Add PREMIS-related activities

ffvalidate/activity_test.go

@@ -101,7 +101,7 @@ fmt/95,"PDF/A "
 		{
 			name:    "Does nothing when no allowlist path configured",
 			params:  ffvalidate.Params{Path: validFormatsDir},
-			wantLog: "INFO validate-file-formats: No allowlist path configured, skipping file format validation ActivityID 0 ActivityType validate-file-formats\n",
+			wantLog: "V[1] Executing activity. ActivityID 0 ActivityType validate-file-formats\nINFO validate-file-formats: No allowlist path configured, skipping file format validation ActivityID 0 ActivityType validate-file-formats\n",
 		},
 		{
 			name:    "Errors when allowlist path doesn't exist",

go.mod

@@ -1,16 +1,20 @@
 module github.com/artefactual-sdps/temporal-activities
 
-go 1.21.4
+go 1.22
+
+toolchain go1.22.2
 
 require (
 	github.com/artefactual-labs/bagit-gython v0.2.0
+	github.com/beevik/etree v1.4.1
 	github.com/google/safeopen v0.0.0-20240125081138-66b54d5181c6
+	github.com/google/uuid v1.6.0
 	github.com/mholt/archiver/v4 v4.0.0-alpha.8
 	github.com/nyudlts/go-bagit v0.3.0-alpha.0.20240515212815-8dab411c23af
 	github.com/otiai10/copy v1.14.0
 	github.com/richardlehane/siegfried v1.11.1
 	github.com/tonglil/buflogr v1.1.1
-	go.artefactual.dev/tools v0.14.0
+	go.artefactual.dev/tools v0.17.0
 	go.temporal.io/sdk v1.26.1
 	gocloud.dev v0.39.0
 	gotest.tools/v3 v3.5.1
@@ -31,7 +35,6 @@ require (
 	github.com/golang/mock v1.6.0 // indirect
 	github.com/golang/snappy v0.0.4 // indirect
 	github.com/google/go-cmp v0.6.0 // indirect
-	github.com/google/uuid v1.6.0 // indirect
 	github.com/googleapis/gax-go/v2 v2.13.0 // indirect
 	github.com/grpc-ecosystem/go-grpc-middleware v1.4.0 // indirect
 	github.com/grpc-ecosystem/grpc-gateway/v2 v2.19.1 // indirect

go.sum

@@ -73,6 +73,8 @@ github.com/aws/aws-sdk-go-v2/service/sts v1.30.3 h1:ZsDKRLXGWHk8WdtyYMoGNO7bTudr
 github.com/aws/aws-sdk-go-v2/service/sts v1.30.3/go.mod h1:zwySh8fpFyXp9yOr/KVzxOl8SRqgf/IDw5aUt9UKFcQ=
 github.com/aws/smithy-go v1.20.3 h1:ryHwveWzPV5BIof6fyDvor6V3iUL7nTfiTKXHiW05nE=
 github.com/aws/smithy-go v1.20.3/go.mod h1:krry+ya/rV9RDcV/Q16kpu6ypI4K2czasz0NC3qS14E=
+github.com/beevik/etree v1.4.1 h1:PmQJDDYahBGNKDcpdX8uPy1xRCwoCGVUiW669MEirVI=
+github.com/beevik/etree v1.4.1/go.mod h1:gPNJNaBGVZ9AwsidazFZyygnd+0pAU38N4D+WemwKNs=
 github.com/benbjohnson/clock v1.1.0/go.mod h1:J11/hYXuz8f4ySSvYwY0FKfm+ezbsZBKZxNJlLklBHA=
 github.com/bodgit/plumbing v1.2.0 h1:gg4haxoKphLjml+tgnecR4yLBV5zo4HAZGCtAh3xCzM=
 github.com/bodgit/plumbing v1.2.0/go.mod h1:b9TeRi7Hvc6Y05rjm8VML3+47n4XTZPtQ/5ghqic2n8=
@@ -279,8 +281,8 @@ github.com/ulikunitz/xz v0.5.10/go.mod h1:nbz6k7qbPmH4IRqmfOplQw/tblSgqTqBwxkY0o
 github.com/yuin/goldmark v1.1.27/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
 github.com/yuin/goldmark v1.2.1/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
 github.com/yuin/goldmark v1.3.5/go.mod h1:mwnBkeHKe2W/ZEtQ+71ViKU8L12m81fl3OWwC1Zlc8k=
-go.artefactual.dev/tools v0.14.0 h1:ESLbemsnkdIPmYXtz0uZTcPqVnTUXIEZd9DSTRyTZqY=
-go.artefactual.dev/tools v0.14.0/go.mod h1:5RJ7ObocHZv/zQFYFv/zG9cW/UVRGPFywcJx/oQ+TG8=
+go.artefactual.dev/tools v0.17.0 h1:7X/qZYKyKT8RxVjBsksqvalQ8F4wcor6jcA0ewjc92M=
+go.artefactual.dev/tools v0.17.0/go.mod h1:lsu0JcKFEJanNdrf5/IFjjzxul4pazG1dDHnLX9Nkvs=
 go.opencensus.io v0.21.0/go.mod h1:mSImk1erAIZhrmZN+AvHh14ztQfjbGwt4TtuofqLduU=
 go.opencensus.io v0.22.0/go.mod h1:+kGneAE2xo2IficOXnaByMWTGM9T73dGwxeWcUqIpI8=
 go.opencensus.io v0.22.2/go.mod h1:yxeiOL68Rb0Xd1ddK5vPZ/oVn4vY4Ynel7k9FzqtOIw=

premis/agent/activity.go

@@ -0,0 +1,46 @@
+package premis_agent
+
+import (
+	"context"
+
+	"github.com/artefactual-sdps/temporal-activities/premis"
+)
+
+const Name = "add-premis-agent"
+
+type (
+	Params struct {
+		PREMISFilePath string
+		Agent          premis.Agent
+	}
+
+	Result struct{}
+
+	Activity struct{}
+)
+
+func New() *Activity {
+	return &Activity{}
+}
+
+func (a *Activity) Execute(
+	ctx context.Context,
+	params *Params,
+) (*Result, error) {
+	doc, err := premis.ParseOrInitialize(params.PREMISFilePath)
+	if err != nil {
+		return nil, err
+	}
+
+	err = premis.AppendAgentXML(doc, params.Agent)
+	if err != nil {
+		return nil, err
+	}
+
+	err = premis.WriteIndentedToFile(doc, params.PREMISFilePath)
+	if err != nil {
+		return nil, err
+	}
+
+	return &Result{}, nil
+}

premis/agent/activity_test.go

@@ -0,0 +1,114 @@
+package premis_agent_test
+
+import (
+	"os"
+	"testing"
+
+	temporalsdk_activity "go.temporal.io/sdk/activity"
+	temporalsdk_testsuite "go.temporal.io/sdk/testsuite"
+	"gotest.tools/v3/assert"
+	"gotest.tools/v3/fs"
+
+	"github.com/artefactual-sdps/temporal-activities/premis"
+	premis_agent "github.com/artefactual-sdps/temporal-activities/premis/agent"
+)
+
+const expectedPREMISWithAgent = `<?xml version="1.0" encoding="UTF-8"?>
+<premis:premis xmlns:premis="http://www.loc.gov/premis/v3" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.loc.gov/premis/v3 https://www.loc.gov/standards/premis/premis.xsd" version="3.0">
+  <premis:agent>
+    <premis:agentIdentifier>
+      <premis:agentIdentifierType valueURI="http://id.loc.gov/vocabulary/identifiers/local">url</premis:agentIdentifierType>
+      <premis:agentIdentifierValue>https://github.com/artefactual-sdps/preprocessing-sfa</premis:agentIdentifierValue>
+    </premis:agentIdentifier>
+    <premis:agentName>Enduro</premis:agentName>
+    <premis:agentType>software</premis:agentType>
+  </premis:agent>
+</premis:premis>
+`
+
+func TestAddPREMISAgent(t *testing.T) {
+	t.Parallel()
+
+	// Transfer that's been deleted (for execution expected to fail).
+	transferDeleted := fs.NewDir(t, "")
+	PREMISFilePathNonExistent := transferDeleted.Join("metadata", "premis.xml")
+	transferDeleted.Remove()
+
+	tests := []struct {
+		name       string
+		params     premis_agent.Params
+		result     premis_agent.Result
+		wantErr    string
+		wantPREMIS string
+	}{
+		{
+			name: "Add PREMIS agent for normal content",
+			params: premis_agent.Params{
+				PREMISFilePath: fs.NewDir(t, "",
+					fs.WithFile("something.txt", "1234567899"),
+					fs.WithDir("metadata"),
+				).Join("metadata", "premis.xml"),
+				Agent: premis.AgentDefault(),
+			},
+			result:     premis_agent.Result{},
+			wantPREMIS: expectedPREMISWithAgent,
+		},
+		{
+			name: "Add PREMIS agent for no content",
+			params: premis_agent.Params{
+				PREMISFilePath: fs.NewDir(t, "",
+					fs.WithDir("metadata"),
+				).Join("metadata", "premis.xml"),
+				Agent: premis.AgentDefault(),
+			},
+			result:     premis_agent.Result{},
+			wantPREMIS: expectedPREMISWithAgent,
+		},
+		{
+			name: "Add PREMIS agent for bad path",
+			params: premis_agent.Params{
+				PREMISFilePath: PREMISFilePathNonExistent,
+				Agent:          premis.AgentDefault(),
+			},
+			result:  premis_agent.Result{},
+			wantErr: "no such file or directory",
+		},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			t.Parallel()
+
+			ts := &temporalsdk_testsuite.WorkflowTestSuite{}
+			env := ts.NewTestActivityEnvironment()
+			env.RegisterActivityWithOptions(
+				premis_agent.New().Execute,
+				temporalsdk_activity.RegisterOptions{Name: premis_agent.Name},
+			)
+
+			var res premis_agent.Result
+			future, err := env.ExecuteActivity(premis_agent.Name, tt.params)
+
+			if tt.wantErr != "" {
+				if err == nil {
+					t.Errorf("error is nil, expecting: %q", tt.wantErr)
+				} else {
+					assert.ErrorContains(t, err, tt.wantErr)
+				}
+
+				return
+			}
+
+			assert.NilError(t, err)
+
+			future.Get(&res)
+			assert.DeepEqual(t, res, tt.result)
+
+			// Compare PREMIS output to what's expected.
+			if tt.wantPREMIS != "" {
+				b, err := os.ReadFile(tt.params.PREMISFilePath)
+				assert.NilError(t, err)
+				assert.Equal(t, string(b), tt.wantPREMIS)
+			}
+		})
+	}
+}

premis/event/activity.go

@@ -0,0 +1,47 @@
+package premis_event
+
+import (
+	"context"
+
+	"github.com/artefactual-sdps/temporal-activities/premis"
+)
+
+const Name = "add-premis-event"
+
+type (
+	Params struct {
+		PREMISFilePath string
+		Agent          premis.Agent
+		Summary        premis.EventSummary
+	}
+
+	Result struct{}
+
+	Activity struct{}
+)
+
+func New() *Activity {
+	return &Activity{}
+}
+
+func (a *Activity) Execute(
+	ctx context.Context,
+	params *Params,
+) (*Result, error) {
+	doc, err := premis.ParseOrInitialize(params.PREMISFilePath)
+	if err != nil {
+		return nil, err
+	}
+
+	err = premis.AppendEventXMLForEachObject(doc, params.Summary, params.Agent)
+	if err != nil {
+		return nil, err
+	}
+
+	err = premis.WriteIndentedToFile(doc, params.PREMISFilePath)
+	if err != nil {
+		return nil, err
+	}
+
+	return &Result{}, nil
+}