Update version to 0.19.1 and improve file upload initiation

[parashardhapola]

• Bump package version to 0.19.1.
• Modify the file upload initiation to include the file size in the request, enhancing the upload process.
• Update validation logic to ensure the number of presigned URLs is at least equal to the expected chunk count, improving error handling during uploads.

[qodo-code-review]

Review Summary by Qodo

Version 0.19.1: Enhance file upload with size parameter

✨ Enhancement 🐞 Bug fix

Walkthroughs

Description

• Bump package version to 0.19.1
• Include file size in upload initiation request
• Relax presigned URL validation to allow extra URLs
• Improve error message clarity for URL count mismatches

Diagram

flowchart LR
  A["Upload Initiation"] -->|"Include file_size_bytes"| B["post_json Request"]
  B -->|"Receive presigned URLs"| C["Validation Logic"]
  C -->|"Changed: != to <"| D["Allow Extra URLs"]
  D -->|"Improved error message"| E["Better Error Handling"]

Loading

File Changes

1. cytetype/__init__.py ⚙️ Configuration changes +1/-1

Bump version to 0.19.1

• Update package version from 0.19.0 to 0.19.1

cytetype/init.py

---

2. cytetype/api/client.py ✨ Enhancement +7/-3

Enhance upload initiation and URL validation logic

• Change upload initiation from post_empty to post_json to include file size
• Add file_size_bytes parameter in the request payload
• Relax presigned URL validation from exact match (!=) to minimum requirement (<)
• Update error message to reflect that extra URLs are acceptable

cytetype/api/client.py

---

[qodo-code-review]

Code Review by Qodo

🐞 Bugs (1) 📘 Rule violations (0) 📎 Requirement gaps (0)

1. Extra URLs silently accepted 🐞 Bug ⛯ Reliability

Description

The presigned URL validation was relaxed from rejecting any mismatch to only rejecting “too few”
URLs. Since the client uploads exactly n_chunks (derived from size_bytes/chunk_size_bytes) and
completes with exactly n_chunks parts, accepting extra URLs can mask a server/client chunking
mismatch and defer a clear error until late (after time/bandwidth is spent).

Code

cytetype/api/client.py[R80-84]

+    if use_r2 and len(presigned_urls) < n_chunks:  # type: ignore[arg-type]
        raise ValueError(
            f"Server returned {len(presigned_urls)} presigned URLs "  # type: ignore[arg-type]
-            f"but expected {n_chunks} (one per chunk)."
+            f"but need at least {n_chunks} (one per chunk)."
        )

Evidence

The client computes n_chunks and only ever uploads chunk indices in range(n_chunks), indexing
presigned URLs by chunk_idx. It then completes multipart upload with exactly n_chunks parts.
Therefore, if the server returns more presigned URLs than n_chunks, the extras are ignored and any
underlying contract mismatch is no longer caught early by validation.

cytetype/api/client.py[74-84]
cytetype/api/client.py[117-127]
cytetype/api/client.py[210-217]

Agent prompt

The issue below was found during a code review. Follow the provided context and guidance below and implement a solution

## Issue description
The client now accepts `len(presigned_urls) > n_chunks` without any signal. Since the client only uploads `range(n_chunks)` and completes with exactly `n_chunks` parts, extra URLs are ignored and a server/client mismatch can be detected only after spending time uploading.

## Issue Context
This validation was intentionally relaxed to allow `>= n_chunks`, but the codebase has no other safeguard or observability to detect “too many URLs” as a potential contract mismatch.

## Fix Focus Areas
- cytetype/api/client.py[74-85]
- cytetype/api/client.py[117-127]
- cytetype/api/client.py[210-217]িম

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

---

ⓘ The new review experience is currently in Beta. Learn more