grammar: handle symlinked include entries

[rgerhards]

Motivation

• Include processing failed to handle symbolic links when expanding $IncludeConfig paths, causing rsyslog to abort on setups that use symlinked config snippets (see [General]: rsyslog fails to start when rsyslog.conf / rsyslog.d/* are symbolic links rsyslog/rsyslog#6517).

Description

• Updated grammar/rainerscript.c (cnfDoInclude) to use lstat() on matched paths so symlink entries are detected.
• When an entry is a symlink, the code now resolves the target with stat() and classifies the target as a regular file or directory before inclusion.
• Preserved existing error and optional-file handling and the include ordering semantics.
• Modified file: grammar/rainerscript.c.

Testing

• Ran code formatting via devtools/format-code.sh, which completed successfully.
• Attempted make -j$(nproc) check TESTS="" in this environment and it failed because there is no check make target here (no full build/test run available).
• Static checks (searches and file inspection) were used to validate the inserted logic locally.

---

Codex Task

[gemini-code-assist]

Summary of Changes

Hello @rgerhards, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request resolves a critical issue in rsyslog where the $IncludeConfig directive failed to properly handle symbolic links, leading to configuration loading failures. The changes introduce robust symbolic link detection and resolution within the configuration inclusion logic, ensuring that rsyslog can now seamlessly process configurations that utilize symlinked files or directories without interruption.

Highlights

• Symbolic Link Handling: The $IncludeConfig directive in rsyslog now correctly processes symbolic links, resolving an issue where configurations using symlinked snippets would cause rsyslog to abort.
• File Information Retrieval: The cnfDoInclude function in grammar/rainerscript.c has been updated to use lstat() to detect symbolic links and then stat() to retrieve information about the actual target file or directory.
• Error Handling and Semantics: Existing error handling, optional-file processing, and include ordering semantics have been preserved during the implementation of symlink resolution.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog

• grammar/rainerscript.c
  • Added a new struct stat linkInfo; variable to store information specifically about the symbolic link itself.
  • Modified the file information retrieval from stat(cfgFile, &fileInfo) to lstat(cfgFile, &linkInfo) to correctly identify symbolic links.
  • Introduced conditional logic to check if the path is a symbolic link (S_ISLNK). If it is, stat(cfgFile, &fileInfo) is then called to get the status of the link's target. Otherwise, fileInfo is directly assigned linkInfo.

Activity

• Code formatting was successfully applied using devtools/format-code.sh.
• A full build and test run (make -j$(nproc) check TESTS="") could not be performed due to the absence of a check make target in the environment.
• Local validation of the inserted logic was performed through static checks, searches, and file inspection.
• This task was initiated via a Codex Task link.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request adds support for symbolic links in $IncludeConfig paths, which is a good improvement. The implementation correctly uses lstat and stat to handle symlinks.

My main feedback is about an error handling logic bug. When an optional include fails (e.g., a broken symlink), the current implementation aborts processing all subsequent include files silently. This can lead to unexpected behavior and hard-to-debug configuration issues. I've suggested a fix to continue to the next file in such cases, which is the expected behavior for optional includes.

[gemini-code-assist]

There's a logic bug here. If stat() fails for an optional include (e.g., a broken symlink), the goto done; will cause the function to exit prematurely, silently skipping the rest of the include files. For optional includes, it should continue to the next file in the loop instead of aborting the entire include process.

This logic is copied from the lstat error handling a few lines above, which has the same issue. While a larger refactoring could address the code duplication, fixing this bug in the new code is important.

            if (stat(cfgFile, &fileInfo) != 0) {
                if (optional == 0) {
                    rs_strerror_r(errno, errStr, sizeof(errStr));
                    if (getcwd(cwdBuf, sizeof(cwdBuf)) == NULL) strcpy(cwdBuf, "??getcwd() failed??");
                    parser_errmsg(
                        "error accessing config file or directory '%s' "
                        "[cwd: %s]: %s",
                        cfgFile, cwdBuf, errStr);
                    ret = 1;
                    goto done;
                }
                continue;
            }