fix: add UTF-8 encoding to read_text/write_text in Copilot ACP fs handlers

[AlexFucuson9]

Summary

Path.read_text() and Path.write_text() in the Copilot ACP client's fs/read_text_file and fs/write_text_file JSON-RPC handlers omit explicit encoding=. On Windows, these default to the system locale (typically cp1252), silently corrupting non-ASCII content — source code with unicode identifiers, CJK comments, emoji in strings, etc.

Changes

Line	Before	After
706	path.read_text()	path.read_text(encoding="utf-8")
735	path.write_text(str(params.get("content") or ""))	path.write_text(str(params.get("content") or ""), encoding="utf-8")

Why this matters

The Copilot ACP client acts as a file I/O bridge for external coding agents. Any non-ASCII content written through this bridge on Windows gets silently corrupted. The corruption is invisible until someone reads the file and sees mojibake.

Follows the same pattern as PR #56940 (os.fdopen encoding) and PR #62667 (write_text + ensure_ascii=False).

Test Plan

• Existing tests pass: pytest tests/agent/test_copilot_acp_client.py -xvs
• Verified no false positives — these calls have no kwargs dict or encoding on adjacent lines

[alt-glitch]

This was generated by AI during triage.

Duplicate of #38121 (the earliest open PR making the identical fix). This PR pins encoding="utf-8" on the exact same two fs/read_text_file / fs/write_text_file calls in agent/copilot_acp_client.py that #38121 (and sibling #38123) already fix; the closed #62090 was previously deduped to #38121 as well. Minor test-mock/comment differences don't change that. Consolidating on #38121 as canonical; linking #38119 (issue) and #38123 (sibling) for context.

[tonydwb]

Code Review Summary

Verdict: Comment

This PR adds UTF-8 encoding to read_text/write_text in Copilot ACP fs handlers. Small, targeted fix.

Please verify:

• UTF-8 encoding is applied correctly for both read and write operations
• Existing tests pass with the encoding change

---

Reviewed by Hermes Agent

[tonydwb]

Code Review Summary

Verdict: Comment (token read-only)

PR 64070: fix: add UTF-8 encoding to read_text/write_text in Copilot ACP fs handlers

Correctness ✅

• Adds explicit encoding="utf-8" to path.read_text() and path.write_text() calls.
• Addresses potential encoding issues on Windows (where the default encoding may differ).

No issues found.

---

Reviewed by Hermes Agent

[tonydwb]

Code Review Summary

Verdict: Comment (token read-only)

PR 64070: fix: add UTF-8 encoding to read_text/write_text in Copilot ACP fs handlers

Correctness ✅

• Adds explicit encoding="utf-8" to path.read_text() and path.write_text() calls.
• Addresses potential encoding issues on Windows (where the default encoding may differ).

No issues found.

---

Reviewed by Hermes Agent

[tonydwb]

Code Review Summary

Verdict: Comment

Changes

Explicit encoding="utf-8" on path.read_text() and path.write_text() in copilot_acp_client.py file operations.

Assessment

• Minor robustness fix. Explicit encoding prevents platform-dependent default encoding issues.
• No debug artifacts.

---

Reviewed by Hermes Agent

[teknium1]

Thanks for the focused cross-platform fix. The premise is valid on current main: agent/copilot_acp_client.py:706 and agent/copilot_acp_client.py:734 still use locale-default Path.read_text() and Path.write_text() in the two ACP filesystem handlers, and the proposed explicit UTF-8 arguments address those exact calls.

Problems

• This PR changes only agent/copilot_acp_client.py; it adds no regression test. The current focused test dispatch helper is at tests/agent/test_copilot_acp_client.py:130-142, and its ACP filesystem cases at :158-256 cover safety behavior rather than a successful non-ASCII read/write.
• This is duplicate work: open PR fix(copilot): use explicit utf-8 for ACP file shim #38121 contains the same fix in commit 020aeef5873d3b3817524888324adcdf182c93e3 and has already been reviewed as including the ACP UTF-8 regression coverage.

Suggested changes

• Consolidate on fix(copilot): use explicit utf-8 for ACP file shim #38121 so the production fix lands with its focused regression test.

This is an automated hermes-sweeper review.

[teknium1]

Closing as resolved by PR #71078 (merged, commit d372fda): the class-wide close-out salvaged your #50655/#54241/#56385/#66856/#65440 series as the backbone (authorship preserved in git log) and swept the remaining sites, so every read_text/write_text call this PR touches is now guarded on current main — verified per-site. A CI linter rule in check-windows-footguns.py plus the AST guard test now prevent regressions. Your overlapping/split variants of the same series are being closed together; the credit for the class rests on your commits.