mcohen01 / amazonica Goto Github PK

I couldn't discern how to do this from the readme and found no examples online.

My use case is setting the public/read permission on newly-uploaded files for the purposes of website hosting.

I'd like to be able to use amazonica in applications running on EC2 instances. Instead of storing the credentials in the code or a config file, I'd like to use an IAM role on the instance to authenticate to the API.

The Java SDK developer guide describes how the SDK will do this:

If your application software constructs a client object for an AWS service using an overload of the constructor that does not take any parameters, the constructor searches the "credentials provider chain." The credentials provider chain is the set of places where the constructor attempts to find credentials if they are not specified explicitly as parameters. For Java, the credentials provider chain is:

• Environment Variables: AWS_ACCESS_KEY_ID and AWS_SECRET_KEY
• Java System Properties: aws.accessKeyId and aws.secretKey
• Instance Metadata Service, which provides the credentials associated with the IAM role for the EC2 instance

I did some experimenting with this yesterday and it wasn't obvious to me. Is it possible to use IAM roles with amazonica?

Thanks,
Dave

Is there any way to change the region endpoint for DynamoDB client? The following code creates table "TestTable" in the US East region, despite I set :endpoint to "eu-west-1":

(def cred {:access-key "aws-access-key"
           :secret-key "aws-secret-key"
           :endpoint "eu-west-1"
           :client-config {:proxy-host "my-proxy"
                           :proxy-port 8080}})

(create-table cred :table-name "TestTable"
  ; ....
  )

amazonica 0.1.22

(Thanks for resolving the put-record issue.)

I am now having an issue consuming the records written to Kinesis. This command:

(kinesis/get-records
  cred
  :shard-iterator shard-iterator
  :limit batch-limit)

Results in this error:

Exception in thread "main" java.lang.IllegalArgumentException: No value supplied for key: 2, compiling:(/private/var/folders/g8/1b2_h6yx7t7csbtr7g4x9qvh0000gn/T/form-init4194254269444512279.clj:1:142)

Where batch-limit is set to "2". When I removed :limit, the same error cited key as the shard-iterator value, so it seems like an argument counting/position issue. Thanks.

I'm having trouble getting delete-objects to work. For delete-object (singular), the following seems to work:

(delete-object {:bucket-name "my-bucket" :key "key1"})

So by analogy between DeleteObjectRequest and DeleteObjectsRequest, I expected this to work:

(delete-objects {:bucket-name "my-bucket" :keys ["key1" "key2"]})

but instead I get:

UnsupportedOperationException: nth not supported on this type: Character
    at clojure.lang.RT.nthFrom(RT.java:857)

Am I wrong to expect delete-objects to work this way?

Only the administrative functions for CloudSearchV2 are exposed. I would like to use the search functions in CloudSearchDomain to actually build and execute a query. Is it possible to add these?

One of the possible exceptions thrown by IRecordProcessorCheckpointer checkpoint() is KinesisClientLibDependencyException according to

https://github.com/awslabs/amazon-kinesis-client/blob/master/src/main/java/com/amazonaws/services/kinesis/clientlibrary/interfaces/IRecordProcessorCheckpointer.java

(NB. This is version 1.1.0 of the Kinesis client library - so possibly the exception is new).

This exception is not currently handled in kinesis.clj mark-checkpoint. The Amazon comments suggest "...the application can backoff and retry."

Should KinesisClientLibDependencyException be handled similar to ThrottlingException?

project.clj says
[com.amazonaws/amazon-kinesis-client "1.0.0"]

So perhaps this doesn't apply just yet.

I ran into something that I think might be a bug, and am interested to know if you have any ideas: https://gist.github.com/anonymous/d4dad5ff47ae7e92f7a5

The problem is that the pending atom always winds up as -1.

I thought this was odd, and so I tested it a bit in the REPL, and it appears that after eval'ing the buffer containing this code in emacs and running a single upload, :started never is not seen by the progress-listener. This only happens the first time the upload function is called.

Once a single upload has been processed, all future uploads seem to see the :started event, and inc the pending atom accordingly.

Any thoughts?

Thanks

Amazonica's section on auth says:

The default authentication scheme is to use the chained Provider class from the AWS SDK, whereby authentication is attempted in the following order:

• Environment Variables - AWS_ACCESS_KEY_ID and AWS_SECRET_KEY
• Java System Properties - aws.accessKeyId and aws.secretKey
• Instance profile credentials delivered through the Amazon EC2 metadata service

Only listing 3 options, but the AWS Default chain (http://docs.aws.amazon.com/AWSJavaSDK/latest/javadoc/com/amazonaws/auth/DefaultAWSCredentialsProviderChain.html) actually lists 4. The missing one is:

• Credential profiles file at the default location (~/.aws/credentials) shared by all AWS SDKs and the AWS CLI

This isn't just a documentation discrepancy, it appears that Amazonica really doesn't look for this file in it's Chain.

worker in kinesis.clj has the code

:or {checkpoint 60000 ...

I'm not clear on exactly what's happening with opts, but is there a mismatched units bug here since processor-factory multiplies checkpoint by 1000 (presumably to convert from seconds to milliseconds) ?

(reset! next-check (+' (System/currentTimeMillis) (*' 1000 checkpoint))))

Hi,

I wanted to create security groups and assign rules to them. Also wanted to create a VPC and subnets. I do not see any api exposed for that. Can u help me how I should proceed ?

Thanks,
Murtaza

Function ex-info does not exist.

http://docs.aws.amazon.com/AWSJavaSDK/latest/javadoc/com/amazonaws/services/elasticmapreduce/AmazonElasticMapReduceClient.html#describeJobFlows(com.amazonaws.services.elasticmapreduce.model.DescribeJobFlowsRequest)

I think they want you to use describe-step or describe-cluster now along with the list methods for those.

I'm calling:

(amazonica.aws.s3/put-object bucket-name key-name (io/input-stream a-file) {:some :metadata} )

com.amazonaws.AmazonClientException: Unable to unmarshall error response (Premature end of file.). Response Code: 400, Response Text: Bad Request
    at com.amazonaws.http.AmazonHttpClient.handleErrorResponse(AmazonHttpClient.java:792)
    at com.amazonaws.http.AmazonHttpClient.executeHelper(AmazonHttpClient.java:421)
    at com.amazonaws.http.AmazonHttpClient.execute(AmazonHttpClient.java:232)
    at com.amazonaws.services.s3.AmazonS3Client.invoke(AmazonS3Client.java:3566)
    at com.amazonaws.services.s3.AmazonS3Client.putObject(AmazonS3Client.java:1434)
    at com.amazonaws.services.s3.AmazonS3Client.putObject(AmazonS3Client.java:1275)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:483)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:483)
    at clojure.lang.Reflector.invokeMatchingMethod(Reflector.java:93)
    at clojure.lang.Reflector.invokeInstanceMethod(Reflector.java:28)
    at amazonica.core$fn_call$fn__1614.invoke(core.clj:718)
    at amazonica.core$intern_function$fn__1633.doInvoke(core.clj:769)
    at clojure.lang.RestFn.invoke(RestFn.java:457)

I'm pretty sure this is caused by the S3Client being GC'd before the file is finished uploading, in amazonica.core/fn-call.

References:

• http://stackoverflow.com/questions/9952815/s3-java-client-fails-a-lot-with-premature-end-of-content-length-delimited-messa
• https://forums.aws.amazon.com/thread.jspa?threadID=83326
• weavejester/clj-aws-s3#3

As an example, see #60.

I don't have experience with reflection in Java, but it seems plausible to me that unhelpful error messages would just come with the territory. If so, feel free to just close this issue as "Won't fix" or something.

P.S. I mean this more as an FYI than as a complaint

I want to use the range feature of the AWS SDK, but I'm not sure if this is supported by Amazonica as it is an option that requires two arguments instead of one.

I tried the following:

(s3/get-object :bucket-name "my-bucket" :key "my-key" :range 1000)
; IllegalArgumentException wrong number of arguments
; sun.reflect.NativeMethodAccessorImpl.invoke0 (NativeMethodAccessorImpl.java:-2)
; sun.reflect.NativeMethodAccessorImpl.invoke (NativeMethodAccessorImpl.java:39)
; sun.reflect.DelegatingMethodAccessorImpl.invoke (DelegatingMethodAccessorImpl.java:25)
; java.lang.reflect.Method.invoke (Method.java:597)
; sun.reflect.GeneratedMethodAccessor15.invoke (:-1)
; sun.reflect.DelegatingMethodAccessorImpl.invoke (DelegatingMethodAccessorImpl.java:25)
; java.lang.reflect.Method.invoke (Method.java:597)
; clojure.lang.Reflector.invokeMatchingMethod (Reflector.java:93)
; clojure.lang.Reflector.invokeInstanceMethod (Reflector.java:28)
; amazonica.core/invoke (core.clj:444)

(s3/get-object :bucket-name "my-bucket" :key "my-key" :range [0 1000])
;; Same error

Hi,

I am trying to add a new rule to a security group I created.

(http://docs.aws.amazon.com/AWSSdkDocsJava/latest/DeveloperGuide/authorize-ingress.html)

(ns aws-infra.security-groups
(:require [amazonica.core :as aws-core :refer [defcredential]]
[amazonica.aws.ec2 :as aws-ec2])))

(aws-ec2/authorize-security-group-ingress
:group-name "test-group"
:ip-permissions [{:cidr-ip "21.21.22.23/32"
:ip-protocol "tcp"
:from-port "22"
:to-port "22"}])

I get the below exception, can you please help ?

java.lang.ClassCastException: java.lang.String cannot be cast to java.lang.Character
RT.java:1087 clojure.lang.RT.intCast
core.clj:846 clojure.core/int
core.clj:314 amazonica.core/coerce-value
core.clj:508 amazonica.core/invoke-method
AFn.java:160 clojure.lang.AFn.applyToHelper

DynamoDB now has support for maps:

http://aws.amazon.com/blogs/aws/dynamodb-update-json-and-more/

With amazonica 0.1.3 running

(get-object db-creds
            :bucket-name "db-bucket"
            :key "db/foo.txt")

doesn't work (invalid arity exception) but

(get-object db-creds
             "db-bucket"
             "db/foo.txt")

works.
Looks like the two-arity string string Java method is getting called in favor of the one-arity GetObjectRequest method.
It's a bit surprising, so either the keyword example should work out of the box or the second example put in the README.

It would be useful to have an AmazonS3EncryptionClient for client-side data encryption prior to upload.

http://aws.amazon.com/articles/2850096021478074/
http://docs.aws.amazon.com/redshift/latest/mgmt/uploading-aws-sdk-for-java-encrypted.html

I'd like to be able to use Amazonica through an HTTP proxy. The Amazon*Clients support using a proxy by passing a suitable configured ClientConfiguration object to their constructors.

I'm not sure what the best interface would be. Perhaps it'd be easiest to read system properties and environment variables to find http proxy configuration, but I'm not sure that's the best solution.

An alternative would be to create a with-configuration macro that sets a dynamic var with config options.

Is this something that you'd consider adding to Amazonica? I'd be happy to come up with a pull request if you are interested.

So Amazonica is a really slick library, particularly the bean-to-Clojure map mapping. I'm writing a s3sync utility that I hope will be more flexible/functional than the commonly used s3sync.rb. Being able to express things as Clojure maps makes things work really well, but user-metadata requires a workaround if you're updating metadata that you pulled from s3.

In particular, :user-metadata is a keyword -> String mapping when downloaded, but must be a String -> String mapping when uploaded.

Example below

user=> (amazonica.aws.s3/get-object-metadata my-credentials my-website "404.html")
{:content-length 6452, :last-modified #<DateTime 2013-04-04T13:21:30.000-05:00>, :content-type "text/html", :raw-metadata {:Content-Type "text/html", :Accept-Ranges "bytes", :Last-Modified #<DateTime 2013-04-04T13:21:30.000-05:00>, :Content-Length 6452, :ETag "2d477e36be6f149b4c559591a6201774"}, :etag "2d477e36be6f149b4c559591a6201774", :user-metadata {:foo "bar"}}

user=> (amazonica.aws.s3/copy-object my-credentials :source-bucket-name my-website :destination-bucket-name my-website :source-key "404.html" :destination-key "404.html" :new-object-metadata {:content-type "text/html" :user-metadata {:foo "bar"}})
ClassCastException clojure.lang.Keyword cannot be cast to java.lang.String  com.amazonaws.services.s3.AmazonS3Client.populateRequestMetadata (AmazonS3Client.java:2634)

user=> (amazonica.aws.s3/copy-object my-credentials :source-bucket-name my-website :destination-bucket-name my-website :source-key "404.html" :destination-key "404.html" :new-object-metadata {:content-type "text/html" :user-metadata {"foo" "bar"}})
{:etag "2d477e36be6f149b4c559591a6201774", :last-modified-date #<DateTime 2013-04-04T13:21:58.000-05:00>}

Using Amazonica 0.1.21.

I'm not sure if this is an issue, or I'm just using the library awkwardly.

If I put my Amazon credentials into environment variable the following works fine:

(use 'amazonica.aws.s3)
(get-object-metadata {:bucket-name "ahjones-test" :key "foo"})

However, if I want to pass in credentials as the first parameter I get a message that says that the best method can't be found.

(def cred {:access-key "key" :access-secret "secret"})
(get-object-metadata cred {:bucket-name "ahjones-test" :key "foo"})

The exception:

IllegalArgumentException Could not determine best method to invoke for get-object-metadata using arguments ({:secret-key "secret", :access-key "key"} {:key "foo", :bucket-name "ahjones-test"})
    amazonica.core/intern-function/fn--1458 (core.clj:705)
    user/eval1649 (form-init5154478138353900686.clj:1)
    clojure.lang.Compiler.eval (Compiler.java:6619)
    clojure.lang.Compiler.eval (Compiler.java:6582)
    clojure.core/eval (core.clj:2852)
    clojure.main/repl/read-eval-print--6588/fn--6591 (main.clj:259)
    clojure.main/repl/read-eval-print--6588 (main.clj:259)
    clojure.main/repl/fn--6597 (main.clj:277)
    clojure.main/repl (main.clj:277)
    clojure.tools.nrepl.middleware.interruptible-eval/evaluate/fn--591 (interruptible_eval.clj:56)
    clojure.core/apply (core.clj:617)
    clojure.core/with-bindings* (core.clj:1788)

However this is OK

(get-object-metadata cred :bucket-name "ahjones-test" :key "foo")

Any idea why I'm getting this exception? :-)

Jul 29, 2013 3:02:06 PM com.amazonaws.http.AmazonHttpClient executeHelper
INFO: Unable to execute HTTP request: peer not authenticated
javax.net.ssl.SSLPeerUnverifiedException: peer not authenticated
at sun.security.ssl.SSLSessionImpl.getPeerCertificates(SSLSessionImpl.java:397)
at org.apache.http.conn.ssl.AbstractVerifier.verify(AbstractVerifier.java:128)
at org.apache.http.conn.ssl.SSLSocketFactory.connectSocket(SSLSocketFactory.java:572)
at org.apache.http.impl.conn.DefaultClientConnectionOperator.openConnection(DefaultClientConnectionOperator.java:180)
at org.apache.http.impl.conn.ManagedClientConnectionImpl.open(ManagedClientConnectionImpl.java:294)
at org.apache.http.impl.client.DefaultRequestDirector.tryConnect(DefaultRequestDirector.java:641)
at org.apache.http.impl.client.DefaultRequestDirector.execute(DefaultRequestDirector.java:480)
at org.apache.http.impl.client.AbstractHttpClient.execute(AbstractHttpClient.java:906)
at org.apache.http.impl.client.AbstractHttpClient.execute(AbstractHttpClient.java:805)
at org.apache.http.impl.client.AbstractHttpClient.execute(AbstractHttpClient.java:784)
at com.amazonaws.http.AmazonHttpClient.executeHelper(AmazonHttpClient.java:315)
at com.amazonaws.http.AmazonHttpClient.execute(AmazonHttpClient.java:199)
at com.amazonaws.services.s3.AmazonS3Client.invoke(AmazonS3Client.java:2994)
at com.amazonaws.services.s3.AmazonS3Client.getObjectMetadata(AmazonS3Client.java:800)
at com.amazonaws.services.s3.AmazonS3Client.getObjectMetadata(AmazonS3Client.java:780)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:613)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:613)
at clojure.lang.Reflector.invokeMatchingMethod(Reflector.java:93)
at clojure.lang.Reflector.invokeInstanceMethod(Reflector.java:28)
at amazonica.core$fn_call$fn__7047.invoke(core.clj:589)
at amazonica.core$intern_function$fn__7059.doInvoke(core.clj:629)
at clojure.lang.RestFn.invoke(RestFn.java:436)
...
SSLPeerUnverifiedException peer not authenticated sun.security.ssl.SSLSessionImpl.getPeerCertificates

I'm cleaning up a lein project's deps, and after shuffling around some namespaces and updating the amazonica project.clj dependency to use 0.1.22, I'm getting a new error when calling s3/get-object:

java.lang.NoClassDefFoundError: org/apache/http/impl/conn/PoolingClientConnectionManager
 at com.amazonaws.http.ConnectionManagerFactory.createPoolingClientConnManager (ConnectionManagerFactory.java:26)
    com.amazonaws.http.HttpClientFactory.createHttpClient (HttpClientFactory.java:87)
    com.amazonaws.http.AmazonHttpClient.<init> (AmazonHttpClient.java:121)
    com.amazonaws.AmazonWebServiceClient.<init> (AmazonWebServiceClient.java:66)
    com.amazonaws.services.s3.AmazonS3Client.<init> (AmazonS3Client.java:304)
    com.amazonaws.services.s3.AmazonS3Client.<init> (AmazonS3Client.java:286)
    sun.reflect.NativeConstructorAccessorImpl.newInstance0 (NativeConstructorAccessorImpl.java:-2)
    sun.reflect.NativeConstructorAccessorImpl.newInstance (NativeConstructorAccessorImpl.java:57)
    sun.reflect.DelegatingConstructorAccessorImpl.newInstance (DelegatingConstructorAccessorImpl.java:45)
    java.lang.reflect.Constructor.newInstance (Constructor.java:526)
    clojure.lang.Reflector.invokeConstructor (Reflector.java:180)
    amazonica.core$create_client.invoke (core.clj:139)
    amazonica.core$amazon_client_STAR_.invoke (core.clj:187)
    clojure.lang.AFn.applyToHelper (AFn.java:167)
    clojure.lang.AFn.applyTo (AFn.java:151)
    clojure.core$apply.invoke (core.clj:617)
    clojure.core$memoize$fn__5049.doInvoke (core.clj:5735)
    clojure.lang.RestFn.invoke (RestFn.java:436)
    amazonica.core$candidate_client.invoke (core.clj:661)
    amazonica.core$fn_call$fn__1670.invoke (core.clj:671)
    clojure.lang.Delay.deref (Delay.java:33)
    clojure.core$deref.invoke (core.clj:2128)
    amazonica.core$fn_call$fn__1672.invoke (core.clj:674)
    amazonica.core$intern_function$fn__1687.doInvoke (core.clj:718)
.
.
.

I tried resetting my amazonica dependency back to the previous version 0.1.15, but I'm still getting the error. It looks like the aws-sdk jar updated the ConnectionManagerFactory code when it bumped to version 1.5.0, which amazonica was including a bit before version 0.1.15 if I'm not mistaken.

Any idea why this is happening now? As part of my project refactoring, I removed some dependencies from the project. Is there any reason why amazonica would be dependent on any other libraries being present in the project?

Hello!

I'm having some issues when deleting a DynamoDB item using amazonica.aws.dynamodbv2.

An example:

(delete-item :table-name "my-table" :key {:id {:s "12345"}})

This fails with a java.lang.IllegalArgumentException: null exception.

Any ideas here? I am wracking my brain trying to understand what I am doing wrong. Thanks!

In the docs it suggests that in order to do explict checkpointing from your kinessis records processor you should set the checkpoint to Long/MAX_VALUE and return true from process_records. However later on in processes-factory checkpoint and System/currentTimeMillis are added together causing an integer overflow error.

https://github.com/mcohen01/amazonica/blob/master/src/amazonica/aws/kinesis.clj#L94

I suspect the quick answer is to update the docs (possibly describing the role that the checkpoint value plays). But I would that there are two follow up issues. First it's not particularly clear the role that :checkpoint plays as a time variable. Second if it's going to be overloaded like this it might make sense top be able to pass in an explicit value for "no automatic checkpointing" like { :checkpoint :no } to make the distinction a little more clear.

Thanks!

I know it's probably not a surprise that SimpleDB doesn't work —given that you don't include a MVS in the readme— but I figured it might be useful for you to track the task in an issue anyway.

(require '[amazonica.core :refer [defcredentials])
(defcredentials "AccessKeyID" "SecretKey")
(require '[amazonica.aws.simpledb :as sdb])
(sdb/put-attributes "domain"
                    "devapiTue Aug 05 23:24:36 UTC 2014"
                    [{:name "args", :value ["Test message"]}
                     {:name "instant", :value #inst "2014-08-05T23:24:36.701-00:00"}
                     {:name "ns", :value "some-api.main"}
                     {:name "file", :value "/tmp/form-init6934682238825582339.clj"}
                     {:name "hostname", :value "devapi"}
                     {:name "output", :value "2014-Aug-05 23:24:36 +0000 devapi WARN [some-api.main] - Test message"}
                     {:name "prefix", :value "2014-Aug-05 23:24:36 +0000 devapi WARN [some-api.main]"}
                     {:name "level", :value :warn}
                     {:name "line", :value nil}
                     {:name "ap-config", :value {}}
                     {:name "error?", :value false}
                     {:name "throwable", :value nil}
                     {:name "timestamp", :value "2014-Aug-05 23:24:36 +0000"}
                     {:name "message", :value "Test message"}])

And the error is:

IllegalArgumentException Could not determine best method to invoke for put-attributes using arguments ("spike_for_logging" "devapiTue Aug 05 23:25:28 UTC 2014" ({:name "args", :value "[\"Test message\"]"} {:name "instant", :value "Tue Aug 05 23:25:28 UTC 2014"} {:name "ns", :value "some-api.main"} {:name "file", :value "/tmp/form-init6934682238825582339.clj"} {:name "hostname", :value "devapi"} {:name "output", :value "2014-Aug-05 23:25:28 +0000 devapi WARN [some-api.main] - Test message"} {:name "prefix", :value "2014-Aug-05 23:25:28 +0000 devapi WARN [some-api.main]"} {:name "level", :value ":warn"} {:name "line", :value ""} {:name "ap-config", :value "{}"} {:name "error?", :value "false"} {:name "throwable", :value ""} {:name "timestamp", :value "2014-Aug-05 23:25:28 +0000"} {:name "message", :value "Test message"}))  amazonica.core/intern-function/fn--11248 (core.clj:780)

This looks like a really awesome library---I like that you're using reflection to solve everything at once rather than try to wrap one service at a time.

I'm just diving in but seem to have hit a snag.
Here's a minimal example:

(ns scratch
  (:require amazonica.core)
  (:use amazonica.aws.identitymanagement))

(def creds {:access-key "root-access-key"
            :secret-key "root-secret-key"})

(create-user creds :user-name "db")
;;This works fine; user is created

(create-access-key creds :user-name "db")
;;This returns a new access key, but it's for the root account (i.e., the same account as creds), not for the new "db" account.

The problem is that the :user-name isn't being taken into account, so access keys are created for the same user that owns the creds used to make the request, not the specified new IAM user.

The docs for the underlying CreateAccessKeyRequest object seem to match the get/set method model that you're reflecting against, so I have no idea why it doesn't work.
It seems like the no-argument CreateAccessKeyRequest is matching first.

I have Kinesis records that contain Snappy-encoded blocks such that when a process that decodes each block yields a sequence of strings like this:

"{:foo 42} "
"{:bar 3.14} "
...
"{:qaz [1 2 3 4]} "

So each string is an EDN value (a basic Clojure literal).

Amazonica seems to assume that data must be nippy-serialized. So when I try:

(get-records :shard-iterator (get-shard-iterator "my-stream"
                                                 "shardId-000000000003"
                                                 "TRIM_HORIZON"))

I get:

CompilerException java.lang.Exception: Thaw failed: Uncompressed data?, compiling:(form-init1883914083661306631.clj:1:9)

Perhaps I am missing it, but I don't see a simple way to tell Amazonica not to do that and instead give me raw bytes, which I could decompress and decode however I like.
So, is there a way to do that?

So far I see that you have unwrap function that is used directly and indirectly by get-records and by processor-factory, respectively. Would be wonderful if I could supply my own version of unwrap, instead of nippy-thawing version of unwrap.

When running the first SimpleDB select query, the library prints the credentials (secret and all):

user> (select cred :select-expression "select count(*) from `Users`")
#<BasicAWSCredentials com.amazonaws.auth.BasicAWSCredentials@72cb56cd>
{:class com.amazonaws.auth.BasicAWSCredentials, :AWSSecretKey XXXXXXXXXXXXXX :AWSAccessKeyId XXXXXXXXXXXXXX}
{:items [{:name "Domain", :attributes [{:name "Count", :value "257"}]}]}

The middle two lines are printed, while the last is the returned value.

Subsequent requests don't print them.

One more for you:

(s3t/upload (bucket-name)
              (UUID/randomUUID)
              (file "/Users/sritchie/Desktop/20131009-TWITTER-ENGINEERS-016edit-660x824.jpg"))

;; CompilerException java.lang.IllegalArgumentException: No matching method found: setRegion for class com.amazonaws.services.s3.transfer.TransferManager, compiling:(form-init6963743079950413614.clj:2:3) 

I'm trying to list the files in an S3 bucket using [amazonica "0.2.24"] and the following code:

(s3/list-objects :bucket-name "mybucket")

I get the exception listed below.

Caused by org.xml.sax.SAXParseException
Premature end of file.

  ErrorHandlerWrapper.java:  203  com.sun.org.apache.xerces.internal.util.ErrorHandlerWrapper/createSAXParseException
  ErrorHandlerWrapper.java:  177  com.sun.org.apache.xerces.internal.util.ErrorHandlerWrapper/fatalError
     XMLErrorReporter.java:  441  com.sun.org.apache.xerces.internal.impl.XMLErrorReporter/reportError
     XMLErrorReporter.java:  368  com.sun.org.apache.xerces.internal.impl.XMLErrorReporter/reportError
           XMLScanner.java: 1436  com.sun.org.apache.xerces.internal.impl.XMLScanner/reportFatalError

XMLDocumentScannerImpl.java: 1019 com.sun.org.apache.xerces.internal.impl.XMLDocumentScannerImpl$PrologDriver/next
XMLDocumentScannerImpl.java: 606 com.sun.org.apache.xerces.internal.impl.XMLDocumentScannerImpl/next
XMLNSDocumentScannerImpl.java: 117 com.sun.org.apache.xerces.internal.impl.XMLNSDocumentScannerImpl/next
XMLDocumentFragmentScannerImpl.java: 510 com.sun.org.apache.xerces.internal.impl.XMLDocumentFragmentScannerImpl/scanDocument
XML11Configuration.java: 848 com.sun.org.apache.xerces.internal.parsers.XML11Configuration/parse
XML11Configuration.java: 777 com.sun.org.apache.xerces.internal.parsers.XML11Configuration/parse
XMLParser.java: 141 com.sun.org.apache.xerces.internal.parsers.XMLParser/parse
AbstractSAXParser.java: 1213 com.sun.org.apache.xerces.internal.parsers.AbstractSAXParser/parse
XmlResponsesSaxParser.java: 145 com.amazonaws.services.s3.model.transform.XmlResponsesSaxParser/parseXmlInputStream
XmlResponsesSaxParser.java: 293 com.amazonaws.services.s3.model.transform.XmlResponsesSaxParser/parseListBucketObjectsResponse
Unmarshallers.java: 76 com.amazonaws.services.s3.model.transform.Unmarshallers$ListObjectsUnmarshaller/unmarshall
Unmarshallers.java: 73 com.amazonaws.services.s3.model.transform.Unmarshallers$ListObjectsUnmarshaller/unmarshall
S3XmlResponseHandler.java: 62 com.amazonaws.services.s3.internal.S3XmlResponseHandler/handle
S3XmlResponseHandler.java: 31 com.amazonaws.services.s3.internal.S3XmlResponseHandler/handle
AmazonHttpClient.java: 795 com.amazonaws.http.AmazonHttpClient/handleResponse
AmazonHttpClient.java: 463 com.amazonaws.http.AmazonHttpClient/executeHelper
AmazonHttpClient.java: 257 com.amazonaws.http.AmazonHttpClient/execute
AmazonS3Client.java: 3623 com.amazonaws.services.s3.AmazonS3Client/invoke
AmazonS3Client.java: 3575 com.amazonaws.services.s3.AmazonS3Client/invoke
AmazonS3Client.java: 620 com.amazonaws.services.s3.AmazonS3Client/listObjects
NativeMethodAccessorImpl.java: -2 sun.reflect.NativeMethodAccessorImpl/invoke0
NativeMethodAccessorImpl.java: 62 sun.reflect.NativeMethodAccessorImpl/invoke
DelegatingMethodAccessorImpl.java: 43 sun.reflect.DelegatingMethodAccessorImpl/invoke
Method.java: 483 java.lang.reflect.Method/invoke
nil: -1 sun.reflect.GeneratedMethodAccessor54/invoke
DelegatingMethodAccessorImpl.java: 43 sun.reflect.DelegatingMethodAccessorImpl/invoke
Method.java: 483 java.lang.reflect.Method/invoke
Reflector.java: 93 clojure.lang.Reflector/invokeMatchingMethod
Reflector.java: 28 clojure.lang.Reflector/invokeInstanceMethod
core.clj: 726 amazonica.core/fn-call/fn
core.clj: 777 amazonica.core/intern-function/fn
RestFn.java: 421 clojure.lang.RestFn/invoke
replutils.clj: 74 unpacker.examples.replutils/unparsable-keys
REPL: 1 unpacker.examples.replutils/eval15366

I'm trying to invoke the following AmazonS3Client method :

http://amzn.to/1hDfA05

public void setObjectAcl(String bucketName, String key,CannedAccessControlList acl)

Here is how I'm invoking it in my code (where creds is a map with my aws credentials and client config):

(require '[amazonica.aws.s3 :as amazonica-s3])
(import com.amazonaws.services.s3.model.CannedAccessControlList)
(amazonica-s3/set-object-acl creds "com.test.bucket" "test/key" CannedAccessControlList/PublicRead)

IllegalArgumentException Don't know how to create ISeq from: com.amazonaws.services.s3.model.CannedAccessControlList  clojure.lang.RT.seqFrom (RT.java:494)

I have a feeling that its trying to invoke a similar method in the s3 client with the following signature instead:

public void setObjectAcl(String bucketName, String key, AccessControlList acl)

Does amazonica API match the underlying client API by type of args or just the count of the args?

When I call (receive-message :queue-url "https://path.to.queue") I get an exception:

IllegalArgumentException No value supplied for key: https://path.to.queue
    clojure.lang.PersistentHashMap.create (PersistentHashMap.java:77)
    clojure.core/hash-map (core.clj:365)
    clojure.core/apply (core.clj:617)
    amazonica.aws.sqs/delete-on-receive (sqs.clj:34)
    clojure.lang.Var.invoke (Var.java:423)
    clojure.lang.Var.applyTo (Var.java:532)
    clojure.core/apply (core.clj:619)
    robert.hooke/compose-hooks/fn--1482 (hooke.clj:40)
    clojure.core/apply (core.clj:617)
    robert.hooke/run-hooks (hooke.clj:46)
    robert.hooke/prepare-for-hooks/fn--1487/fn--1488 (hooke.clj:54)
    clojure.lang.AFunction$1.doInvoke (AFunction.java:29)

Line 34 of delete-on-receive is a hook that deletes messages when they've been received if the :delete option is present.

It's possible to work around the bug with the following:

(use 'robert.hooke)
(with-hooks-disabled receive-message
    (receive-message :queue-url "https://path.to.queue"))

but of course you can't use the delete-on-receive functionality if you do this.

Please since I couldn't figure out where else to make this request I'm using this medium. Please do you have any examples of using cloudformation from amazonica. I'm especially having diffuiculty knowing how to map the exmaple from the amazon website to amazonica. I can't find how to specify an amazon template to amazonica's cloudformation api. Yes I also can't figure the same for the Amazon client .

I'll appreciate any help with this.

Are there plans for implementing support for the new cloudsearchv2 API?

I've tried to run the tests, but I get an exception parse-args already refers to: #'amazonica.aws.glacier/parse-args in namespace: amazonica.test.core

I ran the tests using lein test.

The full stack trace is:

Exception in thread "main" java.lang.IllegalStateException: parse-args already refers to: #'amazonica.aws.glacier/parse-args in namespace: amazonica.test.core
        at clojure.lang.Namespace.warnOrFailOnReplace(Namespace.java:88)
        at clojure.lang.Namespace.reference(Namespace.java:110)
        at clojure.lang.Namespace.refer(Namespace.java:168)
        at clojure.core$refer.doInvoke(core.clj:3850)
        at clojure.lang.RestFn.invoke(RestFn.java:410)
        at clojure.lang.AFn.applyToHelper(AFn.java:161)
        at clojure.lang.RestFn.applyTo(RestFn.java:132)
        at clojure.core$apply.invoke(core.clj:619)
        at clojure.core$load_lib.doInvoke(core.clj:5394)
        at clojure.lang.RestFn.applyTo(RestFn.java:142)
        at clojure.core$apply.invoke(core.clj:619)
        at clojure.core$load_libs.doInvoke(core.clj:5417)
        at clojure.lang.RestFn.applyTo(RestFn.java:137)
        at clojure.core$apply.invoke(core.clj:621)
        at clojure.core$use.doInvoke(core.clj:5507)
        at clojure.lang.RestFn.invoke(RestFn.java:1789)
        at amazonica.test.core$eval496$loading__4910__auto____497.invoke(core.clj:1)
        at amazonica.test.core$eval496.invoke(core.clj:1)
        at clojure.lang.Compiler.eval(Compiler.java:6619)
        at clojure.lang.Compiler.eval(Compiler.java:6608)
        at clojure.lang.Compiler.load(Compiler.java:7064)
        at clojure.lang.RT.loadResourceScript(RT.java:370)
        at clojure.lang.RT.loadResourceScript(RT.java:361)
        at clojure.lang.RT.load(RT.java:440)
        at clojure.lang.RT.load(RT.java:411)
        at clojure.core$load$fn__5018.invoke(core.clj:5530)
        at clojure.core$load.doInvoke(core.clj:5529)
        at clojure.lang.RestFn.invoke(RestFn.java:408)
        at clojure.core$load_one.invoke(core.clj:5336)
        at clojure.core$load_lib$fn__4967.invoke(core.clj:5375)
        at clojure.core$load_lib.doInvoke(core.clj:5374)
        at clojure.lang.RestFn.applyTo(RestFn.java:142)
        at clojure.core$apply.invoke(core.clj:619)
        at clojure.core$load_libs.doInvoke(core.clj:5413)
        at clojure.lang.RestFn.applyTo(RestFn.java:137)
        at clojure.core$apply.invoke(core.clj:619)
        at clojure.core$require.doInvoke(core.clj:5496)
        at clojure.lang.RestFn.applyTo(RestFn.java:137)
        at clojure.core$apply.invoke(core.clj:619)
        at user$eval85.invoke(form-init8712192415606825579.clj:1)
        at clojure.lang.Compiler.eval(Compiler.java:6619)
        at clojure.lang.Compiler.eval(Compiler.java:6609)
        at clojure.lang.Compiler.load(Compiler.java:7064)
        at clojure.lang.Compiler.loadFile(Compiler.java:7020)
        at clojure.main$load_script.invoke(main.clj:294)
        at clojure.main$init_opt.invoke(main.clj:299)
        at clojure.main$initialize.invoke(main.clj:327)
        at clojure.main$null_opt.invoke(main.clj:362)
        at clojure.main$main.doInvoke(main.clj:440)
        at clojure.lang.RestFn.invoke(RestFn.java:421)
        at clojure.lang.Var.invoke(Var.java:419)
        at clojure.lang.AFn.applyToHelper(AFn.java:163)
        at clojure.lang.Var.applyTo(Var.java:532)
        at clojure.main.main(main.java:37)
Tests failed.

I am attempting to test kinesis via:

(put-record cred "beatport-api-test" new-event event-key)

Both new-event and event-key are java.lang.String. I am getting the following error:

Caused by: java.lang.IllegalArgumentException: No coercion is available to turn {"response":{"status":200,"headers":{"link":"</search?q=hee&group-by=kind&page=1>"},"body":{"list":[],"track":[],"release":[],"mix":[],"genre":[],"account":[],"best-match":null}},"ip":"127.0.0.1","user-agent":"","method":"get","events":"","duration":"369ms","http-server":"org.eclipse.jetty.server.HttpInput@3b702644","function-times":{},"id":"2014-07-16-api-usw1a-001-00000000","action":"Request","time-unix":1405563461,"uri":"/search","user":"newport"} into an object of type class java.nio.ByteBuffer

(Where new-event is the {"response ... newport"} portion, a simulated log event encoded as json, which will need be processed as text by the consumer.)

Hopefully, there is something simple that I am missing. But I have reviewed the Readme, and it seems the next troubleshooting step would be to coerce new-event to java.nio.ByteBuffer myself. Any help is appreciated. Thanks.

Almost all clients have an asynchronous counterpart as listed below. It'd be useful to have them supported by amazonica.

Here's an example of using one of them in clojure.

As far as I can see, as a general rule, the async client implements two methods of the form:

Future<Void> methodNameFromTheSyncClientAsync(RequestClass aRequest);
Future<Void> methodNameFromTheSyncClientAsync(RequestClass aRequest, \
  AsyncHandler<RequestClass,ResultClass> asyncHandler);

for each method of the sync version of the client.

• Autoscaling
• CloudFormation
• CloudFront
• CloudSearch
• CloudWatch
• DataPipeline
• DirectConnect
• DynamoDB
• EC2
• ElastiCache
• ElasticBeanstalk
• ElasticLoadBalancing
• ElasticMapReduce
• Glacier
• IdentityManagement
• OpsWorks
• RDS
• Redshift
• Route53
• S3 (seems there's no async version, but there's TransferManager class)
• SimpleDB
• SimpleEmail
• SNS
• SQS
• StorageGateway

Background

There would be a number of benefits for some Clojure apps if a Kinesis shard could be presented as a Clojure core.async channel. Delivering shards as channels would create new options for Clojure stream consumers, beyond the limited Kinesis notion of worker and record processor and the bandwidth and other limits applied to shards.

Here is an idea for one method of doing this in Amazonica, in case it is useful.

(Disclaimer - This is a rough sketch based on an inexpert read of the AWS documentation and what I understand so far of core.async.)

Step 1 - Basic Kinesis shard->core.async channel

• A Kinesis consumer app using core.async creates a buffered shard channel and passes it to Amazonica, along with the usual starting parameters (similar to calling worker in kinesis.clj).
• a modified record processor (similar to processRecords in processor-factory) performs blocking writes to the channel for each Kinesis record (rather than calling a processor function for each record)
• the core.async consumer app reads from the channel to process the shard data

This of course is insufficient on its own - by simply dumping shard records on a channel, we have lost the ability to know when each record is "done". We don't know when the records will be read from the channel, or when they will be processed. So a new mechanism is required to restore the ability to checkpoint sequence numbers in the shard.

Step 2 - Checkpointing

Only the core.async app knows when a record is really done; one method to communicate "doneness" back to Amazonica is by using another channel. The app can write to this channel to send completed sequence numbers back to Amazonica.

• Amazonica creates a checkpoint channel, for example (chan (sliding-buffer 1)). A sliding buffer channel will drop oldest values; in this case only the latest put survives.
• whenever the data for a sequence number is fully processed (or whenever it chooses), the app puts the sequence number on the checkpoint channel; it should never block, and Amazonica is only ever interested in the latest value

Step 3 - Complete version with checkpointing

To combine the previous steps, a variant of processor-factory does the following:

1. startup - app creates buffered shard channel and passes to Amazonica. Amazonica creates and returns checkpoint channel (which app can use as it sees fit, or ignore)
2. receive records pushed from Kinesis (as usual) and do a blocking write for each record to the shard channel
3. if a sufficient time interval has passed since the last checkpoint, do a zero timeout get from the checkpoint channel; if a value is available, call the Kinesis checkpoint(String sequenceNumber) variant of IRecordProcessorCheckpointer
4. repeat from step 2

For step 3: note that checkpoint() (no arguments) currently used in kinesis.clj checkpoints the progress at the last record that was delivered to the record processor; with channels, we want to checkpoint a specific sequence number. This capability is added in the Kinesis Client Library version 1.1.

One way (not sure if this is idiomatic) to get the latest value from the checkpoint channel, without waiting if nothing is available:

(alts!! [checkpoint-channel (timeout 0)] :priority true)

If not nil, the returned value is checkpointed.

See also:

https://github.com/awslabs/amazon-kinesis-client/blob/master/src/main/java/com/amazonaws/services/kinesis/clientlibrary/interfaces/IRecordProcessorCheckpointer.java

https://forums.aws.amazon.com/message.jspa?messageID=531052

Hi,

I would like to run amazonica against Java 1.6.
The latest Clojar release has been compiled with Java 1.7.
I'm not sure if this is due to the Java Amazon bindings enforce this.

Is it is possible, could we get a release that works with Java 1.6?

Thanks,

I'm getting this error when attempting to create a launch configuration. You can recreate the error by doing the following:

(use 'amazonica.aws.autoscaling)
(create-launch-configuration :security-groups (seq ["hello" "world"]))

... results in...

IllegalArgumentException No method in multimethod 'fmap' for dispatch value: class clojure.lang.PersistentVector$ChunkedSeq  clojure.lang.MultiFn.getFn (MultiFn.java:160)

In my case the contents of my :security groups param have been generated by map. This also happens when using create-auto-scaling-group so I have a feeling it's likely to happen in a lot of places.

I'm using Clojure 1.6.0 but have tried with 1.5.1 as well. The stack trace points to the error occurring in org.clojure/algo.generic so I tried updating that from 0.1.0 to 0.1.2 but that didn't help.

I can eliminate the problem by wrapping my sequence in vec but that's not an ideal solution. Am I doing something wrong here or is this something Amazonica could help with?

Does amazonica support IAM and STS services as well?

It's not at all obvious to me from the docs or code how to accomplish this. Using ENV Vars I do this:

(def aws_access_key_id 
  (.getAWSAccessKeyId 
    (.getCredentials (amazonica.core/get-credentials :cred))))
(def aws_secret_key 
  (.getAWSSecretKey 
;   (.getCredentials (amazonica.core/get-credentials :cred))))

(defcredential aws_access_key_id aws_secret_key (:region options))

However, if this happens while on an instance with IAM profile, an error is thrown about a missing security token. Is there some simple alternative that I'm missing?

I think this great library needs some more documentation and examples. E.g. I'm trying (and failing) to write a correct amazonica.aws.dynamodbv2/batch-write-item request for an hour or so already. And the fact that the API is automatically generated from AWS API makes it harder (at least for me) to understand how to do it.

I'd be happy to help with the stuff I'm using (mainly dynamodb and simpledb) but first I have to understand it myself :)

Hey guys, I'm getting an exception using with-credentials:

clojure.lang.ArityException: Wrong number of args (2) passed to: core$amazon-client-STAR-

Here's the code:

(ns paddleguru.util.aws
  (:require [environ.core :refer [env]]
            [amazonica.core :refer [with-credential]]
            [amazonica.aws.s3 :as s3]
            [amazonica.aws.s3transfer :as s3t]))

(with-credential [(:aws-access-key-id env)
                  (:aws-access-key-secret env)
                  "us-west-1"]
  (s3/list-buckets))

The same code works great without the wrapping with-credential.

I'm attempting to do this, which seems to fit the Java SDK spec:

(let [bytes (.getBytes "foo")]
  (put-object
   :bucket-name "my-bucket"
   :key "foo"
   :input (ByteArrayInputStream. bytes)
   :metadata {:content-length (count bytes)
              :conten-type "text/plan"}))

However I receive a cryptic error and cannot figure out how to debug it:

java.lang.NullPointerException: null
         AmazonS3Client.java:1130 com.amazonaws.services.s3.AmazonS3Client.putObject
                 (Unknown Source) sun.reflect.GeneratedMethodAccessor25.invoke
DelegatingMethodAccessorImpl.java:43 sun.reflect.DelegatingMethodAccessorImpl.invoke
                  Method.java:601 java.lang.reflect.Method.invoke
                 (Unknown Source) sun.reflect.GeneratedMethodAccessor8.invoke
DelegatingMethodAccessorImpl.java:43 sun.reflect.DelegatingMethodAccessorImpl.invoke
                  Method.java:601 java.lang.reflect.Method.invoke
                Reflector.java:93 clojure.lang.Reflector.invokeMatchingMethod
                Reflector.java:28 clojure.lang.Reflector.invokeInstanceMethod
                     core.clj:589 amazonica.core/fn-call[fn]
                     core.clj:629 amazonica.core/intern-function[fn]
                  RestFn.java:619 clojure.lang.RestFn.invoke
               NO_SOURCE_FILE:114 canary.sensor/eval5799
               Compiler.java:6619 clojure.lang.Compiler.eval
               Compiler.java:6582 clojure.lang.Compiler.eval
                    core.clj:2852 clojure.core/eval9 lighttable.hub.clj.eval/->result
                     AFn.java:163 clojure.lang.AFn.applyToHelper
                     AFn.java:151 clojure.lang.AFn.applyTo
                     core.clj:619 clojure.core/apply
                    core.clj:2396 clojure.core/partial[fn]
                  RestFn.java:408 clojure.lang.RestFn.invoke
                    core.clj:2485 clojure.core/map[fn]
                  LazySeq.java:42 clojure.lang.LazySeq.sval
                  LazySeq.java:60 clojure.lang.LazySeq.seq
                      RT.java:484 clojure.lang.RT.seq
                     core.clj:133 clojure.core/seq
                    core.clj:2523 clojure.core/filter[fn]
                  LazySeq.java:42 clojure.lang.LazySeq.sval
                  LazySeq.java:60 clojure.lang.LazySeq.seq
                      RT.java:484 clojure.lang.RT.seq
                     core.clj:133 clojure.core/seq
                    core.clj:2780 clojure.core/dorun
                    core.clj:2796 clojure.core/doall
                     eval.clj:150 lighttable.hub.clj.eval/eval-clj[fn]
                    core.clj:1836 clojure.core/binding-conveyor-fn[fn]
                      AFn.java:18 clojure.lang.AFn.call
              FutureTask.java:334 java.util.concurrent.FutureTask$Sync.innerRun
              FutureTask.java:166 java.util.concurrent.FutureTask.run
     ThreadPoolExecutor.java:1145 java.util.concurrent.ThreadPoolExecutor.runWorker
      ThreadPoolExecutor.java:615 java.util.concurrent.ThreadPoolExecutor$Worker.run
                  Thread.java:722 java.lang.Thread.run

I notice the the :file approach to uploading seems to work well, but I need to upload an input stream. I'm mostly trying to understand if this is on my end or something deeper.

Any pointers?

CompilerException java.lang.IllegalStateException: marshall already refers to: #'amazonica.core/marshall in namespace: amazonica.aws.dynamodb, compiling:(amazonica/aws/dynamodb.clj:60:12)

hi,

I came across this problem with the macro with-credential

(defmacro with-credential
  "Per invocation binding of credentials for ad-hoc
  service calls using alternate user/password combos
  (and endpoints)."
  [[a b c] & body]
  `(binding [*credentials* ~(keys->cred a b c)]
    (do ~@body)))

In this macro you destructure the [a b c] as a vector,
and this works fine when used as

(with-credential ["a" "b" "c"]
     (comment "foo"))

however if you try to use in this context

(defn get-my-credentials []
   ["a" "b" "c"])

(with-credential (get-my-credentials)
     (comment "foo"))

it is going to fail because the macro will try to destructure the (get-my-credentials)
as a list with one element instead of evaluate the function call.

This is the macroexpand result.

(clojure.core/binding [amazonica.core/*credentials* {:access-key
                                                     get-my-credentials,
                                                     :secret-key nil}]
  (do (comment "foo")))

To fix this I suggest to remove the destructuring of the vector in the macro,
and call the keys->cred on the unquote (~) of the credential argument.

best regards
Bruno