dsmiley commented on code in PR #4499:
URL: https://github.com/apache/solr/pull/4499#discussion_r3369644216
##########
gradle/validation/rat-sources.gradle:
##########
@@ -28,39 +39,102 @@ configure(rootProject) {
}
}
+/**
+ * Reads all file paths from the git index (DirCache) once and caches the
result.
+ * Returns null if this is not a git repository or the index cannot be read.
+ * Paths use forward slashes and are relative to the repository root.
+ */
+Set<String> loadGitTrackedFiles() {
+ if (rootProject.ext.has('ratGitIndexLoaded')) return
rootProject.ext.ratGitTrackedFiles
+ rootProject.ext.ratGitIndexLoaded = true
+ rootProject.ext.ratGitTrackedFiles = null
+
+ def dotGit = new File(rootProject.projectDir, ".git")
+ if (dotGit.isFile()) {
+ // git worktrees use a .git file — jgit does not fully support them
+ logger.warn("WARNING: git worktrees are not supported by jgit — RAT
git-tracking filter disabled.")
+ return null
+ }
+ if (!dotGit.isDirectory()) {
+ return null // not a git repository
+ }
+
+ def repository = null
+ try {
+ repository = new FileRepositoryBuilder()
+ .setWorkTree(rootProject.projectDir)
+ .setMustExist(true)
+ .build()
+ def tracked = new HashSet<String>()
+ def dirCache = repository.readDirCache()
+ for (int i = 0; i < dirCache.getEntryCount(); i++) {
+ tracked.add(dirCache.getEntry(i).getPathString())
+ }
+ rootProject.ext.ratGitTrackedFiles = tracked
+ return tracked
+ } catch (Exception e) {
+ logger.warn("WARNING: Could not read git index for RAT check:
${e.message}")
+ return null
+ } finally {
+ if (repository != null) repository.close()
+ }
+}
+
+/**
+ * Returns the set of git-tracked file paths relative to the given project
directory,
+ * or null if git tracking is unavailable.
+ */
+Set<String> gitTrackedFiles(File dir) {
+ def allFiles = loadGitTrackedFiles()
+ if (allFiles == null) return null
+
+ def prefix = rootProject.projectDir.toPath().relativize(dir.toPath())
+ .toString().replace('\\', '/')
+ if (prefix.isEmpty()) return allFiles
+
+ prefix += "/"
+ def result = new HashSet<String>()
+ for (String path : allFiles) {
+ if (path.startsWith(prefix)) {
+ result.add(path.substring(prefix.length()))
+ }
+ }
+ return result
+}
+
// Configure the rat validation task and all scanned directories.
allprojects {
- task("rat", type: RatTask) {
+ tasks.register("rat", RatTask) {
group = 'Verification'
description = 'Runs Apache Rat checks.'
def defaultScanFileTree = project.fileTree(projectDir, {
- // Don't check under the project's build folder.
- exclude project.buildDir.name
-
- // Exclude any generated stuff.
- exclude "src/generated"
+ // Only check files tracked by git — skip untracked/gitignored files
+ // (IDE artifacts, AI assistant configs, etc.)
+ def trackedFiles = gitTrackedFiles(projectDir)
+ if (trackedFiles != null) {
+ exclude { element ->
+ // Allow directories through (they are just containers), exclude
untracked files
+ !element.isDirectory() &&
!trackedFiles.contains(element.relativePath.pathString)
+ }
+ }
- // Don't recurse into local Lucene dev repo.
- exclude "/lucene"
+ // Exclude the build directory — even though it's not git-tracked, Gradle
+ // validates input/output overlaps at configuration time before the git
+ // filter closure runs, so this must be excluded explicitly.
+ exclude project.layout.buildDirectory.get().asFile.name
// Don't check any of the subprojects - they have their own rat tasks.
exclude subprojects.collect {it.projectDir.name}
- // At the module scope we only check selected file patterns as folks
have various .gitignore-d resources
- // generated by IDEs, etc.
+ // The git index filter above excludes untracked files. These include
+ // patterns select the file types that should carry license headers.
include "**/*.xml"
include "**/*.md"
include "**/*.py"
include "**/*.sh"
include "**/*.bat"
- // include build.gradle but exclude .gradle directories
include "**/*.gradle"
Review Comment:
I think it's very lame that we have license headers on build files
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
