feat: add retry logic and extended timeouts for large file downloads

[flexiondotorg]

• Add automatic retry mechanism with exponential backoff (3 retries max)
• Implement automatic resume on retry by re-checking file size
• Extend connection timeout to 10 minutes for large file downloads
• Add connection pooling and TCP keepalive settings
• Refactor download_file_content to handle transient network errors
• Simplify function signature by removing explicit file_size parameter

Addresses issues:

• hash error - only less size files downloaded #76: Large files downloading with incorrect size/hash errors
• I keep getting Error: Network("Connection failed: error decoding response body") while downloading #89: Connection errors during download ('error decoding response body')

Changes from PR #86 by @ApfelTeeSaft

---

Summary by cubic

Adds automatic retry with exponential backoff and resume, plus longer connection timeouts and keepalive, to make large file downloads reliable. Resolves #76 (size/hash mismatches) and #89 (“error decoding response body”).

• New Features

  • Retries failed downloads up to 3 times with exponential backoff.
  • Auto-resume on retry by re-checking current file size (Range requests).
  • 10-minute connection timeout, no read timeout, connection pooling, and TCP keepalive.
  • Simplified download_file_content signature (removed file_size param).

• Bug Fixes

  • Prevents truncated downloads that led to incorrect size/hash (hash error - only less size files downloaded #76).
  • Handles transient network/read errors to avoid decode failures (I keep getting Error: Network("Connection failed: error decoding response body") while downloading #89).

^{Written for commit 74bc8ad. Summary will update on new commits.}

[cubic-dev-ai]

1 issue found across 2 files

Prompt for AI agents (all issues)

Check if these issues are valid — if so, understand the root cause of each and fix them.


<file name="src/main.rs">

<violation number="1" location="src/main.rs:25">
P2: Magic number `60` should use the existing `HTTP_TIMEOUT` constant. The PR removed the import but hardcoded the same value, which defeats the purpose of having the constant and makes future maintenance harder.</violation>
</file>

_{Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.}

[cubic-dev-ai]

P2: Magic number 60 should use the existing HTTP_TIMEOUT constant. The PR removed the import but hardcoded the same value, which defeats the purpose of having the constant and makes future maintenance harder.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At src/main.rs, line 25:

<comment>Magic number `60` should use the existing `HTTP_TIMEOUT` constant. The PR removed the import but hardcoded the same value, which defeats the purpose of having the constant and makes future maintenance harder.</comment>

<file context>
@@ -8,18 +8,21 @@
     let response = client
         .head(url)
-        .timeout(std::time::Duration::from_secs(HTTP_TIMEOUT))
+        .timeout(std::time::Duration::from_secs(60))
         .send()
         .await?;
</file context>

[cubic-dev-ai]

I've fixed the magic number issue by replacing the hardcoded 60 with the existing HTTP_TIMEOUT constant from ia_get::constants.

Changes made:

• Added HTTP_TIMEOUT to the import statement on line 11
• Replaced 60 with HTTP_TIMEOUT on line 25

This ensures consistency with the defined constant and makes future maintenance easier.

PR: #92