Merge pull request #198 from tableflowhq/feature/schemaless-data-types

Schemaless data types support

Author: ciminelli

admin-server/docs/docs.go

@@ -321,7 +321,7 @@
                         "schema": {
                             "type": "object",
                             "additionalProperties": {
-                                "type": "string"
+                                "$ref": "#/definitions/types.UploadColumnMapping"
                             }
                         }
                     }
@@ -780,6 +780,15 @@
                 }
             }
         },
+        "types.UploadColumnMapping": {
+            "type": "object",
+            "properties": {
+                "template_column_id": {
+                    "type": "string",
+                    "example": "a1ed136d-33ce-4b7e-a7a4-8a5ccfe54cd5"
+                }
+            }
+        },
         "types.UploadHeaderRowSelection": {
             "type": "object",
             "properties": {

admin-server/docs/swagger.json

@@ -260,6 +260,12 @@ definitions:
         example: a1ed136d-33ce-4b7e-a7a4-8a5ccfe54cd5
         type: string
     type: object
+  types.UploadColumnMapping:
+    properties:
+      template_column_id:
+        example: a1ed136d-33ce-4b7e-a7a4-8a5ccfe54cd5
+        type: string
+    type: object
   types.UploadHeaderRowSelection:
     properties:
       index:
@@ -518,7 +524,7 @@ paths:
         required: true
         schema:
           additionalProperties:
-            type: string
+            $ref: '#/definitions/types.UploadColumnMapping'
           type: object
       responses:
         "200":

admin-server/docs/swagger.yaml

@@ -79,6 +79,10 @@ type UploadHeaderRowSelection struct {
 	Index *int `json:"index" example:"0"`
 }
 
+type UploadColumnMapping struct {
+	TemplateColumnID string `json:"template_column_id" example:"a1ed136d-33ce-4b7e-a7a4-8a5ccfe54cd5"`
+}
+
 type UploadRow struct {
 	Index  int            `json:"index" example:"0"`
 	Values map[int]string `json:"values"`

admin-server/go/pkg/types/importer.go

@@ -344,23 +344,28 @@ func importerSetHeaderRow(c *gin.Context) {
 //	@Success		200	{object}	types.Res
 //	@Failure		400	{object}	types.Res
 //	@Router			/file-import/v1/upload/{id}/set-column-mapping [post]
-//	@Param			id		path	string				true	"Upload ID"
-//	@Param			body	body	map[string]string	true	"Request body"
+//	@Param			id		path	string									true	"Upload ID"
+//	@Param			body	body	map[string]types.UploadColumnMapping	true	"Request body"
 func importerSetColumnMapping(c *gin.Context) {
 	id := c.Param("id")
 	if len(id) == 0 {
 		c.AbortWithStatusJSON(http.StatusBadRequest, types.Res{Err: "No upload ID provided"})
 		return
 	}
 
-	// Non-schemaless: Upload column ID -> Template column ID
-	// Schemaless:     Upload column ID -> User-provided key (i.e. first_name) (only from the request, this will be updated to IDs after the template is generated)
-	columnMapping := make(map[string]string)
-	if err := c.ShouldBindJSON(&columnMapping); err != nil {
+	columnMappingRequest := make(map[string]types.UploadColumnMapping)
+	if err := c.ShouldBindJSON(&columnMappingRequest); err != nil {
 		tf.Log.Warnw("Could not bind JSON", "error", err)
 		c.AbortWithStatusJSON(http.StatusBadRequest, types.Res{Err: err.Error()})
 		return
 	}
+
+	// Non-schemaless: Upload column ID -> Template column ID
+	// Schemaless:     Upload column ID -> User-provided key (i.e. first_name) (only from the request, this will be updated to IDs after the template is generated)
+	columnMapping := make(map[string]string)
+	for k, mappingSelection := range columnMappingRequest {
+		columnMapping[k] = mappingSelection.TemplateColumnID
+	}
 	if len(columnMapping) == 0 {
 		c.AbortWithStatusJSON(http.StatusBadRequest, types.Res{Err: "Please select at least one destination column"})
 		return

admin-server/go/pkg/web/file_import_routes.go

@@ -2,3 +2,5 @@
 title: "Download Import CSV"
 openapi: "GET /import/{id}/download"
 ---
+
+Retrieve an import as a CSV file.

docs/api-reference/download-import.mdx

@@ -3,8 +3,31 @@ title: "Get Import Rows"
 openapi: "GET /import/{id}/rows"
 ---
 
-Retrieve the rows of an import as JSON. This endpoint supports pagination by using a limit/offset. If the limit and offset are not provided, it will return the first 1000 rows of the import.
+Retrieve the rows of an import as JSON. This endpoint supports pagination by using a limit/offset. If the limit and offset are not provided, it will return the first 1,000 rows of the import.
 
 To use the limit/offset, start by setting the offset to 0 and the limit to 100 to get the first 100 rows of data. To get the next 100 rows, set the offset to 100 while keeping the limit the same. Continue increasing the offset by 100 until no more rows are returned.
 
-Note: the max limit is 1000.
+<Info>The maximum `limit` is 1000.</Info>
+
+### Example Response
+
+```json
+[
+  {
+    "index": 0,
+    "values": {
+      "age": 23,
+      "email": "maria@example.com",
+      "first_name": "Maria"
+    }
+  },
+  {
+    "index": 1,
+    "values": {
+      "age": 32,
+      "email": "robert@example.com",
+      "first_name": "Robert"
+    }
+  }
+]
+```

docs/api-reference/get-import-rows.mdx

@@ -1,4 +1,63 @@
 ---
-title: "Get Import Metadata"
+title: "Get Import"
 openapi: "GET /import/{id}"
 ---
+
+Retrieve the row data, column definitions, and other information about the import.
+
+<Info>The number of rows included is limited to 10,000. If there are more than 10,000 rows, an `error` will be set and
+  the data should be retrieved using the [/rows](/api-reference/get-import-rows) endpoint.</Info>
+
+### Example Response
+
+```json
+{
+  "id": "da5554e3-6c87-41b2-9366-5449a2f15b53",
+  "importer_id": "a0fadb1d-9888-4fcb-b185-25b984bcb227",
+  "num_rows": 2,
+  "num_columns": 3,
+  "num_processed_values": 5,
+  "metadata": {
+    "user_id": 1234,
+    "user_email": "user@example.com",
+    "environment": "staging"
+  },
+  "created_at": 1698172312,
+  "error": null,
+  "columns": [
+    {
+      "data_type": "number",
+      "key": "age",
+      "name": "Age"
+    },
+    {
+      "data_type": "string",
+      "key": "email",
+      "name": "Email"
+    },
+    {
+      "data_type": "string",
+      "key": "first_name",
+      "name": "First Name"
+    }
+  ],
+  "rows": [
+    {
+      "index": 0,
+      "values": {
+        "age": 23,
+        "email": "maria@example.com",
+        "first_name": "Maria"
+      }
+    },
+    {
+      "index": 1,
+      "values": {
+        "age": 32,
+        "email": "robert@example.com",
+        "first_name": "Robert"
+      }
+    }
+  ]
+}
+```