When the request is passed straight through to the endpoint the apiaxle_* params should be stripped. There are alternatives if people want the params passed through.

I get several errors trying to install apiaxle under CentOS 7

npm install -g apiaxle-proxy

npm WARN deprecated [email protected]: lodash@<3.0.0 is no longer maintained. Upgrade to lodash@^4.0.0.
...snip...
In file included from ../src/libxmljs.cc:3:0:
/usr/lib/node_modules/apiaxle-proxy/node_modules/libxmljs/.node-gyp/4.4.3/include/node/v8.h:7603:16: note: static v8::Local<v8::Boolean> v8::Boolean::New(v8::Isolate*, bool)
 Local<Boolean> Boolean::New(Isolate* isolate, bool value) {
                ^
/usr/lib/node_modules/apiaxle-proxy/node_modules/libxmljs/.node-gyp/4.4.3/include/node/v8.h:7603:16: note:   candidate expects 2 arguments, 1 provided
../src/libxmljs.cc:137:19: error: ‘NewSymbol’ is not a member of ‘v8::String’
       target->Set(v8::String::NewSymbol("libxml"), target);
                   ^
../src/libxmljs.cc: In function ‘v8::Handle<v8::Value> libxmljs::ThrowError(const char*)’:
../src/libxmljs.cc:97:1: warning: control reaches end of non-void function [-Wreturn-type]
 }
 ^
make: *** [Release/obj.target/xmljs/src/libxmljs.o] Error 1
make: Leaving directory `/usr/lib/node_modules/apiaxle-proxy/node_modules/libxmljs/build'
gyp ERR! build error
gyp ERR! stack Error: `make` failed with exit code: 2
gyp ERR! stack     at ChildProcess.onExit (/usr/lib/node_modules/npm/node_modules/node-gyp/lib/build.js:276:23)
gyp ERR! stack     at emitTwo (events.js:87:13)
gyp ERR! stack     at ChildProcess.emit (events.js:172:7)
gyp ERR! stack     at Process.ChildProcess._handle.onexit (internal/child_process.js:200:12)
gyp ERR! System Linux 3.10.0-327.10.1.el7.x86_64
gyp ERR! command "/usr/bin/node" "/usr/lib/node_modules/npm/node_modules/node-gyp/bin/node-gyp.js" "rebuild"
gyp ERR! cwd /usr/lib/node_modules/apiaxle-proxy/node_modules/libxmljs
gyp ERR! node -v v4.4.3
gyp ERR! node-gyp -v v3.3.1
gyp ERR! not ok
npm ERR! Linux 3.10.0-327.10.1.el7.x86_64
npm ERR! argv "/usr/bin/node" "/usr/bin/npm" "install" "-g" "apiaxle-proxy"
npm ERR! node v4.4.3
npm ERR! npm  v2.15.1
npm ERR! code ELIFECYCLE

npm ERR! [email protected] install: `node-gyp rebuild`
npm ERR! Exit status 1
npm ERR!
npm ERR! Failed at the [email protected] install script 'node-gyp rebuild'.
npm ERR! This is most likely a problem with the hiredis package,
npm ERR! not with npm itself.
npm ERR! Tell the author that this fails on your system:
npm ERR!     node-gyp rebuild
npm ERR! You can get information on how to open an issue for this project with:
npm ERR!     npm bugs hiredis
npm ERR! Or if that isn't available, you can get their info via:
npm ERR!
npm ERR!     npm owner ls hiredis
npm ERR! There is likely additional logging output above.

Here's the RFC: http://tools.ietf.org/html/rfc2616#section-14.9

The whole thing doesn't need implementing, we can cherry-pick the useful aspects.

Hello,

It would be very usefull to have a response template to keep consistency between API responses and apiAxle responses.
e.g. the common apiAxle response:

{
  "meta": {
    "version": 1,
    "status_code": 403
  },
  "results": {
    "error": {
      "type": "KeyError",
      "message": "No api_key specified."
    }
  }
}

Our current response template:

{
  "status": "OK",
  "response": [],
  "error": null
}

I have configured an api name "places" with endpoint "localhost:2100". My API has many routes, now i want to add capture path to the configured "places" api in ApiAxle so tthat i can get the statistics for that particular API.

I am using the following PUT API From the documentation:

curl -H 'content-type: application/json'
-X PUT
'http://localhost:3000/v1/api/places/addcapturepath/localhost:2100/xx/yy/zz'
But i am getting the error:

{"meta":{"version":1,"status_code":404},"results":{"error":{"type":"NotFoundError","message":"'/v1/api/places/addcapturepath/localhost:2100/xx/yy/zz' not found."}}}

In what way i have to specify the capture path.

started apiaxle-proxy from command-line crashes with errors after a number of minutes.
Env: Ubuntu

With following output:

[2013-08-23 13:42:02.857] [INFO] [default] - Loaded configuration from
[2013-08-23 13:42:02.932] [INFO] [default] - Loaded getcatchall, postcatchall, putcatchall, deletecatchall, headcatchall, patchcatchall from '/usr/lib/node_modules/apiaxle-proxy/app/controller/__controller.{js,coffee}'
[2013-08-23 13:42:03.011] [INFO] [default] - Loaded apifactory, apilimits, statcounters, stattimers, cache, counters, keyfactory, keyringfactory, register, stats from '/usr/lib/node_modules/apiaxle-proxy/node_modules/apiaxle-base/lib/../app/model/redis/_.{js,coffee}'
[2013-08-23 13:42:03.018] [INFO] [default] - Listening at 127.0.0.1:3000

/usr/lib/node_modules/apiaxle-proxy/node_modules/redis/index.js:551
throw err;
^
TypeError: Cannot read property 'value' of undefined
at /usr/lib/node_modules/apiaxle-proxy/node_modules/async/lib/async.js:158:23
at iterate (/usr/lib/node_modules/apiaxle-proxy/node_modules/async/lib/async.js:118:13)
at /usr/lib/node_modules/apiaxle-proxy/node_modules/async/lib/async.js:129:25
at /usr/lib/node_modules/apiaxle-proxy/node_modules/async/lib/async.js:160:17
at RedisClient. (/usr/lib/node_modules/apiaxle-proxy/node_modules/async/lib/async.js:458:34)
at RedisClient.EventEmitter.emit (events.js:92:17)
at RedisClient.on_ready (/usr/lib/node_modules/apiaxle-proxy/node_modules/redis/index.js:334:10)
at RedisClient.on_info_cmd (/usr/lib/node_modules/apiaxle-proxy/node_modules/redis/index.js:365:14)
at /usr/lib/node_modules/apiaxle-proxy/node_modules/redis/index.js:389:14
at try_callback (/usr/lib/node_modules/apiaxle-proxy/node_modules/redis/index.js:548:9)
Worker 1515 died.

Unlimited qpd causes headers to show qps in the place of qpd.

To duplicate, set a key's limits to qpd: -1, qps: 20.
First response should give you headers:

< X-ApiaxleProxy-Qpd-Left: 19
< X-ApiaxleProxy-Qps-Left: undefined

According to the help documentation, when ues the configuration file[integration.json] to start the server, but receive ApiUnknown Error.Look up redis database，i guess the key contain 'development' leads to this error,i want to know how to change this “Development” field?

Unless I have missed it, we need a way to append friendly names to api keys in order to know which key belongs to which client and a way to list all keys (without including the keyless access keys) and a way to search for a specific key based on the friendly name.

Right now is impossible (unless like I said I missed the way to do it) to manage a lot of clients.

Means we also need to take account of certs too.

If this date:
https://github.com/apiaxle/apiaxle/blob/develop/api/test/app/controller/v1/stats_controller_test.coffee#L49
is in the past, then redis expires stats immediately during the test, causing the tests to fail.

Expiring being done here:
https://github.com/apiaxle/apiaxle/blob/develop/base/app/model/redis/stats.coffee#L67

API docs are a bit sketchy. Need filling in properly.

Hi,

I think that for performance critical applications it would be nice to have a task that converts the application into a (consolidated) C application. What do you think about this?

Whoa, this project is awesome! I can think of so many useful scenarios..
I really like it!

The forApis parameters only supports adding new Apis, it does not remove Apis e.g. :
if I do a PUT with this array:

{
  forApis: [1,2,3]
}

and then a PUT with this array

{
  forApis: [1,3]
}

the API 2 will still be linked to the key

https://github.com/apiaxle/apiaxle/blob/develop/circle.yml

drwxrwxr-x 11 peter peter 4.0K Sep 23 13:25 .
drwxrwxrwx 53 peter peter 4.0K Sep 23 13:25 ..
drwxrwxr-x  7 peter peter 4.0K Sep 23 13:25 api
drwxrwxr-x  6 peter peter 4.0K Sep 23 13:25 base
drwxrwxr-x  2 peter peter 4.0K Sep 23 13:25 bin
lrwxrwxrwx  1 peter peter   39 Sep 23 13:25 circle.yml -> /home/phil/projects/apiaxle/.travis.yml
-rw-rw-r--  1 peter peter  706 Sep 23 13:25 coffeelint.json

A global TTL (set in configuration file?) for all calls across a single API.

Sometime I need to track, who is consuming my resources. Knowing api_key from logs, I want to find which keyring it belongs to.

In current version of ApiAxle API I did not find the way to list keyring(s), a key belongs to. The only method is from the other side - loop through all keyrings and list their apikeys, checking, where is they key I care about mentioned.

As workaround I could find that information in Redis:

$ redis-cli
127.0.0.1:6379> hgetall "gk:development:key:javl-keyrings"
1) "jan.vlcinsky"
2) "1"
127.0.0.1:6379> 

It would be great to see that option via API too.

(PS: I am a Python guy, so it would take me a bit longer to add such modification in your code, but I may try later on, if there is a need and time)

For example:

http://my.api.com/resource?foo=bar&foo=baz

The above query string is being proxied to the endPoint as ?foo=bar,baz

I've got a couple test APIs configured (one of which is the apiaxle-api), and a test key.

I've made a bunch of requests through both APIs, but when I fetch stats, I get nothing.

I'm putting together our "production" setup, so when I set apiaxle-proxy to start, I'm not using the -q option when I start.

I'm assuming this is why I'm not getting stats.

How does the apiaxle-proxy-event-processor work?

Is it run on a cron? Is it another service to start up?

Or, do you just recommend running with the -q option?

Thanks!

Hi, i have tried to add and delete some keys and to link them to a keyring and see what happens.
And i have found that a key is not unlinked from keyring when it is deleted.

Do you know this already or maybe is a my fault for some configuration problem or so?

Similar to the pattern used in Netflix's Zuul, providing a framework to register filters would allow API Axle developers the ability to create a set of actions that would run on the request where applicable before it's passed to the origin, and likewise more actions to be applied to the response on the way out.

These actions might be to do something simple like adding headers or fields to the request/response body (very easily done with JSON in node/coffeescript) or something more complex like creating callouts to other services to do some sort of authorisation to add a JWT header or grab data from another API necessary for the request or for a mashup response.

See IZuulFilter for the basic interface which is simply boolean shouldFilter() to decide whether to run the filter and run() to run it.

Also see the abstract ZuulFilter to see the base filter which includes some more functions, most importantly ordering where necessary (otherwise filters can run concurrently)

It would be useful to monitor the following things:

• The average, max and min time taken for each HTTP request within the same time periods and RRD style as the HTTP status statistics.

I guess this means we need to store, for each time period:

• current average
• count making up the current score
• minimum value
• maximum value

ApiAxle looks really promising thus far. But I can't find a definite answer if there is ACL (Access Control)?

For example, I want to have user X with it's API key to only be able to access /api/apicall1 and /api/apicall2 nothing else.

Is this possible?

Thanks.

I'm hoping you can help me optimize the configuration of the proxy for our service load test.
We are using to stress test through apiaxle with 4800 connections over 5 minutes and I think we have reached the limit apiaxle, at least as we have it configured.

Our setup:
APIAxle is running in a docker container behind and nginx reverse proxy.
We have a cluster of 6 machines with round robin load balancing for each service via haproxy.
The complete path for an incoming request is:
nginx -> haproxy -> api axle (6 instances 8 forks on 8 core machines) -> our service

When I remove apiaxle from the chain, we have no problems. When apiaxle is included in that chain we quickly run into 504s from apiaxle and 502 from nginx. I'm inclined to believe that the 504 timeouts are not due to slowness in our service, since without apiaxle in the chain, our response times are still ~60ms during the load test

I was hoping you might have some ideas for things I can try for optimization.
Thanks

Some APIs specify the key as part of the path:

http://api-merchants.skimlinks.com/merchants/xml/aabbccddeeffgghhiijjkk1122334455/categories
http://api-merchants.skimlinks.com/merchants/xml/aabbccddeeffgghhiijjkk1122334455/domains/limit/1/start/100
http://api-merchants.skimlinks.com/merchants/xml/aabbccddeeffgghhiijjkk112233445566/search/con/limit/2/start/0?filter_by=preferred
http://api-merchants.skimlinks.com/merchants/xml/aabbccddeeffgghhiijjkk1122334455/category/2/limit/1

We would need to give an API owner a way to extract the key from the URL and use that in place of apiaxle_key or api_key.

The Skimlinks merchant API uses get requests with the public API key baked into the api path:

http://merchantsapi.stuart.skimlinks.com/merchants/xml/1b908e69c96485706095ed1b37bea00a/search/amazon.com

This is not currently supported in axle, perhaps we could add a special place holder:

http://merchantsapi.stuart.skimlinks.com/merchants/xml/:KEY:/search/amazon.com

I tried first with

npm install apiaxle-api

but it failed exactly for the same reason as #81
As I saw the fix on the development branch #82 , so I downloaded the source and compiled with make.
I could compile it without any problems. But after that when I try to npm install <folder-to-apiaxle-base> it fails still:

In file included from ../src/libxmljs.h:7:0,
                 from ../src/libxmljs.cc:7:
../node_modules/nan/nan.h: At global scope:
../node_modules/nan/nan.h:592:20: error: variable or field ‘AddGCEpilogueCallback’ declared void
       v8::Isolate::GCEpilogueCallback callback
                    ^
../node_modules/nan/nan.h:592:7: error: ‘GCEpilogueCallback’ is not a member of ‘v8::Isolate’
       v8::Isolate::GCEpilogueCallback callback
       ^
../node_modules/nan/nan.h:593:18: error: expected primary-expression before ‘gc_type_filter’
     , v8::GCType gc_type_filter = v8::kGCTypeAll) {
                  ^
../node_modules/nan/nan.h:598:20: error: variable or field ‘RemoveGCEpilogueCallback’ declared void
       v8::Isolate::GCEpilogueCallback callback) {
                    ^
../node_modules/nan/nan.h:598:7: error: ‘GCEpilogueCallback’ is not a member of ‘v8::Isolate’
       v8::Isolate::GCEpilogueCallback callback) {
       ^
../node_modules/nan/nan.h:603:20: error: variable or field ‘AddGCPrologueCallback’ declared void
       v8::Isolate::GCPrologueCallback callback
                    ^
../node_modules/nan/nan.h:603:7: error: ‘GCPrologueCallback’ is not a member of ‘v8::Isolate’
       v8::Isolate::GCPrologueCallback callback
       ^
../node_modules/nan/nan.h:604:18: error: expected primary-expression before ‘gc_type_filter’
     , v8::GCType gc_type_filter = v8::kGCTypeAll) {
                  ^
../node_modules/nan/nan.h:609:20: error: variable or field ‘RemoveGCPrologueCallback’ declared void
       v8::Isolate::GCPrologueCallback callback) {
                    ^
../node_modules/nan/nan.h:609:7: error: ‘GCPrologueCallback’ is not a member of ‘v8::Isolate’
       v8::Isolate::GCPrologueCallback callback) {
       ^
../node_modules/nan/nan.h: In function ‘bool Nan::SetAccessor(v8::Local<v8::Object>, v8::Local<v8::String>, Nan::GetterCallback, Nan::SetterCallback, v8::Local<v8::Value>, v8::AccessControl, v8::PropertyAttribute)’:
../node_modules/nan/nan.h:1939:16: warning: ‘bool v8::Object::SetAccessor(v8::Local<v8::Name>, v8::AccessorNameGetterCallback, v8::AccessorNameSetterCallback, v8::Local<v8::Value>, v8::AccessControl, v8::PropertyAttribute)’ is deprecated (declared at /home/gulyas/.node-gyp/6.3.1/include/node/v8.h:2735): Use maybe version [-Wdeprecated-declarations]
     , attribute);
                ^
../src/libxmljs.cc: In function ‘void libxmljs::xmlMemFreeWrap(void*)’:
../src/libxmljs.cc:56:24: warning: ‘static bool v8::V8::IsDead()’ is deprecated (declared at /home/gulyas/.node-gyp/6.3.1/include/node/v8.h:8462): Use isolate version [-Wdeprecated-declarations]
     if (v8::V8::IsDead())
                        ^
make: *** [Release/obj.target/xmljs/src/libxmljs.o] Error 1
make: Leaving directory `/home/gulyas/Projects/gli-solutions/apiaxle/apiaxle-user/node_modules/libxmljs/build'
gyp ERR! build error 
gyp ERR! stack Error: `make` failed with exit code: 2
gyp ERR! stack     at ChildProcess.onExit (/usr/lib/node_modules/npm/node_modules/node-gyp/lib/build.js:276:23)
gyp ERR! stack     at emitTwo (events.js:106:13)
gyp ERR! stack     at ChildProcess.emit (events.js:191:7)
gyp ERR! stack     at Process.ChildProcess._handle.onexit (internal/child_process.js:204:12)
gyp ERR! System Linux 3.16.0-77-generic
gyp ERR! command "/usr/bin/nodejs" "/usr/lib/node_modules/npm/node_modules/node-gyp/bin/node-gyp.js" "rebuild"
gyp ERR! cwd /home/gulyas/Projects/gli-solutions/apiaxle/apiaxle-user/node_modules/libxmljs
gyp ERR! node -v v6.3.1
gyp ERR! node-gyp -v v3.4.0
gyp ERR! not ok 
[email protected] /home/gulyas/Projects/gli-solutions/apiaxle/apiaxle-user
├── [email protected]  extraneous
├── [email protected]  extraneous
├── [email protected]  extraneous
├── [email protected]  extraneous
├── [email protected]  extraneous
├── [email protected]  extraneous
└── [email protected]  extraneous

npm WARN [email protected] No description
npm WARN [email protected] No repository field.
npm ERR! Linux 3.16.0-77-generic
npm ERR! argv "/usr/bin/nodejs" "/usr/bin/npm" "install" "../apiaxle/base"
npm ERR! node v6.3.1
npm ERR! npm  v3.10.6
npm ERR! code ELIFECYCLE

npm ERR! [email protected] install: `node-gyp rebuild`
npm ERR! Exit status 1
npm ERR! 
npm ERR! Failed at the [email protected] install script 'node-gyp rebuild'.
npm ERR! Make sure you have the latest version of node.js and npm installed.
npm ERR! If you do, this is most likely a problem with the libxmljs package,
npm ERR! not with npm itself.
npm ERR! Tell the author that this fails on your system:
npm ERR!     node-gyp rebuild
npm ERR! You can get information on how to open an issue for this project with:
npm ERR!     npm bugs libxmljs
npm ERR! Or if that isn't available, you can get their info via:
npm ERR!     npm owner ls libxmljs
npm ERR! There is likely additional logging output above.

npm-debug.log.zip

For the design of our API we would sometimes like to have no version tag in a URI.

(As a policy choice, the plan is to add a version tag if we do a breaking change, but otherwise omit the tag, in order to keep the URI short and pretty. So /foo would be succeeded by /v2/foo).

There appears to be no way to do this with ApiAxle as it is currently implemented. Is that right?

I have an outside system that is provisioning API keys on ApiAxle, and would like for ApiAxle to expire these keys after a set period of time, similar to the temporary keys that are created in "keyless" mode, but the TTL would be set on a per key basis at the time of key creation. One hitch is these keys are attached to keyrings, so this relationship would also need expiring.

Thanks.

429 responses (and possibly other errors?) do not contain X-ApiaxleProxy-* headers.

Eg. when exceeding a QPS limit, the client cannot see how many QPD they have left

< HTTP/1.1 429 Too Many Requests
< Content-Type: application/json
< Date: Tue, 13 Oct 2015 12:52:32 GMT
* Server nginx/1.4.6 (Ubuntu) is not blacklisted
< Server: nginx/1.4.6 (Ubuntu)
< Content-Length: 159
< Connection: keep-alive
< 

At the moment we have stats for individual keys at /v1/stats/:key/all (this endpoint is wrong, should be at /v1/keys/:key/stats). Would be good to have them for entire APIs, at, say /v1/api/:api/stats/all

Need to decide on which licence(s) to use. When that's done:

• Add a licence to the source code.
• Document on the front page of the website.

Lists keys that belong to a particular api. Tasks are probably:

• Create the aggregation in the database.
• Build a new controller from ListController
  • Tests, docs

Ranges should be specified with query parameters rather than as URL segments.
This will require updates to the node-axle wrapper library
https://github.com/stuartquin/node-axle

The way in which apiaxle returns its own errors is different to our API. For example, when an API key is not valid:

{"meta":{"version":1,"status_code":403},"results":{"error":{"type":"KeyError","message":"'xxx' is not a valid key."}}}

This error message includes information about apiaxle itself. We don't want to show this to users. It would be very useful if it was possible to configure apiaxle to return errors in a particular format.

 { protocol: 'https',
 apiFormat: 'json',
 endPointTimeout: 30,
 disabled: false,
 strictSSL: true,
 sendThroughApiKey: false,
 sendThroughApiSig: false,
 hasCapturePaths: false,
 allowKeylessUse: false,
 keylessQps: 2,
 keylessQpd: 172800,
 endPoint: 'sample.api.com:18201',  // custom port
 createdAt: 1396446648145,
 updatedAt: 1396540278244 }

port is missing

[2014-04-03 09:54:31.924] [DEBUG] [default] - Backend: POST to 'https://sample.api.com/

It would be good to support a broader range of param names for the apikey in the request, or even specify a custom value on api creation.
I have an api that uses api_key, the workaround of also including apikey as a parameter suffices but it could be cleaner.

Thanks

It would be nice to add the following to the stats endpoint:

Just show 200s:

&filter_status=200

Just show uncached:

&filter_type=uncached

So uncached 400s:

&filter_type=uncached&filter_status=400

I am using ApiAxle for my node.js API. In that i followed this tutorial "http://www.cubrid.org/blog/cubrid-appstools/apiaxle-open-source-api-management-analytics-proxy/". After that i tried to get the statistics by using the following url:

"http://localhost:3000/v1/apis/charts"
"http://localhost:3000/v1/keys/charts".

But the result is always coming as empty "results:{}"

Details for this to be set in the configuration file. Allow arbitrary timed caching based on regexps. Config excerpt might look like this:

{
  "^/products/.*$": 20,
  "^/users/.+/details": 200
}

Should be a way of expiring the cache through a header (check RFC for which one).

I can not figure out how to configure my json file so that when apiaxle connects to redis it uses our auth key. I have looked on the site and see the config example which only includes host and port. Does ApiAxle support connecting to a redis instance that requires auth?

It's possible for people to override the 'magic' all keys for models. I would suggest we move the objects to a new namespace eg:

gk:development:apifactory:facebook

becomes:

gk:development:apifactory:api:facebook

Meaning a model gets much more namespace maneuverability. Sound sensible?

Issue with installing apiaxle modules: can't install apiaxle because [email protected] has compilation problems. There is no problems with [email protected]
papandreou/node-cldr#16
Issue on libxmljs's github: libxmljs/libxmljs#238

Hello,

It would be nice to be able to authenticate directly via the API.

Architecturally this would separate the proxy from apiaxle allowing different modes of working.

For example, applications could call the API directly from inside (without the need for a proxy) or proxy servers, such as nginx, could be used to front applications whilst delegating to apiaxle for authentication, limiting, monitoring etc. on the APIs.

Thanks
Peter

My coffee script understand basically only comes from reading ApiAxle source code, but I think the variables here are back to front:

Hi, I am configuring ApiAxle in Ubuntu 12 and i am getting an error when i try to get an url from the api.

The error is "DEPTH_ZERO_SELF_SIGNED_CERT".

I am using Node.js and restify. The protocol used is https, and when i write my custom api url in the browser, i get a correct answerd, but when i use apiaxle i receive that error.

If i change https to http, it works!!

Commands i used:

api “myapi” create endPoint=”localhost:3001” protocol=”https”
key “1234” create
api myapi linkkey “1234”

And used this command to test it

curl 'http://myapi.api.localhost:3000/user/new?api_key=1234'

Could you give some advice or knowledgement to fix it?

Thanks, best regards

Both the Api and the Proxy handle well Redis disconnection and try to reconnect, it is not the case of the event queue processor.

Step to reproduce:
restart Redis when the queue processor is running,
after Redis restart, the queue just grow.

Making a call to /v1/api/facebook/keys?from=0&to=10&resolve=true seems to return all keys regardless of the API they belong to.

Will try to get a look at it today

This is probably a bit silly, but it would be nice if there is an exit/quit command instead of using Ctrl+C to get out of the api repl.

It's just a command that you usually expect a shell to have :)

Cheeers,

Alex

Somewhat low priority, but why not. Appears to happen with any nonexistant argument after help, exits back to shell.

axle> help xoxo

/usr/lib/node_modules/apiaxle-repl/lib/repl.js:99
        command_object = new this.constructor.all_commands[subcommand](this.ap
                         ^
TypeError: undefined is not a function
    at ReplHelper.exports.ReplHelper.ReplHelper.runCommands (/usr/lib/node_modules/apiaxle-repl/lib/repl.js:99:26)
    at ReplHelper.exports.ReplHelper.ReplHelper.processLine (/usr/lib/node_modules/apiaxle-repl/lib/repl.js:137:17)
    at /usr/lib/node_modules/apiaxle-repl/lib/repl.js:144:20
    at Interface._onLine (readline.js:200:5)
    at Interface._line (readline.js:531:8)
    at Interface._ttyWrite (readline.js:760:14)
    at ReadStream.onkeypress (readline.js:99:10)
    at ReadStream.EventEmitter.emit (events.js:98:17)
    at emitKey (readline.js:1095:12)
    at ReadStream.onData (readline.js:840:14)