What is Hoverfly?¶
Hoverfly is a lightweight, open source API simulation tool. Using Hoverfly, you can create realistic simulations of the APIs your application depends on.
Replace slow, flaky API dependencies with realistic, re-usable simulations
Simulate network latency, random failures or rate limits to test edge-cases
Extend and customize with any programming language
Export, share, edit and import API simulations
REST API
Lightweight, high-performance, run anywhere
Apache 2 license
Source¶
The Hoverfly source code is available on GitHub.
Hoverfly is developed and maintained by SpectoLabs.
Contents¶
Introduction¶
Motivation¶
Developing and testing interdependent applications is difficult. Maybe you’re working on a mobile application that needs to talk to a legacy API. Or a microservice that relies on two other services that are still in development.
The problem is the same: how do you develop and test against external dependencies which you cannot control?
You could use mocking libraries as substitutes for external dependencies. But mocks are intrusive, and do not allow you to test all the way to the architectural boundary of your application.
Stubbed services are better, but they often involve too much configuration or may not be transparent to your application.
Then there is the problem of managing test data. Often, to write proper tests, you need fine-grained control over the data in your mocks or stubs. Managing test data across large projects with multiple teams introduces bottlenecks that impact delivery times.
Integration testing “over the wire” is problematic too. When stubs or mocks are swapped out for real services (in a continuous integration environment for example) new variables are introduced. Network latency and random outages can cause integration tests to fail unexpectedly.
Hoverfly was designed to provide you with the means to create your own “dependency sandbox”: a simulated development and test environment that you control.
Hoverfly grew out of an effort to build “the smallest service virtualization tool possible”.
Download and installation¶
Hoverfly comes with a command line interface called hoverctl. Archives containing the Hoverfly and hoverctl binaries are available for the major operating systems and architectures.
Download the correct archive, extract the binaries and place them in a directory on your PATH.
Homebrew (MacOS)¶
If you have homebrew, you can install Hoverfly using the brew command.
brew install SpectoLabs/tap/hoverfly
To upgrade your existing hoverfly to the latest release:
brew upgrade hoverfly
To show which versions are installed in your machine:
brew list --versions hoverfly
You can switch to a previously installed version as well:
brew switch hoverfly <version>
To remove old versions :
brew cleanup hoverfly
Docker¶
If you have Docker, you can run Hoverfly using the docker command.
docker run -d -p 8888:8888 -p 8500:8500 spectolabs/hoverfly:latest
This will run the latest version of the Hoverfly Docker image.
You can also pass Hoverfly configuration flags when starting with Docker. For example if you need to run Hoverfly in webserver mode:
docker run -d -p 8888:8888 -p 8500:8500 spectolabs/hoverfly:latest -webserver
This Docker image does not contain hoverctl. Our recommendation is to have hoverctl on your host machine and then configure hoverctl to use the newly started Hoverfly Docker instance as a new target.
See also
For a tutorial of creating a new target in hoverctl, see Controlling a remote Hoverfly instance with hoverctl.
Kubernetes¶
You can use Helm to install Hoverfly directly to your Kubernetes cluster. Hoverfly chart is available from the official Helm incubator repo.
Use helm repo add to add the incubator repo if you haven’t done so:
helm repo add incubator https://kubernetes-charts-incubator.storage.googleapis.com/
Here is the command for a basic Hoverfly installation with the default settings:
helm install incubator/hoverfly
The default installation create a ClusterIP type service, which makes it only reachable from within the cluster.
After the installation, you can use port forwarding to access the application on localhost:
kubectl port-forward $HOVERFLY_POD_NAME 8888 8500
You can find the HOVERFLY_POD_NAME by doing kubectl get pod
See more details on Helm Charts project.
Getting Started¶
Hoverfly is composed of two binaries: Hoverfly and hoverctl.
hoverctl is a command line tool that can be used to configure and control Hoverfly. It allows you to run Hoverfly as a daemon.
Hoverfly is the application that does the bulk of the work. It provides the proxy server or webserver, and the API endpoints.
Once you have extracted both Hoverfly and hoverctl into a directory on your PATH, you can run hoverctl and Hoverfly.
hoverctl version
hoverfly -version
Both of these commands should return a version number. Now you can run an instance of Hoverfly:
hoverctl start
Check whether Hoverfly is running with the following command:
hoverctl logs
The logs should contain the string serving proxy. This indicates that Hoverfly is running.
Finally, stop Hoverfly with:
hoverctl stop
Key Concepts¶
Hoverfly’s functionality is quite broad. You are encouraged to take the time to understand these key concepts before jumping into the Tutorials.
Hoverfly as a proxy server¶
A proxy server passes requests between a client and server.
It is sometimes necessary to use a proxy server to reach a network (as a security measure, for example). Because of this, all network-enabled software can be configured to use a proxy server.
The relationship between clients and servers via a proxy server can be one-to-one, one-to-many, many-to-one, or many-to-many.
By default Hoverfly starts as a proxy server.
Using a proxy server¶
Applications can usually be configured to use a proxy server by setting environment variables:
export HTTP_PROXY="http://proxy-address:port"
export HTTPS_PROXY="https://proxy-address:port"
Launching network-enabled software within an environment containing these variables should make the application use the specified proxy server. The term should is used as not all software respects these environment variables for security reasons.
Alternatively, applications themselves can usually be configured to use a proxy. Curl can be configured to use a proxy via flags.
curl http://hoverfly.io --proxy http://proxy-ip:port
Note
The proxy configuration methods described here are intended to help you use the code examples in this documentation. The method of configuring an application or operating system to use a proxy varies depending on the environment.
The difference between a proxy server and a webserver¶
A proxy server is a type of webserver. The main difference is that when a webserver receives a request from a client, it is expected to respond with whatever the intended response is (an HTML page, for example). The data it responds with is generally expected to reside on that server, or within the same network.
A proxy server is expected to pass the incoming request on to another server (the “destination”). It is also expected to set some appropriate headers along the way, such as X-Forwarded-For, X-Real-IP, X-Forwarded-Proto etc. Once the proxy server receives a response from the destination, it is expected to pass it back to the client.
Hoverfly as a webserver¶
Sometimes you may not be able to configure your client to use a proxy, or you may want to explicitly point your application at Hoverfly. For this reason, Hoverfly can run as a webserver.
Note
When running as a webserver, Hoverfly cannot capture traffic (see Capture mode) - it can only be used to simulate and synthesize APIs (see Simulate mode and Synthesize mode). For this reason, when you use Hoverfly as a webserver, you should have Hoverfly simulations ready to be loaded.
When running as a webserver, Hoverfly strips the domain from the endpoint URL. For example, if you made requests to the following URL while capturing traffic with Hoverfly running as a proxy:
http://echo.jsontest.com/key/value
And Hoverfly is running in simulate mode as a webserver on:
http://localhost:8888
Then the URL you would use to retrieve the data from Hoverfly would be:
http://localhost:8500/key/value
See also
Please refer to the Running Hoverfly as a webserver tutorial for a step-by-step example.
Hoverfly modes¶
Hoverfly has six different modes. It can only run in one mode at any one time.
Capture mode¶
Capture mode is used for creating API simulations.
Note
Hoverfly cannot be set to Capture mode when running as a webserver (see Hoverfly as a webserver).
In Capture mode, Hoverfly (running as a proxy server - see Hoverfly as a proxy server) intercepts communication between the client application and the external service. It transparently records outgoing requests from the client and the incoming responses from the service API.
Most commonly, requests to the external service API are triggered by running automated tests against the application that consumes the API. During subsequent test runs, Hoverfly can be set to run in Simulate mode, removing the dependency on the real external service API. Alternatively, requests can be generated using a manual process.
Usually, Capture mode is used as the starting point in the process of creating an API simulation. Captured data is then exported and modified before being re-imported into Hoverfly for use as a simulation.
By default, Hoverfly will ignore duplicate requests if the request has not changed. This can be a problem when trying to capture a stateful endpoint that may return a different response each time you make a request.
- Using the stateful mode argument when setting Hoverfly to capture mode will disable the duplicate request check,
enabling you to capture sequences of responses and play them back in Simulate mode in order.
See also
This functionality is best understood via a practical example: see Capturing a stateful sequence of responses in the Tutorials section.
Simulate mode¶
In this mode, Hoverfly uses its simulation data in order to simulate external APIs. Each time Hoverfly receives a request, rather than forwarding it on to the real API, it will respond instead. No network traffic will ever reach the real external API.
The simulation can be produced automatically via by running Hoverfly in Capture mode, or created manually. See Simulations for information.
Spy mode¶
In this mode, Hoverfly simulates external APIs if a request match is found in simulation data (See Simulate mode), otherwise, the request will be passed through to the real API.
Synthesize mode¶
This mode is similar to Simulate mode, but instead of looking for a response in stored simulation data, the request is passed directly to a user-supplied executable file. These files are known as Middleware.
In Synthesize mode, the middleware executable is expected to generate a response to the incoming request “on the fly”. Hoverfly will then return the generated response to the application. For Hoverfly to operate in Synthesize mode, a middleware executable must be specified.
Note
You might use this mode to simulate an API that may be too difficult to record correctly via Capture mode. An example would be an API that uses state to change the responses. You could create middleware that manages this state and produces the desired response based on the data in the request.
Modify mode¶
Note
Hoverfly cannot be set to Modify mode when running as a webserver (see Hoverfly as a webserver).
Modify mode is similar to Capture mode, except it does not save the requests and responses. In Modify mode, Hoverfly will pass each request to a Middleware executable before forwarding it to the destination. Responses will also be passed to middleware before being returned to the client.
You could use this mode to “man in the middle” your own requests and responses. For example, you could change the API key you are using to authenticate against a third-party API.
Diff mode¶
In this mode, Hoverfly forwards a request to an external service and compares a response with currently stored simulation. With both the stored simulation response and the real response from the external service, Hoverfly is able to detect differences between the two. When Hoverfly has finished comparing the two responses, the difference is stored and the incoming request is served the real response from the external service.
The differences can be retrieved from Hoverfly using the API (GET /api/v2/diff). The response contains a list of differences, containing the request and the differences of the response.
{
"diff": [{
"request": {
"method": "GET",
"host": "time.jsontest.com",
"path": "/",
"query": ""
},
"diffReports": [{
"timestamp": "2018-03-16T17:45:40Z",
"diffEntries": [{
"field": "header/X-Cloud-Trace-Context",
"expected": "[ec6c455330b682c3038ba365ade6652a]",
"actual": "[043c9bb2eafa1974bc09af654ef15dc3]"
}, {
"field": "header/Date",
"expected": "[Fri, 16 Mar 2018 17:45:34 GMT]",
"actual": "[Fri, 16 Mar 2018 17:45:41 GMT]"
}, {
"field": "body/time",
"expected": "05:45:34 PM",
"actual": "05:45:41 PM"
}, {
"field": "body/milliseconds_since_epoch",
"expected": "1.521222334104e+12",
"actual": "1.521222341017e+12"
}]
}]
}]
}
This data is stored and kept until the Hoverfly instance is stopped or the the storage is cleaned by calling the API (DELETE /api/v2/diff).
See also
For more information on the API to retrieve differences, see REST API.
Simulations¶
The core functionality of Hoverfly is to capture HTTP(S) traffic to create API simulations which can be used in testing. Hoverfly stores captured traffic as simulations.
Simulation JSON can be exported, edited and imported in and out of Hoverfly, and can be shared among Hoverfly users or instances. Simulation JSON files must adhere to the Hoverfly Simulation schema.
Simulations consist of Request Matchers and Responses, Delays and Metadata (“Meta”).
Request Responses Pairs¶
Hoverfly simulates APIs by matching incoming requests from the client to stored requests. Stored requests have an associated stored response which is returned to the client if the match is successful.
The matching logic that Hoverfly uses to compare incoming requests with stored requests can be configured using Request Matchers.
Request Matchers¶
When Hoverfly captures a request, it creates a Request Matcher for each field in the request. A Request Matcher consists of the request field name, the type of match which will be used to compare the field in the incoming request to the field in the stored request, and the request field value.
By default, Hoverfly will set the type of match to exact for each field.
See also
There are many types of Request Matcher. Please refer to Request matchers for a list of the types available, and examples of how to use them.
There alse are two different matching strategies: strongest match (default) and first match (legacy). Please refer to Matching strategies for more information.
An example Request Matcher Set might look like this:
Field |
Matcher Type |
Value |
|---|---|---|
scheme |
exact |
“https” |
method |
exact |
“GET” |
destination |
exact |
“docs.hoverfly.io” |
path |
exact |
“/pages/keyconcepts/templates.html” |
query |
exact |
“query=true” |
body |
exact |
“” |
headers |
exact |
In the Hoverfly simulation JSON file, this Request Matcher Set would be represented like this:
1 "request": {
2 "path": [
3 {
4 "matcher": "exact",
5 "value": "/pages/keyconcepts/templates.html"
6 }
7 ],
8 "method": [
9 {
10 "matcher": "exact",
11 "value": "GET"
12 }
13 ],
14 "destination": [
15 {
16 "matcher": "exact",
17 "value": "docs.hoverfly.io"
18 }
19 ],
20 "scheme": [
21 {
22 "matcher": "exact",
23 "value": "http"
24 }
25 ],
26 "body": [
27 {
28 "matcher": "exact",
29 "value": ""
30 }
31 ],
32 "query": {
33 "query": [
34 {
35 "matcher": "exact",
36 "value": "true"
37 }
38 ]
39 }
40 },
The matching logic that Hoverfly uses to compare an incoming request to a stored request can be changed by editing the Request Matchers in the simulation JSON file.
It is not necessary to have a Request Matcher for every request field. By omitting Request Matchers, it is possible to implement partial matching - meaning that Hoverfly will return one stored response for multiple incoming requests.
For example, this Request Matcher will match any incoming request to the docs.hoverfly.io destination:
1 "destination": [
2 {
3 "matcher": "exact",
4 "value": "docs.hoverfly.io"
5 }
6 ]
In the example below, the globMatch Request Matcher type is used to match any subdomain of hoverfly.io:
1 "destination": [
2 {
3 "matcher": "glob",
4 "value": "*.hoverfly.io"
5 }
6 ]
It is also possible to use more than one Request Matcher for each field.
In the example below, a regexMatch and a globMatch are used on the destination field.
This will match on any subdomain of hoverfly.io which begins with the letter d. This means that
incoming requests to docs.hoverfly.io and dogs.hoverfly.io will be matched, but requests to
cats.hoverfly.io will not be matched.
1 "destination": [
2 {
3 "matcher": "glob",
4 "value": "*.hoverfly.io"
5 },
6 {
7 "matcher": "regex",
8 "value": "(\\Ad)"
9 }
10 ]
See also
There are many types of Request Matcher. Please refer to Request matchers for a list of the types available, and examples of how to use them.
For a practical example of how to use a Request Matcher, please refer to Loose request matching using a Request Matcher in the tutorials section.
There alse are two different matching strategies: strongest match (default) and first match (legacy). Please refer to Matching strategies for more information.
Responses¶
Each Request Matcher Set has a response associated with is. If the request match is successful, Hoverfly will return the response to the client.
1 "response": {
2 "status": 200,
3 "body": "Response from docs.hoverfly.io/pages/keyconcepts/templates.html",
4 "encodedBody": false,
5 "headers": {
6 "Hoverfly": [
7 "Was-Here"
8 ]
9 },
10 "templated": false
11 }
Editing the fields in response, combined with editing the Request Matcher set, makes it possible to configure complex request/response logic.
JSON is a text-based file format so it has no intrinsic support for binary data. Therefore if Hoverfly a response body contains binary data (images, gzipped, etc), the response body will be base64 encoded and the encodedBody field set to true.
1 "body": "YmFzZTY0IGVuY29kZWQ=",
2 "encodedBody": true,
Starting from version 1.3.0 hoverfly can return response body from a specific file:
"response": {
"status": 200,
"encodedBody": false,
"templated": false,
"bodyFile": "responses/200-success.json"
}
response/200-success.json is resolved against the directory specified in -response-body-files-path
which is your current working directory by default.
When both body and bodyFile are specified, body takes precedence.
bodyFile is read into memory only on simulation import, not in runtime.
Reading response bodies sometimes might not be comfortable. Imagine a developer team that needs a single updating “files provider”. Syncing to this provider can be a challenge and that’s why hoverfly supports downloading response bodies from external urls:
"response": {
"status": 200,
"encodedBody": false,
"templated": false,
"bodyFile": "https://raw.githubusercontent.com/SpectoLabs/hoverfly/master/core/handlers/v2/schema.json"
}
Like local files, this feature is supported only on simulation import. To escape security issues there’s another
mandatory option to specify -response-body-files-allow-origin that lets you explicitly set the collections of
urls to allow body files to be downloaded from. For the above to work you need to run hoverfly like this:
hoverfly -response-body-files-allow-origin="https://raw.githubusercontent.com/"
Delays¶
Once you have created a simulated service by capturing traffic between your application and an external service, you may wish to make the simulation more “realistic” by applying latency to the responses returned by Hoverfly.
Hoverfly can be configured to apply delays to responses based on URL pattern matching or HTTP method. This is done using a regular expression to match against the URL, a delay value in milliseconds, and an optional HTTP method value.
See also
This functionality is best understood via a practical example: see Adding delays to a simulation in the Tutorials section.
You can use Log-normal distributions of delay for realistic delay.
You can also apply delays to simulations using Middleware (see the Using middleware to simulate network latency tutorial). Using middleware to apply delays sacrifices performance for flexibility.
Meta¶
The last part of the simulation schema is the meta object. Its purpose is to store metadata that is relevant to your simulation. This includes the simulation schema version, the version of Hoverfly used to export the simulation and the date and time at which the simulation was exported.
1 },
2 "meta": {
3 "schemaVersion": "v5.2",
4 "hoverflyVersion": "v1.2.0",
5 "timeExported": "2020-04-25T17:56:32+03:00"
See also
For a hands-on tutorial of creating and editing simulations, see Creating and exporting a simulation.
Matching strategies¶
Hoverfly has two matching strategies. Each has advantages and trade-offs.
Note
In order to fully understand Hoverfly’s matching strategies, it is recommended that you read the Simulations section first.
Strongest Match¶
This is the default matching strategy for Hoverfly. If Hoverfly finds multiple Request Response Pairs that match an incoming request, it will return the Response from the pair which has the highest matching score.
To set “strongest” as the matching strategy, simply run:
hoverctl mode simulate
Or to be explicit run:
hoverctl mode simulate --matching-strategy=strongest
Matching scores¶
This example shows how matching scores are calculated.
Let’s assume Hoverfly is running in simulate mode, and the simulation data contains four Request Responses Pairs. Each Request Response Pair contains one or more Request Matchers.
Hoverfly then receives a GET request to the destination www.destination.com. The incoming request contains the following
fields.
Request
Field |
Value |
|---|---|
method |
GET |
destination |
www.destination.com |
Request Response Pair 1
Field |
Matcher Type |
Value |
Score |
Total Score |
Matched? |
|---|---|---|---|---|---|
method |
exact |
DELETE |
+0 |
1 |
false |
destination |
exact |
www.destination.com |
+1 |
This pair contains two Request Matchers. The method value in the incoming request (GET) does not match
the value for the method matcher (DELETE). However the destination value does match.
This gives the Request Response Pair a total score of 1, but since one match failed, it
is treated as unmatched (Matched? = false).
Request Response Pair 2
Field |
Matcher Type |
Value |
Score |
Total Score |
Matched? |
|---|---|---|---|---|---|
method |
exact |
GET |
+1 |
1 |
true |
This pair contains one Request Matcher. The method value in the incoming request (GET) matches
the value for the method matcher. This gives the pair a total score of 1, and since no matches
failed, it is treated as matched.
Request Response Pair 3
Field |
Matcher Type |
Value |
Score |
Total Score |
Matched? |
|---|---|---|---|---|---|
method |
exact |
GET |
+1 |
2 |
true |
destination |
exact |
www.destination.com |
+1 |
In this pair, the method and destination values in the incoming request both match the corresponding Request Matcher values. This gives the pair a total score of 2, and it treated as matched.
Request Response Pair 4
Field |
Matcher Type |
Value |
Score |
Total Score |
Matched? |
|---|---|---|---|---|---|
method |
exact |
GET |
+1 |
1 |
false |
destination |
exact |
www.miss.com |
+0 |
This pair is treated as unmatched because the destination matcher failed.
Request Response Pair 3 has the highest score, and is therefore the strongest match.
This means that Hoverfly will return the Response contained within Request Response Pair 3.
Note
When there are multiple matches all with the same score, Hoverfly will pick the last one in the simulation.
The strongest match strategy makes it much easier to identify why Hoverfly has not returned a Response to an incoming Request. If Hoverfly is not able to match an incoming Request to a Request Response Pair, it will return the closest match. For more information see Troubleshooting.
However, the additional logic required to calculate matching scores does affect Hoverfly’s performance.
First Match¶
First match is the alternative (legacy) mechanism of matching. There is no scoring, and Hoverfly simply returns the first match it finds in the simulation data.
To set first match as the matching strategy, run:
hoverctl mode simulate --matching-strategy=first
The main advantage of this strategy is performance - although it makes debugging matching errors harder.
Caching¶
In Simulate mode, Hoverfly uses caching in order to retain strong performance characteristics, even in the event of complex matching. The cache is a key-value store of request hashes (hash of all the request excluding headers) to responses.
Caching matches¶
When Hoverfly receives a request in simulate mode, it will first hash it and then look for it in the cache. If a cache entry is found, it will send the cached response back to Hoverfly. If it is not found, it will look for a match in the list of matchers. Whenever a new match is found, it will be added to the cache.
Caches misses¶
Hoverfly also caches misses. This means that repeating a request which was not matched will return a cached miss, avoiding the need to perform matching. The closest miss is also cached, so Hoverfly will not lose any useful information about which matchers came closest.
Header caching¶
Currently, headers are not included in the hash for a request. This is because headers tend to change across time and clients, impacting ordinary and eager caching respectively.
Eager caching¶
The cache is automatically pre-populated whenever switching to simulate mode. This only works on certain matchers (such as matchers where every field is an “exactMatch”), but it means the initial cache population does not happen during simulate mode.
Cache invalidation¶
Cache invalidation is a straightforward process in Hoverfly. It only occurs when a simulation is modified.
Templating¶
Hoverfly can build responses dynamically through templating. This is particularly useful when combined with loose matching, as it allows a single matcher to represent an unlimited combination of responses.
Enabling Templating¶
By default templating is disabled. In order to enable it, set the templated field to true in the response of a simulation.
Getting data from the request¶
Currently, you can get the following data from request to the response via templating:
Field |
Example |
Request |
Result |
|---|---|---|---|
Request scheme |
|
http |
|
Query parameter value |
|
bar |
|
Query parameter value (list) |
|
bar2 |
|
Path parameter value |
|
one |
|
Method |
|
GET |
|
Host |
|
www.foo.com |
|
jsonpath on body |
|
{ “id”: 123, “username”: “hoverfly” } |
123 |
xpath on body |
|
<root><id>123</id></root> |
123 |
From data |
|
||
Header value |
|
{ “X-Header-Id”: [“bar”] } |
bar |
Header value (list) |
|
{ “X-Header-Id”: [“bar1”, “bar2”] } |
bar2 |
State |
|
State Store = {“basket”:”eggs”} |
eggs |
Helper Methods¶
Additional data can come from helper methods. These are the ones Hoverfly currently support:
Description |
Example |
Result |
|---|---|---|
The current date time with offset, in the given format. For example:
|
|
|
A random string |
|
hGfclKjnmwcCds |
A random string with a specified length |
|
KC |
A random boolean |
|
true |
A random integer |
|
42 |
A random integer within a range |
|
7 |
A random float |
|
42 |
A random float within a range |
|
7.4563213423 |
A random email address |
|
|
A random IPv4 address |
|
224.36.27.8 |
A random IPv6 address |
|
41d7:daa0:6e97:6fce:411e:681:f86f:e557 |
A random UUID |
|
7b791f3d-d7f4-4635-8ea1-99568d821562 |
Replace all occurrences of the old value with the new value in the target string |
(where Request.Body has the value of
|
to mock or not to mock |
Generate random data using go-fakeit |
|
John Smith |
Query CSV data source where ID = 3 and return its name |
|
John Smith |
|
|
John Smith |
Time offset¶
When using template helper method now, time offset must be formatted using the following syntax.
Shorthand |
Type |
|---|---|
ns |
Nanosecond |
us/µs |
Microsecond |
ms |
Millisecond |
s |
Second |
m |
Minute |
h |
Hour |
d |
Day |
y |
Year |
Prefix an offset with - to subtract the duration from the current date time.
Example time offset¶
5m |
5 minutes |
1h30m |
1 hour 5 minutes |
1y10d |
1 year 10 days |
Date time formats¶
When using template helper method now, date time formats must follow the Golang syntax.
More can be found out here https://golang.org/pkg/time/#Parse
Example date time formats¶
2006-01-02T15:04:05Z07:00 |
Mon, 02 Jan 2006 15:04:05 MST |
Jan _2 15:04:05 |
Note
If you leave the format string empty, the default format to be used is ISO 8601 (2006-01-02T15:04:05Z07:00).
You can also get an UNIX timestamp by setting the format to:
unix: UNIX timestamp in secondsepoch: UNIX timestamp in milliseconds
Faker¶
Support for go-fakeit was added in order to extend the
templating capabilities of Hoverfly. Faker covers many different test data requirements and it can be used within
Hoverfly templated responses by using the faker helper followed by the faker type (e.g. Name, Email)
For example, you can generate a random name using the following expression:
{
"body": "{\"name\": \"{{faker 'Name'}}\"}"
}
Fakers that require arguments are currently not supported.
Templating Data Source¶
User can upload CSV data file using hoverfly/hoverctl CLI or Admin API that can be queried via templating function.
{
"body": "{\"name\": \"{{csv '(data-source-name)' '(column-name)' '(query-value)' '(selected-column)' }}\"}"
}
Note
Data source name is case sensitive whereas other parameters in this function are case insensitive. Secondly, you can refer hoverfly/hoverctl options or Admin API docs in order to upload CSV data source to running hoverfly instance.
Example: Start hoverfly with templating CSV datasource(student-marks.csv) provided below.
hoverfly -templating-data-source "student-marks <path to below CSV file>"
ID |
Name | Marks |
1 |
Test1 | 55 |
2 |
Test2 | 65 |
3 |
Test3 | 98 |
4 |
Test4 | 23 |
5 |
Test5 | 15 |
NA | 0 |
Description |
Example |
Result |
|---|---|---|
Search where ID = 3 and return name |
csv ‘student-marks’ ‘Id’ ‘3’ ‘Name’ |
Test3 |
Search where ID = 4 and return its marks |
csv ‘student-marks’ ‘Id’ ‘4’ ‘Marks’ |
Test23 |
Search where Name = Test1 and return marks |
csv ‘student-marks’ ‘Name’ ‘Test1’ ‘Marks’ |
55 |
Search where Id is not match and return marks (in this scenario, it matches wildcard * and returns) |
csv ‘student-marks’ ‘Id’ ‘Test100’ ‘Marks’ |
0 |
Search where Id = first path param and return marks URL looks like - http://test.com/students/5/marks |
csv ‘student-marks’ ‘Id’ ‘Request.Path.[0]’ ‘Marks’ |
15 |
Journal¶
Journal Entry can be queried using its index and its extracted value.
Syntax
journal "index name" "extracted value" "request/response" "xpath/jsonpath" "lookup query"
index name should be the same key expression you have specified when you enable the journal index.
extracted value is for doing a key lookup for the journal entry from that index.
request/response specifies if you want to get data from the request or response.
xpath/jsonpath specifies whether you want to extract it using xpath or json path expression.
lookup query is either jsonpath or xpath expressions to parse the request/response data.
Example:
{
"body": "{\"name\": \"{{ journal 'Request.QueryParam.id' '1' 'response' 'jsonpath' '$.name' }}\"}"
}
In the above example, we are querying the name from JSON response in the journal entry where index Request.QueryParam.id has a key value of 1.
Conditional Templating, Looping and More¶
Hoverfly uses the https://github.com/aymerick/raymond library for templating, which is based on http://handlebarsjs.com/
To learn about more advanced templating functionality, such as looping and conditionals, read the documentation for these projects.
Global Literals and Variables¶
You can define global literals and variables for templated response. This comes in handy when you have a lot of templated responses that share the same constant values or helper methods.
Literals¶
Literals are constant values. You can declare literals as follows and then reference it in templated response as {{ Literals.<literal name> }}.
{
"data": {
...
"literals": [
{
"name":"literal1",
"value":"value1"
},
{
"name":"literal2",
"value":["value1", "value2", "value3"]
},
{
"name":"literal3",
"value": {
"key": "value"
}
}
]
}
Variables¶
Variable lets you define a helper method that can be shared among templated responses.
You can associate the helper method with a name and then reference it in templated response as {{ Vars.<variable name> }}.
{
"data": {
...
"variables": [
{
"name":"<variable name>",
"function":"<helper method name>",
"arguments":["arg1", "arg2"]
}
]
}
{
"data": {
...
"variables": [
{
"name":"varOne",
"function":"faker",
"arguments":["Name"]
},
{
"name":"idFromJSONRequestBody",
"function":"requestBody",
"arguments":["jsonpath", "$.id"]
},
{
"name":"idFromXMLRequestBody",
"function":"requestBody",
"arguments":["xpath", "/root/id"]
}
]
}
State¶
Hoverfly contains a map of keys and values which it uses to store it’s internal state. Some Request matchers can be made to only match when Hoverfly is in a certain state, and other matchers can be set to mutate Hoverfly’s state.
Setting State when Performing a Match¶
A response includes two fields, transitionsState and removesState which alter Hoverflies internal state during a match:
"request": {
"path": [
{
"matcher": "exact",
"value": "/pay"
}
]
},
"response": {
"status": 200,
"body": "eggs and large bacon",
"transitionsState" : {
"payment-flow" : "complete"
},
"removesState" : [
"basket"
]
}
In the above case, the following changes to Hoverflies internal state would be made on a match:
Current State of Hoverfly |
New State of Hoverfly? |
reason |
|---|---|---|
payment-flow=pending,basket=full |
payment-flow=complete |
Payment value transitions, basket deleted by key |
basket=full |
payment-flow=complete |
Payment value created, basket deleted by key |
payment-flow=complete |
Payment value created, basket already absent |
Requiring State in order to Match¶
A matcher can include a field requiresState, which dictates the state Hoverfly must be in for there to be a match:
"request": {
"path": [
{
"matcher": "exact",
"value": "/basket"
}
]
"requiresState": {
"eggs": "present",
"bacon" : "large"
}
},
"response": {
"status": 200,
"body": "eggs and large bacon"
}
In the above case, the following matches results would occur when making a request to /basket:
Current State of Hoverfly |
matches? |
reason |
|---|---|---|
eggs=present,bacon=large |
true |
Required and current state are equal |
eggs=present,bacon=large,f=x |
true |
Additional state ‘f=x’ is not used by this matcher |
eggs=present |
false |
Bacon is missing |
eggs=present,bacon=small |
false |
Bacon is has the wrong value |
Managing state via Hoverctl¶
It could be tricky to reason about the current state of Hoverfly, or to get Hoverfly in a state that you desire for testing. This is why Hoverctl comes with commands that let you orchestrate it’s state. Some useful commands are:
$ hoverctl state --help
$ hoverctl state get-all
$ hoverctl state get key
$ hoverctl state set key value
$ hoverctl state delete-all
Sequences¶
Using state, it is possible to recreate a sequence of different responses that may come back given a single request. This can be useful when trying to test stateful endpoints.
When defining state for request response pairs, if you prefix your state key with the string sequence:, Hoverfly
will acknowledge the pair as being part of a stateful sequence. When simulating this sequence, Hoverfly will keep track
of the user’s position in the sequence and move them forwards.
Once Hoverfly has reached the end of the sequence, it will continue to return the final response.
{
"data": {
"pairs": [{
"request": {
"requiresState": {
"sequence:1": "1"
}
},
"response": {
"status": 200,
"body": "First response",
"transitionsState": {
"sequence:1": "2"
}
}
},
{
"request": {
"requiresState": {
"sequence:1": "2"
}
},
"response": {
"status": 200,
"body": "Second response"
}
}
]
},
"meta": {
"schemaVersion": "v5.2"
}
}
Destination filtering¶
By default, Hoverfly will process every request it receives. However, you may wish to control which URLs Hoverfly processes.
This is done by filtering the destination URLs using either a string or a regular expression. The destination string or regular expression will be compared against the host and the path of a URL.
For example, specifying hoverfly.io as the destination value will tell Hoverfly to process only URLs on the hoverfly.io host.
hoverctl destination "hoverfly.io"
Specifying api as the destination value during Capture mode will tell Hoverfly to capture only URLs that contain the string api. This would include both api.hoverfly.io/endpoint and hoverfly.io/api/endpoint.
See also
This functionality is best understood via a practical example: see Capturing or simulating specific URLs in the Tutorials section.
Note
The destination setting applies to all Hoverfly modes. If a destination value is set while Hoverfly is running in Simulate mode, requests that are excluded by the destination setting will be passed through to the real URLs. This makes it possible to return both real and simulated responses.
Middleware¶
Middleware intercepts traffic between the client and the API (whether real or simulated), and allowing you to manipulate it.
You can use middleware to manipulate data in simulated responses, or to inject unpredictable performance characteristics into your simulation.
Middleware works differently depending on the Hoverfly mode.
Capture mode: middleware affects only outgoing requests
Simulate/Spy mode: middleware affects only responses (cache contents remain untouched)
Synthesize mode: middleware creates responses
Modify mode: middleware affects requests and responses
Note
Middleware is applied after rendering the templating functions (see Templating) in the response body.
You can write middleware in any language. There are two different types of middleware.
Local Middleware¶
Hoverfly has the ability to invoke middleware by executing a script or binary file on a host operating system. The only requires are that the provided middleware can be executed and sends the Middleware JSON schema to stdout when the Middleware JSON schema is received on stdin.
HTTP Middleware¶
Hoverfly can also send middleware requests to a HTTP server instead of running a process locally. The benefits of this are that Hoverfly does not initiate the process, giving more control to the user. The only requirements are that Hoverfly can POST the Middleware JSON schema to middleware URL provided and the middleware HTTP server responses with a 200 and the Middleware JSON schema is in the response.
Middleware Interface¶
When middleware is called by Hoverfly, it expects to receive and return JSON (see Simulation schema). Middleware can be used to modify the values in the JSON but must not modify the schema itself.
Hoverfly will send the JSON object to middleware via the standard input stream. Hoverfly will then listen to the standard output stream and wait for the JSON object to be returned.
See also
Middleware examples are covered in the tutorials section. See Using middleware to simulate network latency and Using middleware to modify response payload and status code.
Post Serve Action¶
Overview¶
PostServeAction allows you to execute custom code after a response has been served in simulate or spy mode.
It is custom script that can be written in any language. Hoverfly has the ability to invoke a script or binary file on a host operating system. Custom code is execute after a provided delay(in ms) once simulated response is served.
We can register multiple post serve actions.
In order to register post serve action, it takes mainly four parameters - binary to invoke script, script content/location, delay(in ms) post which it will be executed and name of that action.
Ways to register a Post Serve Action¶
At time of startup by passing single/multiple -post-serve-action flag(s) as mentioned in the hoverfly command page.
Via PUT API to register new post serve action as mentioned in the API page.
Using command line hoverctl post-serve-action set command as mentioned in the hoverctl command page.
Once post serve action is registered, we can trigger particular post serve action by putting it in response part of request-response pair in simulation JSON.
Example Simulation JSON
{
"response": {
"postServeAction": "<name of post serve action we want to invoke>"
}
}
Hoverctl¶
Hoverctl is a command line utility that is shipped with Hoverfly. Its purpose is to make it easier to interact with Hoverfly APIs and your local filesystem, while also providing a way to start and stop instances of Hoverfly.
Hoverctl also has the ability to work with multiple instances of Hoverfly, through the use of the target option. Configuration is stored against each target meaning it is possible to start an instance of Hoverfly locally and remotely and still be able to interact with each separately.
See also
Please refer to hoverctl commands for more information about hoverctl.
See also
Please refer to the Controlling a remote Hoverfly instance with hoverctl tutorial for a step by step example of using targets.
Native language bindings¶
Native language bindings are available for Hoverfly to make it easy to integrate into different environments.
HoverPy¶
To get started:
sudo pip install hoverpy
python
And in Python you can simply get started with:
import hoverpy
import requests
# capture mode
with hoverpy.HoverPy(capture=True) as hp:
data = requests.get("http://time.jsontest.com/").json()
# simulation mode
with hoverpy.HoverPy() as hp:
simData = requests.get("http://time.jsontest.com/").json()
print(simData)
assert(data["milliseconds_since_epoch"] == simData["milliseconds_since_epoch"])
For more information, read the HoverPy documentation.
Hoverfly Java¶
Strict or loose HTTP request matching based on URL, method, body and header combinations
Fluent and expressive DSL for easy generation of simulated APIs
Automatic marshalling of objects into JSON during request/response body generation
HTTPS automatically supported, no extra configuration required
Download via Maven or Gradle
To get started, read the Hoverfly Java documentation.
Tutorials¶
In these examples, we will use the hoverctl (CLI tool for Hoverfly) to interact with Hoverfly.
Hoverfly can also be controlled via its REST API, or via Native language bindings.
Basic tutorials¶
Creating and exporting a simulation¶
Note
If you are running Hoverfly on a machine that accesses the internet via a proxy (for example if you are on a corporate network), please follow the Using Hoverfly behind a proxy tutorial before proceeding.
Start Hoverfly and set it to Capture mode
hoverctl start
hoverctl mode capture
Make a request with cURL, using Hoverfly as a proxy server:
curl --proxy http://localhost:8500 http://time.jsontest.com
View the Hoverfly logs
hoverctl logs
Export the simulation to a JSON file
hoverctl export simulation.json
Stop hoverfly
hoverctl stop
You’ll now see a simulation.json file in your current working directory, which contains all your simulation data.
In case you are curious, the sequence diagram for this process looks like this:
Note
By default, request headers are not captured. If you want to capture headers, you will need to specify them when setting capture mode.
hoverctl mode capture --headers "User-Agent,Content-Type,Authorization"
hoverctl mode capture --all-headers
Note
It is possible to filter and export simulation to separate JSON files by providing a plain text or regex string to the --url-pattern flag:
hoverctl export echo.json --url-pattern "echo.jsontest.com" // export simulations for echo.jsontest.com only
hoverctl export api.json --url-pattern "(.+).jsontest.com" // export simulations for all jsontest.com subdomains
Importing and using a simulation¶
In this tutorial we are going to import the simulation we created in the previous tutorial.
hoverctl start
hoverctl import simulation.json
Hoverfly can also import simulation data that is stored on a remote host via HTTP:
hoverctl import https://example.com/example.json
Make a request with cURL, using Hoverfly as a proxy.
curl --proxy localhost:8500 http://time.jsontest.com
This outputs the time at the time the request was captured.
{
"time": "02:07:28 PM",
"milliseconds_since_epoch": 1482242848562,
"date": "12-20-2016"
}
Stop Hoverfly:
hoverctl stop
Note
Importing multiple simulations:
The above example shows importing one simulation into Hoverfly. You can also import multiple simulations:
hoverctl simulation add foo.json bar.json
You can specify one or more simulations when starting Hoverfly using hoverfly or hoverctl commands:
hoverfly -import foo.json -import bar.json
hoverctl start --import foo.json --import bar.json
Hoverfly appends any unique pair to the existing simulation by comparing the equality of the request JSON objects. If a conflict occurs, the pair is not added.
Adding delays to a simulation¶
Simulating API latency during development allows you to write code that will deal with it gracefully.
In Hoverfly, this is done by applying “delays” or “delaysLogNormal” to responses in a simulation.
Delays are applied by editing the Hoverfly simulation JSON file. Delays can be applied selectively according to request URL pattern and/or HTTP method.
Applying a delay to all responses¶
Let’s apply a 2 second delay to all responses. First, we need to create and export a simulation.
hoverctl start
hoverctl mode capture
curl --proxy localhost:8500 http://time.jsontest.com
curl --proxy localhost:8500 http://date.jsontest.com
hoverctl export simulation.json
hoverctl stop
Take a look at the "globalActions" property within the simulation.json file you exported. It
should look like this:
1 "globalActions": {
2 "delays": [],
3 "delaysLogNormal": []
Edit the file so the "globalActions" property looks like this:
1 "globalActions": {
2 "delays": [
3 {
4 "urlPattern": ".",
5 "httpMethod": "",
6 "delay": 2000
7 }
8 ],
9 "delaysLogNormal": []
Hoverfly will apply a delay of 2000ms to all URLs that match the "urlPattern" value. We want
the delay to be applied to all URLs, so we set the "urlPattern" value to the regular expression ".".
Now import the edited simulation.json file, switch Hoverfly to Simulate mode and make the requests
again.
hoverctl start
hoverctl import simulation.json
curl --proxy localhost:8500 http://time.jsontest.com
curl --proxy localhost:8500 http://date.jsontest.com
hoverctl stop
The responses to both requests are delayed by 2 seconds.
Applying a delay to specific responses¶
When you want to add a delay to specific responses and global host / uri / http method matching is not the case, you can specify fixedDelay in response. Here we apply a delay of 3000ms only to request /api/profile with X-API-Version: v1 header:
{
"request": {
"path": [
{"matcher": "exact", "value": "/api/profile"}
],
"headers": {
"X-API-Version": [
{"matcher": "exact", "value": "v1"}
]
}
},
"response": {
"status": 404,
"body": "Page not found",
"fixedDelay": 3000
}
}
It’s also possible to apply a logNormal delay:
{
"request": {
"path": [
{"matcher": "exact", "value": "/api/profile"}
],
"headers": {
"X-API-Version": [
{"matcher": "exact", "value": "v1"}
]
}
},
"response": {
"status": 404,
"body": "Page not found",
"logNormalDelay": {
"min": 100,
"max": 10000,
"mean": 5000,
"median": 500
}
}
}
Like global delays, when both fixedDelay and logNormalDelay are provided they are applied one after another.
Applying different delays based on host¶
Now let’s apply a delay of 1 second on responses from time.jsontest.com and a delay of 2 seconds on responses from date.jsontest.com.
Run the following to create and export a simulation.
hoverctl start
hoverctl mode capture
curl --proxy localhost:8500 http://time.jsontest.com
curl --proxy localhost:8500 http://date.jsontest.com
hoverctl export simulation.json
hoverctl stop
Edit the simulation.json file so that the "globalActions" property looks like this:
1 "globalActions": {
2 "delays": [
3 {
4 "urlPattern": "time\\.jsontest\\.com",
5 "httpMethod": "",
6 "delay": 1000
7 },
8 {
9 "urlPattern": "date\\.jsontest\\.com",
10 "httpMethod": "",
11 "delay": 2000
12 }
13 ],
14 "delaysLogNormal": []
Now run the following to import the edited simulation.json file and run the simulation:
hoverctl start
hoverctl import simulation.json
curl --proxy localhost:8500 http://time.jsontest.com
curl --proxy localhost:8500 http://date.jsontest.com
hoverctl stop
You should notice a 1 second delay on responses from time.jsontest.com, and a 2 second delay on responses from date.jsontest.com.
Note
You can easily get into a situation where your request URL has multiple matches. In this case, the first successful match wins.
Applying different delays based on URI¶
Now let’s apply different delays based on location. Run the following to create and export a simulation.
hoverctl start
hoverctl mode capture
curl --proxy localhost:8500 http://echo.jsontest.com/a/b
curl --proxy localhost:8500 http://echo.jsontest.com/b/c
curl --proxy localhost:8500 http://echo.jsontest.com/c/d
hoverctl export simulation.json
hoverctl stop
Edit the simulation.json file so that the "globalActions" property looks like this:
1 "globalActions": {
2 "delays": [
3 {
4 "urlPattern": "echo\\.jsontest\\.com\\/a\\/b",
5 "httpMethod": "",
6 "delay": 2000
7 },
8 {
9 "urlPattern": "echo\\.jsontest\\.com\\/b\\/c",
10 "httpMethod": "",
11 "delay": 2000
12 },
13 {
14 "urlPattern": "echo\\.jsontest\\.com\\/c\\/d",
15 "httpMethod": "",
16 "delay": 3000
17 }
18 ],
19 "delaysLogNormal": []
Now run the following to import the edited simulation.json file and run the simulation:
hoverctl start
hoverctl import simulation.json
hoverctl mode simulate
curl --proxy localhost:8500 http://echo.jsontest.com/a/b
curl --proxy localhost:8500 http://echo.jsontest.com/b/c
curl --proxy localhost:8500 http://echo.jsontest.com/c/d
hoverctl stop
You should notice a 2 second delay on responses from echo.jsontest.com/a/b and echo.jsontest.com/b/c, and a 3 second delay on the response from echo.jsontest.com/c/d.
Note
When you use Hoverfly as a webserver, you should replace or remove any original host pattern in the urlPattern filter, because the endpoint URL is based on Hoverfly webserver address (eg. http://localhost:8888).
Applying different delays based on HTTP method¶
Let’s apply a delay of 2 seconds on responses to GET requests only made to echo.jsontest.com/b/c.
Run the following to create and export a simulation.
hoverctl start
hoverctl mode capture
curl --proxy localhost:8500 http://echo.jsontest.com/b/c
curl --proxy localhost:8500 -X POST http://echo.jsontest.com/b/c
hoverctl export simulation.json
hoverctl stop
Edit the simulation.json file so that the "globalActions" property looks like this:
1 "globalActions": {
2 "delays": [
3 {
4 "urlPattern": "echo\\.jsontest\\.com\\/b\\/c",
5 "httpMethod": "GET",
6 "delay": 2000
7 }
8 ],
9 "delaysLogNormal": []
Now run the following to import the edited simulation.json file and run the simulation:
hoverctl start
hoverctl import simulation.json
curl --proxy localhost:8500 http://echo.jsontest.com/b/c
curl -X POST --proxy localhost:8500 http://echo.jsontest.com/b/c
hoverctl stop
You should notice a 2 second delay on the response to the GET request and no delay on the response to the POST request.
Log-normal distributions of delay¶
The Log-normal distribution is a pretty accurate description of a server latency. The Log-normal distribution defines by 2 parameters μ and σ. We will compute these parameters from mean and median of a server response time. These values you can see in your monitoring of the production server. If need you can adjust response time by min and max parameters.
Let’s apply a random log-normal distributed delay to all responses. First, we need to create and export a simulation.
hoverctl start
hoverctl mode capture
curl --proxy localhost:8500 http://echo.jsontest.com/b/c
curl --proxy localhost:8500 -X POST http://echo.jsontest.com/b/c
hoverctl export simulation.json
hoverctl stop
Take a look at the "globalActions" property within the simulation.json file you exported. It
should look like this:
1 "globalActions": {
2 "delays": [],
3 "delaysLogNormal": []
Edit the file so the "globalActions" property looks like this:
1 "globalActions": {
2 "delaysLogNormal": [
3 {
4 "urlPattern": ".",
5 "httpMethod": "",
6 "min": 100,
7 "max": 10000,
8 "mean": 5000,
9 "median": 500
10 }
11 ]
12 }
Hoverfly will apply a delay to all URLs that match the "urlPattern" value. We want
the delay to be applied to all URLs, so we set the "urlPattern" value to the regular expression ".".
Now import the edited simulation.json file, switch Hoverfly to Simulate mode and make the requests
again.
hoverctl start
hoverctl import simulation.json
curl --proxy localhost:8500 http://echo.jsontest.com/b/c
curl -X POST --proxy localhost:8500 http://echo.jsontest.com/b/c
hoverctl stop
Loose request matching using a Request Matcher¶
See also
Please carefully read through Request Responses Pairs alongside this tutorial to gain a high-level understanding of what we are about to cover.
In some cases you may want Hoverfly to return the same stored response for more than one incoming request. This can be done using Request Matchers.
Let’s begin by capturing some traffic and exporting a simulation. This step saves us having to manually create a simulation ourselves and gives us a request to work with.
hoverctl start
hoverctl mode capture
curl --proxy http://localhost:8500 http://echo.jsontest.com/foo/baz/bar/spam
hoverctl export simulation.json
hoverctl stop
If you take a look at your simulation.json you should notice these lines in your request.
"path": [
{
"matcher": "exact",
"value": "/foo/baz/bar/spam"
}
]
Modify them to:
"path": [
{
"matcher": "glob",
"value": "/foo/*/bar/spam"
}
]
Save the file as simulationimport.json and run the following command to import it and cURL the simulated endpoint:
hoverctl start
hoverctl mode simulate
hoverctl import simulationimport.json
curl --proxy http://localhost:8500 http://echo.jsontest.com/foo/QUX/bar/spam
hoverctl stop
The same response is returned, even though we created our simulation with a request to http://echo.jsontest.com/foo/baz/bar/spam in Capture mode and then sent a request to http://echo.jsontest.com/foo/QUX/bar/spam in Simulate mode.
See also
In this example we used the globMatch Request Matcher type. For a list of other Request Matcher types and examples
of how to use them, please see the Request matchers section.
Note
Key points:
To change how incoming requests are matched to stored responses, capture a simulation, export it, edit it
While editing, choose a request field to match on, select a Request Matcher type and a matcher value
Re-import the simulation
Requests can be manually added without capturing the request
Using middleware to simulate network latency¶
See also
Please carefully read through Middleware alongside these tutorials to gain a high-level understanding of what we are about to cover.
We will use a Python script to apply a random delay of less than one second to every response in a simulation.
Before you proceed, please ensure that you have Python installed.
Let’s begin by writing our middleware. Save the following as middleware.py:
#!/usr/bin/env python
import sys
import logging
import random
from time import sleep
logging.basicConfig(filename='random_delay_middleware.log', level=logging.DEBUG)
logging.debug('Random delay middleware is called')
# set delay to random value less than one second
SLEEP_SECS = random.random()
def main():
data = sys.stdin.readlines()
# this is a json string in one line so we are interested in that one line
payload = data[0]
logging.debug("sleeping for %s seconds" % SLEEP_SECS)
sleep(SLEEP_SECS)
# do not modifying payload, returning same one
print(payload)
if __name__ == "__main__":
main()
The middleware script delays each response by a random value of less than one second.
hoverctl start
hoverctl mode capture
curl --proxy http://localhost:8500 http://time.jsontest.com
hoverctl mode simulate
hoverctl middleware --binary python --script middleware.py
curl --proxy http://localhost:8500 http://time.jsontest.com
hoverctl stop
Middleware gives you control over the behaviour of a simulation, as well as the data.
Note
Middleware gives you flexibility when simulating network latency - allowing you to randomize the delay value for example - but a new process is spawned every time the middleware script is executed. This can impact Hoverfly’s performance under load.
If you need to simulate latency during a load test, it is recommended that you use Hoverfly’s native Delays functionality to simulate network latency (see Adding delays to a simulation) instead of writing middleware. The delays functionality sacrifices flexibility for performance.
Using middleware to modify response payload and status code¶
See also
Please carefully read through Middleware alongside these tutorials to gain a high-level understanding of what we are about to cover.
We will use a python script to modify the body of a response and randomly change the status code.
Let’s begin by writing our middleware. Save the following as middleware.py:
#!/usr/bin/env python
import sys
import json
import logging
import random
logging.basicConfig(filename='middleware.log', level=logging.DEBUG)
logging.debug('Middleware "modify_request" called')
def main():
payload = sys.stdin.readlines()[0]
logging.debug(payload)
payload_dict = json.loads(payload)
payload_dict['response']['status'] = random.choice([200, 201])
if "response" in payload_dict and "body" in payload_dict["response"]:
payload_dict["response"]["body"] = "{'foo': 'baz'}\n"
print(json.dumps(payload_dict))
if __name__ == "__main__":
main()
The middleware script randomly toggles the status code between 200 and 201, and changes the response body to a dictionary containing {'foo':'baz'}.
hoverctl start
hoverctl mode capture
curl --proxy http://localhost:8500 http://time.jsontest.com
hoverctl mode simulate
hoverctl middleware --binary python --script middleware.py
curl --proxy http://localhost:8500 http://time.jsontest.com
hoverctl stop
As you can see, middleware allows you to completely modify the content of a simulated HTTP response.
Simulating HTTPS APIs¶
To capture HTTPS traffic, you need to use Hoverfly’s SSL certificate.
First, download the certificate:
wget https://raw.githubusercontent.com/SpectoLabs/hoverfly/master/core/cert.pem
We can now run Hoverfly with the standard capture then simulate workflow.
hoverctl start
hoverctl mode capture
curl --proxy localhost:8500 https://example.com --cacert cert.pem
hoverctl mode simulate
curl --proxy localhost:8500 https://example.com --cacert cert.pem
hoverctl stop
Curl makes the HTTPS request by first establishing a TLS tunnel to the destination with Hoverfly using the HTTP CONNECT method. As curl has supplied Hoverfly’s SSL certificate, Hoverfly is then able to intercept and capture the traffic. Effectively SSL-encrypted communication (HTTPS) is established through an unencrypted HTTP proxy.
Note
This example uses cURL. If you are using Hoverfly in another environment, you will need to add the certificate to your trust store. This is done automatically by the Hoverfly Java library (see Hoverfly Java).
See also
This example uses Hoverfly’s default SSL certificate. Alternatively, you can use Hoverfly to generate a new certificate. For more information, see Configuring SSL in Hoverfly.
Running Hoverfly as a webserver¶
See also
Please carefully read through Hoverfly as a webserver alongside this tutorial to gain a high-level understanding of what we are about to cover.
Below is a complete example how to capture data with Hoverfly running as a proxy, and how to save it in a simulation file.
hoverctl start
hoverctl mode capture
curl --proxy http://localhost:8500 http://echo.jsontest.com/a/b
hoverctl export simulation.json
hoverctl stop
Now we can use Hoverfly as a webserver in Simulate mode.
hoverctl start webserver
hoverctl import simulation.json
curl http://localhost:8500/a/b
hoverctl stop
Hoverfly returned a response to our request while running as a webserver, not as a proxy.
Notice that we issued a cURL command to http://localhost:8500/a/b instead of http://echo.jsontest.com/a/b.
This is because when running as a webserver, Hoverfly strips the domain from the endpoint URL in
the simulation.
This is explained in more detail in the Hoverfly as a webserver section.
Note
Hoverfly starts in Simulate mode by default.
Enable CORS support on Hoverfly¶
By enabling CORS (Cross-Origin Resource Sharing) support, your web application running on the browser can make requests to Hoverfly even it’s not on the same domain.
Starting Hoverfly with CORS enabled is simple:
hoverfly -cors
Or using hoverctl:
hoverctl start --cors
You can check if CORS is enabled on the Hoverfly by querying the status:
hoverctl status
When CORS is enabled, Hoverfly intercepts any pre-flight request, and returns an empty 200 response with the following default CORS headers:
Access-Control-Allow-Origin: (same value as the
Originheader from the request)Access-Control-Allow-Methods: GET,POST,PUT,PATCH,DELETE,HEAD,OPTIONS
Access-Control-Allow-Headers: (same value as the
Access-Control-Request-Headersheader from the request)Access-Control-Max-Age: 1800
Access-Control-Allow-Credentials: true
Hoverfly also intercepts the actual CORS requests, and add the following default CORS headers to the response:
Access-Control-Allow-Origin: (same value as the
Originheader from the request)Access-Control-Allow-Credentials: true
Support for customizing the CORS headers will be added in the future release.
Note
Two points to notice when Hoverfly is in capture mode and CORS is enabled: 1. Pre-flight requests handling and CORS headers provided by Hoverfly are not recorded in the simulation. 2. Hoverfly preserves the CORS headers from the remote server if they are present.
Capturing or simulating specific URLs¶
We can use the hoverctl destination command to specify which URLs to capture or simulate. The destination
setting can be tested using the --dry-run flag. This makes it easy to check whether the destination setting
will filter out URLs as required.
hoverctl start
hoverctl destination "ip" --dry-run http://ip.jsontest.com
hoverctl destination "ip" --dry-run http://time.jsontest.com
hoverctl stop
This tells us that setting the destination to ip will allow the URL http://ip.jsontest.com to be captured or simulated, while the URL
http://time.jsontest.com will be ignored.
Now we have checked the destination setting, we can apply it to filter out the URLs we don’t want to capture.
hoverctl start
hoverctl destination "ip"
hoverctl mode capture
curl --proxy http://localhost:8500 http://ip.jsontest.com
curl --proxy http://localhost:8500 http://time.jsontest.com
hoverctl logs
hoverctl export simulation.json
hoverctl stop
If we examine the logs and the simulation.json file, we can see that only a request response pair to the http://ip.jsontest.com
URL has been captured.
The destination setting can be either a string or a regular expression.
hoverctl start
hoverctl destination "^.*api.*com" --dry-run https://api.github.com
hoverctl destination "^.*api.*com" --dry-run https://api.slack.com
hoverctl destination "^.*api.*com" --dry-run https://github.com
hoverctl stop
Here, we can see that setting the destination to ^.*api.*com will allow the https://api.github.com and https://api.slack.com
URLs to be captured or simulated, while the https://github.com URL will be ignored.
Capturing a stateful sequence of responses¶
By default Hoverfly will store a given request/response pair once only. If the same request returns different responses you may want capture the sequence of changing request/response pairs. You may want to do this because the API is stateful rather than stateless.
A simple example of this is an API that returns the time.
To record a sequence of request/responses where the request is the same but the response is different, we need to enable stateful recording in capture mode.
hoverctl start
hoverctl mode capture --stateful
Now that we have enabled stateful recording, we can capture several request/response pairs.
curl --proxy http://localhost:8500 http://time.jsontest.com
curl --proxy http://localhost:8500 http://time.jsontest.com
Once we have finished capturing requests, we can switch Hoverfly back to simulate mode.
hoverctl mode simulate
Now we are in simulate, we can make the same requests again, and we will see the time update each request we make until we reach the end of our recorded sequence.
{
"time": "01:59:21 PM",
"milliseconds_since_epoch": 1528120761743,
"date": "06-04-2018"
}
{
"time": "01:59:23 PM",
"milliseconds_since_epoch": 1528120763647,
"date": "06-04-2018"
}
If we look at the simulation captured, can see that the requests have the requiresState fields set.
You will see Hoverfly has added a state variable called sequence:1 that acts as a counter.
"requiresState": {
"sequence:1": "1"
}
"requiresState": {
"sequence:1": "2"
}
We can also see that the first response has transitionsState` field set.
"transitionsState": {
"sequence:1": "2"
}
Note that Hoverfly will automatically set the state for any “sequence:” key to “1” on import. If you want to use a more meaningful key name you will need to initialise the state as follows:
hoverctl state set shopping-basket empty
See also
For a more detailed explaination of how sequences work in hoverfly: see Sequences in the Key Concepts section.
Adding Post Serve Action to particular response¶
See also
Please carefully read through Post Serve Action alongside this tutorial to gain a high-level understanding of what we are about to cover.
PostServeAction allows you to execute custom code after a response has been served in simulate or spy mode.
In this tutorial, we will help you to setup post serve action and execute it after particular response is served.
Let’s begin by writing our post serve action. The following python script makes an HTTP call to http://ip.jsontest.com and prints out your IP address.
We can make a call to any other URL as well in order to send webhook or initiate some processing once response is served. Although the example script is written in python, you can also write an action using any other language as long as you have provided a binary in your local environment to execute the code.
Save the following as post_serve_action.py:
#!/usr/bin/env python
import sys
import requests
def get_ip_info():
url = "http://ip.jsontest.com/"
try:
response = requests.get(url)
response.raise_for_status() # Raise an exception if the request was not successful (HTTP status code >= 400)
data = response.json() # Parse the JSON response
return data
except requests.exceptions.RequestException as e:
print(f"An error occurred: {e}")
return None
def main():
data = sys.stdin.readlines()
# this is a request-response pair json string. We can use this pair to make outbound call or do any operations
payload = data[0]
ip_info = get_ip_info()
if ip_info:
print(f"HTTP call invoked from IP Address: {ip_info['ip']}")
else:
print("Failed to retrieve IP information.")
if __name__ == "__main__":
main()
Start Hoverfly and register a post serve action:
hoverctl start
hoverctl post-serve-action set --binary python3 --name callback-script --script <path to script directory>/post_serve_action.py --delay 3000
Once the post serve action is registered, you can confirm using below hoverctl command.
hoverctl post-serve-action get-all
Sample output
+-----------------+---------+--------------------------------+-----------+
| ACTION NAME | BINARY | SCRIPT | DELAY(MS) |
+-----------------+---------+--------------------------------+-----------+
| callback-script | python3 | #!/usr/bin/env python import | 3000 |
| | | sys import logging import | |
| | | random from time import sleep | |
| | | ... | |
+-----------------+---------+--------------------------------+-----------+
Copy this simulation JSON content to a file called simulation.json:
{
"data": {
"pairs": [
{
"request": {
"path": [
{
"matcher": "exact",
"value": "/"
}
],
"method": [
{
"matcher": "exact",
"value": "GET"
}
],
"destination": [
{
"matcher": "exact",
"value": "date.jsontest.com"
}
],
"scheme": [
{
"matcher": "exact",
"value": "http"
}
],
"body": [
{
"matcher": "exact",
"value": ""
}
]
},
"response": {
"status": 200,
"body": "01-01-1111",
"encodedBody": false,
"templated": false,
"postServeAction": "callback-script"
}
}
]
},
"meta": {
"schemaVersion": "v5.2",
"hoverflyVersion": "v1.5.3",
"timeExported": "2023-09-04T11:50:40+05:30"
}
}
Run this hoverctl command to import the simulation file.
hoverctl import <path-to-simulation-file>
The simulation sets hoverfly to return a successful response and 3 seconds after that invokes the “callback-script” action. You can try it out by making the following request to http://date.jsontest.com using cURL.
curl --proxy http://localhost:8500 http://date.jsontest.com
You should see the message in the hoverfly logs - Output from post serve action HTTP call invoked from IP Address.
hoverctl logs
Advanced tutorials¶
Using Hoverfly behind a proxy¶
In some environments, you may only be able to access the internet via a proxy. For example, your organization may route all traffic through a proxy for security reasons.
If this is the case, you will need to configure Hoverfly to work with the ‘upstream’ proxy.
This configuration value can be easily set when starting an instance of Hoverfly.
For example, if the ‘upstream’ proxy is running on port 8080 on host corp.proxy:
hoverctl start --upstream-proxy http://corp.proxy:8080
Configuring Hoverfly with a PAC file¶
If you have a PAC file for configuring clients with the proxy set up in your environment, you can use this to configure Hoverfly.
For example:
hoverctl start --pac-file path/to/file.pac
Upstream proxy authentication¶
If the proxy you are using uses HTTP basic authentication, you can provide the authentication credentials as part of the upstream proxy configuration setting.
For example:
hoverctl start --upstream-proxy http://my-user:my-pass@corp.proxy:8080
Currently, HTTP basic authentication is the only supported authentication method for an authenticated proxy.
Controlling a remote Hoverfly instance with hoverctl¶
So far, the tutorials have shown how hoverctl can be used to control an instance of Hoverfly running on the same machine.
In some cases, you may wish to use hoverctl to control an instance of Hoverfly running on a remote host. With hoverctl, you can do this using the targets feature.
In this example, we assume that the remote host is reachable at hoverfly.example.com, and that
ports 8880 and 8555 are available. We will also assume that the Hoverfly binary is installed on the remote host.
On the remote host, start Hoverfly using flags to override the default admin port (-ap) and proxy port (-pp).
hoverfly -ap 8880 -pp 8555
See also
For a full list of all Hoverfly flags, please refer to Hoverfly commands in the Reference section.
On your local machine, you can create a target named remote using hoverctl. This target will be configured to communicate
with Hoverfly.
hoverctl targets create remote \
--host hoverfly.example.com \
--admin-port 8880 \
--proxy-port 8555
Now that hoverctl knows the location of the remote Hoverfly instance, run the following commands
on your local machine to capture and simulate a URL using this instance:
hoverctl -t remote mode capture
curl --proxy http://hoverfly.example.com:8555 http://ip.jsontest.com
hoverctl -t remote mode simulate
curl --proxy http://hoverfly.example.com:8555 http://ip.jsontest.com
You will now need to specify the remote target every time you want to interact with this Hoverfly instance.
If you are only working with this remote instance, you can set it to be the default target instance for hoverctl.
hoverctl targets default remote
Note
The --host value of the hoverctl target allows hoverctl to interact with the admin API
of the remote Hoverfly instance.
The application that is making the request (in this case, cURL), also needs to be configured to
use the remote Hoverfly instance as a proxy. In this example, it is done using cURL’s --proxy flag.
If you are running Hoverfly on a remote host, you may wish to enable authentication on the Hoverfly proxy and admin API. This is described in the Enabling authentication for the Hoverfly proxy and API tutorial.
Enabling authentication for the Hoverfly proxy and API¶
Sometimes, you may need authentication on Hoverfly. An example use case would be when running Hoverfly on a remote host.
Hoverfly can provide authentication for both the admin API (using a JWT token) and the proxy (using HTTP basic auth).
Setting Hoverfly authentication credentials¶
To start a Hoverfly instance with authentication enabled, you need to run
the hoverctl start command with the authentication (--auth) flag.
hoverctl start --auth
Running this command will prompt you to set a username and password for the Hoverfly instance.
This can be bypassed by providing the --username
and --password flags, although this will leave credentials
in your terminal history.
Warning
By default, hoverctl will start Hoverfly with authentication disabled. If you require authentication
you must make sure the --auth flag are supplied every time Hoverfly is started.
Logging in to a Hoverfly instance with hoverctl¶
Now that a Hoverfly instance has started with authentication enabled, you will need to login to the instance using hoverctl.
hoverctl login
Running this command will prompt you to enter the username and
password you set for the Hoverfly instance. Again, this can be bypassed by providing the --username
and --password flags.
There may be situations in which you need to log into to a Hoverfly instance
that is already running. In this case, it is best practice to create a new target
for the instance (please see Controlling a remote Hoverfly instance with hoverctl for more information on targets). You can do this using
the --new-target flag.
In this example, a remote Hoverfly instance is already running on the host `hoverfly.example.com`, with
the ports set to 8880 and 8555 and authentication enabled (the example from Controlling a remote Hoverfly instance with hoverctl).
You will need to create a new target (named remote) for the instance and log in with it.
hoverctl login --new-target remote \
--host hoverfly.example.io \
--admin-port 8880 \
--proxy-port 8550
You will be prompted to enter the username and password for the instance.
Now run the following commands to capture and simulate a URL using the remote Hoverfly:
hoverctl -t remote mode capture
curl --proxy http://my-user:my-pass@hoverfly.example.com:8555 http://ip.jsontest.com
hoverctl -t remote mode simulate
curl --proxy http://my-user:my-pass@hoverfly.example.com:8555 http://ip.jsontest.com
Configuring SSL in Hoverfly¶
Hoverfly supports both one-way and two-way SSL authentication.
Hoverfly uses default certificate which you should add to your HTTPS client’s trust store for one-way SSL authentication. You have options to provide your own certificate, please see below.
Override default certificate for one-way SSL authentication¶
In some cases, you may not wish to use Hoverfly’s default SSL certificate. Hoverfly allows you to generate a new certificate and key.
The following command will start a Hoverfly process and create new cert.pem and key.pem
files in the current working directory. These newly-created files will be loaded into the
running Hoverfly instance.
hoverfly -generate-ca-cert
Optionally, you can provide a custom certificate name and authority:
hoverfly -generate-ca-cert -cert-name tutorial.cert -cert-org "Tutorial Certificate Authority"
Once you have generated cert.pem and key.pem files with Hoverfly, you can use hoverctl
to start an instance of Hoverfly using these files.
hoverctl start --certificate cert.pem --key key.pem
Note
Both a certificate and a key file must be supplied. The files must be in unencrypted PEM format.
Configure Hoverfly for two-way SSL authentication¶
For two-way or mutual SSL authentication, you should provide Hoverfly with a client certificate and a certificate key that you use to authenticate with the remote server.
Two-way SSL authentication is only enabled for request hosts that match the value you provided to the --client-authentication-destination flag. You can also pass a regex pattern if you need to match multiple hosts.
hoverctl start --client-authentication-client-cert cert.pem --client-authentication-client-key key.pem --client-authentication-destination <host name of the remote server>
If you need to provide a CA cert, you can do so using the --client-authentication-ca-cert flag.
Troubleshooting¶
Why isn’t Hoverfly matching my request?¶
When Hoverfly cannot match a response to an incoming request, it will return information on the closest match:
Hoverfly Error!
There was an error when matching
Got error: Could not find a match for request, create or record a valid matcher first!
The following request was made, but was not matched by Hoverfly:
{
"Path": "/closest-miss",
"Method": "GET",
"Destination": "destination.com",
"Scheme": "http",
"Query": "",
"Body": "",
"Headers": {
"Accept-Encoding": [
"gzip"
],
"User-Agent": [
"Go-http-client/1.1"
]
}
}
The matcher which came closest was:
{
"path": [
{
"matcher": "exact",
"value": "/closest-miss"
}
],
"destination": [
{
"matcher": "exact",
"value": "destination.com"
}
],
"body": [
{
"matcher": "exact",
"value": "body"
}
]
}
But it did not match on the following fields:
[body]
Which if hit would have given the following response:
{
"status": 200,
"body": "",
"encodedBody": false,
"headers": null
}`
Here, you can see which fields did not match. In this case, it was the body.
You can also view this information by running hoverctl logs.
Why isn’t Hoverfly returning the closest match when it cannot match a request?¶
Hoverfly will only provide this information when the matching strategy is set to strongest match (the default). If you are using the first match matching strategy, the closet match information will not be returned.
How can I view the Hoverfly logs?¶
hoverctl logs
Why does my simulation have a deprecatedQuery field?¶
Older simulations that have been upgraded through newer versions of Hoverfly may now contain a field
on requests called deprecatedQuery. With the v5 simulation schema, the request query field was
updated to more fully represent request query paramters. This involves storing queries based on
query keys, similarly to how headers are stored in a simulation.
Currently the deprecatedQuery field will work and works alongside the query field and support
for this field will eventually be dropped.
If you have deprecatedQuery field, you should remove it by splitting it by query keys.
"deprecatedQuery": "page=20&pageSize=15"
"query": {
"page": [
{
"matcher": "exact",
"value": "20"
}
],
"pageSize": [
{
"matcher": "exact",
"value": "15"
}
],
}
If you cannot update your deprecatedQuery from your simulation for a technical reason, feel free to
raise an issue on Hoverfly.
Why am I not able to access my Hoverfly remotely?¶
That’s because Hoverfly is bind to loopback interface by default, meaning that you can only access
to it on localhost. To access it remotely, you can specify the IP address it listens on. For example,
setting 0.0.0.0 to listen on all network interfaces.
hoverfly -listen-on-host 0.0.0.0
My simulation file is very large because of response bodies, what can I do with that?¶
You can move those response bodies into separate files and specify bodyFile in the response instead of
body. Please refer to Request Responses Pairs.
Reference¶
These reference documents contain information regarding invoking the hoverctl command, hoverfly command, and interacting with the APIs.
hoverctl commands¶
This page contains the output of:
hoverctl --help
The command’s help content has been placed here for convenience.
hoverctl is the command line tool for Hoverfly
Usage:
hoverctl [command]
Available Commands:
completion Create Bash completion file for hoverctl
config Show hoverctl configuration information
delete Delete Hoverfly simulation
destination Get and set Hoverfly destination
diff Manage the diffs for Hoverfly
export Export a simulation from Hoverfly
flush Flush the internal cache in Hoverfly
import Import a simulation into Hoverfly
login Login to Hoverfly
logs Get the logs from Hoverfly
middleware Get and set Hoverfly middleware
mode Get and set the Hoverfly mode
post-serve-action Manage the post-serve-action for Hoverfly
simulation Manage the simulation for Hoverfly
start Start Hoverfly
state Manage the state for Hoverfly
status Get the current status of Hoverfly
stop Stop Hoverfly
targets Get the current targets registered with hoverctl
templating-data-source Manage the templating data source for Hoverfly
version Get the version of hoverctl
Flags:
-f, --force Bypass any confirmation when using hoverctl
-h, --help help for hoverctl
--set-default Sets the current target as the default target for hoverctl
-t, --target string A name for an instance of Hoverfly you are trying to communicate with. Overrides the default target (default)
-v, --verbose Verbose logging from hoverctl
Use "hoverctl [command] --help" for more information about a command.
hoverctl auto completion¶
hoverctl supplies auto completion for Bash. Run the following command to install the completions.
hoverctl completion
This will create the completion file in your hoverfly directory and create a symbolic link in your bash_completion.d folder.
Optionally you can supply a location for the symbolic link as an argument to the completion command.
hoverctl completion /usr/local/etc/bash_completion.d/hoverctl
Hoverfly commands¶
This page contains the output of:
hoverfly --help
The command’s help content has been placed here for convenience.
Usage of hoverfly:
-add
Add new user '-add -username hfadmin -password hfpass'
-admin
Supply '-admin=false' to make this non admin user (default true)
-ap string
Admin port - run admin interface on another port (i.e. '-ap 1234' to run admin UI on port 1234)
-auth
Enable authentication
-cache-size int
Set the size of request/response cache (default 1000)
-capture
Start Hoverfly in capture mode - transparently intercepts and saves requests/response
-cert string
CA certificate used to sign MITM certificates
-cert-name string
Cert name (default "hoverfly.proxy")
-cert-org string
Organisation name for new cert (default "Hoverfly Authority")
-client-authentication-ca-cert string
Path to the ca cert file used for authentication
-client-authentication-client-cert string
Path to the client certification file used for authentication
-client-authentication-client-key string
Path to the client key file used for authentication
-client-authentication-destination string
Regular expression of destination with client authentication
-cors
Enable CORS support
-db string
Storage to use - 'boltdb' or 'memory' which will not write anything to disk (DEPRECATED) (default "memory")
-db-path string
A path to a BoltDB file with persisted user and token data for authentication (DEPRECATED)
-dest value
Specify which hosts to process (i.e. '-dest fooservice.org -dest barservice.org -dest catservice.org') - other hosts will be ignored will passthrough'
-destination string
Control which URLs Hoverfly should intercept and process, it can be string or regex (default ".")
-dev
Enable CORS headers to allow Hoverfly Admin UI development
-dev-cors-origin string
Custom CORS origin for dev mode (default "http://localhost:4200")
-diff
Start Hoverfly in diff mode - calls real server and compares the actual response with the expected simulation config if present
-disable-cache
Disable the request/response cache (the cache that sits in front of matching)
-generate-ca-cert
Generate CA certificate and private key for MITM
-import value
Import from file or from URL (i.e. '-import my_service.json' or '-import http://mypage.com/service_x.json'
-journal-size int
Set the size of request/response journal (default 1000)
-journal-indexing-key string
Specify the index key using which you want to index journal. Index shares same syntax as the one for templating, such as Request.QueryParam.myParam or Request.Header.X-Header-Id.[1].
It is used for extracting the data from the journal entry to use as a key for that entry.
-key string
Private key of the CA used to sign MITM certificates
-listen-on-host string
Specify which network interface to bind to, eg. 0.0.0.0 will bind to all interfaces. By default hoverfly will only bind ports to loopback interface
-log-level string
Set log level (panic, fatal, error, warn, info or debug) (default "info")
-log-no-color
Disable colors for logging
-log-no-quotes
Disable quoting and escaping of logged fields
-logs string
Specify format for logs, options are "plaintext" and "json" (default "plaintext")
-logs-file string
Specify log file name for output logs (default "hoverfly.log")
-logs-output value
Specify locations for output logs, options are "console" and "file" (default "console")
-logs-size int
Set the amount of logs to be stored in memory (default 1000)
-metrics
Enable metrics logging to stdout
-post-serve-action string
Set post serve action by passing the action name, binary and the path of the action script and delay in Ms separated by space.
(i.e. -post-serve-action "<action name> python3 <script path to load> 1000" -post-serve-action "<action name> python3 <script path to load> 3000")
We can set multiple post serve actions and use in our simulation schema file.
-middleware string
Set middleware by passing the name of the binary and the path of the middleware script separated by space. (i.e. '-middleware "python script.py"')
-modify
Start Hoverfly in modify mode - applies middleware (required) to both outgoing and incoming HTTP traffic
-no-import-check
Skip duplicate request check when importing simulations
-password string
Password for new user
-password-hash string
Password hash for new user instead of password
-plain-http-tunneling
Use plain http tunneling to host with non-443 port
-pp string
Proxy port - run proxy on another port (i.e. '-pp 9999' to run proxy on port 9999)
-response-body-files-allow-origin value
When a response contains a url in bodyFile, it will be loaded only if the origin is allowed
-response-body-files-path string
When a response contains a relative bodyFile, it will be resolved against this path (default is CWD)
-spy
Start Hoverfly in spy mode, similar to simulate but calls real server when cache miss
-synthesize
Start Hoverfly in synthesize mode (middleware is required)
-templating-data-source
Set templating CSV data source by passing data source name and CSV data file path separated by space.
(i.e. -templating-data-source "<data source name> <file path>")
We can set multiple template CSV data source and then we can query data by using templating csv helper method
-tls-verification
Turn on/off tls verification for outgoing requests (will not try to verify certificates) (default true)
-upstream-proxy string
Specify an upstream proxy for hoverfly to route traffic through
-username string
Username for new user
-v Should every proxy request be logged to stdout
-version
Get the version of hoverfly
-webserver
Start Hoverfly in webserver mode (simulate mode)
Request matchers¶
A Request Matcher is used to define the desired value for a specific request field when matching against incoming requests. Given a matcher value and string to match, each matcher will transform and compare the values in a different way.
Exact matcher¶
Evaluates the equality of the matcher value and the string to match. There are no transformations. This is the default Request Matcher type which is set by Hoverfly when requests and responses are captured.
Example¶
"matcher": "exact"
"value": "?"
| String to match | Matcher value | Match |
|---|---|---|
| docs.hoverfly.io | docs.hoverfly.io | |
| specto.io | docs.hoverfly.io | |
Glob matcher¶
Allows wildcard matching (similar to BASH) using the * character.
Example¶
"matcher": "glob"
"value": "?"
| String to match | Matcher value | Match |
|---|---|---|
| docs.hoverfly.io | *.hoverfly.io | |
| docs.specto.io | *.hoverfly.io | |
| docs.hoverfly.io | h*verfly.* | |
| hooverfly.com | h*verfly.* | |
Regex matcher¶
Parses the matcher value as a regular expression which is then executed against the string to match. This will pass only if the regular expression successfully returns a result.
Example¶
"matcher": "regex"
"value": "?"
| String to match | Matcher value | Match |
|---|---|---|
| docs.hoverfly.io | (\\Ad) | |
| hoverfly.io | (\\Ad) | |
| docs.hoverfly.io | (.*).(.*).(io|com|biz) | |
| buy.stuff.biz | (.*).(.*).(io|com|biz) | |
XML matcher¶
Transforms both the matcher value and string to match into XML objects and then evaluates their equality.
Example¶
"matcher": "xml"
"value": "?"
| String to match | Matcher value | Match |
|---|---|---|
| <?xml version="1.0" encoding="UTF-8"?> <document type="book"> Hoverfly Documentation </document> | <?xml version="1.0" encoding="UTF-8"?> <document type="book"> Hoverfly Documentation </document> | |
| <?xml version="1.0" encoding="UTF-8"?> <documents type="book"> <document type="book"> Hoverfly Documentation </document> </document> | <?xml version="1.0" encoding="UTF-8"?> <document type="book"> Hoverfly Documentation </document> | |
XPath matcher¶
Parses the matcher value as an XPath expression, transforms the string to match into an XML object and then executes the expression against it. This will pass only if the expression successfully returns a result.
Example¶
"matcher": "xpath"
"value": "?"
| String to match | Matcher value | Match |
|---|---|---|
| <?xml version="1.0" encoding="UTF-8"?> <documents> <document> Hoverfly Documentation </document> </documents> | /documents | |
| <?xml version="1.0" encoding="UTF-8"?> <document> Hoverfly Documentation </document> | /documents | |
| <?xml version="1.0" encoding="UTF-8"?> <documents> <document type="book"> Hoverfly Documentation </document> </documents> | /documents/document[2] | |
| <?xml version="1.0" encoding="UTF-8"?> <documents type="book"> <document> Someone Else's Documentation </document> <document> Hoverfly Documentation </document> </documents> | /documents/document[2] | |
JSON matcher¶
Transforms both the matcher value and string to match into JSON objects and then evaluates their equality.
Example¶
"matcher": "json"
"value": "?"
| String to match | Matcher value | Match |
|---|---|---|
| { "objects": [ { "name": "Object 1", "set": true },{ "name": "Object 2", "set": false, "age": 400 }] } | { "objects": [ { "name": "Object 1", "set": true },{ "name": "Object 2", "set": false, "age": 400 }] } | |
| { "objects": [ { "name": "Object 1", "set": true }] } | { "objects": [ { "name": "Object 1", "set": true },{ "name": "Object 2", "set": false, "age": 400 }] } | |
JSON partial matcher¶
Unlike a JSON matcher which does the full matching of two JSON documents, this matcher evaluates if the matcher value is a subset of the incoming JSON document. The matcher ignores any absent fields and lets you match only the part of JSON document you care about.
Example¶
"matcher": "jsonPartial"
"value": "?"
| String to match | Matcher value | Match |
|---|---|---|
| { "objects": [ { "name": "Object 1", },{ "name": "Object 2", "set": false, "age": 400 }] } | { "objects": [ { "name": "Object 1" },{ "name": "Object 2" }] } | |
| { "objects": [ { "name": "Object 1", },{ "name": "Object 2", "set": false, "age": 400 }] } | { "name": "Object 2", "set": false, "age": 400 } | |
| { "objects": [ { "name": "Object 1", "set": true }] } | { "objects": [ { "name": "Object 1", "set": true },{ "name": "Object 2", "set": false, "age": 400 }] } | |
JSONPath matcher¶
Parses the matcher value as a JSONPath expression, transforms the string to match into a JSON object and then executes the expression against it. This will pass only if the expression successfully returns a result.
Example¶
"matcher": "jsonpath"
"value": "?"
| String to match | Matcher value | Match |
|---|---|---|
| { "objects": [ { "name": "Object 1", "set": true }] } | $.objects | |
| { "name": "Object 1", "set": true } | $.objects | |
| { "objects": [ { "name": "Object 1", "set": true }] } | $.objects[1].name | |
| { "objects": [ { "name": "Object 1", "set": true }, { "name": "Object 2", "set": false }] } | $.objects[1].name | |
Form matcher¶
Matches form data posted in the request payload with content type application/x-www-form-urlencoded.
You can match only the form params you are interested in regardless of the order. You can also leverage
jwt or jsonpath matchers if your form params contains JWT tokens or JSON document.
Please note that this matcher only works for body field.
Example¶
"matcher": "form",
"value": {
"grant_type": [
{
"matcher": "exact",
"value": "authorization_code"
}
],
"client_assertion": [
{
"matcher": "jwt",
"value": "{\"header\":{\"alg\":\"HS256\"},\"payload\":{\"sub\":\"1234567890\",\"name\":\"John Doe\"}}"
}
]
}
Array matcher¶
Matches an array contains exactly the given values and nothing else. This can be used to match multi-value query param or header in the request data.
The following configuration options are available to change the behaviour of the matcher:
ignoreOrder - ignore the order of the values.
ignoreUnknown - ignore any extra values.
ignoreOccurrences - ignore any duplicated values.
Example¶
"matcher": "array",
"config": {
"ignoreUnknown": "<true/false>",
"ignoreOrder": "<true/false>",
"ignoreOccurrences": "<true/false>"
},
"value": [
"access:vod",
"order:latest",
"profile:vd"
]
JWT matcher¶
This matcher is primarily used for matching JWT tokens. This matcher converts base64 encoded JWT to JSON document ({“header”: {}, “payload”: “”}) and does JSON partial match with the matcher value.
Matcher value contains only keys that they want to match in JWT.
Example¶
"matcher": "jwt"
"value": "{\"header\":{\"alg\":\"HS256\"},\"payload\":{\"sub\":\"1234567890\",\"name\":\"John Doe\"}}"
Matcher chaining¶
Matcher chaining allows you to pass a matched value into another matcher to do further matching.
It typically removes the stress of composing and testing complex expressions and make matchers more readable.
For an example, one can use JSONPath to get a JSON node, then use another matcher to match the JSON node value as follows.
Example¶
"matcher": "jsonpath",
"value": "$.user.id",
"doMatch": {
"matcher": "exact",
"value": "1"
}
REST API¶
GET /api/v2/simulation¶
Gets all simulation data. The simulation JSON contains all the information Hoverfly can hold; this includes recordings, templates, delays and metadata.
Example response body
{
"data": {
"pairs": [
{
"request": {
"path": [
{
"matcher": "exact",
"value": "/"
}
],
"method": [
{
"matcher": "exact",
"value": "GET"
}
],
"destination": [
{
"matcher": "exact",
"value": "myhost.io"
}
],
"scheme": [
{
"matcher": "exact",
"value": "https"
}
],
"body": [
{
"matcher": "exact",
"value": ""
}
],
"headers": {
"Accept": [
{
"matcher": "glob",
"value": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8"
}
],
"Content-Type": [
{
"matcher": "glob",
"value": "text/plain; charset=utf-8"
}
],
"User-Agent": [
{
"matcher": "glob",
"value": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/52.0.2743.116 Safari/537.36"
}
]
},
"query": {
"status": [
{
"matcher": "exact",
"value": "available"
}
]
}
},
"response": {
"status": 200,
"body": "<h1>Matched on recording</h1>",
"encodedBody": false,
"headers": {
"Content-Type": [
"text/html; charset=utf-8"
]
},
"templated": false
}
},
{
"request": {
"path": [
{
"matcher": "exact",
"value": "/template"
}
]
},
"response": {
"status": 200,
"body": "<h1>Matched on template</h1>",
"encodedBody": false,
"headers": {
"Content-Type": [
"text/html; charset=utf-8"
]
},
"templated": false
}
}
],
"globalActions": {
"delays": [],
"delaysLogNormal": []
}
},
"meta": {
"schemaVersion": "v5",
"hoverflyVersion": "v1.0.0",
"timeExported": "2019-05-30T22:14:24+01:00"
}
}
PUT /api/v2/simulation¶
This puts the supplied simulation JSON into Hoverfly, overwriting any existing simulation data.
Example request body
{
"data": {
"pairs": [
{
"request": {
"path": [
{
"matcher": "exact",
"value": "/"
}
],
"method": [
{
"matcher": "exact",
"value": "GET"
}
],
"destination": [
{
"matcher": "exact",
"value": "myhost.io"
}
],
"scheme": [
{
"matcher": "exact",
"value": "https"
}
],
"body": [
{
"matcher": "exact",
"value": ""
}
],
"headers": {
"Accept": [
{
"matcher": "glob",
"value": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8"
}
],
"Content-Type": [
{
"matcher": "glob",
"value": "text/plain; charset=utf-8"
}
],
"User-Agent": [
{
"matcher": "glob",
"value": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/52.0.2743.116 Safari/537.36"
}
]
},
"query": {
"status": [
{
"matcher": "exact",
"value": "available"
}
]
}
},
"response": {
"status": 200,
"body": "<h1>Matched on recording</h1>",
"encodedBody": false,
"headers": {
"Content-Type": [
"text/html; charset=utf-8"
]
},
"templated": false
}
},
{
"request": {
"path": [
{
"matcher": "exact",
"value": "/template"
}
]
},
"response": {
"status": 200,
"body": "<h1>Matched on template</h1>",
"encodedBody": false,
"headers": {
"Content-Type": [
"text/html; charset=utf-8"
]
},
"templated": false
}
}
],
"globalActions": {
"delays": [],
"delaysLogNormal": []
}
},
"meta": {
"schemaVersion": "v5",
"hoverflyVersion": "v1.0.0",
"timeExported": "2019-05-30T22:14:24+01:00"
}
}
POST /api/v2/simulation¶
This appends the supplied simulation JSON to the existing simulation data in Hoverfly. Any pair that has request data identical to the existing ones will not be added.
Example request body
{
"data": {
"pairs": [
{
"request": {
"path": [
{
"matcher": "exact",
"value": "/"
}
],
"method": [
{
"matcher": "exact",
"value": "GET"
}
],
"destination": [
{
"matcher": "exact",
"value": "myhost.io"
}
],
"scheme": [
{
"matcher": "exact",
"value": "https"
}
],
"body": [
{
"matcher": "exact",
"value": ""
}
],
"headers": {
"Accept": [
{
"matcher": "glob",
"value": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8"
}
],
"Content-Type": [
{
"matcher": "glob",
"value": "text/plain; charset=utf-8"
}
],
"User-Agent": [
{
"matcher": "glob",
"value": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/52.0.2743.116 Safari/537.36"
}
]
},
"query": {
"status": [
{
"matcher": "exact",
"value": "available"
}
]
}
},
"response": {
"status": 200,
"body": "<h1>Matched on recording</h1>",
"encodedBody": false,
"headers": {
"Content-Type": [
"text/html; charset=utf-8"
]
},
"templated": false
}
},
{
"request": {
"path": [
{
"matcher": "exact",
"value": "/template"
}
]
},
"response": {
"status": 200,
"body": "<h1>Matched on template</h1>",
"encodedBody": false,
"headers": {
"Content-Type": [
"text/html; charset=utf-8"
]
},
"templated": false
}
}
],
"globalActions": {
"delays": [],
"delaysLogNormal": []
}
},
"meta": {
"schemaVersion": "v5",
"hoverflyVersion": "v1.0.0",
"timeExported": "2019-05-30T22:14:24+01:00"
}
}
DELETE /api/v2/simulation¶
Unsets the simulation data for Hoverfly.
GET /api/v2/simulation/schema¶
Gets the JSON Schema used to validate the simulation JSON.
GET /api/v2/hoverfly¶
Gets configuration information from the running instance of Hoverfly.
Example response body
{
"cors": {
"enabled": true,
"allowOrigin": "*",
"allowMethods": "GET,POST,PUT,PATCH,DELETE,HEAD,OPTIONS",
"allowHeaders": "Content-Type,Origin,Accept,Authorization,Content-Length,X-Requested-With",
"preflightMaxAge": 1800,
"allowCredentials": true
},
"destination": ".",
"middleware": {
"binary": "python",
"script": "# a python script would go here",
"remote": ""
},
"mode": "simulate",
"arguments": {
"matchingStrategy": "strongest"
},
"isWebServer": false,
"usage": {
"counters": {
"capture": 0,
"modify": 0,
"simulate": 0,
"synthesize": 0
}
},
"version": "v1.3.3",
"upstreamProxy": ""
}
GET /api/v2/hoverfly/cors¶
Gets CORS configuration information from the running instance of Hoverfly.
Example response body
{
"enabled": true,
"allowOrigin": "*",
"allowMethods": "GET,POST,PUT,PATCH,DELETE,HEAD,OPTIONS",
"allowHeaders": "Content-Type, Origin, Accept, Authorization, Content-Length, X-Requested-With",
"preflightMaxAge": 1800,
"allowCredentials": true
}
GET /api/v2/hoverfly/destination¶
Gets the current destination setting for the running instance of Hoverfly.
Example response body
{
destination: "."
}
PUT /api/v2/hoverfly/destination¶
Sets a new destination for the running instance of Hoverfly, overwriting the existing destination setting.
Example request body
{
destination: "new-destination"
}
GET /api/v2/hoverfly/middleware¶
Gets the middleware settings for the running instance of Hoverfly. This could be either an executable binary, a script that can be executed with a binary or a URL to remote middleware.
Example response body
{
"binary": "python",
"script": "#python code goes here",
"remote": ""
}
PUT /api/v2/hoverfly/middleware¶
Sets new middleware, overwriting the existing middleware for the running instance of Hoverfly. The middleware being set can be either an executable binary located on the host, a script and the binary to execute it or the URL to a remote middleware.
Example request body
{
"binary": "python",
"script": "#python code goes here",
"remote": ""
}
GET /api/v2/hoverfly/post-serve-action¶
Get all the post serve actions for the running instance of Hoverfly. It will return list of scripts that can be executed with a binary after response is served back.
Example response body
{
"actions": [
{
"actionName": "<post serve action name>",
"binary": "python",
"script": "#python code goes here",
"delayInMs":"#delay(in ms) post which script will be executed after serving the request"
}
]
}
PUT /api/v2/hoverfly/post-serve-action¶
Sets new post serve action, overwriting the existing post serve action for the running instance of Hoverfly. The post serve action being set is an executable binary located on the host. We can set multiple post serve actions. It returns all the post serve actions.
Example request body
{
"actionName": "#post serve action names goes here",
"binary": "python",
"script": "#python code goes here",
"delayInMs": "#delay(in ms) post which script will be executed after serving the request"
}
DELETE /api/v2/hoverfly/post-serve-action/:actionName¶
Delete a particular post serve action for the running instance of Hoverfly. It returns all the remaining post serve actions.
GET /api/v2/hoverfly/mode¶
Gets the mode for the running instance of Hoverfly.
Example response body
{
"mode": "capture",
"arguments": {
"headersWhitelist": [
"*"
],
"stateful": true,
"overwriteDuplicate": true
}
}
PUT /api/v2/hoverfly/mode¶
Changes the mode of the running instance of Hoverfly. Pass additional arguments to set the mode options.
Example request body
{
"mode": "capture",
"arguments": {
"headersWhitelist": [
"*"
],
"stateful": true,
"overwriteDuplicate": true
}
}
GET /api/v2/hoverfly/usage¶
Gets metrics information for the running instance of Hoverfly.
Example response body
{
"metrics": {
"counters": {
"capture": 0,
"modify": 0,
"simulate": 0,
"synthesize": 0
}
}
}
GET /api/v2/hoverfly/version¶
Gets the version of Hoverfly.
Example response body
{
"version": "v0.10.1"
}
GET /api/v2/hoverfly/upstream-proxy¶
Gets the upstream proxy configured for Hoverfly.
Example response body
{
"upstreamProxy": "proxy.corp.big-it-company.org:8080"
}
GET /api/v2/hoverfly/pac¶
Gets the PAC file configured for Hoverfly. The response contains plain text with PAC file. If no PAC was provided before, the response is 404 with contents:
{
"error": "Not found"
}
PUT /api/v2/hoverfly/pac¶
Sets the PAC file for Hoverfly.
DELETE /api/v2/hoverfly/pac¶
Unsets the PAC file configured for Hoverfly.
GET /api/v2/cache¶
Gets the requests and responses stored in the cache.
Example response body
{
"cache": [
{
"key": "2fc8afceec1b6bcf99ff1f547c1f5b11",
"matchingPair": {
"request": {
"path": [{
"matcher": "exact",
"value": "hoverfly.io"
}]
},
"response": {
"status": 200,
"body": "response body",
"encodedBody": false,
"headers": {
"Hoverfly": [
"Was-Here"
]
}
}
},
"headerMatch": false,
"closestMiss": null
}
]
}
DELETE /api/v2/cache¶
Delete all requests and responses stored in the cache.
GET /api/v2/logs¶
Gets the logs from Hoverfly.
It supports multiple parameters to limit the amount of entries returned:
limit- Maximum amount of entries. 500 by default;from- Timestamp to start filtering from.
Running hoverfly with -logs-size=0 disables logging and 500 response is returned with body:
{
"error": "Logs disabled"
}
Example response body
{
"logs": [
{
"level": "info",
"msg": "serving proxy",
"time": "2017-03-13T12:22:39Z"
},
{
"destination": ".",
"level": "info",
"mode": "simulate",
"msg": "current proxy configuration",
"port": "8500",
"time": "2017-03-13T12:22:39Z"
},
{
"destination": ".",
"Mode": "simulate",
"ProxyPort": "8500",
"level": "info",
"msg": "Proxy prepared...",
"time": "2017-03-13T12:22:39Z"
},
]
}
GET /api/v2/journal¶
Gets the journal from Hoverfly. Each journal entry contains both the request Hoverfly received and the response it served along with the mode Hoverfly was in, the time the request was received and the time taken for Hoverfly to process the request. Latency is in milliseconds. It also returns its corresponding indexes.
It supports paging using the offset and limit query parameters.
It supports multiple parameters to limit the amount of entries returned:
limit- Maximum amount of entries. 500 by default;offset- Offset of the first element;to- Timestamp to start filtering to;from- Timestamp to start filtering from;sort- Sort results in format “field:order”. Supported fields:timestartedandlatency. Supported orders:ascanddesc.
Running hoverfly with -journal-size=0 disables logging and 500 response is returned with body:
{
"error": "Journal disabled"
}
Example response body
{
"journal": [
{
"request": {
"path": "/",
"method": "GET",
"destination": "hoverfly.io",
"scheme": "http",
"query": "",
"body": "",
"headers": {
"Accept": [
"*/*"
],
"Proxy-Connection": [
"Keep-Alive"
],
"User-Agent": [
"curl/7.50.2"
]
}
},
"response": {
"status": 502,
"body": "Hoverfly Error!\n\nThere was an error when matching\n\nGot error: Could not find a match for request, create or record a valid matcher first!",
"encodedBody": false,
"headers": {
"Content-Type": [
"text/plain"
]
}
},
"id":"mOBdPSIIBbjNqBvpZ8H-",
"mode": "simulate",
"timeStarted": "2017-07-17T10:41:59.168+01:00",
"latency": 0.61334
}
],
"indexes": [
{
"name": "Request.destination",
"entries": [
{
"key": "hoverfly.io",
"journalEntryId": "mOBdPSIIBbjNqBvpZ8H-"
}
]
}],
"offset": 0,
"limit": 25,
"total": 1
}
DELETE /api/v2/journal¶
Delete all entries stored in the journal.
POST /api/v2/journal¶
Filter and search entries stored in the journal.
Example request body
{
"request": {
"destination": [{
"matcher": "exact",
"value": "hoverfly.io"
}]
}
}
GET /api/v2/journal/index¶
Gets all the journal indexes from Hoverfly. Each Index contains key, extracted value for that particular key and journal index id to which it is pointing to.
Example response body
[
{
"name": "Request.QueryParam.id",
"entries": [
{
"key": "100",
"journalEntryId": "ZCyiQtamEtwi-NNU9RT1"
},
{
"key": "101",
"journalEntryId": "YFU5dm2uDZ4UStX3ldkX"
}
]
},
{
"name": "Request.QueryParam.name",
"entries": [
{
"key": "Test1",
"journalEntryId": "ZCyiQtamEtwi-NNU9RT1"
},
{
"key": "Test2",
"journalEntryId": "YFU5dm2uDZ4UStX3ldkX"
}
]
},
]
POST /api/v2/journal/index¶
Allow a user to set journal indexing by specifying index key/name. Index name is “request-query” shares the same syntax as the one for templating, such as Request.QueryParam.myParam or Request.Header.X-Header-Id.[1] It’s used for extracting the data from the journal entry to use as a key for that entry. It returns all the journal indexes that have been set. It indexes pre-existing or new journal entries.
Example request body
{
"name":"Request.QueryParam.myParam"
}
DELETE /api/v2/journal/index/:index-name¶
Deletes journal index from hoverfly.
GET /api/v2/state¶
Gets the state from Hoverfly. State is represented as a set of key value pairs.
Example response body
{
"state": {
"page_state": "CHECKOUT"
}
}
DELETE /api/v2/state¶
Deletes all state from Hoverfly.
PUT /api/v2/state¶
Deletes all state from Hoverfly and then sets the state to match the state in the request body.
Example request body
{
"state": {
"page_state": "CHECKOUT"
}
}
PATCH /api/v2/state¶
Updates state in Hoverfly. Will update each state key referenced in the request body.
Example request body
{
"state": {
"page_state": "CHECKOUT"
}
}
GET /api/v2/diff¶
Gets all reports containing response differences from Hoverfly. The diffs are represented as lists of strings grouped by the same requests.
Example response body
{
"diff": [{
"request": {
"method": "GET",
"host": "time.jsontest.com",
"path": "/",
"query": ""
},
"diffReports": [{
"timestamp": "2018-03-16T17:45:40Z",
"diffEntries": [{
"field": "header/X-Cloud-Trace-Context",
"expected": "[ec6c455330b682c3038ba365ade6652a]",
"actual": "[043c9bb2eafa1974bc09af654ef15dc3]"
}, {
"field": "header/Date",
"expected": "[Fri, 16 Mar 2018 17:45:34 GMT]",
"actual": "[Fri, 16 Mar 2018 17:45:41 GMT]"
}, {
"field": "body/time",
"expected": "05:45:34 PM",
"actual": "05:45:41 PM"
}, {
"field": "body/milliseconds_since_epoch",
"expected": "1.521222334104e+12",
"actual": "1.521222341017e+12"
}]
}]
}]
}
POST /api/v2/diff¶
Gets reports containing response differences from Hoverfly filtered on basis of excluded criteria provided(i.e. headers and response keys in jsonpath format to exclude). The diffs are in same format as we receive in GET request.
Example request body
{
"excludedHeaders":["Date"],
"excludedResponseFields":["$.time"]
}
DELETE /api/v2/diff¶
Deletes all reports containing differences from Hoverfly.
GET /api/v2/hoverfly/templating-data-source/csv¶
Get all the templating data source for the running instance of Hoverfly. It will return list of data sources that has been uploaded by the user to be queried by the template function in the response.
Example response body
{
"csvDataSources": [
{
"name": "student-marks",
"data": "id,name,marks\n1,Test1,300\n2,Test2,600\n"
}
]
}
PUT /api/v2/hoverfly/templating-data-source/csv¶
Sets new template data source, overwriting the existing template data source for the running instance of Hoverfly. The template CSV data source being set is being queried via template function to generate the response. This API call returns back all the templating data source that have been set.
Example request body
{
"name":"student-marks",
"data":"id,name,marks\n1,Test1,55\n2,Test2,56\n*,Dummy,ABSENT"
}
DELETE /api/v2/hoverfly/templating-data-source/csv/:data-source-name¶
Delete a particular data source for the running instance of Hoverfly. It returns all the remaining template data sources.
DELETE /api/v2/shutdown¶
Shuts down the hoverfly instance.
Simulation schema¶
This is the JSON schema for v5 Hoverfly simulations.
{
"additionalProperties": false,
"definitions": {
"delay": {
"properties": {
"delay": {
"type": "integer"
},
"httpMethod": {
"type": "string"
},
"urlPattern": {
"type": "string"
}
},
"type": "object"
},
"delay-log-normal": {
"properties": {
"httpMethod": {
"type": "string"
},
"max": {
"type": "integer"
},
"mean": {
"type": "integer"
},
"median": {
"type": "integer"
},
"min": {
"type": "integer"
},
"urlPattern": {
"type": "string"
}
},
"type": "object"
},
"field-matchers": {
"properties": {
"matcher": {
"type": "string"
},
"value": {},
"config": {
"properties": {
"ignoreUnknown": {
"type": "boolean"
},
"ignoreOrder": {
"type": "boolean"
},
"ignoreOccurrences": {
"type": "boolean"
}
},
"type": "object"
},
"doMatch": {
"$ref": "#/definitions/field-matchers"
}
},
"type": "object"
},
"headers": {
"additionalProperties": {
"items": {
"type": "string"
},
"type": "array"
},
"type": "object"
},
"literals": {
"properties": {
"name": {
"type": "string"
},
"value": {}
},
"required": ["name", "value"],
"type": "object"
},
"meta": {
"properties": {
"hoverflyVersion": {
"type": "string"
},
"schemaVersion": {
"type": "string"
},
"timeExported": {
"type": "string"
}
},
"required": ["schemaVersion"],
"type": "object"
},
"request": {
"properties": {
"body": {
"items": {
"$ref": "#/definitions/field-matchers"
},
"type": "array"
},
"destination": {
"items": {
"$ref": "#/definitions/field-matchers"
},
"type": "array"
},
"headers": {
"$ref": "#/definitions/request-headers"
},
"path": {
"items": {
"$ref": "#/definitions/field-matchers"
},
"type": "array"
},
"query": {
"$ref": "#/definitions/request-queries"
},
"requiresState": {
"patternProperties": {
".{1,}": {
"type": "string"
}
},
"type": "object"
},
"scheme": {
"items": {
"$ref": "#/definitions/field-matchers"
},
"type": "array"
}
},
"type": "object"
},
"request-headers": {
"additionalProperties": {
"items": {
"$ref": "#/definitions/field-matchers"
},
"type": "array"
},
"type": "object"
},
"request-queries": {
"additionalProperties": {
"items": {
"$ref": "#/definitions/field-matchers"
},
"type": "array"
},
"type": "object"
},
"request-response-pair": {
"properties": {
"request": {
"$ref": "#/definitions/request"
},
"response": {
"$ref": "#/definitions/response"
}
},
"required": ["request", "response"],
"type": "object"
},
"response": {
"properties": {
"body": {
"type": "string"
},
"bodyFile": {
"type": "string"
},
"encodedBody": {
"type": "boolean"
},
"fixedDelay": {
"type": "integer"
},
"headers": {
"$ref": "#/definitions/headers"
},
"logNormalDelay": {
"properties": {
"max": {
"type": "integer"
},
"mean": {
"type": "integer"
},
"median": {
"type": "integer"
},
"min": {
"type": "integer"
}
}
},
"postServeAction": {
"type": "string"
},
"removesState": {
"type": "array"
},
"status": {
"type": "integer"
},
"templated": {
"type": "boolean"
},
"transitionsState": {
"patternProperties": {
".{1,}": {
"type": "string"
}
},
"type": "object"
}
},
"type": "object"
},
"variables": {
"properties": {
"name": {
"type": "string"
},
"function": {
"type": "string"
},
"arguments": {
"type": "array"
}
},
"required": ["name", "function"],
"type": "object"
}
},
"description": "Hoverfly simulation schema",
"properties": {
"data": {
"properties": {
"globalActions": {
"properties": {
"delays": {
"items": {
"$ref": "#/definitions/delay"
},
"type": "array"
},
"delaysLogNormal": {
"items": {
"$ref": "#/definitions/delay-log-normal"
},
"type": "array"
}
},
"type": "object"
},
"literals": {
"items": {
"$ref": "#/definitions/literals"
},
"type": "array"
},
"pairs": {
"items": {
"$ref": "#/definitions/request-response-pair"
},
"type": "array"
},
"variables": {
"items": {
"$ref": "#/definitions/variables"
},
"type": "array"
}
},
"type": "object"
},
"meta": {
"$ref": "#/definitions/meta"
}
},
"required": ["data", "meta"],
"type": "object"
}
Contributing¶
Contributions are welcome! To contribute, please:
Fork the repository
Create a feature branch on your fork
Commit your changes, and create a pull request against Hoverfly’s master branch
In your pull request, please include details regarding your change (why you made the change, how to test it etc).
Learn more about the forking workflow here.
Building, running & testing¶
You will need Go 1.18 . Instructions on how to set up your Go environment can be found here.
git clone https://github.com/SpectoLabs/hoverfly.git
# or: git clone https://github.com/<your_username>/hoverfly.git
cd hoverfly
make build
Notice the binaries are in the target directory.
Finally, to test your build:
make test
Community¶
Chat on the Hoverfly Gitter channel
Join the Hoverfly mailing list
Raise a GitHub Issue
Get in touch with SpectoLabs on Twitter