Merge remote-tracking branch 'origin/dev' into dev

vitrivr · Oct 31, 2024 · c27e605 · c27e605
2 parents c696273 + f314526
commit c27e605
Show file tree

Hide file tree

Showing 7 changed files with 101 additions and 75 deletions.
diff --git a/example-configs/ingestion/dense/dense.json b/example-configs/ingestion/dense/dense.json
@@ -8,7 +8,7 @@
         "path": "../cache"
       },
       "enumerator": {
-        "path": "../objects_dense",,
+        "path": "../objects_dense",
         "depth": "5"
       },
       "selector":{

diff --git a/example-configs/ingestion/example/video.json b/example-configs/ingestion/example/video.json
@@ -1,55 +1,79 @@
 {
   "schema": "example",
   "context": {
-    "contentFactory": "InMemoryContentFactory",
+    "contentFactory": "CachedContentFactory",
     "resolverName":"disk",
+
     "local": {
+      "clip": {
+        "contentSources": "selector"
+      },
+      "averagecolor": {
+        "contentSources": "selector"
+      },
+      "content": {
+        "path": "../cache"
+      },
+      "thumbnail": {
+        "contentSources": "selector",
+        "maxSideResolution": "400",
+        "mimeType": "JPG"
+      },
+      "selector":{
+        "contentSources": "decoder"
+      },
       "enumerator": {
-        "path": "./example/media/",
+        "path": "D:\\v3c",
         "depth": "3"
       },
+      "decoder": {
+        "timeWindowMs": "6000"
+      },
       "filter": {
-        "type": "SOURCE:IMAGE"
+        "type": "SOURCE:VIDEO"
+      },
+      "path": {
+        "field": "file"
       }
     }
   },
   "operators": {
     "enumerator": {
       "type": "ENUMERATOR",
       "factory": "FileSystemEnumerator",
-      "mediaTypes": ["IMAGE"]
+      "mediaTypes": ["VIDEO"]
     },
     "decoder": {
       "type": "DECODER",
-      "factory": "ImageDecoder"
+      "factory": "VideoDecoder"
     },
-    "averagecolor": {
+    "metadata":{
       "type": "EXTRACTOR",
-      "fieldName": "averagecolor"
+      "fieldName": "video"
     },
-    "clip": {
+    "source":{
       "type": "EXTRACTOR",
-      "fieldName": "clip"
+      "fieldName": "file"
     },
-    "dino": {
-      "type": "EXTRACTOR",
-      "fieldName": "dino"
+    "path": {
+      "type": "TRANSFORMER",
+      "factory":"DescriptorAsContentTransformer"
     },
-    "ocr": {
-      "type": "EXTRACTOR",
-      "fieldName": "ocr"
+    "selector": {
+      "type": "TRANSFORMER",
+      "factory": "LastContentAggregator"
     },
-    "meta-file": {
+    "time": {
       "type": "EXTRACTOR",
-      "fieldName": "file"
+      "fieldName": "time"
     },
-    "meta-video": {
+    "averagecolor": {
       "type": "EXTRACTOR",
-      "fieldName": "video"
+      "fieldName": "averagecolor"
     },
-    "meta-time": {
+    "clip": {
       "type": "EXTRACTOR",
-      "fieldName": "time"
+      "fieldName": "clip"
     },
     "thumbnail": {
       "type": "EXPORTER",
@@ -61,22 +85,23 @@
     }
   },
   "operations": {
-    "stage-0-0": {"operator": "enumerator"},
-    "stage-1-0": {"operator": "decoder","inputs": ["stage-0-0"]},
-    "stage-2-0": {"operator": "clip","inputs": ["stage-1-0"]},
-    "stage-2-1": {"operator": "dino","inputs": ["stage-1-0"]},
-    "stage-2-2": {"operator": "ocr","inputs": ["stage-1-0"]},
-    "stage-2-3": {"operator": "averagecolor","inputs": ["stage-1-0"]},
-    "stage-2-4": {"operator": "thumbnail","inputs": ["stage-1-0"]},
-    "stage-3-0": {"operator": "filter","inputs": ["stage-2-3","stage-2-4"], "merge": "COMBINE"},
-    "stage-4-0": {"operator": "meta-file", "inputs": ["stage-3-0"]}
+    "enumerator": {"operator": "enumerator"},
+    "decoder": {"operator": "decoder", "inputs": ["enumerator"]},
+    "path": {"operator": "path", "inputs": ["decoder"]},
+    "metadata": {"operator": "metadata", "inputs": ["path"], "merge": "COMBINE"},
+
+    "source": {"operator": "source", "inputs": ["metadata"]},
+
+    "time": {"operator": "time", "inputs": ["source"]},
+    "selector": {"operator": "selector", "inputs": ["time"]},
+
+    "thumbnail": {"operator": "thumbnail", "inputs": ["selector"]},
+
+    "clip": {"operator": "clip", "inputs": ["thumbnail"]},
+    "filter": {"operator": "filter","inputs": ["clip"],"merge": "MERGE"}
   },
   "output": [
-    "stage-2-0",
-    "stage-2-1",
-    "stage-2-2",
-    "stage-2-3",
-    "stage-4-0"
+    "filter"
   ],
-  "mergeType": "COMBINE"
+  "mergeType": "MERGE"
 }
diff --git a/example-configs/schema/dense.json b/example-configs/schema/dense.json
@@ -11,6 +11,9 @@
         }
       },
       "fields": {
+        "imageFilePathContent": {
+          "field": "file"
+        },
         "averagecolor": {
           "factory": "AverageColor"
         },

diff --git a/example-configs/schema/example.json b/example-configs/schema/example.json
@@ -2,10 +2,12 @@
   "schemas": {
     "example": {
       "connection": {
-        "database": "CottontailConnectionProvider",
+        "database": "PgVectorConnectionProvider",
         "parameters": {
           "Host": "127.0.0.1",
-          "port": "1865"
+          "port": "5432",
+          "username": "postgres",
+          "password": "vitrivr"
         }
       },
       "fields": {
@@ -18,31 +20,11 @@
         "clip": {
           "factory": "DenseEmbedding",
           "parameters": {
-            "host": "http://127.0.0.1:8888",
+            "host": "http://10.34.64.84:8888/",
             "model": "open-clip-vit-b32",
-            "length": "512"
-          }
-        },
-        "dino": {
-          "factory": "DenseEmbedding",
-          "parameters": {
-            "host": "http://127.0.0.1:8888/",
-            "model": "dino-v2-vits14",
-            "length": "384"
-          }
-        },
-        "whisper": {
-          "factory": "ASR",
-          "parameters": {
-            "host": "http://127.0.0.1:8888/",
-            "model": "whisper"
-          }
-        },
-        "ocr": {
-          "factory": "OCR",
-          "parameters": {
-            "host": "http://127.0.0.1:8888/",
-            "model": "tesseract"
+            "length": "512",
+            "timeoutSeconds": "100",
+            "retries": "1000"
           }
         },
         "time": {
@@ -72,10 +54,7 @@
       },
       "extractionPipelines": {
         "video": {
-          "path": "./example/video.json"
-        },
-        "image": {
-          "path": "./example/image.json"
+          "path": "./example-configs/ingestion/example/video.json"
         }
       }
     }

diff --git a/settings.gradle b/settings.gradle
@@ -1,5 +1,5 @@
 plugins {
-    id 'org.gradle.toolchains.foojay-resolver-convention' version '0.5.0'
+    id 'org.gradle.toolchains.foojay-resolver-convention' version '0.8.0'
 }
 rootProject.name = 'vitrivr-engine'
 

diff --git a/...n/kotlin/org/vitrivr/engine/core/util/extension/IngestionPipelineBuilderPrintExtension.kt b/...n/kotlin/org/vitrivr/engine/core/util/extension/IngestionPipelineBuilderPrintExtension.kt
@@ -0,0 +1,18 @@
+package org.vitrivr.engine.core.util.extension
+
+import org.vitrivr.engine.core.config.ingest.IngestionPipelineBuilder
+import org.vitrivr.engine.core.model.retrievable.Retrievable
+import org.vitrivr.engine.core.operators.Operator
+
+public enum class Flavor {
+    MERMAID
+}
+
+fun IngestionPipelineBuilder.build(flavor: Flavor) : List<Operator.Sink<Retrievable>> {
+    val nodes = this.build()
+    nodes.forEach {
+        it.input
+
+    }
+    return nodes
+}
diff --git a/vitrivr-engine-server/src/main/kotlin/org/vitrivr/engine/server/Main.kt b/vitrivr-engine-server/src/main/kotlin/org/vitrivr/engine/server/Main.kt
@@ -44,13 +44,14 @@ fun main(args: Array<String>) {
         c.jsonMapper(KotlinxJsonMapper)
 
         /* Registers Open API plugin. */
-        c.registerPlugin(OpenApiPlugin{
+        c.registerPlugin(OpenApiPlugin {
             it.withDocumentationPath("/openapi.json")
-                .withDefinitionConfiguration{ _, def ->
-                    def.withInfo{i ->
+                .withDefinitionConfiguration { _, def ->
+                    def.withInfo { i ->
                         i.title = "vitrivr engine API"
                         i.version = "0.1.0"
-                        i.description = "Rest API for the vitrivr engine project. Provides query (runtime) and extraction (ingestion) endpoints"
+                        i.description =
+                            "Rest API for the vitrivr engine project. Provides query (runtime) and extraction (ingestion) endpoints"
                     }
                         .withSecurity(
                             SecurityComponentConfiguration().withSecurityScheme("CookieAuth", CookieAuth("SESSIONID"))
@@ -59,9 +60,9 @@ fun main(args: Array<String>) {
         })
         c.http.maxRequestSize = 1024 * 1024 * 1024 /* 1GB */
 
-        c.bundledPlugins.enableCors{cors ->
+        c.bundledPlugins.enableCors { cors ->
             /* https://javalin.io/plugins/cors#getting-started */
-            cors.addRule{
+            cors.addRule {
                 it.reflectClientOrigin = true // might be a little too loose
                 it.allowCredentials = true
             }
@@ -74,7 +75,7 @@ fun main(args: Array<String>) {
             swaggerConfig.uiPath = "/swagger-ui"
         })
 
-        c.router.apiBuilder{
+        c.router.apiBuilder {
             configureApiRoutes(config.api, manager, executor)
         }
     }.exception(ErrorStatusException::class.java) { e, ctx ->