Validate document content length by extracted text, not JSON size

Add Node.TextLength() to walk the ProseMirror tree and sum actual user
text. Introduce ProseMirrorDocumentMaxTextLength validator that enforces
a 50k character limit on extracted text, paired with a 500k byte safety
cap on the raw JSON string. Extend e2e tests for content length.

Signed-off-by: Émile Ré <emile@getprobo.com>
This commit is contained in:
Émile Ré
2026-04-09 12:50:46 +04:00
parent 303455ded6
commit 7cd4606278
6 changed files with 228 additions and 6 deletions

View File

@@ -966,7 +966,7 @@ func TestDocument_MaxLength_Validation(t *testing.T) {
longTitle := strings.Repeat("a", 1001)
t.Run("create", func(t *testing.T) {
t.Run("create with long title", func(t *testing.T) {
query := `
mutation CreateDocument($input: CreateDocumentInput!) {
createDocument(input: $input) {
@@ -990,7 +990,7 @@ func TestDocument_MaxLength_Validation(t *testing.T) {
assert.Contains(t, err.Error(), "title")
})
t.Run("update", func(t *testing.T) {
t.Run("update with long title", func(t *testing.T) {
documentID := factory.NewDocument(owner).WithTitle("Max Length Test").Create()
query := `
@@ -1010,6 +1010,57 @@ func TestDocument_MaxLength_Validation(t *testing.T) {
require.Error(t, err)
assert.Contains(t, err.Error(), "title")
})
t.Run("create with long content", func(t *testing.T) {
query := `
mutation CreateDocument($input: CreateDocumentInput!) {
createDocument(input: $input) {
documentEdge {
node { id }
}
}
}
`
longContent := testutil.ProseMirrorTextDoc(strings.Repeat("a", 50_001))
_, err := owner.Do(query, map[string]any{
"input": map[string]any{
"organizationId": owner.GetOrganizationID().String(),
"title": "Content Length Test",
"content": longContent,
"documentType": "POLICY",
"classification": "INTERNAL",
},
})
require.Error(t, err)
assert.Contains(t, err.Error(), "content")
})
t.Run("update version with long content", func(t *testing.T) {
docID, versionID := createTestDocument(t, owner)
require.NotEmpty(t, docID)
require.NotEmpty(t, versionID)
query := `
mutation UpdateDocumentVersion($input: UpdateDocumentVersionInput!) {
updateDocumentVersion(input: $input) {
documentVersion { id }
}
}
`
longContent := testutil.ProseMirrorTextDoc(strings.Repeat("a", 50_001))
_, err := owner.Do(query, map[string]any{
"input": map[string]any{
"documentVersionId": versionID,
"content": longContent,
},
})
require.Error(t, err)
assert.Contains(t, err.Error(), "content")
})
}
func TestDocument_Pagination(t *testing.T) {