Skip to content

json : support enum values within allOf#15830

Merged
aldehir merged 2 commits intoggml-org:masterfrom
aldehir:fix-allOf-enum
Sep 8, 2025
Merged

json : support enum values within allOf#15830
aldehir merged 2 commits intoggml-org:masterfrom
aldehir:fix-allOf-enum

Conversation

@aldehir
Copy link
Collaborator

@aldehir aldehir commented Sep 6, 2025

fix #15823

The allOf logic only supports object schemas with properties, this PR extends it to support enum as well.

Strictly speaking, combining allOf with enum should only consider the intersect of all sub-schemas. This PR violates that by aggregating them to minimize complexity. I can add the intersection if desired, I just didn't think it was worth the hassle.

@deiteris
Copy link

deiteris commented Sep 6, 2025

I'm not sure the current implementation considers the intersection even for properties (with additionalProperties: false there must be an intersection as well, which can be a pita). From the correctness standpoint, intersection is a must here, otherwise the resulting grammar will not be equivalent to the rule you define, but ofc it depends on expected usage. For example the following schema:

{
  "allOf": [
      {"$ref": "#/definitions/foo"},
      {"$ref": "#/definitions/bar"}
  ],
  "definitions": {
      "foo": {
          "type": "string",
          "enum": ["a", "b"]
      },
      "bar": {
          "type": "string",
          "enum": ["b", "c"]
      }
  }
}

In your case will produce an aggregate of enums and your resulting rule will be able to match "a", "b", "b", and "c". But if you validate the same data against the original schema using a compliant JSON Schema validator, only "b" will pass.

@directrix1
Copy link

This fixed my issue with using Zed editor which was defining tool calls with this kind of schema.

@aldehir
Copy link
Collaborator Author

aldehir commented Sep 6, 2025

For the sake of correctness, I added the intersection logic. That said, I highly doubt clients are going to add multiple enum schemas.

@aldehir aldehir merged commit 7057faf into ggml-org:master Sep 8, 2025
50 checks passed
SamuelOliveirads pushed a commit to SamuelOliveirads/llama.cpp that referenced this pull request Dec 29, 2025
* grammar : fix JSON Schema for string regex with top-level alt. (ggml-org#9903)

Prior to this commit, using a JSON Schema containing a string
with `pattern` regular expression that uses top-level alternation
(e.g. `"pattern": "^A|B|C|D$"`) would result in invalid JSON
output from the constrained sampling grammar, because it
ended up creating a grammar rule like this for the string:

```
thing ::= "\"" "A" | "B" | "C" | "D" "\"" space
```

Note that this rule will only match a starting quote for the "A" case,
and will only match an ending quote for the "D" case,
so this rule will always produce invalid JSON when used for sampling
(that is, the JSON will always be lacking the starting quote,
the ending quote, or both).

This was fixed in a simple way by adding parentheses to the
generated rule (for all string pattern rules, to keep it simple),
such that the new generated rule looks like this (correct):

```
thing ::= "\"" ("A" | "B" | "C" | "D") "\"" space
```

* grammars : add English-only grammar (ggml-org#10612)

* grammar : handle maxItems == 0 in JSON schema (ggml-org#13117)

Co-authored-by: Richard Lyons <frob@cloudstaff.com>

* grammar-parser : fix possible null-deref (ggml-org#9004)

Fixes: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=70680

Signed-off-by: David Korczynski <david@adalogics.com>

* llama : fix typo in llama-grammar.h [no ci] (ggml-org#11816)

* * server: fix "--grammar-file" parameter (ggml-org#12285)

* common : use std::string_view now that we target c++17 (ggml-org#14319)

* json : support `enum` values within `allOf` (ggml-org#15830)

* grammar : use int64_t to avoid int overflows in int schema to grammar conversion logic (ggml-org#16626)

* grammar : support array references in json schema (ggml-org#16792)

* grammar : support array references in json schema

* Update json-schema-to-grammar.cpp

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* grammar : improve regex when naming ref derived rules

* grammar : replace non-conformant definitions array with anyOf test case

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
# Conflicts:
#	tests/test-json-schema-to-grammar.cpp

* merge fix

* llama : minor grammar refactor (ggml-org#10897)

* llama: fix error on bad grammar (ggml-org#12628)

* grammar : fix integer overflow (ggml-org#17381)

* Fix DoS / integer overflow

* Remove optional, use INT64_MAX instead as placeholder value (it's technically -1, so it fits :)

* White space

* Actually, since it's unsigned, use UINT64_MAX
# Conflicts:
#	src/llama-grammar.cpp

* grammar: fix regression caused by ggml-org#17381 (ggml-org#17412)

* grammar: fix regression caused by ggml-org#17381

* more readable
# Conflicts:
#	src/llama-grammar.cpp

* Merge Fix

* Fix warnings

---------

Signed-off-by: David Korczynski <david@adalogics.com>
Co-authored-by: Joe Eli McIlvain <joe.eli.mac@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Co-authored-by: frob <rick+github@frob.com.au>
Co-authored-by: Richard Lyons <frob@cloudstaff.com>
Co-authored-by: DavidKorczynski <david@adalogics.com>
Co-authored-by: Daniel Bevenius <daniel.bevenius@gmail.com>
Co-authored-by: firecoperana <firecoperana>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
Co-authored-by: Aldehir Rojas <hello@alde.dev>
Co-authored-by: Olivier Chafik <olivier.chafik@gmail.com>
Co-authored-by: Piotr Wilkin (ilintar) <piotr.wilkin@syndatis.com>
Co-authored-by: Xuan-Son Nguyen <son@huggingface.co>
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
blime4 referenced this pull request in blime4/llama.cpp Feb 5, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

examples python python script changes server testing Everything test related

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Eval bug: "edit_file" tool call does not allow "mode" argument to be set

4 participants