rgraff

rgraff

Ash Core Team

I have a validation based on use Ash.Resource.Validation that when invalid returns

{:error,
  InvalidAttribute.exception(
    field: :variant_key,
    path: [:feature_override],
    message: "feature variant not found"
   )}

The graphql response is missing the path

%{"data" => %{"addFeatureOverrideToAccount" => %{"errors" => [%{"code" => "invalid_attribute", "fields" => ["variant_key"], "message" => "feature variant not found"}], "result" => nil}}}

I expect the path to be ["feature_override","variant_key"]

I’m seeing the same issue in liveview with the error not getting rendered in forms.

Showing Posts 1 to 10

zachdaniel

zachdaniel

Creator of Ash

Can you try this:

{:error,
  InvalidAttribute.exception(
    field: :variant_key,
    message: "feature variant not found"
   ) |> Ash.Error.set_path([:feature_override])}

We should make the path key work for all exceptions by default I think, as that is a pretty natural way to set it.

rgraff

rgraff OP

Ash Core Team

The path is being set correctly.

      {:error,
       InvalidAttribute.exception(
         field: :variant_key,
         path: [:feature_override],
         message: "feature variant not found"
       )
       |> IO.inspect()}

%Ash.Error.Changes.InvalidAttribute{
  field: :variant_key,
  message: "feature variant not found",
  private_vars: nil,
  value: nil,
  changeset: nil,
  query: nil,
  error_context: [],
  vars: [],
  path: [:feature_override],
  stacktrace: #Stacktrace<>,
  class: :invalid
}

Using Ash.Error.set_path(...) adds to the path but it’s still not in the graphql fields.

Yesterday when I was digging I thought I remember seeing builtin errors from constraints also have keys attribute: and resource: but now I can’t recreate those.

rgraff

rgraff OP

Ash Core Team

May be relevant: The argument (feature_override) is an embedded resource that has an attribute (variant_key). The update action being called is on the parent. The action appends the argument into a list of embedded resources (feature_overrides). The validation is on the parent resource because attributes of the parent affect the validation of the embedded children.

zachdaniel

zachdaniel

Creator of Ash

:thinking: Maybe this is an issue in graphql? How is the error rendering vs how you want to? It would be a separate improvement to make the path: ... option work for ash exceptions by default, but it makes sense to do and would be easy.

rgraff

rgraff OP

Ash Core Team

The issue isn’t specific to graphql. I have similar issues in Ash Phoenix Forms. The error gets added to the changeset with the path, but something else must not be set correctly because the path gets dropped by graphql and Ash Phoenix Form too.

I’ll see if I can find a easy way to reproduce.

rgraff

rgraff OP

Ash Core Team

@zachdaniel, I hope this is a better explanation.

Given a action with an argument this is an embedded resource:

update :add_feature_override do
  argument :feature_override, Environments.FeatureOverride, allow_nil?: false
  validate {FeatureVariantExists, []}
  ...
end

And when invalid FeatureVariantExists returns:

  {:error,
    InvalidArgument.exception(
      field: :variant_key,  # this is an attribute on the FeatureOverride resource
      path: [:feature_override],
      message: "variant key not found"
    )}

And I have a form like this:

AshPhoenix.Form.for_update(socket.assigns.account, action,
    as: "account",
    api: Environments,
    forms: [auto?: true]
  )
  |> to_form()
  |> AshPhoenix.Form.add_form(:feature_override, params: feature_override_params)

When the form is invalid, the error is as expected in form.source.source.errors[] but is not in the nested form for feature_override

form.source.source.errors[
  %Ash.Error.Changes.InvalidArgument{
    field: :variant_key,
    message: "variant key not found",
    changeset: nil,
    query: nil,
    error_context: [],
    vars: [],
    path: [:feature_override],
    stacktrace: #Stacktrace<>,
    class: :invalid
  }
] 

form.forms.feature_override.source.errors = [] # I expect my error to be here on the variant_key field with no path
form.forms.feature_override.submit_errors = [] # I expect [variant_key: {"variant key not found", []}] 

When getting the error from a graphql mutation (instead of form)

%{"data" => %{"addFeatureOverrideToAccount" => %{"errors" => [%{"code" => "invalid_argument", "fields" => ["variant_key"], "message" => "variant key not found"}], "result" => nil}}}

# I expect fields to ["feature_override", "variant_key"] or even better["input", "featureOverride", "variantKey"]
zachdaniel

zachdaniel

Creator of Ash

So, for ash_phoenix that is definitely an issue, a reproduction test would be greatly appreciated. I can hunt down what might be going wrong. As for ash_graphql, it looks like we actually just don’t have path handling logic set up at all. It should be an addition to the existing logic. The fields is a pointer to multiple fields, not a path to the fields.

rgraff

rgraff OP

Ash Core Team

I’ll do a PR for an ash_phoenix test to reproduce this. :pray:

For ash_graphql, it’s less of an issue for me right now and I’m not sure what the conventions/specs are for errors. I personally would expect a path with camelCase.

zachdaniel

zachdaniel

Creator of Ash

Yeah, I agree we should add a path to the graphql errors and camelcase them :slight_smile:

Neophen

Neophen

@rgraff did you manage to solve it? As i’m running into an issue where using the AshPhoenix form i’m not getting the error for a path, doing an inspect like so:

form = AshPhoenix.Form.validate(socket.assigns.form, edition_params)
IO.inspect(form, label: "____form")

the source.source.errors does contain both errors

 errors: [
        %Ash.Error.Changes.InvalidAttribute{
          field: :en,
          message: "length must be greater than or equal to %{min}",
          private_vars: nil,
          value: %{"_unused_lt" => "", "en" => "a", "lt" => ""},
          splode: nil,
          bread_crumbs: [],
          vars: [min: 2],
          path: [:title],
          stacktrace: #Splode.Stacktrace<>,
          class: :invalid
        },
        %Ash.Error.Changes.InvalidAttribute{
          field: :content_types,
          message: "must have %{min} or more items",
          private_vars: nil,
          value: [],
          splode: nil,
          bread_crumbs: [],
          vars: [min: 1],
          path: [],
          stacktrace: #Splode.Stacktrace<>,
          class: :invalid
        }
      ],

but the actual errors doesn’t contain the title.en error:

  errors: [content_types: {"must have %{min} or more items", [min: 1]}],

This is a custom ash type based on map. here’s a post with more details:

— All posts loaded —

Where Next? Top

Trending in Questions Top

Hellowman
Hello everyone! I am new to coding with Elixir/Erlang. And i am looking for some utilities for fuzzing testing like AFL++. Does anyone ...
New
yarrichar
I’m very new to Elixir, so sorry if this is a silly question: I am wondering what happens to the mailbox of a process when the machine i...
New
pragdave
How do I tell Hologram not to escape HTML when substituting into templates? {Textile.render(@product.long_description)} escapes the HTM...
New
Fl4m3Ph03n1x
Background When trying to execute mix release on a Windows 11 machine for a Phoenix project I get the following error: * assembling mark...
New
mandar1045
I load tested a plain Phoenix chat app backed by Postgres with 100 million messages in it, everything on default settings, to see what br...
New
rlipscombe
I’d like to create a file with 600 permissions (rw-------), but there’s a race condition if I create it and then change the permissions. ...
New
saleyn
Is there a way to import record definitions in the iex shell for pretty printing (similar to the rr command in the erl shell)? This woul...
New

Other Trending Topics Top

GenericJam
Edit: 2026 May 15 - This post is archived. Mob is alive!! Main docs: mob v0.7.11 — Documentation A bit of explanation for the slightly c...
261 13177 112
New
mudasobwa
I fully migrated to my own harness from Anthropic/Gemini and I think it’s time to share it. Welcome DSH, the DeepSeek Harness, fully writ...
New
OndrejValenta
(I just needed to vent somewhere and LinkedIn is full of hope, or hype, I’m not sure which exactly) AI dream has many faces, but general...
New
kip
Localize is the next generation localisation library for Elixir. Think of it as ex_cldr version 3.0. The first version will be released ...
New
jimsynz
Beam Bots (or just BB for short) is a framework for building fault-tolerant robotics applications in Elixir using familiar OTP patterns. ...
New
mcass19
ExRatatui lets you cook up rich terminal UIs in Elixir, powered by Rust’s ratatui via Rustler NIFs. Build interactive terminal applicatio...
New

We're in Beta

About us Mission Statement

Options

Thread Display Mode




Thread Preview

Skip Thread Previews