|
| 1 | +(ns cljdoc.s3 |
| 2 | + (:require [cljdoc.server.log-init] ;; to quiet odd jetty DEBUG logging |
| 3 | + [clojure.java.io :as io]) |
| 4 | + (:import (java.lang AutoCloseable) |
| 5 | + (software.amazon.awssdk.auth.credentials AwsBasicCredentials AwsCredentialsProvider StaticCredentialsProvider) |
| 6 | + (software.amazon.awssdk.core.sync RequestBody ResponseTransformer) |
| 7 | + (software.amazon.awssdk.regions Region) |
| 8 | + (software.amazon.awssdk.services.s3 S3Client) |
| 9 | + (software.amazon.awssdk.services.s3.model CopyObjectRequest DeleteObjectRequest GetObjectRequest ListObjectsV2Request ObjectCannedACL PutObjectRequest S3Object))) |
| 10 | + |
| 11 | +(set! *warn-on-reflection* true) |
| 12 | + |
| 13 | +(defprotocol IObjectStore |
| 14 | + "Use a protocol to make switching to different implementation a bit easier |
| 15 | + We are currently using aws sdk, but only because aws-api currently blows our heap |
| 16 | + by loading entire objects into RAM. |
| 17 | +
|
| 18 | + implement specific to our use case: |
| 19 | + - always cotained to a single bucket |
| 20 | + - public-read acl |
| 21 | + - only expose data we care about |
| 22 | + - put and get at granularity of file only (no streams or strings, etc)" |
| 23 | + (list-objects [object-store]) |
| 24 | + (put-object [object-store object-key from-file]) |
| 25 | + (get-object [object-store object-key to-file]) |
| 26 | + (delete-object [object-store object-key]) |
| 27 | + (copy-object [object-store source-key dest-key])) |
| 28 | + |
| 29 | +(defrecord AwsSdkObjectStore [^S3Client s3 opts] |
| 30 | + IObjectStore AutoCloseable |
| 31 | + (list-objects [_] |
| 32 | + (let [{:keys [bucket-name]} opts |
| 33 | + ^ListObjectsV2Request request (-> (ListObjectsV2Request/builder) |
| 34 | + (.bucket bucket-name) |
| 35 | + .build)] |
| 36 | + (->> (.listObjectsV2 s3 request) |
| 37 | + .contents |
| 38 | + (mapv (fn [^S3Object o] {:key (.key o)}))))) |
| 39 | + (put-object [_ object-key from-file] |
| 40 | + (let [{:keys [bucket-name]} opts |
| 41 | + ^PutObjectRequest request (-> (PutObjectRequest/builder) |
| 42 | + (.bucket bucket-name) |
| 43 | + (.key object-key) |
| 44 | + (.acl ObjectCannedACL/PUBLIC_READ) |
| 45 | + .build)] |
| 46 | + (.putObject s3 request (RequestBody/fromFile (io/file from-file))))) |
| 47 | + (get-object [_ object-key to-file] |
| 48 | + (let [{:keys [bucket-name]} opts |
| 49 | + ^GetObjectRequest request (-> (GetObjectRequest/builder) |
| 50 | + (.bucket bucket-name) |
| 51 | + (.key object-key) |
| 52 | + .build)] |
| 53 | + (.getObject s3 request (ResponseTransformer/toFile (io/file to-file))))) |
| 54 | + (delete-object [_ object-key] |
| 55 | + (let [{:keys [bucket-name]} opts |
| 56 | + ^DeleteObjectRequest request (-> (DeleteObjectRequest/builder) |
| 57 | + (.bucket bucket-name) |
| 58 | + (.key object-key) |
| 59 | + .build)] |
| 60 | + (.deleteObject s3 request))) |
| 61 | + (copy-object [_ source-key dest-key] |
| 62 | + (let [{:keys [bucket-name]} opts |
| 63 | + ^CopyObjectRequest request (-> (CopyObjectRequest/builder) |
| 64 | + (.sourceBucket bucket-name) |
| 65 | + (.sourceKey source-key) |
| 66 | + (.destinationBucket bucket-name) |
| 67 | + (.destinationKey dest-key) |
| 68 | + .build)] |
| 69 | + (.copyObject s3 request))) |
| 70 | + (close [_] (.close s3))) |
| 71 | + |
| 72 | +(defn s3-exo-client [{:keys [bucket-key bucket-secret bucket-region]}] |
| 73 | + (let [endpoint (format "https://sos-%s.exo.io" bucket-region) |
| 74 | + ^AwsCredentialsProvider creds-provider (StaticCredentialsProvider/create |
| 75 | + (AwsBasicCredentials/create bucket-key bucket-secret))] |
| 76 | + (.build (doto (S3Client/builder) |
| 77 | + (.region Region/AWS_GLOBAL) ;; AWS SDK requires this even though we are not using AWS services |
| 78 | + (.endpointOverride (java.net.URI. endpoint)) |
| 79 | + (.credentialsProvider creds-provider))))) |
| 80 | + |
| 81 | +(defn make-exo-object-store [opts] |
| 82 | + (let [s3 (s3-exo-client opts)] |
| 83 | + (AwsSdkObjectStore. s3 opts))) |
| 84 | + |
| 85 | +(comment |
| 86 | + (require '[cljdoc.config :as cfg]) |
| 87 | + |
| 88 | + ;; assumes you've loaded up secrets to a working exo endpoint |
| 89 | + (def opts (cfg/db-backup (cfg/config))) |
| 90 | + |
| 91 | + (:bucket-region opts) |
| 92 | + |
| 93 | + (:bucket-name opts) |
| 94 | + |
| 95 | + (spit "target/dummy-file.txt" "foobar") |
| 96 | + |
| 97 | + (def object-store (make-exo-object-store opts)) |
| 98 | + |
| 99 | + (list-objects object-store) |
| 100 | + ;; => [{:key "daily/cljdoc-db-2024-09-03_2024-09-03T20-22-00.tar.zst"} |
| 101 | + ;; {:key "daily/cljdoc-db-2024-09-17_2024-09-17T18-01-44.tar.zst"}] |
| 102 | + |
| 103 | + (put-object object-store "daily/dummy-file" "target/dummy-file.txt") |
| 104 | + ;; => #object[software.amazon.awssdk.services.s3.model.PutObjectResponse 0x67c79dcd "PutObjectResponse(ETag=\"3858f62230ac3c915f300c664312c63f\")"] |
| 105 | + |
| 106 | + (list-objects object-store) |
| 107 | + ;; => [{:key "daily/cljdoc-db-2024-09-03_2024-09-03T20-22-00.tar.zst"} |
| 108 | + ;; {:key "daily/cljdoc-db-2024-09-17_2024-09-17T18-01-44.tar.zst"} |
| 109 | + ;; {:key "daily/dummy-file"}] |
| 110 | + |
| 111 | + (get-object object-store "daily/dummy-file" "target/dummy-file.down.txt") |
| 112 | + ;; => #object[software.amazon.awssdk.services.s3.model.GetObjectResponse 0x7e2790ce "GetObjectResponse(AcceptRanges=bytes, LastModified=2024-09-21T14:12:04Z, ContentLength=6, ETag=\"3858f62230ac3c915f300c664312c63f\", ContentType=text/plain, Metadata={})"] |
| 113 | + |
| 114 | + (slurp "target/dummy-file.down.txt") |
| 115 | + ;; => "foobar" |
| 116 | + |
| 117 | + (delete-object object-store "daily/dummy-file") |
| 118 | + ;; => #object[software.amazon.awssdk.services.s3.model.DeleteObjectResponse 0x4465db91 "DeleteObjectResponse()"] |
| 119 | + |
| 120 | + (list-objects object-store) |
| 121 | + ;; => [{:key "daily/cljdoc-db-2024-09-03_2024-09-03T20-22-00.tar.zst"} |
| 122 | + ;; {:key "daily/cljdoc-db-2024-09-17_2024-09-17T18-01-44.tar.zst"}] |
| 123 | + |
| 124 | + (put-object object-store "daily/dummy-file" "target/dummy-file.txt") |
| 125 | + ;; => #object[software.amazon.awssdk.services.s3.model.PutObjectResponse 0x517e713b "PutObjectResponse(ETag=\"3858f62230ac3c915f300c664312c63f\")"] |
| 126 | + |
| 127 | + (copy-object object-store "daily/dummy-file" "daily/dummy-file-copy") |
| 128 | + ;; => #object[software.amazon.awssdk.services.s3.model.CopyObjectResponse 0x4488e7e1 "CopyObjectResponse(CopyObjectResult=CopyObjectResult(ETag=3858f62230ac3c915f300c664312c63f, LastModified=2024-09-21T14:17:12.018Z))"] |
| 129 | + |
| 130 | + (list-objects object-store) |
| 131 | + ;; => [{:key "daily/cljdoc-db-2024-09-03_2024-09-03T20-22-00.tar.zst"} |
| 132 | + ;; {:key "daily/cljdoc-db-2024-09-17_2024-09-17T18-01-44.tar.zst"} |
| 133 | + ;; {:key "daily/dummy-file"} |
| 134 | + ;; {:key "daily/dummy-file-copy"}] |
| 135 | + |
| 136 | + (get-object object-store "daily/dummy-file-copy" "target/dummy-file-copy.down.txt") |
| 137 | + ;; => #object[software.amazon.awssdk.services.s3.model.GetObjectResponse 0x437659bf "GetObjectResponse(AcceptRanges=bytes, LastModified=2024-09-21T14:17:12Z, ContentLength=6, ETag=\"3858f62230ac3c915f300c664312c63f\", ContentType=text/plain, Metadata={})"] |
| 138 | + |
| 139 | + (slurp "target/dummy-file-copy.down.txt") |
| 140 | + ;; => "foobar" |
| 141 | + |
| 142 | + (delete-object object-store "daily/dummy-file-copy") |
| 143 | + ;; => #object[software.amazon.awssdk.services.s3.model.DeleteObjectResponse 0x7235fc92 "DeleteObjectResponse()"] |
| 144 | + |
| 145 | + (delete-object object-store "daily/dummy-file") |
| 146 | + ;; => #object[software.amazon.awssdk.services.s3.model.DeleteObjectResponse 0x2507c9f0 "DeleteObjectResponse()"] |
| 147 | + |
| 148 | + (list-objects object-store) |
| 149 | + ;; => [{:key "daily/cljdoc-db-2024-09-03_2024-09-03T20-22-00.tar.zst"} |
| 150 | + ;; {:key "daily/cljdoc-db-2024-09-17_2024-09-17T18-01-44.tar.zst"}] |
| 151 | + |
| 152 | + (.close object-store) |
| 153 | + |
| 154 | + (list-objects object-store) |
| 155 | + ;; => Execution error (IllegalStateException) at org.apache.http.util.Asserts/check (Asserts.java:34). |
| 156 | + ;; Connection pool shut down |
| 157 | + |
| 158 | + :eoc) |
0 commit comments