Walkthrough

The Range API and related types were widened from 32-bit to 64-bit: Range(size int) was changed to Range(size int64) in Ctx and Req interfaces, RangeSet fields became int64, parsing now validates bounds with a new helper, tests and docs updated for large-file ranges.

Changes

Sequence Diagram(s)

sequenceDiagram
    participant Client
    participant Ctx
    participant DefaultReq
    participant parseBound

    Client->>Ctx: Range(size: int64)
    Ctx->>DefaultReq: Range(size: int64)
    DefaultReq->>parseBound: validate start/end strings
    alt bounds valid
        parseBound-->>DefaultReq: start:int64, end:int64
        DefaultReq-->>Ctx: Range{Start:int64, End:int64}
        Ctx-->>Client: return Range
    else malformed bounds / overflow
        parseBound-->>DefaultReq: ErrRangeMalformed
        DefaultReq-->>Ctx: return error
        Ctx-->>Client: return error
    end

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

• Verify all interface and implementation signatures changed consistently.
• Inspect parseBound logic for correct int64 conversion, overflow handling, and negative-value behavior.
• Confirm Content-Range string formatting and any uses that assume 32-bit ints.
• Run and validate updated tests for large-file behavior.

Possibly related PRs

• 🐛 bug: Fix Range() handling of HTTP 416 per RFC 9110 #3552 — touches Range handling and tests for unsatisfiable/edge-case range behavior; closely related to Range parsing/signature changes.

Suggested labels

🧹 Updates

Suggested reviewers

• sixcolors
• efectn
• ReneWerner87

Poem

🐇
I hopped through bytes both wide and tall,
From int to int64 I stretch my sprawl.
Ranges grow long and errors shrink,
This rabbit tests each boundary link. 🥕

Pre-merge checks and finishing touches

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.

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

Summary of Changes

Hello @gaby, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request significantly improves the robustness of HTTP Range header parsing by addressing potential integer overflow issues when dealing with very large file sizes. It refactors the internal Range handling to consistently use int64 for size and range bounds, introduces a clamping mechanism to prevent overflows during parsing, and updates relevant interfaces and documentation. These changes ensure more reliable and secure processing of range requests, especially in environments with large data transfers.

Highlights

• Range Parsing Robustness: The Range header parsing logic has been enhanced to prevent potential integer overflows by clamping parsed range header bounds against math.MaxUint64 >> 1 before converting them to int64. This ensures that the overflow check remains within the math package's capabilities.
• Type Signature Updates: The Range method in both the Ctx and Req interfaces, as well as its implementation in DefaultReq, now accepts int64 for the size parameter instead of int. Additionally, the Start and End fields within the RangeSet struct have been updated from int to int64 to accommodate larger values.
• New Helper Function and Error Handling: A new internal helper function, parseBound, has been introduced in req.go to encapsulate the logic for parsing string values into int64 and applying the math.MaxUint64 >> 1 clamp. This function also includes specific error handling for malformed range values.
• Expanded Test Coverage: New test cases (Test_Ctx_Range_LargeFile) have been added to ctx_test.go to specifically validate the Range method's behavior when dealing with large file sizes and range values, confirming the correct handling of int64 conversions and the new clamping mechanism.
• Documentation Alignment: The API documentation for the Ctx.Range method in docs/api/ctx.md has been updated to reflect the change in the size parameter's type from int to int64.

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 91.58%. Comparing base (5d91977) to head (25e844a).
⚠️ Report is 43 commits behind head on main.

@@            Coverage Diff             @@
##             main    #3874      +/-   ##
==========================================
- Coverage   91.64%   91.58%   -0.06%     
==========================================
  Files         115      115              
  Lines        9848     9857       +9     
==========================================
+ Hits         9025     9028       +3     
- Misses        523      528       +5     
- Partials      300      301       +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.