MAINT Support relative blob paths in AzureBlobStorageIO

[biefan]

Summary

This updates AzureBlobStorageIO so that relative blob paths are handled consistently across blob operations.

Problem

Several AzureBlobStorageIO methods document or imply support for either:

• a full blob URL, or
• a relative blob path inside the configured container

In particular, read_file() explicitly documents relative-path support. However, the implementation always routed the input through parse_blob_url(), which only accepts full URLs and raises ValueError for relative paths.

That means calls such as:

await storage.read_file("dir1/dir2/sample.png")

fail immediately with ValueError("Invalid blob URL") even though the method contract says the relative path should be used as-is.

The same issue also affects path_exists() and is_file(), and it unnecessarily constrains write_file() to URL-only inputs as well.

Fix

• Added a _resolve_blob_name helper that:
  • extracts the blob name from a full blob URL (via parse_blob_url), or
  • passes relative blob paths through unchanged
  • uses try/except around parse_blob_url to avoid redundant urlparse calls
  • normalizes backslashes to forward slashes so that Path objects created on Windows produce valid blob names
• Used that helper in read_file(), write_file(), path_exists(), and is_file()
• Updated write_file() docstring to document that both full URLs and relative paths are accepted
• Documented intentional container name discard behavior in _resolve_blob_name docstring

This keeps the existing full-URL behavior intact while making the relative-path behavior match the documented API.

Tests

Added regression coverage for relative-path handling in all four affected methods:

• read_file()
• write_file()
• path_exists()
• is_file()

Added edge case tests for _resolve_blob_name:

• Full URL resolution
• Relative path passthrough
• Simple filename
• Backslash normalization
• Path object handling

Validation command:

pytest tests/unit/models/test_storage_io.py -q

Validation result:

22 passed in 0.26s

[Copilot]

Pull request overview

Updates AzureBlobStorageIO so callers can pass either a full Azure Blob URL or a relative blob path (within the configured container) consistently across blob operations.

Changes:

• Added a _resolve_blob_name() helper to accept either full blob URLs or relative blob paths.
• Updated read_file(), write_file(), path_exists(), and is_file() to use the helper instead of requiring URL parsing.
• Added regression tests for relative-path handling in read_file() and path_exists().

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.

File	Description
pyrit/models/storage_io.py	Adds blob-name resolution helper and applies it across Azure blob operations to support relative paths.
tests/unit/models/test_storage_io.py	Adds unit tests validating relative-path behavior for read_file() and path_exists().

---

You can also share your feedback on Copilot code review. Take the survey.

[romanlutz]

It's a theoretical problem since we're not using relative blob paths but nevertheless not a bad idea.