If non-null, specifies manually-added classloaders that should be searched after the context classloader(s).

If non-null, specifies manually-added ModuleLayers that should be searched after the visible ModuleLayers.

URL schemes that are allowed in classpath elements (not counting the optional "jar:" prefix and/or "file:", which are automatically allowed).

Class accept/reject criteria (fully-qualified class names, with separator '.').

Classfile accept/reject criteria (path to classfiles, with separator '/', ending in ".class").

Package containing accept/reject criteriaed classes (with separator '.').

Path to accept/reject criteriaed classes (with separator '/').

If non-null, a list of filter operations to apply to classpath elements.

Classpath element resource path accept/reject criteria.

If true, don't scan runtime-invisible annotations (only scan annotations with RetentionPolicy.RUNTIME).

If true, enables the saving of annotation info (for class, field, method or method parameter annotations) during the scan. This information can be obtained using ClassInfo#getAnnotationInfo() etc. By default, annotation info is not scanned, for efficiency.

If true, scan classfile bytecodes, producing ClassInfo objects.

If true, allow external classes (classes outside of accepted packages) to be returned in the ScanResult, if they are directly referred to by an accepted class, as a superclass, implemented interface or annotation. Disabled by default.

If true, enables the saving of field info during the scan. This information can be obtained using ClassInfo#getFieldInfo(). By default, field info is not scanned, for efficiency.

If true, enables the determination of inter-class dependencies.

If true, use a MappedByteBuffer rather than the FileChannel API to access file content.

If true, enables the saving of method info during the scan. This information can be obtained using ClassInfo#getMethodInfo(). By default, method info is not scanned, for efficiency.

If true, all multi-release versions of a resource are found.

Enable the storing of constant initializer values for static final fields in ClassInfo objects.

If true, system jarfiles (rt.jar) and system packages and modules (java.*, jre.*, etc.) should be scanned .

If true, when classes have superclasses, implemented interfaces or annotations that are external classes, those classes are also scanned. (Even though this slows down scanning a bit, there is no API for disabling this currently, since disabling it can lead to problems -- see #261.)

If true, ignore class visibility. If false, classes must be public to be scanned.

If true, ignore field visibility. If false, fields must be public to be scanned.

If true, ignore method visibility. If false, methods must be public to be scanned.

If true, do not fetch paths from parent classloaders.

If true, do not scan module layers that are the parent of other module layers.

Whether to initialize classes when loading them.

Jar accept/reject criteria (leafname only, ending in ".jar").

lib/ext jar accept/reject criteria (leafname only, ending in ".jar").

The maximum size of an inner (nested) jar that has been deflated (i.e. compressed, not stored) within an outer jar, before it has to be spilled to disk rather than stored in a RAM-backed ByteBuffer when it is deflated, in order for the inner jar's entries to be read. (Note that this situation of having to deflate a nested jar to RAM or disk in order to read it is rare, because normally adding a jarfile to another jarfile will store the inner jar, rather than deflate it, because deflating a jarfile does not usually produce any further compression gains. If an inner jar is stored, not deflated, then its zip entries can be read directly using ClassGraph's own zipfile central directory parser, which can use file slicing to extract entries directly from stored nested jars.)

This is also the maximum size of a jar downloaded from an http:// or https:// classpath URL to RAM. Once this many bytes have been read from the URL's InputStream, then the RAM contents are spilled over to a temporary file on disk, and the rest of the content is downloaded to the temporary file. (This is also rare, because normally there are no http:// or https:// classpath entries.)

Default: 64MB (i.e. writing to disk is avoided wherever possible). Setting a lower max RAM size value will decrease ClassGraph's memory usage if either of the above rare situations occurs.

Module accept/reject criteria (with separator '.').

Commandline module path parameters.

If non-null, this list of ClassLoaders will be searched instead of the visible/context ClassLoader(s). In particular, this causes ClassGraph to ignore the java.class.path system property.

If non-null, specifies a list of classpath elements (String, URL or URI to use to override the default classpath.

If non-null, this list of ModuleLayers will be searched instead of the visible ModuleLayers.

Package accept/reject criteria (with separator '.').

Package prefix accept/reject criteria, for recursive scanning (with separator '.', ending in '.').

Path accept/reject criteria (with separator '/').

Path prefix accept/reject criteria, for recursive scanning (with separator '/', ending in '/').

If true, nested jarfiles (jarfiles within jarfiles) that are extracted during scanning are removed from their temporary directory (e.g. /tmp/ClassGraph-8JX2u4w) after the scan has completed. If false, temporary files are removed by the ScanResult finalizer, or on JVM exit.

If true, scan directories.

If true, scan jarfiles.

If true, scan modules.

If true, scan nested jarfiles (jarfiles within jarfiles).

Constructor for deserialization.