Skip to content

FCOMBO triplets are read one field too early - #2290

Open
denislauri1999 wants to merge 2 commits into
NosCoreIO:masterfrom
denislauri1999:pr/fcombo-offset
Open

FCOMBO triplets are read one field too early#2290
denislauri1999 wants to merge 2 commits into
NosCoreIO:masterfrom
denislauri1999:pr/fcombo-offset

Conversation

@denislauri1999

Copy link
Copy Markdown
Contributor

The first field of an FCOMBO row is not a step, it is a switch: 1 when the skill has a combo chain and 0 when it has not — true of all 1958 skills in the file. Five triplets of (hit, animation, effect) follow it, and AddCombos counted them from the switch instead of from after it.

220 Basic Slash   FCOMBO 1 | 3 40 513 | 4 25 525 | 5 13 524 | 0 0 0 | 0 0 0
read as           (hit 1, anim 3, eff 40), (hit 513, ...), (hit 525, ...)

Hit 513 never arrives — the consecutive-hit counter stays in single figures — so the chain could not fire on any of the twenty-one skills that carry one. Nothing reported it: no exception, just the third blow’s special animation that never plays.

The raw row starts at index 2 (field 0 is empty, field 1 is the section name), so the switch sits at 2 and triplet j starts at 3 + j*3.

Tests

Two. The first parses the real row above and asserts all three steps, ending with a check that no hit number reaches double figures — that is the shape this bug had, and the shape any future off-by-one would have. The second holds a skill with no chain to zero saved steps, so the switch cannot be mistaken for a step again.

Zero warnings, all tests pass.

The first field of an FCOMBO row is not a step, it is a switch: 1 when the
skill has a combo chain and 0 when it has not, which holds for all 1958 skills
in the file. Five triplets of (hit, animation, effect) follow it, and the
parser counted them from the switch instead of from after it.

    220 Basic Slash   FCOMBO 1 | 3 40 513 | 4 25 525 | 5 13 524 | 0 0 0 | 0 0 0
    read as           (hit 1, anim 3, eff 40), (hit 513, ...), (hit 525, ...)

Hit 513 never arrives - the consecutive-hit counter stays in single figures -
so the chain could not fire on any of the twenty-one skills that carry one.
Nothing reported it: no exception, just the third blow's special animation that
never plays.

Two tests. The first parses the real row above and asserts the three steps,
with a final check that no hit number reaches double figures - which is the
shape the bug had, and the shape any future off-by-one would have. The second
holds a skill with no chain to zero saved steps, so the switch cannot be
mistaken for a step again.
@coderabbitai

coderabbitai Bot commented Aug 23, 2026

Copy link
Copy Markdown

Warning

Review limit reached

Next included review available in 25 minutes.

View limit details

Limit details: You’ve used all 2 included reviews currently available.

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Learn how review limits work.

Review configuration:

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 282216c8-05b2-4bcd-928b-25dacac63a44

📥 Commits

Reviewing files that changed from the base of the PR and between 55dc38c and cc49f51.

📒 Files selected for processing (2)
  • src/NosCore.Parser/Parsers/SkillParser.cs
  • test/NosCore.Parser.Tests/SkillParserTests.cs

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

Three lines for the offset, one for the test fixture. The rest was retelling
what the assertions already check.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant