req` and starts a new Span called
// `operationName` accordingly. If no trace could be found in `req`, the Span
// will be a trace root. The Span is incorporated in the returned Context and
// can be retrieved with opentracing.SpanFromContext(ctx).
func HTTPToContext(tracer opentracing.Tracer, operationName string, logger log.Logger) kithttp.RequestFunc {
return func(ctx context.Context, req *http.Request) context.Context {
// Try to join to a trace propagated in `req`.
var span opentracing.Span
wireContext, err := tracer.Extract(
opentracing.HTTPHeaders,
opentracing.HTTPHeadersCarrier(req.Header),
)
if err != nil && err != opentracing.ErrSpanContextNotFound {
logger.Log("err", err)
}
span = tracer.StartSpan(operationName, ext.RPCServerOption(wireContext))
ext.HTTPMethod.Set(span, req.Method)
ext.HTTPUrl.Set(span, req.URL.String())
return opentracing.ContextWithSpan(ctx, span)
}
}
+2
View File
@@ -0,0 +1,2 @@
// Package transport contains helpers applicable to all supported transports.
package transport
+39
View File
@@ -0,0 +1,39 @@
package transport
import (
"context"
"github.com/go-kit/log"
)
// ErrorHandler receives a transport error to be processed for diagnostic purposes.
// Usually this means logging the error.
type ErrorHandler interface {
Handle(ctx context.Context, err error)
}
// LogErrorHandler is a transport error handler implementation which logs an error.
type LogErrorHandler struct {
logger log.Logger
}
func NewLogErrorHandler(logger log.Logger) *LogErrorHandler {
return &LogErrorHandler{
logger: logger,
}
}
func (h *LogErrorHandler) Handle(ctx context.Context, err error) {
h.logger.Log("err", err)
}
// The ErrorHandlerFunc type is an adapter to allow the use of
// ordinary function as ErrorHandler. If f is a function
// with the appropriate signature, ErrorHandlerFunc(f) is a
// ErrorHandler that calls f.
type ErrorHandlerFunc func(ctx context.Context, err error)
// Handle calls f(ctx, err).
func (f ErrorHandlerFunc) Handle(ctx context.Context, err error) {
f(ctx, err)
}
+219
View File
@@ -0,0 +1,219 @@
package http
import (
"bytes"
"context"
"encoding/json"
"encoding/xml"
"io"
"io/ioutil"
"net/http"
"net/url"
"github.com/go-kit/kit/endpoint"
)
// HTTPClient is an interface that models *http.Client.
type HTTPClient interface {
Do(req *http.Request) (*http.Response, error)
}
// Client wraps a URL and provides a method that implements endpoint.Endpoint.
type Client struct {
client HTTPClient
req CreateRequestFunc
dec DecodeResponseFunc
before []RequestFunc
after []ClientResponseFunc
finalizer []ClientFinalizerFunc
bufferedStream bool
}
// NewClient constructs a usable Client for a single remote method.
func NewClient(method string, tgt *url.URL, enc EncodeRequestFunc, dec DecodeResponseFunc, options ...ClientOption) *Client {
return NewExplicitClient(makeCreateRequestFunc(method, tgt, enc), dec, options...)
}
// NewExplicitClient is like NewClient but uses a CreateRequestFunc instead of a
// method, target URL, and EncodeRequestFunc, which allows for more control over
// the outgoing HTTP request.
func NewExplicitClient(req CreateRequestFunc, dec DecodeResponseFunc, options ...ClientOption) *Client {
c := &Client{
client: http.DefaultClient,
req: req,
dec: dec,
}
for _, option := range options {
option(c)
}
return c
}
// ClientOption sets an optional parameter for clients.
type ClientOption func(*Client)
// SetClient sets the underlying HTTP client used for requests.
// By default, http.DefaultClient is used.
func SetClient(client HTTPClient) ClientOption {
return func(c *Client) { c.client = client }
}
// ClientBefore adds one or more RequestFuncs to be applied to the outgoing HTTP
// request before it's invoked.
func ClientBefore(before ...RequestFunc) ClientOption {
return func(c *Client) { c.before = append(c.before, before...) }
}
// ClientAfter adds one or more ClientResponseFuncs, which are applied to the
// incoming HTTP response prior to it being decoded. This is useful for
// obtaining anything off of the response and adding it into the context prior
// to decoding.
func ClientAfter(after ...ClientResponseFunc) ClientOption {
return func(c *Client) { c.after = append(c.after, after...) }
}
// ClientFinalizer adds one or more ClientFinalizerFuncs to be executed at the
// end of every HTTP request. Finalizers are executed in the order in which they
// were added. By default, no finalizer is registered.
func ClientFinalizer(f ...ClientFinalizerFunc) ClientOption {
return func(s *Client) { s.finalizer = append(s.finalizer, f...) }
}
// BufferedStream sets whether the HTTP response body is left open, allowing it
// to be read from later. Useful for transporting a file as a buffered stream.
// That body has to be drained and closed to properly end the request.
func BufferedStream(buffered bool) ClientOption {
return func(c *Client) { c.bufferedStream = buffered }
}
// Endpoint returns a usable Go kit endpoint that calls the remote HTTP endpoint.
func (c Client) Endpoint() endpoint.Endpoint {
return func(ctx context.Context, request interface{}) (interface{}, error) {
ctx, cancel := context.WithCancel(ctx)
var (
resp *http.Response
err error
)
if c.finalizer != nil {
defer func() {
if resp != nil {
ctx = context.WithValue(ctx, ContextKeyResponseHeaders, resp.Header)
ctx = context.WithValue(ctx, ContextKeyResponseSize, resp.ContentLength)
}
for _, f := range c.finalizer {
f(ctx, err)
}
}()
}
req, err := c.req(ctx, request)
if err != nil {
cancel()
return nil, err
}
for _, f := range c.before {
ctx = f(ctx, req)
}
resp, err = c.client.Do(req.WithContext(ctx))
if err != nil {
cancel()
return nil, err
}
// If the caller asked for a buffered stream, we don't cancel the
// context when the endpoint returns. Instead, we should call the
// cancel func when closing the response body.
if c.bufferedStream {
resp.Body = bodyWithCancel{ReadCloser: resp.Body, cancel: cancel}
} else {
defer resp.Body.Close()
defer cancel()
}
for _, f := range c.after {
ctx = f(ctx, resp)
}
response, err := c.dec(ctx, resp)
if err != nil {
return nil, err
}
return response, nil
}
}
// bodyWithCancel is a wrapper for an io.ReadCloser with also a
// cancel function which is called when the Close is used
type bodyWithCancel struct {
io.ReadCloser
cancel context.CancelFunc
}
func (bwc bodyWithCancel) Close() error {
bwc.ReadCloser.Close()
bwc.cancel()
return nil
}
// ClientFinalizerFunc can be used to perform work at the end of a client HTTP
// request, after the response is returned. The principal
// intended use is for error logging. Additional response parameters are
// provided in the context under keys with the ContextKeyResponse prefix.
// Note: err may be nil. There maybe also no additional response parameters
// depending on when an error occurs.
type ClientFinalizerFunc func(ctx context.Context, err error)
// EncodeJSONRequest is an EncodeRequestFunc that serializes the request as a
// JSON object to the Request body. Many JSON-over-HTTP services can use it as
// a sensible default. If the request implements Headerer, the provided headers
// will be applied to the request.
func EncodeJSONRequest(c context.Context, r *http.Request, request interface{}) error {
r.Header.Set("Content-Type", "application/json; charset=utf-8")
if headerer, ok := request.(Headerer); ok {
for k := range headerer.Headers() {
r.Header.Set(k, headerer.Headers().Get(k))
}
}
var b bytes.Buffer
r.Body = ioutil.NopCloser(&b)
return json.NewEncoder(&b).Encode(request)
}
// EncodeXMLRequest is an EncodeRequestFunc that serializes the request as a
// XML object to the Request body. If the request implements Headerer,
// the provided headers will be applied to the request.
func EncodeXMLRequest(c context.Context, r *http.Request, request interface{}) error {
r.Header.Set("Content-Type", "text/xml; charset=utf-8")
if headerer, ok := request.(Headerer); ok {
for k := range headerer.Headers() {
r.Header.Set(k, headerer.Headers().Get(k))
}
}
var b bytes.Buffer
r.Body = ioutil.NopCloser(&b)
return xml.NewEncoder(&b).Encode(request)
}
//
//
//
func makeCreateRequestFunc(method string, target *url.URL, enc EncodeRequestFunc) CreateRequestFunc {
return func(ctx context.Context, request interface{}) (*http.Request, error) {
req, err := http.NewRequest(method, target.String(), nil)
if err != nil {
return nil, err
}
if err = enc(ctx, req, request); err != nil {
return nil, err
}
return req, nil
}
}
+2
View File
@@ -0,0 +1,2 @@
// Package http provides a general purpose HTTP binding for endpoints.
package http
+36
View File
@@ -0,0 +1,36 @@
package http
import (
"context"
"net/http"
)
// DecodeRequestFunc extracts a user-domain request object from an HTTP
// request object. It's designed to be used in HTTP servers, for server-side
// endpoints. One straightforward DecodeRequestFunc could be something that
// JSON decodes from the request body to the concrete request type.
type DecodeRequestFunc func(context.Context, *http.Request) (request interface{}, err error)
// EncodeRequestFunc encodes the passed request object into the HTTP request
// object. It's designed to be used in HTTP clients, for client-side
// endpoints. One straightforward EncodeRequestFunc could be something that JSON
// encodes the object directly to the request body.
type EncodeRequestFunc func(context.Context, *http.Request, interface{}) error
// CreateRequestFunc creates an outgoing HTTP request based on the passed
// request object. It's designed to be used in HTTP clients, for client-side
// endpoints. It's a more powerful version of EncodeRequestFunc, and can be used
// if more fine-grained control of the HTTP request is required.
type CreateRequestFunc func(context.Context, interface{}) (*http.Request, error)
// EncodeResponseFunc encodes the passed response object to the HTTP response
// writer. It's designed to be used in HTTP servers, for server-side
// endpoints. One straightforward EncodeResponseFunc could be something that
// JSON encodes the object directly to the response body.
type EncodeResponseFunc func(context.Context, http.ResponseWriter, interface{}) error
// DecodeResponseFunc extracts a user-domain response object from an HTTP
// response object. It's designed to be used in HTTP clients, for client-side
// endpoints. One straightforward DecodeResponseFunc could be something that
// JSON decodes from the response body to the concrete response type.
type DecodeResponseFunc func(context.Context, *http.Response) (response interface{}, err error)
+133
View File
@@ -0,0 +1,133 @@
package http
import (
"context"
"net/http"
)
// RequestFunc may take information from an HTTP request and put it into a
// request context. In Servers, RequestFuncs are executed prior to invoking the
// endpoint. In Clients, RequestFuncs are executed after creating the request
// but prior to invoking the HTTP client.
type RequestFunc func(context.Context, *http.Request) context.Context
// ServerResponseFunc may take information from a request context and use it to
// manipulate a ResponseWriter. ServerResponseFuncs are only executed in
// servers, after invoking the endpoint but prior to writing a response.
type ServerResponseFunc func(context.Context, http.ResponseWriter) context.Context
// ClientResponseFunc may take information from an HTTP request and make the
// response available for consumption. ClientResponseFuncs are only executed in
// clients, after a request has been made, but prior to it being decoded.
type ClientResponseFunc func(context.Context, *http.Response) context.Context
// SetContentType returns a ServerResponseFunc that sets the Content-Type header
// to the provided value.
func SetContentType(contentType string) ServerResponseFunc {
return SetResponseHeader("Content-Type", contentType)
}
// SetResponseHeader returns a ServerResponseFunc that sets the given header.
func SetResponseHeader(key, val string) ServerResponseFunc {
return func(ctx context.Context, w http.ResponseWriter) context.Context {
w.Header().Set(key, val)
return ctx
}
}
// SetRequestHeader returns a RequestFunc that sets the given header.
func SetRequestHeader(key, val string) RequestFunc {
return func(ctx context.Context, r *http.Request) context.Context {
r.Header.Set(key, val)
return ctx
}
}
// PopulateRequestContext is a RequestFunc that populates several values into
// the context from the HTTP request. Those values may be extracted using the
// corresponding ContextKey type in this package.
func PopulateRequestContext(ctx context.Context, r *http.Request) context.Context {
for k, v := range map[contextKey]string{
ContextKeyRequestMethod: r.Method,
ContextKeyRequestURI: r.RequestURI,
ContextKeyRequestPath: r.URL.Path,
ContextKeyRequestProto: r.Proto,
ContextKeyRequestHost: r.Host,
ContextKeyRequestRemoteAddr: r.RemoteAddr,
ContextKeyRequestXForwardedFor: r.Header.Get("X-Forwarded-For"),
ContextKeyRequestXForwardedProto: r.Header.Get("X-Forwarded-Proto"),
ContextKeyRequestAuthorization: r.Header.Get("Authorization"),
ContextKeyRequestReferer: r.Header.Get("Referer"),
ContextKeyRequestUserAgent: r.Header.Get("User-Agent"),
ContextKeyRequestXRequestID: r.Header.Get("X-Request-Id"),
ContextKeyRequestAccept: r.Header.Get("Accept"),
} {
ctx = context.WithValue(ctx, k, v)
}
return ctx
}
type contextKey int
const (
// ContextKeyRequestMethod is populated in the context by
// PopulateRequestContext. Its value is r.Method.
ContextKeyRequestMethod contextKey = iota
// ContextKeyRequestURI is populated in the context by
// PopulateRequestContext. Its value is r.RequestURI.
ContextKeyRequestURI
// ContextKeyRequestPath is populated in the context by
// PopulateRequestContext. Its value is r.URL.Path.
ContextKeyRequestPath
// ContextKeyRequestProto is populated in the context by
// PopulateRequestContext. Its value is r.Proto.
ContextKeyRequestProto
// ContextKeyRequestHost is populated in the context by
// PopulateRequestContext. Its value is r.Host.
ContextKeyRequestHost
// ContextKeyRequestRemoteAddr is populated in the context by
// PopulateRequestContext. Its value is r.RemoteAddr.
ContextKeyRequestRemoteAddr
// ContextKeyRequestXForwardedFor is populated in the context by
// PopulateRequestContext. Its value is r.Header.Get("X-Forwarded-For").
ContextKeyRequestXForwardedFor
// ContextKeyRequestXForwardedProto is populated in the context by
// PopulateRequestContext. Its value is r.Header.Get("X-Forwarded-Proto").
ContextKeyRequestXForwardedProto
// ContextKeyRequestAuthorization is populated in the context by
// PopulateRequestContext. Its value is r.Header.Get("Authorization").
ContextKeyRequestAuthorization
// ContextKeyRequestReferer is populated in the context by
// PopulateRequestContext. Its value is r.Header.Get("Referer").
ContextKeyRequestReferer
// ContextKeyRequestUserAgent is populated in the context by
// PopulateRequestContext. Its value is r.Header.Get("User-Agent").
ContextKeyRequestUserAgent
// ContextKeyRequestXRequestID is populated in the context by
// PopulateRequestContext. Its value is r.Header.Get("X-Request-Id").
ContextKeyRequestXRequestID
// ContextKeyRequestAccept is populated in the context by
// PopulateRequestContext. Its value is r.Header.Get("Accept").
ContextKeyRequestAccept
// ContextKeyResponseHeaders is populated in the context whenever a
// ServerFinalizerFunc is specified. Its value is of type http.Header, and
// is captured only once the entire response has been written.
ContextKeyResponseHeaders
// ContextKeyResponseSize is populated in the context whenever a
// ServerFinalizerFunc is specified. Its value is of type int64.
ContextKeyResponseSize
)
+244
View File
@@ -0,0 +1,244 @@
package http
import (
"context"
"encoding/json"
"net/http"
"github.com/go-kit/kit/endpoint"
"github.com/go-kit/kit/transport"
"github.com/go-kit/log"
)
// Server wraps an endpoint and implements http.Handler.
type Server struct {
e endpoint.Endpoint
dec DecodeRequestFunc
enc EncodeResponseFunc
before []RequestFunc
after []ServerResponseFunc
errorEncoder ErrorEncoder
finalizer []ServerFinalizerFunc
errorHandler transport.ErrorHandler
}
// NewServer constructs a new server, which implements http.Handler and wraps
// the provided endpoint.
func NewServer(
e endpoint.Endpoint,
dec DecodeRequestFunc,
enc EncodeResponseFunc,
options ...ServerOption,
) *Server {
s := &Server{
e: e,
dec: dec,
enc: enc,
errorEncoder: DefaultErrorEncoder,
errorHandler: transport.NewLogErrorHandler(log.NewNopLogger()),
}
for _, option := range options {
option(s)
}
return s
}
// ServerOption sets an optional parameter for servers.
type ServerOption func(*Server)
// ServerBefore functions are executed on the HTTP request object before the
// request is decoded.
func ServerBefore(before ...RequestFunc) ServerOption {
return func(s *Server) { s.before = append(s.before, before...) }
}
// ServerAfter functions are executed on the HTTP response writer after the
// endpoint is invoked, but before anything is written to the client.
func ServerAfter(after ...ServerResponseFunc) ServerOption {
return func(s *Server) { s.after = append(s.after, after...) }
}
// ServerErrorEncoder is used to encode errors to the http.ResponseWriter
// whenever they're encountered in the processing of a request. Clients can
// use this to provide custom error formatting and response codes. By default,
// errors will be written with the DefaultErrorEncoder.
func ServerErrorEncoder(ee ErrorEncoder) ServerOption {
return func(s *Server) { s.errorEncoder = ee }
}
// ServerErrorLogger is used to log non-terminal errors. By default, no errors
// are logged. This is intended as a diagnostic measure. Finer-grained control
// of error handling, including logging in more detail, should be performed in a
// custom ServerErrorEncoder or ServerFinalizer, both of which have access to
// the context.
// Deprecated: Use ServerErrorHandler instead.
func ServerErrorLogger(logger log.Logger) ServerOption {
return func(s *Server) { s.errorHandler = transport.NewLogErrorHandler(logger) }
}
// ServerErrorHandler is used to handle non-terminal errors. By default, non-terminal errors
// are ignored. This is intended as a diagnostic measure. Finer-grained control
// of error handling, including logging in more detail, should be performed in a
// custom ServerErrorEncoder or ServerFinalizer, both of which have access to
// the context.
func ServerErrorHandler(errorHandler transport.ErrorHandler) ServerOption {
return func(s *Server) { s.errorHandler = errorHandler }
}
// ServerFinalizer is executed at the end of every HTTP request.
// By default, no finalizer is registered.
func ServerFinalizer(f ...ServerFinalizerFunc) ServerOption {
return func(s *Server) { s.finalizer = append(s.finalizer, f...) }
}
// ServeHTTP implements http.Handler.
func (s Server) ServeHTTP(w http.ResponseWriter, r *http.Request) {
ctx := r.Context()
if len(s.finalizer) > 0 {
iw := &interceptingWriter{w, http.StatusOK, 0}
defer func() {
ctx = context.WithValue(ctx, ContextKeyResponseHeaders, iw.Header())
ctx = context.WithValue(ctx, ContextKeyResponseSize, iw.written)
for _, f := range s.finalizer {
f(ctx, iw.code, r)
}
}()
w = iw
}
for _, f := range s.before {
ctx = f(ctx, r)
}
request, err := s.dec(ctx, r)
if err != nil {
s.errorHandler.Handle(ctx, err)
s.errorEncoder(ctx, err, w)
return
}
response, err := s.e(ctx, request)
if err != nil {
s.errorHandler.Handle(ctx, err)
s.errorEncoder(ctx, err, w)
return
}
for _, f := range s.after {
ctx = f(ctx, w)
}
if err := s.enc(ctx, w, response); err != nil {
s.errorHandler.Handle(ctx, err)
s.errorEncoder(ctx, err, w)
return
}
}
// ErrorEncoder is responsible for encoding an error to the ResponseWriter.
// Users are encouraged to use custom ErrorEncoders to encode HTTP errors to
// their clients, and will likely want to pass and check for their own error
// types. See the example shipping/handling service.
type ErrorEncoder func(ctx context.Context, err error, w http.ResponseWriter)
// ServerFinalizerFunc can be used to perform work at the end of an HTTP
// request, after the response has been written to the client. The principal
// intended use is for request logging. In addition to the response code
// provided in the function signature, additional response parameters are
// provided in the context under keys with the ContextKeyResponse prefix.
type ServerFinalizerFunc func(ctx context.Context, code int, r *http.Request)
// NopRequestDecoder is a DecodeRequestFunc that can be used for requests that do not
// need to be decoded, and simply returns nil, nil.
func NopRequestDecoder(ctx context.Context, r *http.Request) (interface{}, error) {
return nil, nil
}
// EncodeJSONResponse is a EncodeResponseFunc that serializes the response as a
// JSON object to the ResponseWriter. Many JSON-over-HTTP services can use it as
// a sensible default. If the response implements Headerer, the provided headers
// will be applied to the response. If the response implements StatusCoder, the
// provided StatusCode will be used instead of 200.
func EncodeJSONResponse(_ context.Context, w http.ResponseWriter, response interface{}) error {
w.Header().Set("Content-Type", "application/json; charset=utf-8")
if headerer, ok := response.(Headerer); ok {
for k, values := range headerer.Headers() {
for _, v := range values {
w.Header().Add(k, v)
}
}
}
code := http.StatusOK
if sc, ok := response.(StatusCoder); ok {
code = sc.StatusCode()
}
w.WriteHeader(code)
if code == http.StatusNoContent {
return nil
}
return json.NewEncoder(w).Encode(response)
}
// DefaultErrorEncoder writes the error to the ResponseWriter, by default a
// content type of text/plain, a body of the plain text of the error, and a
// status code of 500. If the error implements Headerer, the provided headers
// will be applied to the response. If the error implements json.Marshaler, and
// the marshaling succeeds, a content type of application/json and the JSON
// encoded form of the error will be used. If the error implements StatusCoder,
// the provided StatusCode will be used instead of 500.
func DefaultErrorEncoder(_ context.Context, err error, w http.ResponseWriter) {
contentType, body := "text/plain; charset=utf-8", []byte(err.Error())
if marshaler, ok := err.(json.Marshaler); ok {
if jsonBody, marshalErr := marshaler.MarshalJSON(); marshalErr == nil {
contentType, body = "application/json; charset=utf-8", jsonBody
}
}
w.Header().Set("Content-Type", contentType)
if headerer, ok := err.(Headerer); ok {
for k, values := range headerer.Headers() {
for _, v := range values {
w.Header().Add(k, v)
}
}
}
code := http.StatusInternalServerError
if sc, ok := err.(StatusCoder); ok {
code = sc.StatusCode()
}
w.WriteHeader(code)
w.Write(body)
}
// StatusCoder is checked by DefaultErrorEncoder. If an error value implements
// StatusCoder, the StatusCode will be used when encoding the error. By default,
// StatusInternalServerError (500) is used.
type StatusCoder interface {
StatusCode() int
}
// Headerer is checked by DefaultErrorEncoder. If an error value implements
// Headerer, the provided headers will be applied to the response writer, after
// the Content-Type is set.
type Headerer interface {
Headers() http.Header
}
type interceptingWriter struct {
http.ResponseWriter
code int
written int64
}
// WriteHeader may not be explicitly called, so care must be taken to
// initialize w.code to its default value of http.StatusOK.
func (w *interceptingWriter) WriteHeader(code int) {
w.code = code
w.ResponseWriter.WriteHeader(code)
}
func (w *interceptingWriter) Write(p []byte) (int, error) {
n, err := w.ResponseWriter.Write(p)
w.written += int64(n)
return n, err
}
+15
View File
@@ -0,0 +1,15 @@
# Binaries for programs and plugins
*.exe
*.exe~
*.dll
*.so
*.dylib
# Test binary, built with `go test -c`
*.test
# Output of the go coverage tool, specifically when used with LiteIDE
*.out
# Dependency directories (remove the comment below to include it)
# vendor/
+21
View File
@@ -0,0 +1,21 @@
MIT License
Copyright (c) 2021 Go kit
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
+151
View File
@@ -0,0 +1,151 @@
# package log
`package log` provides a minimal interface for structured logging in services.
It may be wrapped to encode conventions, enforce type-safety, provide leveled
logging, and so on. It can be used for both typical application log events,
and log-structured data streams.
## Structured logging
Structured logging is, basically, conceding to the reality that logs are
_data_, and warrant some level of schematic rigor. Using a stricter,
key/value-oriented message format for our logs, containing contextual and
semantic information, makes it much easier to get insight into the
operational activity of the systems we build. Consequently, `package log` is
of the strong belief that "[the benefits of structured logging outweigh the
minimal effort involved](https://www.thoughtworks.com/radar/techniques/structured-logging)".
Migrating from unstructured to structured logging is probably a lot easier
than you'd expect.
```go
// Unstructured
log.Printf("HTTP server listening on %s", addr)
// Structured
logger.Log("transport", "HTTP", "addr", addr, "msg", "listening")
```
## Usage
### Typical application logging
```go
w := log.NewSyncWriter(os.Stderr)
logger := log.NewLogfmtLogger(w)
logger.Log("question", "what is the meaning of life?", "answer", 42)
// Output:
// question="what is the meaning of life?" answer=42
```
### Contextual Loggers
```go
func main() {
var logger log.Logger
logger = log.NewLogfmtLogger(log.NewSyncWriter(os.Stderr))
logger = log.With(logger, "instance_id", 123)
logger.Log("msg", "starting")
NewWorker(log.With(logger, "component", "worker")).Run()
NewSlacker(log.With(logger, "component", "slacker")).Run()
}
// Output:
// instance_id=123 msg=starting
// instance_id=123 component=worker msg=running
// instance_id=123 component=slacker msg=running
```
### Interact with stdlib logger
Redirect stdlib logger to Go kit logger.
```go
import (
"os"
stdlog "log"
kitlog "github.com/go-kit/log"
)
func main() {
logger := kitlog.NewJSONLogger(kitlog.NewSyncWriter(os.Stdout))
stdlog.SetOutput(kitlog.NewStdlibAdapter(logger))
stdlog.Print("I sure like pie")
}
// Output:
// {"msg":"I sure like pie","ts":"2016/01/01 12:34:56"}
```
Or, if, for legacy reasons, you need to pipe all of your logging through the
stdlib log package, you can redirect Go kit logger to the stdlib logger.
```go
logger := kitlog.NewLogfmtLogger(kitlog.StdlibWriter{})
logger.Log("legacy", true, "msg", "at least it's something")
// Output:
// 2016/01/01 12:34:56 legacy=true msg="at least it's something"
```
### Timestamps and callers
```go
var logger log.Logger
logger = log.NewLogfmtLogger(log.NewSyncWriter(os.Stderr))
logger = log.With(logger, "ts", log.DefaultTimestampUTC, "caller", log.DefaultCaller)
logger.Log("msg", "hello")
// Output:
// ts=2016-01-01T12:34:56Z caller=main.go:15 msg=hello
```
## Levels
Log levels are supported via the [level package](https://godoc.org/github.com/go-kit/log/level).
## Supported output formats
- [Logfmt](https://brandur.org/logfmt) ([see also](https://blog.codeship.com/logfmt-a-log-format-thats-easy-to-read-and-write))
- JSON
## Enhancements
`package log` is centered on the one-method Logger interface.
```go
type Logger interface {
Log(keyvals ...interface{}) error
}
```
This interface, and its supporting code like is the product of much iteration
and evaluation. For more details on the evolution of the Logger interface,
see [The Hunt for a Logger Interface](http://go-talks.appspot.com/github.com/ChrisHines/talks/structured-logging/structured-logging.slide#1),
a talk by [Chris Hines](https://github.com/ChrisHines).
Also, please see
[#63](https://github.com/go-kit/kit/issues/63),
[#76](https://github.com/go-kit/kit/pull/76),
[#131](https://github.com/go-kit/kit/issues/131),
[#157](https://github.com/go-kit/kit/pull/157),
[#164](https://github.com/go-kit/kit/issues/164), and
[#252](https://github.com/go-kit/kit/pull/252)
to review historical conversations about package log and the Logger interface.
Value-add packages and suggestions,
like improvements to [the leveled logger](https://godoc.org/github.com/go-kit/log/level),
are of course welcome. Good proposals should
- Be composable with [contextual loggers](https://godoc.org/github.com/go-kit/log#With),
- Not break the behavior of [log.Caller](https://godoc.org/github.com/go-kit/log#Caller) in any wrapped contextual loggers, and
- Be friendly to packages that accept only an unadorned log.Logger.
## Benchmarks & comparisons
There are a few Go logging benchmarks and comparisons that include Go kit's package log.
- [imkira/go-loggers-bench](https://github.com/imkira/go-loggers-bench) includes kit/log
- [uber-common/zap](https://github.com/uber-common/zap), a zero-alloc logging library, includes a comparison with kit/log
+116
View File
@@ -0,0 +1,116 @@
// Package log provides a structured logger.
//
// Structured logging produces logs easily consumed later by humans or
// machines. Humans might be interested in debugging errors, or tracing
// specific requests. Machines might be interested in counting interesting
// events, or aggregating information for off-line processing. In both cases,
// it is important that the log messages are structured and actionable.
// Package log is designed to encourage both of these best practices.
//
// Basic Usage
//
// The fundamental interface is Logger. Loggers create log events from
// key/value data. The Logger interface has a single method, Log, which
// accepts a sequence of alternating key/value pairs, which this package names
// keyvals.
//
// type Logger interface {
// Log(keyvals ...interface{}) error
// }
//
// Here is an example of a function using a Logger to create log events.
//
// func RunTask(task Task, logger log.Logger) string {
// logger.Log("taskID", task.ID, "event", "starting task")
// ...
// logger.Log("taskID", task.ID, "event", "task complete")
// }
//
// The keys in the above example are "taskID" and "event". The values are
// task.ID, "starting task", and "task complete". Every key is followed
// immediately by its value.
//
// Keys are usually plain strings. Values may be any type that has a sensible
// encoding in the chosen log format. With structured logging it is a good
// idea to log simple values without formatting them. This practice allows
// the chosen logger to encode values in the most appropriate way.
//
// Contextual Loggers
//
// A contextual logger stores keyvals that it includes in all log events.
// Building appropriate contextual loggers reduces repetition and aids
// consistency in the resulting log output. With, WithPrefix, and WithSuffix
// add context to a logger. We can use With to improve the RunTask example.
//
// func RunTask(task Task, logger log.Logger) string {
// logger = log.With(logger, "taskID", task.ID)
// logger.Log("event", "starting task")
// ...
// taskHelper(task.Cmd, logger)
// ...
// logger.Log("event", "task complete")
// }
//
// The improved version emits the same log events as the original for the
// first and last calls to Log. Passing the contextual logger to taskHelper
// enables each log event created by taskHelper to include the task.ID even
// though taskHelper does not have access to that value. Using contextual
// loggers this way simplifies producing log output that enables tracing the
// life cycle of individual tasks. (See the Contextual example for the full
// code of the above snippet.)
//
// Dynamic Contextual Values
//
// A Valuer function stored in a contextual logger generates a new value each
// time an event is logged. The Valuer example demonstrates how this feature
// works.
//
// Valuers provide the basis for consistently logging timestamps and source
// code location. The log package defines several valuers for that purpose.
// See Timestamp, DefaultTimestamp, DefaultTimestampUTC, Caller, and
// DefaultCaller. A common logger initialization sequence that ensures all log
// entries contain a timestamp and source location looks like this:
//
// logger := log.NewLogfmtLogger(log.NewSyncWriter(os.Stdout))
// logger = log.With(logger, "ts", log.DefaultTimestampUTC, "caller", log.DefaultCaller)
//
// Concurrent Safety
//
// Applications with multiple goroutines want each log event written to the
// same logger to remain separate from other log events. Package log provides
// two simple solutions for concurrent safe logging.
//
// NewSyncWriter wraps an io.Writer and serializes each call to its Write
// method. Using a SyncWriter has the benefit that the smallest practical
// portion of the logging logic is performed within a mutex, but it requires
// the formatting Logger to make only one call to Write per log event.
//
// NewSyncLogger wraps any Logger and serializes each call to its Log method.
// Using a SyncLogger has the benefit that it guarantees each log event is
// handled atomically within the wrapped logger, but it typically serializes
// both the formatting and output logic. Use a SyncLogger if the formatting
// logger may perform multiple writes per log event.
//
// Error Handling
//
// This package relies on the practice of wrapping or decorating loggers with
// other loggers to provide composable pieces of functionality. It also means
// that Logger.Log must return an error because some
// implementations—especially those that output log data to an io.Writer—may
// encounter errors that cannot be handled locally. This in turn means that
// Loggers that wrap other loggers should return errors from the wrapped
// logger up the stack.
//
// Fortunately, the decorator pattern also provides a way to avoid the
// necessity to check for errors every time an application calls Logger.Log.
// An application required to panic whenever its Logger encounters
// an error could initialize its logger as follows.
//
// fmtlogger := log.NewLogfmtLogger(log.NewSyncWriter(os.Stdout))
// logger := log.LoggerFunc(func(keyvals ...interface{}) error {
// if err := fmtlogger.Log(keyvals...); err != nil {
// panic(err)
// }
// return nil
// })
package log
+91
View File
@@ -0,0 +1,91 @@
package log
import (
"encoding"
"encoding/json"
"fmt"
"io"
"reflect"
)
type jsonLogger struct {
io.Writer
}
// NewJSONLogger returns a Logger that encodes keyvals to the Writer as a
// single JSON object. Each log event produces no more than one call to
// w.Write. The passed Writer must be safe for concurrent use by multiple
// goroutines if the returned Logger will be used concurrently.
func NewJSONLogger(w io.Writer) Logger {
return &jsonLogger{w}
}
func (l *jsonLogger) Log(keyvals ...interface{}) error {
n := (len(keyvals) + 1) / 2 // +1 to handle case when len is odd
m := make(map[string]interface{}, n)
for i := 0; i < len(keyvals); i += 2 {
k := keyvals[i]
var v interface{} = ErrMissingValue
if i+1 < len(keyvals) {
v = keyvals[i+1]
}
merge(m, k, v)
}
enc := json.NewEncoder(l.Writer)
enc.SetEscapeHTML(false)
return enc.Encode(m)
}
func merge(dst map[string]interface{}, k, v interface{}) {
var key string
switch x := k.(type) {
case string:
key = x
case fmt.Stringer:
key = safeString(x)
default:
key = fmt.Sprint(x)
}
// We want json.Marshaler and encoding.TextMarshaller to take priority over
// err.Error() and v.String(). But json.Marshall (called later) does that by
// default so we force a no-op if it's one of those 2 case.
switch x := v.(type) {
case json.Marshaler:
case encoding.TextMarshaler:
case error:
v = safeError(x)
case fmt.Stringer:
v = safeString(x)
}
dst[key] = v
}
func safeString(str fmt.Stringer) (s string) {
defer func() {
if panicVal := recover(); panicVal != nil {
if v := reflect.ValueOf(str); v.Kind() == reflect.Ptr && v.IsNil() {
s = "NULL"
} else {
panic(panicVal)
}
}
}()
s = str.String()
return
}
func safeError(err error) (s interface{}) {
defer func() {
if panicVal := recover(); panicVal != nil {
if v := reflect.ValueOf(err); v.Kind() == reflect.Ptr && v.IsNil() {
s = nil
} else {
panic(panicVal)
}
}
}()
s = err.Error()
return
}
+179
View File
@@ -0,0 +1,179 @@
package log
import "errors"
// Logger is the fundamental interface for all log operations. Log creates a
// log event from keyvals, a variadic sequence of alternating keys and values.
// Implementations must be safe for concurrent use by multiple goroutines. In
// particular, any implementation of Logger that appends to keyvals or
// modifies or retains any of its elements must make a copy first.
type Logger interface {
Log(keyvals ...interface{}) error
}
// ErrMissingValue is appended to keyvals slices with odd length to substitute
// the missing value.
var ErrMissingValue = errors.New("(MISSING)")
// With returns a new contextual logger with keyvals prepended to those passed
// to calls to Log. If logger is also a contextual logger created by With,
// WithPrefix, or WithSuffix, keyvals is appended to the existing context.
//
// The returned Logger replaces all value elements (odd indexes) containing a
// Valuer with their generated value for each call to its Log method.
func With(logger Logger, keyvals ...interface{}) Logger {
if len(keyvals) == 0 {
return logger
}
l := newContext(logger)
kvs := append(l.keyvals, keyvals...)
if len(kvs)%2 != 0 {
kvs = append(kvs, ErrMissingValue)
}
return &context{
logger: l.logger,
// Limiting the capacity of the stored keyvals ensures that a new
// backing array is created if the slice must grow in Log or With.
// Using the extra capacity without copying risks a data race that
// would violate the Logger interface contract.
keyvals: kvs[:len(kvs):len(kvs)],
hasValuer: l.hasValuer || containsValuer(keyvals),
sKeyvals: l.sKeyvals,
sHasValuer: l.sHasValuer,
}
}
// WithPrefix returns a new contextual logger with keyvals prepended to those
// passed to calls to Log. If logger is also a contextual logger created by
// With, WithPrefix, or WithSuffix, keyvals is prepended to the existing context.
//
// The returned Logger replaces all value elements (odd indexes) containing a
// Valuer with their generated value for each call to its Log method.
func WithPrefix(logger Logger, keyvals ...interface{}) Logger {
if len(keyvals) == 0 {
return logger
}
l := newContext(logger)
// Limiting the capacity of the stored keyvals ensures that a new
// backing array is created if the slice must grow in Log or With.
// Using the extra capacity without copying risks a data race that
// would violate the Logger interface contract.
n := len(l.keyvals) + len(keyvals)
if len(keyvals)%2 != 0 {
n++
}
kvs := make([]interface{}, 0, n)
kvs = append(kvs, keyvals...)
if len(kvs)%2 != 0 {
kvs = append(kvs, ErrMissingValue)
}
kvs = append(kvs, l.keyvals...)
return &context{
logger: l.logger,
keyvals: kvs,
hasValuer: l.hasValuer || containsValuer(keyvals),
sKeyvals: l.sKeyvals,
sHasValuer: l.sHasValuer,
}
}
// WithSuffix returns a new contextual logger with keyvals appended to those
// passed to calls to Log. If logger is also a contextual logger created by
// With, WithPrefix, or WithSuffix, keyvals is appended to the existing context.
//
// The returned Logger replaces all value elements (odd indexes) containing a
// Valuer with their generated value for each call to its Log method.
func WithSuffix(logger Logger, keyvals ...interface{}) Logger {
if len(keyvals) == 0 {
return logger
}
l := newContext(logger)
// Limiting the capacity of the stored keyvals ensures that a new
// backing array is created if the slice must grow in Log or With.
// Using the extra capacity without copying risks a data race that
// would violate the Logger interface contract.
n := len(l.sKeyvals) + len(keyvals)
if len(keyvals)%2 != 0 {
n++
}
kvs := make([]interface{}, 0, n)
kvs = append(kvs, keyvals...)
if len(kvs)%2 != 0 {
kvs = append(kvs, ErrMissingValue)
}
kvs = append(l.sKeyvals, kvs...)
return &context{
logger: l.logger,
keyvals: l.keyvals,
hasValuer: l.hasValuer,
sKeyvals: kvs,
sHasValuer: l.sHasValuer || containsValuer(keyvals),
}
}
// context is the Logger implementation returned by With, WithPrefix, and
// WithSuffix. It wraps a Logger and holds keyvals that it includes in all
// log events. Its Log method calls bindValues to generate values for each
// Valuer in the context keyvals.
//
// A context must always have the same number of stack frames between calls to
// its Log method and the eventual binding of Valuers to their value. This
// requirement comes from the functional requirement to allow a context to
// resolve application call site information for a Caller stored in the
// context. To do this we must be able to predict the number of logging
// functions on the stack when bindValues is called.
//
// Two implementation details provide the needed stack depth consistency.
//
// 1. newContext avoids introducing an additional layer when asked to
// wrap another context.
// 2. With, WithPrefix, and WithSuffix avoid introducing an additional
// layer by returning a newly constructed context with a merged keyvals
// rather than simply wrapping the existing context.
type context struct {
logger Logger
keyvals []interface{}
sKeyvals []interface{} // suffixes
hasValuer bool
sHasValuer bool
}
func newContext(logger Logger) *context {
if c, ok := logger.(*context); ok {
return c
}
return &context{logger: logger}
}
// Log replaces all value elements (odd indexes) containing a Valuer in the
// stored context with their generated value, appends keyvals, and passes the
// result to the wrapped Logger.
func (l *context) Log(keyvals ...interface{}) error {
kvs := append(l.keyvals, keyvals...)
if len(kvs)%2 != 0 {
kvs = append(kvs, ErrMissingValue)
}
if l.hasValuer {
// If no keyvals were appended above then we must copy l.keyvals so
// that future log events will reevaluate the stored Valuers.
if len(keyvals) == 0 {
kvs = append([]interface{}{}, l.keyvals...)
}
bindValues(kvs[:(len(l.keyvals))])
}
kvs = append(kvs, l.sKeyvals...)
if l.sHasValuer {
bindValues(kvs[len(kvs)-len(l.sKeyvals):])
}
return l.logger.Log(kvs...)
}
// LoggerFunc is an adapter to allow use of ordinary functions as Loggers. If
// f is a function with the appropriate signature, LoggerFunc(f) is a Logger
// object that calls f.
type LoggerFunc func(...interface{}) error
// Log implements Logger by calling f(keyvals...).
func (f LoggerFunc) Log(keyvals ...interface{}) error {
return f(keyvals...)
}
+62
View File
@@ -0,0 +1,62 @@
package log
import (
"bytes"
"io"
"sync"
"github.com/go-logfmt/logfmt"
)
type logfmtEncoder struct {
*logfmt.Encoder
buf bytes.Buffer
}
func (l *logfmtEncoder) Reset() {
l.Encoder.Reset()
l.buf.Reset()
}
var logfmtEncoderPool = sync.Pool{
New: func() interface{} {
var enc logfmtEncoder
enc.Encoder = logfmt.NewEncoder(&enc.buf)
return &enc
},
}
type logfmtLogger struct {
w io.Writer
}
// NewLogfmtLogger returns a logger that encodes keyvals to the Writer in
// logfmt format. Each log event produces no more than one call to w.Write.
// The passed Writer must be safe for concurrent use by multiple goroutines if
// the returned Logger will be used concurrently.
func NewLogfmtLogger(w io.Writer) Logger {
return &logfmtLogger{w}
}
func (l logfmtLogger) Log(keyvals ...interface{}) error {
enc := logfmtEncoderPool.Get().(*logfmtEncoder)
enc.Reset()
defer logfmtEncoderPool.Put(enc)
if err := enc.EncodeKeyvals(keyvals...); err != nil {
return err
}
// Add newline to the end of the buffer
if err := enc.EndRecord(); err != nil {
return err
}
// The Logger interface requires implementations to be safe for concurrent
// use by multiple goroutines. For this implementation that means making
// only one call to l.w.Write() for each call to Log.
if _, err := l.w.Write(enc.buf.Bytes()); err != nil {
return err
}
return nil
}
+8
View File
@@ -0,0 +1,8 @@
package log
type nopLogger struct{}
// NewNopLogger returns a logger that doesn't do anything.
func NewNopLogger() Logger { return nopLogger{} }
func (nopLogger) Log(...interface{}) error { return nil }
+151
View File
@@ -0,0 +1,151 @@
package log
import (
"bytes"
"io"
"log"
"regexp"
"strings"
)
// StdlibWriter implements io.Writer by invoking the stdlib log.Print. It's
// designed to be passed to a Go kit logger as the writer, for cases where
// it's necessary to redirect all Go kit log output to the stdlib logger.
//
// If you have any choice in the matter, you shouldn't use this. Prefer to
// redirect the stdlib log to the Go kit logger via NewStdlibAdapter.
type StdlibWriter struct{}
// Write implements io.Writer.
func (w StdlibWriter) Write(p []byte) (int, error) {
log.Print(strings.TrimSpace(string(p)))
return len(p), nil
}
// StdlibAdapter wraps a Logger and allows it to be passed to the stdlib
// logger's SetOutput. It will extract date/timestamps, filenames, and
// messages, and place them under relevant keys.
type StdlibAdapter struct {
Logger
timestampKey string
fileKey string
messageKey string
prefix string
joinPrefixToMsg bool
}
// StdlibAdapterOption sets a parameter for the StdlibAdapter.
type StdlibAdapterOption func(*StdlibAdapter)
// TimestampKey sets the key for the timestamp field. By default, it's "ts".
func TimestampKey(key string) StdlibAdapterOption {
return func(a *StdlibAdapter) { a.timestampKey = key }
}
// FileKey sets the key for the file and line field. By default, it's "caller".
func FileKey(key string) StdlibAdapterOption {
return func(a *StdlibAdapter) { a.fileKey = key }
}
// MessageKey sets the key for the actual log message. By default, it's "msg".
func MessageKey(key string) StdlibAdapterOption {
return func(a *StdlibAdapter) { a.messageKey = key }
}
// Prefix configures the adapter to parse a prefix from stdlib log events. If
// you provide a non-empty prefix to the stdlib logger, then your should provide
// that same prefix to the adapter via this option.
//
// By default, the prefix isn't included in the msg key. Set joinPrefixToMsg to
// true if you want to include the parsed prefix in the msg.
func Prefix(prefix string, joinPrefixToMsg bool) StdlibAdapterOption {
return func(a *StdlibAdapter) { a.prefix = prefix; a.joinPrefixToMsg = joinPrefixToMsg }
}
// NewStdlibAdapter returns a new StdlibAdapter wrapper around the passed
// logger. It's designed to be passed to log.SetOutput.
func NewStdlibAdapter(logger Logger, options ...StdlibAdapterOption) io.Writer {
a := StdlibAdapter{
Logger: logger,
timestampKey: "ts",
fileKey: "caller",
messageKey: "msg",
}
for _, option := range options {
option(&a)
}
return a
}
func (a StdlibAdapter) Write(p []byte) (int, error) {
p = a.handlePrefix(p)
result := subexps(p)
keyvals := []interface{}{}
var timestamp string
if date, ok := result["date"]; ok && date != "" {
timestamp = date
}
if time, ok := result["time"]; ok && time != "" {
if timestamp != "" {
timestamp += " "
}
timestamp += time
}
if timestamp != "" {
keyvals = append(keyvals, a.timestampKey, timestamp)
}
if file, ok := result["file"]; ok && file != "" {
keyvals = append(keyvals, a.fileKey, file)
}
if msg, ok := result["msg"]; ok {
msg = a.handleMessagePrefix(msg)
keyvals = append(keyvals, a.messageKey, msg)
}
if err := a.Logger.Log(keyvals...); err != nil {
return 0, err
}
return len(p), nil
}
func (a StdlibAdapter) handlePrefix(p []byte) []byte {
if a.prefix != "" {
p = bytes.TrimPrefix(p, []byte(a.prefix))
}
return p
}
func (a StdlibAdapter) handleMessagePrefix(msg string) string {
if a.prefix == "" {
return msg
}
msg = strings.TrimPrefix(msg, a.prefix)
if a.joinPrefixToMsg {
msg = a.prefix + msg
}
return msg
}
const (
logRegexpDate = `(?P<date>[0-9]{4}/[0-9]{2}/[0-9]{2})?[ ]?`
logRegexpTime = `(?P<time>[0-9]{2}:[0-9]{2}:[0-9]{2}(\.[0-9]+)?)?[ ]?`
logRegexpFile = `(?P<file>.+?:[0-9]+)?`
logRegexpMsg = `(: )?(?P<msg>(?s:.*))`
)
var (
logRegexp = regexp.MustCompile(logRegexpDate + logRegexpTime + logRegexpFile + logRegexpMsg)
)
func subexps(line []byte) map[string]string {
m := logRegexp.FindSubmatch(line)
if len(m) < len(logRegexp.SubexpNames()) {
return map[string]string{}
}
result := map[string]string{}
for i, name := range logRegexp.SubexpNames() {
result[name] = strings.TrimRight(string(m[i]), "\n")
}
return result
}
+113
View File
@@ -0,0 +1,113 @@
package log
import (
"io"
"sync"
"sync/atomic"
)
// SwapLogger wraps another logger that may be safely replaced while other
// goroutines use the SwapLogger concurrently. The zero value for a SwapLogger
// will discard all log events without error.
//
// SwapLogger serves well as a package global logger that can be changed by
// importers.
type SwapLogger struct {
logger atomic.Value
}
type loggerStruct struct {
Logger
}
// Log implements the Logger interface by forwarding keyvals to the currently
// wrapped logger. It does not log anything if the wrapped logger is nil.
func (l *SwapLogger) Log(keyvals ...interface{}) error {
s, ok := l.logger.Load().(loggerStruct)
if !ok || s.Logger == nil {
return nil
}
return s.Log(keyvals...)
}
// Swap replaces the currently wrapped logger with logger. Swap may be called
// concurrently with calls to Log from other goroutines.
func (l *SwapLogger) Swap(logger Logger) {
l.logger.Store(loggerStruct{logger})
}
// NewSyncWriter returns a new writer that is safe for concurrent use by
// multiple goroutines. Writes to the returned writer are passed on to w. If
// another write is already in progress, the calling goroutine blocks until
// the writer is available.
//
// If w implements the following interface, so does the returned writer.
//
// interface {
// Fd() uintptr
// }
func NewSyncWriter(w io.Writer) io.Writer {
switch w := w.(type) {
case fdWriter:
return &fdSyncWriter{fdWriter: w}
default:
return &syncWriter{Writer: w}
}
}
// syncWriter synchronizes concurrent writes to an io.Writer.
type syncWriter struct {
sync.Mutex
io.Writer
}
// Write writes p to the underlying io.Writer. If another write is already in
// progress, the calling goroutine blocks until the syncWriter is available.
func (w *syncWriter) Write(p []byte) (n int, err error) {
w.Lock()
defer w.Unlock()
return w.Writer.Write(p)
}
// fdWriter is an io.Writer that also has an Fd method. The most common
// example of an fdWriter is an *os.File.
type fdWriter interface {
io.Writer
Fd() uintptr
}
// fdSyncWriter synchronizes concurrent writes to an fdWriter.
type fdSyncWriter struct {
sync.Mutex
fdWriter
}
// Write writes p to the underlying io.Writer. If another write is already in
// progress, the calling goroutine blocks until the fdSyncWriter is available.
func (w *fdSyncWriter) Write(p []byte) (n int, err error) {
w.Lock()
defer w.Unlock()
return w.fdWriter.Write(p)
}
// syncLogger provides concurrent safe logging for another Logger.
type syncLogger struct {
mu sync.Mutex
logger Logger
}
// NewSyncLogger returns a logger that synchronizes concurrent use of the
// wrapped logger. When multiple goroutines use the SyncLogger concurrently
// only one goroutine will be allowed to log to the wrapped logger at a time.
// The other goroutines will block until the logger is available.
func NewSyncLogger(logger Logger) Logger {
return &syncLogger{logger: logger}
}
// Log logs keyvals to the underlying Logger. If another log is already in
// progress, the calling goroutine blocks until the syncLogger is available.
func (l *syncLogger) Log(keyvals ...interface{}) error {
l.mu.Lock()
defer l.mu.Unlock()
return l.logger.Log(keyvals...)
}
+110
View File
@@ -0,0 +1,110 @@
package log
import (
"runtime"
"strconv"
"strings"
"time"
)
// A Valuer generates a log value. When passed to With, WithPrefix, or
// WithSuffix in a value element (odd indexes), it represents a dynamic
// value which is re-evaluated with each log event.
type Valuer func() interface{}
// bindValues replaces all value elements (odd indexes) containing a Valuer
// with their generated value.
func bindValues(keyvals []interface{}) {
for i := 1; i < len(keyvals); i += 2 {
if v, ok := keyvals[i].(Valuer); ok {
keyvals[i] = v()
}
}
}
// containsValuer returns true if any of the value elements (odd indexes)
// contain a Valuer.
func containsValuer(keyvals []interface{}) bool {
for i := 1; i < len(keyvals); i += 2 {
if _, ok := keyvals[i].(Valuer); ok {
return true
}
}
return false
}
// Timestamp returns a timestamp Valuer. It invokes the t function to get the
// time; unless you are doing something tricky, pass time.Now.
//
// Most users will want to use DefaultTimestamp or DefaultTimestampUTC, which
// are TimestampFormats that use the RFC3339Nano format.
func Timestamp(t func() time.Time) Valuer {
return func() interface{} { return t() }
}
// TimestampFormat returns a timestamp Valuer with a custom time format. It
// invokes the t function to get the time to format; unless you are doing
// something tricky, pass time.Now. The layout string is passed to
// Time.Format.
//
// Most users will want to use DefaultTimestamp or DefaultTimestampUTC, which
// are TimestampFormats that use the RFC3339Nano format.
func TimestampFormat(t func() time.Time, layout string) Valuer {
return func() interface{} {
return timeFormat{
time: t(),
layout: layout,
}
}
}
// A timeFormat represents an instant in time and a layout used when
// marshaling to a text format.
type timeFormat struct {
time time.Time
layout string
}
func (tf timeFormat) String() string {
return tf.time.Format(tf.layout)
}
// MarshalText implements encoding.TextMarshaller.
func (tf timeFormat) MarshalText() (text []byte, err error) {
// The following code adapted from the standard library time.Time.Format
// method. Using the same undocumented magic constant to extend the size
// of the buffer as seen there.
b := make([]byte, 0, len(tf.layout)+10)
b = tf.time.AppendFormat(b, tf.layout)
return b, nil
}
// Caller returns a Valuer that returns a file and line from a specified depth
// in the callstack. Users will probably want to use DefaultCaller.
func Caller(depth int) Valuer {
return func() interface{} {
_, file, line, _ := runtime.Caller(depth)
idx := strings.LastIndexByte(file, '/')
// using idx+1 below handles both of following cases:
// idx == -1 because no "/" was found, or
// idx >= 0 and we want to start at the character after the found "/".
return file[idx+1:] + ":" + strconv.Itoa(line)
}
}
var (
// DefaultTimestamp is a Valuer that returns the current wallclock time,
// respecting time zones, when bound.
DefaultTimestamp = TimestampFormat(time.Now, time.RFC3339Nano)
// DefaultTimestampUTC is a Valuer that returns the current time in UTC
// when bound.
DefaultTimestampUTC = TimestampFormat(
func() time.Time { return time.Now().UTC() },
time.RFC3339Nano,
)
// DefaultCaller is a Valuer that returns the file and line where the Log
// method was invoked. It can only be used with log.With.
DefaultCaller = Caller(3)
)
+1
View File
@@ -0,0 +1 @@
.vscode/
+48
View File
@@ -0,0 +1,48 @@
# Changelog
All notable changes to this project will be documented in this file.
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
## [0.5.0] - 2020-01-03
### Changed
- Remove the dependency on github.com/kr/logfmt by [@ChrisHines]
- Move fuzz code to github.com/go-logfmt/fuzzlogfmt by [@ChrisHines]
## [0.4.0] - 2018-11-21
### Added
- Go module support by [@ChrisHines]
- CHANGELOG by [@ChrisHines]
### Changed
- Drop invalid runes from keys instead of returning ErrInvalidKey by [@ChrisHines]
- On panic while printing, attempt to print panic value by [@bboreham]
## [0.3.0] - 2016-11-15
### Added
- Pool buffers for quoted strings and byte slices by [@nussjustin]
### Fixed
- Fuzz fix, quote invalid UTF-8 values by [@judwhite]
## [0.2.0] - 2016-05-08
### Added
- Encoder.EncodeKeyvals by [@ChrisHines]
## [0.1.0] - 2016-03-28
### Added
- Encoder by [@ChrisHines]
- Decoder by [@ChrisHines]
- MarshalKeyvals by [@ChrisHines]
[0.5.0]: https://github.com/go-logfmt/logfmt/compare/v0.4.0...v0.5.0
[0.4.0]: https://github.com/go-logfmt/logfmt/compare/v0.3.0...v0.4.0
[0.3.0]: https://github.com/go-logfmt/logfmt/compare/v0.2.0...v0.3.0
[0.2.0]: https://github.com/go-logfmt/logfmt/compare/v0.1.0...v0.2.0
[0.1.0]: https://github.com/go-logfmt/logfmt/commits/v0.1.0
[@ChrisHines]: https://github.com/ChrisHines
[@bboreham]: https://github.com/bboreham
[@judwhite]: https://github.com/judwhite
[@nussjustin]: https://github.com/nussjustin
+22
View File
@@ -0,0 +1,22 @@
The MIT License (MIT)
Copyright (c) 2015 go-logfmt
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
+33
View File
@@ -0,0 +1,33 @@
[![Go Reference](https://pkg.go.dev/badge/github.com/go-logfmt/logfmt.svg)](https://pkg.go.dev/github.com/go-logfmt/logfmt)
[![Go Report Card](https://goreportcard.com/badge/go-logfmt/logfmt)](https://goreportcard.com/report/go-logfmt/logfmt)
[![Github Actions](https://github.com/go-logfmt/logfmt/actions/workflows/test.yml/badge.svg)](https://github.com/go-logfmt/logfmt/actions/workflows/test.yml)
[![Coverage Status](https://coveralls.io/repos/github/go-logfmt/logfmt/badge.svg?branch=master)](https://coveralls.io/github/go-logfmt/logfmt?branch=master)
# logfmt
Package logfmt implements utilities to marshal and unmarshal data in the [logfmt
format](https://brandur.org/logfmt). It provides an API similar to
[encoding/json](http://golang.org/pkg/encoding/json/) and
[encoding/xml](http://golang.org/pkg/encoding/xml/).
The logfmt format was first documented by Brandur Leach in [this
article](https://brandur.org/logfmt). The format has not been formally
standardized. The most authoritative public specification to date has been the
documentation of a Go Language [package](http://godoc.org/github.com/kr/logfmt)
written by Blake Mizerany and Keith Rarick.
## Goals
This project attempts to conform as closely as possible to the prior art, while
also removing ambiguity where necessary to provide well behaved encoder and
decoder implementations.
## Non-goals
This project does not attempt to formally standardize the logfmt format. In the
event that logfmt is standardized this project would take conforming to the
standard as a goal.
## Versioning
Package logfmt publishes releases via [semver](http://semver.org/) compatible Git tags prefixed with a single 'v'.
+237
View File
@@ -0,0 +1,237 @@
package logfmt
import (
"bufio"
"bytes"
"fmt"
"io"
"unicode/utf8"
)
// A Decoder reads and decodes logfmt records from an input stream.
type Decoder struct {
pos int
key []byte
value []byte
lineNum int
s *bufio.Scanner
err error
}
// NewDecoder returns a new decoder that reads from r.
//
// The decoder introduces its own buffering and may read data from r beyond
// the logfmt records requested.
func NewDecoder(r io.Reader) *Decoder {
dec := &Decoder{
s: bufio.NewScanner(r),
}
return dec
}
// ScanRecord advances the Decoder to the next record, which can then be
// parsed with the ScanKeyval method. It returns false when decoding stops,
// either by reaching the end of the input or an error. After ScanRecord
// returns false, the Err method will return any error that occurred during
// decoding, except that if it was io.EOF, Err will return nil.
func (dec *Decoder) ScanRecord() bool {
if dec.err != nil {
return false
}
if !dec.s.Scan() {
dec.err = dec.s.Err()
return false
}
dec.lineNum++
dec.pos = 0
return true
}
// ScanKeyval advances the Decoder to the next key/value pair of the current
// record, which can then be retrieved with the Key and Value methods. It
// returns false when decoding stops, either by reaching the end of the
// current record or an error.
func (dec *Decoder) ScanKeyval() bool {
dec.key, dec.value = nil, nil
if dec.err != nil {
return false
}
line := dec.s.Bytes()
// garbage
for p, c := range line[dec.pos:] {
if c > ' ' {
dec.pos += p
goto key
}
}
dec.pos = len(line)
return false
key:
const invalidKeyError = "invalid key"
start, multibyte := dec.pos, false
for p, c := range line[dec.pos:] {
switch {
case c == '=':
dec.pos += p
if dec.pos > start {
dec.key = line[start:dec.pos]
if multibyte && bytes.ContainsRune(dec.key, utf8.RuneError) {
dec.syntaxError(invalidKeyError)
return false
}
}
if dec.key == nil {
dec.unexpectedByte(c)
return false
}
goto equal
case c == '"':
dec.pos += p
dec.unexpectedByte(c)
return false
case c <= ' ':
dec.pos += p
if dec.pos > start {
dec.key = line[start:dec.pos]
if multibyte && bytes.ContainsRune(dec.key, utf8.RuneError) {
dec.syntaxError(invalidKeyError)
return false
}
}
return true
case c >= utf8.RuneSelf:
multibyte = true
}
}
dec.pos = len(line)
if dec.pos > start {
dec.key = line[start:dec.pos]
if multibyte && bytes.ContainsRune(dec.key, utf8.RuneError) {
dec.syntaxError(invalidKeyError)
return false
}
}
return true
equal:
dec.pos++
if dec.pos >= len(line) {
return true
}
switch c := line[dec.pos]; {
case c <= ' ':
return true
case c == '"':
goto qvalue
}
// value
start = dec.pos
for p, c := range line[dec.pos:] {
switch {
case c == '=' || c == '"':
dec.pos += p
dec.unexpectedByte(c)
return false
case c <= ' ':
dec.pos += p
if dec.pos > start {
dec.value = line[start:dec.pos]
}
return true
}
}
dec.pos = len(line)
if dec.pos > start {
dec.value = line[start:dec.pos]
}
return true
qvalue:
const (
untermQuote = "unterminated quoted value"
invalidQuote = "invalid quoted value"
)
hasEsc, esc := false, false
start = dec.pos
for p, c := range line[dec.pos+1:] {
switch {
case esc:
esc = false
case c == '\\':
hasEsc, esc = true, true
case c == '"':
dec.pos += p + 2
if hasEsc {
v, ok := unquoteBytes(line[start:dec.pos])
if !ok {
dec.syntaxError(invalidQuote)
return false
}
dec.value = v
} else {
start++
end := dec.pos - 1
if end > start {
dec.value = line[start:end]
}
}
return true
}
}
dec.pos = len(line)
dec.syntaxError(untermQuote)
return false
}
// Key returns the most recent key found by a call to ScanKeyval. The returned
// slice may point to internal buffers and is only valid until the next call
// to ScanRecord. It does no allocation.
func (dec *Decoder) Key() []byte {
return dec.key
}
// Value returns the most recent value found by a call to ScanKeyval. The
// returned slice may point to internal buffers and is only valid until the
// next call to ScanRecord. It does no allocation when the value has no
// escape sequences.
func (dec *Decoder) Value() []byte {
return dec.value
}
// Err returns the first non-EOF error that was encountered by the Scanner.
func (dec *Decoder) Err() error {
return dec.err
}
func (dec *Decoder) syntaxError(msg string) {
dec.err = &SyntaxError{
Msg: msg,
Line: dec.lineNum,
Pos: dec.pos + 1,
}
}
func (dec *Decoder) unexpectedByte(c byte) {
dec.err = &SyntaxError{
Msg: fmt.Sprintf("unexpected %q", c),
Line: dec.lineNum,
Pos: dec.pos + 1,
}
}
// A SyntaxError represents a syntax error in the logfmt input stream.
type SyntaxError struct {
Msg string
Line int
Pos int
}
func (e *SyntaxError) Error() string {
return fmt.Sprintf("logfmt syntax error at pos %d on line %d: %s", e.Pos, e.Line, e.Msg)
}
+6
View File
@@ -0,0 +1,6 @@
// Package logfmt implements utilities to marshal and unmarshal data in the
// logfmt format. The logfmt format records key/value pairs in a way that
// balances readability for humans and simplicity of computer parsing. It is
// most commonly used as a more human friendly alternative to JSON for
// structured logging.
package logfmt
+322
View File
@@ -0,0 +1,322 @@
package logfmt
import (
"bytes"
"encoding"
"errors"
"fmt"
"io"
"reflect"
"strings"
"unicode/utf8"
)
// MarshalKeyvals returns the logfmt encoding of keyvals, a variadic sequence
// of alternating keys and values.
func MarshalKeyvals(keyvals ...interface{}) ([]byte, error) {
buf := &bytes.Buffer{}
if err := NewEncoder(buf).EncodeKeyvals(keyvals...); err != nil {
return nil, err
}
return buf.Bytes(), nil
}
// An Encoder writes logfmt data to an output stream.
type Encoder struct {
w io.Writer
scratch bytes.Buffer
needSep bool
}
// NewEncoder returns a new encoder that writes to w.
func NewEncoder(w io.Writer) *Encoder {
return &Encoder{
w: w,
}
}
var (
space = []byte(" ")
equals = []byte("=")
newline = []byte("\n")
null = []byte("null")
)
// EncodeKeyval writes the logfmt encoding of key and value to the stream. A
// single space is written before the second and subsequent keys in a record.
// Nothing is written if a non-nil error is returned.
func (enc *Encoder) EncodeKeyval(key, value interface{}) error {
enc.scratch.Reset()
if enc.needSep {
if _, err := enc.scratch.Write(space); err != nil {
return err
}
}
if err := writeKey(&enc.scratch, key); err != nil {
return err
}
if _, err := enc.scratch.Write(equals); err != nil {
return err
}
if err := writeValue(&enc.scratch, value); err != nil {
return err
}
_, err := enc.w.Write(enc.scratch.Bytes())
enc.needSep = true
return err
}
// EncodeKeyvals writes the logfmt encoding of keyvals to the stream. Keyvals
// is a variadic sequence of alternating keys and values. Keys of unsupported
// type are skipped along with their corresponding value. Values of
// unsupported type or that cause a MarshalerError are replaced by their error
// but do not cause EncodeKeyvals to return an error. If a non-nil error is
// returned some key/value pairs may not have be written.
func (enc *Encoder) EncodeKeyvals(keyvals ...interface{}) error {
if len(keyvals) == 0 {
return nil
}
if len(keyvals)%2 == 1 {
keyvals = append(keyvals, nil)
}
for i := 0; i < len(keyvals); i += 2 {
k, v := keyvals[i], keyvals[i+1]
err := enc.EncodeKeyval(k, v)
if err == ErrUnsupportedKeyType {
continue
}
if _, ok := err.(*MarshalerError); ok || err == ErrUnsupportedValueType {
v = err
err = enc.EncodeKeyval(k, v)
}
if err != nil {
return err
}
}
return nil
}
// MarshalerError represents an error encountered while marshaling a value.
type MarshalerError struct {
Type reflect.Type
Err error
}
func (e *MarshalerError) Error() string {
return "error marshaling value of type " + e.Type.String() + ": " + e.Err.Error()
}
// ErrNilKey is returned by Marshal functions and Encoder methods if a key is
// a nil interface or pointer value.
var ErrNilKey = errors.New("nil key")
// ErrInvalidKey is returned by Marshal functions and Encoder methods if, after
// dropping invalid runes, a key is empty.
var ErrInvalidKey = errors.New("invalid key")
// ErrUnsupportedKeyType is returned by Encoder methods if a key has an
// unsupported type.
var ErrUnsupportedKeyType = errors.New("unsupported key type")
// ErrUnsupportedValueType is returned by Encoder methods if a value has an
// unsupported type.
var ErrUnsupportedValueType = errors.New("unsupported value type")
func writeKey(w io.Writer, key interface{}) error {
if key == nil {
return ErrNilKey
}
switch k := key.(type) {
case string:
return writeStringKey(w, k)
case []byte:
if k == nil {
return ErrNilKey
}
return writeBytesKey(w, k)
case encoding.TextMarshaler:
kb, err := safeMarshal(k)
if err != nil {
return err
}
if kb == nil {
return ErrNilKey
}
return writeBytesKey(w, kb)
case fmt.Stringer:
ks, ok := safeString(k)
if !ok {
return ErrNilKey
}
return writeStringKey(w, ks)
default:
rkey := reflect.ValueOf(key)
switch rkey.Kind() {
case reflect.Array, reflect.Chan, reflect.Func, reflect.Map, reflect.Slice, reflect.Struct:
return ErrUnsupportedKeyType
case reflect.Ptr:
if rkey.IsNil() {
return ErrNilKey
}
return writeKey(w, rkey.Elem().Interface())
}
return writeStringKey(w, fmt.Sprint(k))
}
}
// keyRuneFilter returns r for all valid key runes, and -1 for all invalid key
// runes. When used as the mapping function for strings.Map and bytes.Map
// functions it causes them to remove invalid key runes from strings or byte
// slices respectively.
func keyRuneFilter(r rune) rune {
if r <= ' ' || r == '=' || r == '"' || r == utf8.RuneError {
return -1
}
return r
}
func writeStringKey(w io.Writer, key string) error {
k := strings.Map(keyRuneFilter, key)
if k == "" {
return ErrInvalidKey
}
_, err := io.WriteString(w, k)
return err
}
func writeBytesKey(w io.Writer, key []byte) error {
k := bytes.Map(keyRuneFilter, key)
if len(k) == 0 {
return ErrInvalidKey
}
_, err := w.Write(k)
return err
}
func writeValue(w io.Writer, value interface{}) error {
switch v := value.(type) {
case nil:
return writeBytesValue(w, null)
case string:
return writeStringValue(w, v, true)
case []byte:
return writeBytesValue(w, v)
case encoding.TextMarshaler:
vb, err := safeMarshal(v)
if err != nil {
return err
}
if vb == nil {
vb = null
}
return writeBytesValue(w, vb)
case error:
se, ok := safeError(v)
return writeStringValue(w, se, ok)
case fmt.Stringer:
ss, ok := safeString(v)
return writeStringValue(w, ss, ok)
default:
rvalue := reflect.ValueOf(value)
switch rvalue.Kind() {
case reflect.Array, reflect.Chan, reflect.Func, reflect.Map, reflect.Slice, reflect.Struct:
return ErrUnsupportedValueType
case reflect.Ptr:
if rvalue.IsNil() {
return writeBytesValue(w, null)
}
return writeValue(w, rvalue.Elem().Interface())
}
return writeStringValue(w, fmt.Sprint(v), true)
}
}
func needsQuotedValueRune(r rune) bool {
return r <= ' ' || r == '=' || r == '"' || r == utf8.RuneError
}
func writeStringValue(w io.Writer, value string, ok bool) error {
var err error
if ok && value == "null" {
_, err = io.WriteString(w, `"null"`)
} else if strings.IndexFunc(value, needsQuotedValueRune) != -1 {
_, err = writeQuotedString(w, value)
} else {
_, err = io.WriteString(w, value)
}
return err
}
func writeBytesValue(w io.Writer, value []byte) error {
var err error
if bytes.IndexFunc(value, needsQuotedValueRune) != -1 {
_, err = writeQuotedBytes(w, value)
} else {
_, err = w.Write(value)
}
return err
}
// EndRecord writes a newline character to the stream and resets the encoder
// to the beginning of a new record.
func (enc *Encoder) EndRecord() error {
_, err := enc.w.Write(newline)
if err == nil {
enc.needSep = false
}
return err
}
// Reset resets the encoder to the beginning of a new record.
func (enc *Encoder) Reset() {
enc.needSep = false
}
func safeError(err error) (s string, ok bool) {
defer func() {
if panicVal := recover(); panicVal != nil {
if v := reflect.ValueOf(err); v.Kind() == reflect.Ptr && v.IsNil() {
s, ok = "null", false
} else {
s, ok = fmt.Sprintf("PANIC:%v", panicVal), false
}
}
}()
s, ok = err.Error(), true
return
}
func safeString(str fmt.Stringer) (s string, ok bool) {
defer func() {
if panicVal := recover(); panicVal != nil {
if v := reflect.ValueOf(str); v.Kind() == reflect.Ptr && v.IsNil() {
s, ok = "null", false
} else {
s, ok = fmt.Sprintf("PANIC:%v", panicVal), true
}
}
}()
s, ok = str.String(), true
return
}
func safeMarshal(tm encoding.TextMarshaler) (b []byte, err error) {
defer func() {
if panicVal := recover(); panicVal != nil {
if v := reflect.ValueOf(tm); v.Kind() == reflect.Ptr && v.IsNil() {
b, err = nil, nil
} else {
b, err = nil, fmt.Errorf("panic when marshalling: %s", panicVal)
}
}
}()
b, err = tm.MarshalText()
if err != nil {
return nil, &MarshalerError{
Type: reflect.TypeOf(tm),
Err: err,
}
}
return
}
+277
View File
@@ -0,0 +1,277 @@
package logfmt
import (
"bytes"
"io"
"strconv"
"sync"
"unicode"
"unicode/utf16"
"unicode/utf8"
)
// Taken from Go's encoding/json and modified for use here.
// Copyright 2010 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
var hex = "0123456789abcdef"
var bufferPool = sync.Pool{
New: func() interface{} {
return &bytes.Buffer{}
},
}
func getBuffer() *bytes.Buffer {
return bufferPool.Get().(*bytes.Buffer)
}
func poolBuffer(buf *bytes.Buffer) {
buf.Reset()
bufferPool.Put(buf)
}
// NOTE: keep in sync with writeQuotedBytes below.
func writeQuotedString(w io.Writer, s string) (int, error) {
buf := getBuffer()
buf.WriteByte('"')
start := 0
for i := 0; i < len(s); {
if b := s[i]; b < utf8.RuneSelf {
if 0x20 <= b && b != '\\' && b != '"' {
i++
continue
}
if start < i {
buf.WriteString(s[start:i])
}
switch b {
case '\\', '"':
buf.WriteByte('\\')
buf.WriteByte(b)
case '\n':
buf.WriteByte('\\')
buf.WriteByte('n')
case '\r':
buf.WriteByte('\\')
buf.WriteByte('r')
case '\t':
buf.WriteByte('\\')
buf.WriteByte('t')
default:
// This encodes bytes < 0x20 except for \n, \r, and \t.
buf.WriteString(`\u00`)
buf.WriteByte(hex[b>>4])
buf.WriteByte(hex[b&0xF])
}
i++
start = i
continue
}
c, size := utf8.DecodeRuneInString(s[i:])
if c == utf8.RuneError {
if start < i {
buf.WriteString(s[start:i])
}
buf.WriteString(`\ufffd`)
i += size
start = i
continue
}
i += size
}
if start < len(s) {
buf.WriteString(s[start:])
}
buf.WriteByte('"')
n, err := w.Write(buf.Bytes())
poolBuffer(buf)
return n, err
}
// NOTE: keep in sync with writeQuoteString above.
func writeQuotedBytes(w io.Writer, s []byte) (int, error) {
buf := getBuffer()
buf.WriteByte('"')
start := 0
for i := 0; i < len(s); {
if b := s[i]; b < utf8.RuneSelf {
if 0x20 <= b && b != '\\' && b != '"' {
i++
continue
}
if start < i {
buf.Write(s[start:i])
}
switch b {
case '\\', '"':
buf.WriteByte('\\')
buf.WriteByte(b)
case '\n':
buf.WriteByte('\\')
buf.WriteByte('n')
case '\r':
buf.WriteByte('\\')
buf.WriteByte('r')
case '\t':
buf.WriteByte('\\')
buf.WriteByte('t')
default:
// This encodes bytes < 0x20 except for \n, \r, and \t.
buf.WriteString(`\u00`)
buf.WriteByte(hex[b>>4])
buf.WriteByte(hex[b&0xF])
}
i++
start = i
continue
}
c, size := utf8.DecodeRune(s[i:])
if c == utf8.RuneError {
if start < i {
buf.Write(s[start:i])
}
buf.WriteString(`\ufffd`)
i += size
start = i
continue
}
i += size
}
if start < len(s) {
buf.Write(s[start:])
}
buf.WriteByte('"')
n, err := w.Write(buf.Bytes())
poolBuffer(buf)
return n, err
}
// getu4 decodes \uXXXX from the beginning of s, returning the hex value,
// or it returns -1.
func getu4(s []byte) rune {
if len(s) < 6 || s[0] != '\\' || s[1] != 'u' {
return -1
}
r, err := strconv.ParseUint(string(s[2:6]), 16, 64)
if err != nil {
return -1
}
return rune(r)
}
func unquoteBytes(s []byte) (t []byte, ok bool) {
if len(s) < 2 || s[0] != '"' || s[len(s)-1] != '"' {
return
}
s = s[1 : len(s)-1]
// Check for unusual characters. If there are none,
// then no unquoting is needed, so return a slice of the
// original bytes.
r := 0
for r < len(s) {
c := s[r]
if c == '\\' || c == '"' || c < ' ' {
break
}
if c < utf8.RuneSelf {
r++
continue
}
rr, size := utf8.DecodeRune(s[r:])
if rr == utf8.RuneError {
break
}
r += size
}
if r == len(s) {
return s, true
}
b := make([]byte, len(s)+2*utf8.UTFMax)
w := copy(b, s[0:r])
for r < len(s) {
// Out of room? Can only happen if s is full of
// malformed UTF-8 and we're replacing each
// byte with RuneError.
if w >= len(b)-2*utf8.UTFMax {
nb := make([]byte, (len(b)+utf8.UTFMax)*2)
copy(nb, b[0:w])
b = nb
}
switch c := s[r]; {
case c == '\\':
r++
if r >= len(s) {
return
}
switch s[r] {
default:
return
case '"', '\\', '/', '\'':
b[w] = s[r]
r++
w++
case 'b':
b[w] = '\b'
r++
w++
case 'f':
b[w] = '\f'
r++
w++
case 'n':
b[w] = '\n'
r++
w++
case 'r':
b[w] = '\r'
r++
w++
case 't':
b[w] = '\t'
r++
w++
case 'u':
r--
rr := getu4(s[r:])
if rr < 0 {
return
}
r += 6
if utf16.IsSurrogate(rr) {
rr1 := getu4(s[r:])
if dec := utf16.DecodeRune(rr, rr1); dec != unicode.ReplacementChar {
// A valid pair; consume.
r += 6
w += utf8.EncodeRune(b[w:], dec)
break
}
// Invalid surrogate; fall back to replacement rune.
rr = unicode.ReplacementChar
}
w += utf8.EncodeRune(b[w:], rr)
}
// Quote, control characters are invalid.
case c == '"', c < ' ':
return
// ASCII
case c < utf8.RuneSelf:
b[w] = c
r++
w++
// Coerce to well-formed UTF-8.
default:
rr, size := utf8.DecodeRune(s[r:])
r += size
w += utf8.EncodeRune(b[w:], rr)
}
}
return b[0:w], true
}
+1
View File
@@ -0,0 +1 @@
tags
+7
View File
@@ -0,0 +1,7 @@
language: go
go:
- tip
script:
- go test -v ./...
+56
View File
@@ -0,0 +1,56 @@
## Changelog
#### Update 10 September 2016
- Add support for go1.7 net.Context
#### Update 25 September 2015
- Add support for Sub router
Example :
``` go
func main() {
mux := bone.New()
sub := mux.NewRouter()
sub.GetFunc("/test/example", func(rw http.ResponseWriter, req *http.Request) {
rw.Write([]byte("From sub router !"))
})
mux.SubRoute("/api", sub)
http.ListenAndServe(":8080", mux)
}
```
#### Update 26 April 2015
- Add Support for REGEX parameters, using ` # ` instead of ` : `.
- Add Mux method ` mux.GetFunc(), mux.PostFunc(), etc ... `, takes ` http.HandlerFunc ` instead of ` http.Handler `.
Example :
``` go
func main() {
mux.GetFunc("/route/#var^[a-z]$", handler)
}
func handler(rw http.ResponseWriter, req *http.Request) {
bone.GetValue(req, "var")
}
```
#### Update 29 january 2015
- Speed improvement for url Parameters, from ```~ 1500 ns/op ``` to ```~ 1000 ns/op ```.
#### Update 25 december 2014
After trying to find a way of using the default url.Query() for route parameters, i decide to change the way bone is dealing with this. url.Query() is too slow for good router performance.
So now to get the parameters value in your handler, you need to use
` bone.GetValue(req, key) ` instead of ` req.Url.Query().Get(key) `.
This change give a big speed improvement for every kind of application using route parameters, like ~80x faster ...
Really sorry for breaking things, but i think it's worth it.
+46
View File
@@ -0,0 +1,46 @@
# Contributor Covenant Code of Conduct
## Our Pledge
In the interest of fostering an open and welcoming environment, we as contributors and maintainers pledge to making participation in our project and our community a harassment-free experience for everyone, regardless of age, body size, disability, ethnicity, gender identity and expression, level of experience, nationality, personal appearance, race, religion, or sexual identity and orientation.
## Our Standards
Examples of behavior that contributes to creating a positive environment include:
* Using welcoming and inclusive language
* Being respectful of differing viewpoints and experiences
* Gracefully accepting constructive criticism
* Focusing on what is best for the community
* Showing empathy towards other community members
Examples of unacceptable behavior by participants include:
* The use of sexualized language or imagery and unwelcome sexual attention or advances
* Trolling, insulting/derogatory comments, and personal or political attacks
* Public or private harassment
* Publishing others' private information, such as a physical or electronic address, without explicit permission
* Other conduct which could reasonably be considered inappropriate in a professional setting
## Our Responsibilities
Project maintainers are responsible for clarifying the standards of acceptable behavior and are expected to take appropriate and fair corrective action in response to any instances of unacceptable behavior.
Project maintainers have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, or to ban temporarily or permanently any contributor for other behaviors that they deem inappropriate, threatening, offensive, or harmful.
## Scope
This Code of Conduct applies both within project spaces and in public spaces when an individual is representing the project or its community. Examples of representing a project or community include using an official project e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event. Representation of a project may be further defined and clarified by project maintainers.
## Enforcement
Instances of abusive, harassing, or otherwise unacceptable behavior may be reported by contacting the project team at piinky05@gmail.com. The project team will review and investigate all complaints, and will respond in a way that it deems appropriate to the circumstances. The project team is obligated to maintain confidentiality with regard to the reporter of an incident. Further details of specific enforcement policies may be posted separately.
Project maintainers who do not follow or enforce the Code of Conduct in good faith may face temporary or permanent repercussions as determined by other members of the project's leadership.
## Attribution
This Code of Conduct is adapted from the [Contributor Covenant][homepage], version 1.4, available at [http://contributor-covenant.org/version/1/4][version]
[homepage]: http://contributor-covenant.org
[version]: http://contributor-covenant.org/version/1/4/
+8
View File
@@ -0,0 +1,8 @@
## Contributing
1. Fork it
2. Create your feature branch (git checkout -b my-new-feature)
3. Write Tests!
4. Commit your changes (git commit -am 'Add some feature')
5. Push to the branch (git push origin my-new-feature)
6. Create new Pull Request
+22
View File
@@ -0,0 +1,22 @@
The MIT License (MIT)
Copyright (c) 2014 CodingFerret
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, Subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or Substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
+91
View File
@@ -0,0 +1,91 @@
bone [![GoDoc](https://godoc.org/github.com/squiidz/bone?status.png)](http://godoc.org/github.com/go-zoo/bone) [![Build Status](https://travis-ci.org/go-zoo/bone.svg)](https://travis-ci.org/go-zoo/bone) [![Go Report Card](https://goreportcard.com/badge/go-zoo/bone)](https://goreportcard.com/report/go-zoo/bone)
=======
## What is bone ?
Bone is a lightweight and lightning fast HTTP Multiplexer for Golang. It support :
- URL Parameters
- REGEX Parameters
- Wildcard routes
- Router Prefix
- Route params validators
- Sub Router, `mux.SubRoute()`, support most standard router (bone, gorilla/mux, httpRouter etc...)
- Http method declaration
- Support for `http.Handler` and `http.HandlerFunc`
- Custom NotFound handler
- Respect the Go standard `http.Handler` interface
![alt tag](https://c2.staticflickr.com/2/1070/540747396_5542b42cca_z.jpg)
## Speed
```
- BenchmarkBoneMux 10000000 118 ns/op
- BenchmarkZeusMux 100000 144 ns/op
- BenchmarkHttpRouterMux 10000000 134 ns/op
- BenchmarkNetHttpMux 3000000 580 ns/op
- BenchmarkGorillaMux 300000 3333 ns/op
- BenchmarkGorillaPatMux 1000000 1889 ns/op
```
These tests are just for fun, all these routers are great and efficient.
Bone isn't the fastest router for every job.
## Example
``` go
package main
import(
"net/http"
"github.com/go-zoo/bone"
)
func main () {
mux := bone.New()
mux.RegisterValidatorFunc("isNum", func(s string) bool {
if _, err := strconv.Atoi(s); err == nil {
return true
}
return false
})
// mux.Get, Post, etc ... takes http.Handler
// validator for route parameter
mux.Get("/home/:id|isNum", http.HandlerFunc(HomeHandler))
// multiple parameter
mux.Get("/profil/:id/:var", http.HandlerFunc(ProfilHandler))
mux.Post("/data", http.HandlerFunc(DataHandler))
// Support REGEX Route params
mux.Get("/index/#id^[0-9]$", http.HandlerFunc(IndexHandler))
// Handle take http.Handler
mux.Handle("/", http.HandlerFunc(RootHandler))
// GetFunc, PostFunc etc ... takes http.HandlerFunc
mux.GetFunc("/test", Handler)
http.ListenAndServe(":8080", mux)
}
func Handler(rw http.ResponseWriter, req *http.Request) {
// Get the value of the "id" parameters.
val := bone.GetValue(req, "id")
rw.Write([]byte(val))
}
```
## Blog Posts
- http://www.peterbe.com/plog/my-favorite-go-multiplexer
- https://harshladha.xyz/my-first-library-in-go-language-hasty-791b8e2b9e69
## Libs
- Errors dump for Go : [Trash](https://github.com/go-zoo/trash)
- Middleware Chaining module : [Claw](https://github.com/go-zoo/claw)
+91
View File
@@ -0,0 +1,91 @@
/********************************
*** Multiplexer for Go ***
*** Bone is under MIT license ***
*** Code by CodingFerret ***
*** github.com/go-zoo ***
*********************************/
package bone
import (
"net/http"
"strings"
)
// Mux have routes and a notFound handler
// Route: all the registred route
// notFound: 404 handler, default http.NotFound if not provided
type Mux struct {
Routes map[string][]*Route
prefix string
notFound http.Handler
Validators map[string]Validator
Serve func(rw http.ResponseWriter, req *http.Request)
CaseSensitive bool
}
var (
static = "static"
method = []string{"GET", "POST", "PUT", "DELETE", "HEAD", "PATCH", "OPTIONS"}
)
type adapter func(*Mux) *Mux
// New create a pointer to a Mux instance
func New(adapters ...adapter) *Mux {
m := &Mux{Routes: make(map[string][]*Route), Serve: nil, CaseSensitive: true}
for _, adap := range adapters {
adap(m)
}
if m.Serve == nil {
m.Serve = m.DefaultServe
}
return m
}
// RegisterValidatorFunc makes the provided function available to the routes register on that mux as a validator
func (m *Mux) RegisterValidatorFunc(name string, validator func(string) bool) {
if m.Validators == nil {
m.Validators = make(map[string]Validator)
}
m.Validators[name] = newValidatorFunc(validator)
}
// RegisterValidator makes the provided validator available to the routes register on that mux
func (m *Mux) RegisterValidator(name string, validator Validator) {
if m.Validators == nil {
m.Validators = make(map[string]Validator)
}
m.Validators[name] = validator
}
// Prefix set a default prefix for all routes registred on the router
func (m *Mux) Prefix(p string) *Mux {
m.prefix = strings.TrimSuffix(p, "/")
return m
}
// DefaultServe is the default http request handler
func (m *Mux) DefaultServe(rw http.ResponseWriter, req *http.Request) {
// Check if a route match
if !m.parse(rw, req) {
// Check if it's a static ressource
if !m.staticRoute(rw, req) {
// Check if the request path doesn't end with /
if !m.validate(rw, req) {
// Check if same route exists for another HTTP method
if !m.otherMethods(rw, req) {
m.HandleNotFound(rw, req)
}
}
}
}
}
// ServeHTTP pass the request to the serve method of Mux
func (m *Mux) ServeHTTP(rw http.ResponseWriter, req *http.Request) {
if !m.CaseSensitive {
req.URL.Path = strings.ToLower(req.URL.Path)
}
m.Serve(rw, req)
}

Some files were not shown because too many files have changed in this diff Show More