improve error logging with file paths and context

[Sahil-u07]

Enhanced error reporting in concore.hpp so I/O failures include contextual details (file path/port/name and retry count) and are written to standard error. This makes debugging read/write failures much clearer while preserving existing behavior.

[Copilot]

Pull request overview

Enhances I/O failure logging in the Concore C++ implementation (concore.hpp) by emitting contextual read/write error messages (including file paths/ports/names and retry count) to standard error to aid debugging.

Changes:

• Replaced generic "Read error" / "skipping ..." stdout messages with contextual stderr messages for file-based reads/writes.
• Added contextual stderr messages for shared-memory read/write failures, including retry count for reads.
• Removed a stale comment about observed retry counts.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

This log line is inside a retry loop; using endl forces a flush on every iteration and can significantly slow down retries / spam stderr. Prefer writing a newline without flushing (e.g., '\n') and consider throttling the log (e.g., only every N retries) to avoid unbounded output if the read keeps failing.

Suggested change

	cerr << "Error reading " << inpath << port << "/" << name << " (retry " << retrycount << ")" << endl;
	if (retrycount % 100 == 0) {
	cerr << "Error reading " << inpath << port << "/" << name << " (retry " << retrycount << ")" << '\n';
	}

[Copilot]

This retry-loop log also uses endl (flush) on every iteration; that can materially degrade performance and overwhelm logs if shared memory stays unavailable. Prefer a non-flushing newline and consider rate-limiting the message (e.g., every N retries).

Suggested change

	cerr << "Error reading from shared memory port " << port << " name " << name << " (retry " << retrycount << ")" << endl;
	if (retrycount % 100 == 0) {
	cerr << "Error reading from shared memory port " << port << " name " << name << " (retry " << retrycount << ")" << '\n';
	}

[Copilot]

Pull request overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated 4 comments.

Comments suppressed due to low confidence (2)

concore.hpp:367

• If the input stays empty after MAX_RETRY attempts, the loop exits with ins == "" and the code later calls parser(ins) / accesses inval[0], which will throw or crash (empty string indexing / out-of-bounds). Add an explicit post-loop fallback (e.g., set ins = initstr or return an empty/default vector) before parsing.

                cerr << "Error reading " << inpath << port << "/" << name << " (retry " << retry << ")" << endl;
            }
            retry++;
        }
        s += ins;

concore.hpp:426

• If shared memory remains empty through MAX_RETRY, the loop exits with ins == "" and the code later calls parser(ins) / accesses inval[0], which will throw or crash. Add a post-loop guard/fallback before parsing when ins is still empty.

                cerr << "Error reading from shared memory port " << port << " name " << name << " (retry " << retry << ")" << endl;
            }
            retry++;
        }
        s += ins;

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

retrycount is a public field used elsewhere in the repo (e.g., scripts print concore.retrycount), but it is no longer incremented during retry attempts here. If the retry count is still intended to be observable, increment retrycount on each failed retry (or remove the field and update callers/docs).

Suggested change

	retry++;
	retry++;
	retrycount++;

[Copilot]

retrycount is a public field used elsewhere in the repo (callers print concore.retrycount), but it is no longer incremented during retry attempts in the shared-memory read loop. If this metric is still relied upon, increment it on each failed retry (or remove the field and update callers/docs).

Suggested change

	cerr << "Error reading from shared memory port " << port << " name " << name << " (retry " << retry << ")" << endl;
	cerr << "Error reading from shared memory port " << port << " name " << name << " (retry " << retry << ")" << endl;
	retrycount++;

[Copilot]

std::strncpy(sharedData_create, result.c_str(), 256 - 1) does not guarantee null-termination when the source is >= 255 bytes, which can leave stale bytes in shared memory and cause readers to see corrupted/concatenated messages. Explicitly null-terminate the buffer (and consider clearing it before copy) after the strncpy.

[Copilot]

The new error log in this catch(...) still drops the underlying failure reason. Prefer catching const std::exception& where possible and including e.what() (and/or errno/std::strerror(errno) for I/O failures) so the log is actionable beyond just the path.

[pradeeban]

Removing retrycount++; makes the code hard to read and understand by humans.