Add _ENCODE_FILE_EXISTING variable

[sabi789]

No description provided.

[augmentcode]

🤖 Augment PR Summary

Summary: Adds an environment override for how existing files are opened via _ENCODE_FILE_EXISTING, and expands UTF-8 support coverage.

Changes:

• Introduce __tag_existing_file() and apply it in the ASCII open/fopen overrides for regular files
• Update the manpage to document _ENCODE_FILE_NEW=UTF-8 and the new _ENCODE_FILE_EXISTING options
• Add unit tests covering UTF-8 new-file tagging and existing-file overrides (IBM-1047/BINARY/ISO8859-1/UTF-8)

_{🤖 Was this summary useful? React with 👍 or 👎}

[augmentcode]

Review completed. 3 suggestions posted.

Comment augment review to trigger a new review at any time.

[augmentcode]

man/zoslib.1:26: This .TP starts a new term immediately after _ENCODE_FILE_NEW=BINARY, which leaves the BINARY option with no description and makes the “tagged as BINARY” paragraph apply to the UTF-8 entry instead.

Severity: medium

_{🤖 Was this useful? React with 👍 or 👎, or 🚀 if it prevented an incident/outage.}

[augmentcode]

src/zos-io.cc:894: The return value from __tag_existing_file(fd) is ignored and errno is reset unconditionally, so failures (e.g., invalid _ENCODE_FILE_EXISTING or a failing tag operation) can be silently dropped while still returning a usable fd.

Severity: medium

Other Locations

• src/zos-io.cc:939

_{🤖 Was this useful? React with 👍 or 👎, or 🚀 if it prevented an incident/outage.}

[augmentcode]

test/test-clib-override.cc:191: This fd isn’t closed before remove(temp_path) and reassigning fd, which can leak a descriptor and make the subsequent remove() behavior platform-dependent. It also means the final close(fd) later in the test may double-close or close the wrong descriptor.

Severity: medium

_{🤖 Was this useful? React with 👍 or 👎, or 🚀 if it prevented an incident/outage.}