feat: add allow_missing for fields in DeserializeRow #1416

Xyedo · 2025-08-04T10:33:37Z

Add #[allow_missing] in the field attribute for DeserializeRow macro both on enforce_order flavor and match_by_name flavor

Pre-review checklist

I have split my patch into logically separate commits.
All commit messages clearly explain what they change and why.
I added relevant tests for new features and bug fixes.
All commits compile, pass static checks and pass test.
PR description sums up the changes and reasons why they should be introduced.
I have provided docstrings for the public items that I want to introduce.
I have adjusted the documentation in ./docs/source/.
I added appropriate Fixes: annotations to PR description.

github-actions · 2025-08-04T10:36:50Z

cargo semver-checks found no API-breaking changes in this PR.
Checked commit: fd9db3f

Copilot

Pull Request Overview

This PR adds an allow_missing attribute to struct fields for both row and value deserialization, allowing fields to be initialized with Default::default() when missing from the database instead of failing. This feature provides compatibility with sqlx's #[sqlx(default)] attribute for smoother migration to scylla-rust-driver.

Implements #[scylla(allow_missing)] attribute for both struct-level and field-level usage
Adds support for the attribute in both row and value deserialization contexts
Includes comprehensive test coverage for the new functionality

Reviewed Changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
scylla-macros/src/lib.rs	Adds documentation for the new `allow_missing` attribute
scylla-macros/src/serialize/value.rs	Adds ignored `allow_missing` attribute parsing for value serialization
scylla-macros/src/serialize/row.rs	Adds ignored `allow_missing` attribute parsing for row serialization
scylla-macros/src/deserialize/value.rs	Implements `allow_missing` logic for value deserialization
scylla-macros/src/deserialize/row.rs	Implements `allow_missing` logic for row deserialization
scylla/tests/integration/macros/hygiene.rs	Adds hygiene tests for the new attribute
scylla-cql/src/deserialize/row_tests.rs	Adds comprehensive tests for `allow_missing` functionality

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

scylla-macros/src/lib.rs

Lorak-mmk

Sorry for the delay! Please rebase on main to make semver checks work again.

Lorak-mmk · 2025-10-16T11:05:35Z

scylla-cql/src/deserialize/row_tests.rs

+#[test]
+fn test_struct_deserialization_loose_ordering_allow_missing() {
+    #[derive(DeserializeRow, PartialEq, Eq, Debug)]
+    #[scylla(crate = "crate")]
+    #[scylla(allow_missing)]


Please let's do one thing in this PR. Other field attributes can't be used on a struct, so let's remove this ability from this one as well, at least for now. It will also simplify this PR a lot.

sorry, i don't understand what your intention is, is it start with scylla[allow_missing] field attributes in this PR, and put scylla[allow_missing] in struct attributes in another (next) PR?

This is how understand this. Also, if we choose to allow this attribute to be put on a struct, then we want to allow others as well. Therefore, such another PR would necessarily by larger.
For now, not to impede the per-field attribute, let's strip this PR and get it merged.

sorry, i don't understand what your intention is, is it start with scylla[allow_missing] field attributes in this PR, and put scylla[allow_missing] in struct attributes in another (next) PR?

In another PR, or maybe not at all - we will have to decide if we want to include such functionality in the driver.
I see "using field attributes on a struct" as a totally separate matter from "using allow_missing in DeserializeRow", so I think it should not be put in a single commit or PR.

Hi! sorry for the delay, removing struct attribute and add enforce_order functionalities based on DeserializeValue code is done in my latest commit

scylla-macros/src/lib.rs

Copilot

Pull Request Overview

Copilot reviewed 8 out of 8 changed files in this pull request and generated 3 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2025-10-31T03:31:18Z

scylla-macros/src/lib.rs

+///
+/// #[scylla(allow_missing)]
+///
+/// if set, implementation will not fail if some columns are missing.


Inconsistent capitalization: sentence should start with a capital letter. Change 'if set' to 'If set'.

Suggested change

/// if set, implementation will not fail if some columns are missing.

/// If set, implementation will not fail if some columns are missing.

Copilot · 2025-10-31T03:31:19Z

scylla-macros/src/lib.rs

+///
+/// #[scylla(allow_missing)]
+///
+/// if set, implementation will not fail if some columns are missing.


Inconsistent capitalization: sentence should start with a capital letter. Change 'if set' to 'If set'.

Suggested change

/// if set, implementation will not fail if some columns are missing.

/// If set, implementation will not fail if some columns are missing.

Copilot · 2025-10-31T03:31:19Z

scylla-macros/src/lib.rs

+/// If the value of the field received from DB is null, the field will be
+/// initialized with `Default::default()`.


The documentation is incorrect. For DeserializeValue, the allow_missing attribute should handle missing fields in the UDT definition, not null values. The description should say 'If the UDT definition does not contain this field' similar to the field-level documentation at line 528.

Suggested change

/// If the value of the field received from DB is null, the field will be

/// initialized with `Default::default()`.

/// If the UDT definition does not contain this field, it will be initialized

/// with `Default::default()`.

🌱 Hmm, perhaps Copilot is right? Perhaps a better name would be allow_null? cc @Lorak-mmk

this is allow_missing struct attributes docs that i forgot to delete, mb!

add allow_missing field attribute with macro DeserializeRow to handle partial deserialization

Lorak-mmk · 2025-11-06T10:48:46Z

scylla-macros/src/deserialize/row.rs

+    // If true, then - if this field is missing from the UDT fields metadata
+    // - it will be initialized to Default::default().
+    #[darling(default)]
+    #[darling(rename = "allow_missing")]
+    default_when_missing: bool,
+


This is DeserializeRow, not DeserializeValue, so UDTs are not applicable here. It should be changed to "if this field is missing from the query result metadata"

Lorak-mmk · 2025-11-06T10:50:38Z

scylla-macros/src/deserialize/value.rs

            let cql_name_literal = field.cql_name_literal();
-            let decrement_if_required: Option<syn::Stmt> = field
-                .is_required()
+            let decrement_if_required: Option<syn::Stmt> = field.is_required()


nitpick: changes in this file are only formatting (and btw its surprising that rustfmt accepts both formattings). Those changes are not necessary here.

Lorak-mmk · 2025-11-06T10:53:34Z

scylla-macros/src/serialize/value.rs

+
+    // Used for deserialization only. Ignored in serialization.
+    #[darling(default)]
+    #[darling(rename = "allow_missing")]
+    _default_when_missing: bool,


I don't see why this is necessary. The PR adds support for allow_missing in DeserializeRow, it should not need to touch SerializeValue. Even before this PR we had the following struct in tests:

// Test attributes for value struct with name flavor #[derive( _scylla::DeserializeValue, _scylla::SerializeValue, PartialEq, Debug, )] #[scylla(crate = _scylla)] #[allow(dead_code)] // TODO: Change to expect after bumping MSRV to 1.89 struct TestStructByName { a: ::core::primitive::i32, #[scylla(allow_missing)] b: ::core::primitive::i32, #[scylla(default_when_null)] c: ::core::primitive::i32, #[scylla(skip)] d: ::core::primitive::i32, #[scylla(rename = "f")] e: ::core::primitive::i32, g: ::core::primitive::i32, }

Ah, those are struct attributes, not fields attributes. It makes this change even more unnecessary.

Lorak-mmk · 2025-11-06T10:54:24Z

scylla-macros/src/serialize/row.rs

+
+    // Used for deserialization only. Ignored in serialization.
+    #[darling(default)]
+    #[darling(rename = "allow_missing")]
+    _default_when_missing: bool,


This is struct with struct-level attributes, this should not be required here.

Lorak-mmk · 2025-11-06T15:01:26Z

scylla-macros/src/deserialize/row.rs

I tried to understand changes in this file but I'm not able to. It looks like apart from adding support for new attribute, you also performed a bunch of refactors. This is fine in general, but such refactors should be split into commits. Each of those commits should be working, and should be easy to understand for reviewers.
We have limited review throughput, so we always try to make sure that review is reasonably easy, even if it means more work for the contributor.

the code structure on deserializeRow on enforce order is similiar like deserializeValue on enforce order flavor, but i think i rush it too much, i will convert it into draft, apply the suggestion and split the commit,
thank you for your time to checking my PR.

Xyedo force-pushed the partial-DeserializeValue branch from 67e73b9 to 820386a Compare August 4, 2025 10:34

github-actions bot added the semver-checks-breaking cargo-semver-checks reports that this PR introduces breaking API changes label Aug 4, 2025

wprzytula requested review from Lorak-mmk, Copilot and wprzytula August 13, 2025 10:32

wprzytula assigned Xyedo Aug 13, 2025

Copilot AI reviewed Aug 13, 2025

View reviewed changes

scylla-macros/src/lib.rs Outdated Show resolved Hide resolved

scylla-macros/src/lib.rs Outdated Show resolved Hide resolved

Xyedo force-pushed the partial-DeserializeValue branch from 820386a to 7091295 Compare September 1, 2025 09:52

Xyedo force-pushed the partial-DeserializeValue branch from 7091295 to 74184aa Compare September 9, 2025 02:56

Lorak-mmk requested changes Oct 16, 2025

View reviewed changes

Lorak-mmk added the waiting-on-author Waiting for a response from issue/PR author label Oct 16, 2025

wprzytula requested changes Oct 17, 2025

View reviewed changes

scylla-macros/src/lib.rs Outdated Show resolved Hide resolved

Xyedo force-pushed the partial-DeserializeValue branch from 74184aa to 27a4fa3 Compare October 31, 2025 03:15

github-actions bot removed the semver-checks-breaking cargo-semver-checks reports that this PR introduces breaking API changes label Oct 31, 2025

Xyedo force-pushed the partial-DeserializeValue branch from 27a4fa3 to 4c8defe Compare October 31, 2025 03:21

Xyedo requested review from Lorak-mmk, Copilot and wprzytula October 31, 2025 03:30

Copilot AI reviewed Oct 31, 2025

View reviewed changes

feat: add allow_missing in the field attribute

fd9db3f

add allow_missing field attribute with macro DeserializeRow to handle partial deserialization

Xyedo force-pushed the partial-DeserializeValue branch from 4c8defe to fd9db3f Compare October 31, 2025 07:40

Lorak-mmk removed the waiting-on-author Waiting for a response from issue/PR author label Nov 3, 2025

Lorak-mmk changed the title ~~feat: add allow_missing in the struct_field~~ feat: add allow_missing for fields in DeserializeRow Nov 6, 2025

Lorak-mmk requested changes Nov 6, 2025

View reviewed changes

Xyedo marked this pull request as draft November 6, 2025 15:31

	/// if set, implementation will not fail if some columns are missing.
	/// If set, implementation will not fail if some columns are missing.

		/// If the value of the field received from DB is null, the field will be
		/// initialized with `Default::default()`.

feat: add allow_missing for fields in DeserializeRow #1416

Are you sure you want to change the base?

feat: add allow_missing for fields in DeserializeRow #1416

Uh oh!

Conversation

Xyedo commented Aug 4, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Pre-review checklist

Uh oh!

github-actions bot commented Aug 4, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Reviewed Changes

Uh oh!

Uh oh!

Uh oh!

Lorak-mmk left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Lorak-mmk Oct 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Uh oh!

Copilot AI Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Xyedo commented Aug 4, 2025 •

edited

Loading

github-actions bot commented Aug 4, 2025 •

edited

Loading

Lorak-mmk left a comment •

edited

Loading

Lorak-mmk Oct 16, 2025 •

edited

Loading