___ _ _ ____ / | __________(_|_) __ \____ _ ____ __ / /| | / ___/ ___/ / / / / / __ \| |/_/ / / / / ___ |(__ ) /__/ / / /_/ / /_/ /> </ /_/ / /_/ |_/____/\___/_/_/_____/\____/_/|_|\__, / /____/
GENERATE_XML = YES
GENERATE_XML = YES JAVADOC_AUTOBRIEF = YES OPTIMIZE_OUTPUT_JAVA = YES
Doxygen does not support all kinds of annotations. For example the annotations of return types are
removed. To trick Doxygen into passing annotations to AsciiDoxy, you need to transform them into a
name that does not contain
@. AsciiDoxy supports changing type prefixes like
__AT__Annotation__ back into
@Annotation. For example to pass through all annotations
with information about nullability, add the following to the Doxygen file:
FILTER_PATTERNS = *.java="sed -E 's/@(Nullable|NonNull|CheckForNull|PossiblyNull)([[:space:]]+[[:alnum:]_])/__AT__\1__\2/g'"
GENERATE_XML = YES EXTENSION_MAPPING = h=objective-c MACRO_EXPANSION = YES EXPAND_ONLY_PREDEF = YES PREDEFINED = NS_ASSUME_NONNULL_BEGIN= NS_UNAVAILABLE= NS_DESIGNATED_INITIALIZER=
To fully support python docstrings formats like the Google Python Style Guide it is recommended to use doxypypy as a filter for Doxygen.
GENERATE_XML = YES FILTER_PATTERNS = *.py="./py_filter.sh"
The contents of py_filter.sh should be:
#!/bin/bash doxypypy -a -c $1 | sed 's/# /# /'
sed command is a workaround for a bug in
Alternatively you can disable MarkDown support and run doxypypy directly:
GENERATE_XML = YES FILTER_PATTERNS = *.py="doxypypy -a -c" MARKDOWN_SUPPORT = NO
Support for type hints in Doxygen is still limited: https://github.com/doxygen/doxygen/issues/7320.
Doxygen will ignore PlantUML diagrams if the path to PlantUML is not set. This is also the case when only the XML format is enabled and Doxygen has no need to call PlantUML. To make sure PlantUML diagrams are exported to XML, make sure to set the path:
PLANTUML_JAR_PATH = /usr/share/java/plantuml/plantuml.jar