Doxygen EXCLUDE_PATTERNS regex - doxygen

I am attempting to exclude certain files from my doxygen generated documentation. I am using version 1.8.14.
My files come in this naming convention:
/Path2/OtherFile.cs
/Path/DAL.Entity/Source.cs
/Path/DAL.Entity/SourceBase.generated.cs
I want to exclude all files that do NOT end in Base.generated.cs, and are located inside of /Path/.
Since it appears doxygen claims to use regex for the exclude_patterns variable, I eventually came up with this:
.*\\Path\\DAL\..{4,15}\\((?<!Base\.generated).)*
Needless to say, it did not work. Nor did multiple other variations. So far a simple wildcard * is the only regex character I have gotten to actually work.
doxygen uses QRegExp for a lot of things, so I assumed that was the library used for this variable as well, but even several variations of a pattern that that library claims to support did not work; granted apparently that library is full of bugs, but I would expect some things to work.
Does doxygen actually use a regex library for this variable?
If so, which library is it?
In either case, is there a method of achieving my goal?

My conclusion is; No... Doxygen Doxyfile does not support real regex. Even though they claim that it do. It's just standard wildcards that work.
We ended up with a really awkward solution to work around this.
What we did is that we added a macro in our CMakeLists.txt that creates a string with everything we want to include in INPUT instead. Manually excluding the parts we don't want.
The sad part is that CMakes regex also is crippled. So we couldn't use advanced regex such as negative lookahead in LIST(FILTER EXLUDE) similar to LIST(FILTER children EXCLUDE REGEX "^((?!autogen/public).)*$")... So even this solution is not really what we wanted.
Our CMakeLists.txt ended up looking something like this
cmake_minimum_required(VERSION 3.9)
project(documentation_html LANGUAGES CXX)
find_package(Doxygen REQUIRED dot)
# Custom macros
## Macro for getting all relevant directories when creating HTML documentain.
## This was created cause the regex matching in Doxygen and CMake are lacking support for more
## advanced syntax.
MACRO(SUBDIRS result current_dir include_regex)
FILE(GLOB_RECURSE children ${current_dir} ${current_dir}/*)
LIST(FILTER children INCLUDE REGEX "${include_regex}")
SET(dir_list "")
FOREACH(child ${children})
get_filename_component(path ${child} DIRECTORY)
IF(${path} MATCHES ".*autogen/public.*$" OR NOT ${path} MATCHES ".*build.*$") # If we have the /source/build/autogen/public folder available we create the doxygen for those interfaces also.
LIST(APPEND dir_list ${path})
ENDIF()
ENDFOREACH()
LIST(REMOVE_DUPLICATES dir_list)
string(REPLACE ";" " " dirs "${dir_list}")
SET(${result} ${dirs})
ENDMACRO()
SUBDIRS(DOCSDIRS "${CMAKE_SOURCE_DIR}/docs" ".*.plantuml$|.*.puml$|.*.md$|.*.txt$|.*.sty$|.*.tex$|")
SUBDIRS(SOURCEDIRS "${CMAKE_SOURCE_DIR}/source" ".*.cpp$|.*.hpp$|.*.h$|.*.md$")
# Common config
set(DOXYGEN_CONFIG_PATH ${CMAKE_SOURCE_DIR}/docs/doxy_config)
set(DOXYGEN_IN ${DOXYGEN_CONFIG_PATH}/Doxyfile.in)
set(DOXYGEN_IMAGE_PATH ${CMAKE_SOURCE_DIR}/docs)
set(DOXYGEN_PLANTUML_INCLUDE_PATH ${CMAKE_SOURCE_DIR}/docs)
set(DOXYGEN_OUTPUT_DIRECTORY docs)
# HTML config
set(DOXYGEN_INPUT "${DOCSDIRS} ${SOURCEDIRS}")
set(DOXYGEN_EXCLUDE_PATTERNS "*/tests/* */.*/*")
set(DOXYGEN_FILE_PATTERNS "*.cpp *.hpp *.h *.md")
set(DOXYGEN_RECURSIVE NO)
set(DOXYGEN_GENERATE_LATEX NO)
set(DOXYGEN_GENERATE_HTML YES)
set(DOXYGEN_HTML_DYNAMIC_MENUS NO)
configure_file(${DOXYGEN_IN} ${CMAKE_BINARY_DIR}/DoxyHTML #ONLY)
add_custom_target(docs
COMMAND ${DOXYGEN_EXECUTABLE} ${CMAKE_BINARY_DIR}/DoxyHTML -d Markdown
WORKING_DIRECTORY ${CMAKE_BINARY_DIR}
COMMENT "Generating documentation"
VERBATIM)
and in the Doxyfile we added the environment variables for those fields
OUTPUT_DIRECTORY = #DOXYGEN_OUTPUT_DIRECTORY#
INPUT = #DOXYGEN_INPUT#
FILE_PATTERNS = #DOXYGEN_FILE_PATTERNS#
RECURSIVE = #DOXYGEN_RECURSIVE#
EXCLUDE_PATTERNS = #DOXYGEN_EXCLUDE_PATTERNS#
IMAGE_PATH = #DOXYGEN_IMAGE_PATH#
GENERATE_HTML = #DOXYGEN_GENERATE_HTML#
HTML_DYNAMIC_MENUS = #DOXYGEN_HTML_DYNAMIC_MENUS#
GENERATE_LATEX = #DOXYGEN_GENERATE_LATEX#
PLANTUML_INCLUDE_PATH = #DOXYGEN_PLANTUML_INCLUDE_PATH#
After this we can run cd ./build && cmake ../ && make docs to create our html documentation and have it include the autogenerated interfaces in our source folder without including all the other directories in the build folder.
Quick description of what actually happens in the CMakeLists.txt
# Macro that gets all directories from current_dir recursively and returns the result to result as a space separated string
MACRO(SUBDIRS result current_dir include_regex)
# Gets all files recursively from current_dir
FILE(GLOB_RECURSE children ${current_dir} ${current_dir}/*)
# Filter files so we only keep the files that match the include_regex (can't be to advanced regex)
LIST(FILTER children INCLUDE REGEX "${include_regex}")
SET(dir_list "")
# Let us act on all files... :)
FOREACH(child ${children})
# We're only interested in the path. So we get the path part from the file
get_filename_component(path ${child} DIRECTORY)
# Since CMakes regex also is crippled we can't do nice things such as LIST(FILTER children EXCLUDE REGEX "^((?!autogen/public).)*$") which would have been preferred (CMake regex does not understand negative lookahead/lookbehind)... So we ended up with this ugly thing instead... Adding all build/autogen/public paths and not adding any other paths inside build. I guess it would be possible to write this expression in regex without negative lookahead. But I'm both not really fluent in regex (who are... right?) and a bit lazy in this case. We just needed to get this one pointer task done... :P
IF(${path} MATCHES ".*autogen/public.*$" OR NOT ${path} MATCHES ".*build.*$")
LIST(APPEND dir_list ${path})
ENDIF()
ENDFOREACH()
# Remove all duplicates... Since we GLOBed all files there are a lot of them. So this is important or Doxygen INPUT will overflow... I know... I tested...
LIST(REMOVE_DUPLICATES dir_list)
# Convert the dir_list to a space seperated string
string(REPLACE ";" " " dirs "${dir_list}")
# Return the result! Coffee and cinnamon buns for everyone!
SET(${result} ${dirs})
ENDMACRO()
# Get all the pathes that we want to include in our documentation ... this is also where the build folders for the different applications are going to be... with our autogenerated interfaces which we want to keep.
SUBDIRS(SOURCEDIRS "${CMAKE_SOURCE_DIR}/source" ".*.cpp$|.*.hpp$|.*.h$|.*.md$")
# Add the dirs we want to the Doxygen INPUT
set(DOXYGEN_INPUT "${SOURCEDIRS}")
# Normal exlude patterns for stuff we don't want to add. This thing does not support regex... even though it should.
set(DOXYGEN_EXCLUDE_PATTERNS "*/tests/* */.*/*")
# Normal use of the file patterns that we want to keep in the documentation
set(DOXYGEN_FILE_PATTERNS "*.cpp *.hpp *.h *.md")
# IMPORTANT! Since we are creating all the INPUT paths our self we don't want Doxygen to do any recursion for us
set(DOXYGEN_RECURSIVE NO)
# Write the config
configure_file(${DOXYGEN_IN} ${CMAKE_BINARY_DIR}/DoxyHTML #ONLY)
# Create the target that will use that config to create the html documentation
add_custom_target(docs
COMMAND ${DOXYGEN_EXECUTABLE} ${CMAKE_BINARY_DIR}/DoxyHTML -d Markdown
WORKING_DIRECTORY ${CMAKE_BINARY_DIR}
COMMENT "Generating documentation"
VERBATIM)
I know this isn't the answer anyone who stumbles in on this question wants... unfortunately it seems to be the only reasonable solution...
... you all have my deepest condolences...

Related

Select files using include and exclude array of recursive glob patterns

I've been given two file glob parameters in JSON, include and exclude, in the following format:
{
include: ['**/*.md', '**/swagger/*.json', '**/*.yml', 'somedir/*.yml'],
exclude: ['**/obj/**', 'otherdir/**', '**/includes/**']
}
I'm tasked with walking a directory tree to select files according to the include and exclude rules in these formats; this has to be written as a Powershell script.
I've been trying to find a built-in command that supports the double-asterisk, recursive file glob pattern; additionally, since Powershell is converting the JSON to an object, it would be nice if the command parameters could accept an array as input.
I've looked at Get-ChildItem, but I'm not sure that I can mimic the glob resolution behavior using -include, -exclude, and/or -filter. I've also looked at Resolve-Path, but I'm not sure if the wildcards will work correctly (and I might have to manually exclude paths).
How can I select paths using multiple recursive wildcard file globs in Powershell while excluding other globs? Is there a Powershell command that supports this?
Thank you!
EDIT:
In these glob patterns, the single asterisk is a regular wildcard. The double asterisk (**), however, is a known standard which denotes a recursive directory search.
For example: the pattern dir1/*/file.txt would match:
dir1/dir2/file.txt
dir1/dir3/file.txt
...but not:
dir1/dir2/dir3/file.txt
The pattern dir1/**/file.txt would match everything that the above selector would, but it would also match:
dir1/dir3/dir4/file.txt
dir1/dir7/dir9/dir23/dir47/file.txt
and so on. So, an exclude glob pattern like **/obj/** basically means "exclude anything found in any obj folder found at any point in the directory hierarchy, no matter how deep".

Uppercasing filename in Makefile using sed

I try to convert a filename such as foo/bar/baz.proto into something like foo/bar/Baz.java in my Makefile. For this purpose, I thought I could use sed. However, it seems that the command does not work as expected:
uppercase_file = $(shell echo "$(1)" | sed 's/\(.*\/\)\(.*\)/\1\u\2/')
# generated Java sources
PROTO_JAVA_TARGETS := ${PROTO_SPECS:$(SRCDIR)/%.proto=$(JAVAGEN)/$(call uppercase_file,%).java}
When I try to run the sed command on the command line it seems to work:
~$ echo "foo/bar/baz" | sed 's/\(.*\/\)\(.*\)/\1\u\2/'
foo/bar/Baz
Any ideas why this does not work inside the Makefile?
UPDATE:
The java files are generated with the following target:
$(JAVAGEN)/%.java: $(SRCDIR)/%.proto
How can I apply the substitution also for targets?
GNU Make does not replace % character in the replacement part of a substitution reference (which is basically a syntactic sugar for patsubst) if it is part of a variable reference. I have not found this behavior described in the documentation, but you can look it implemented in the source code (the relevant function I believe is find_char_unquote).
I suggest moving the call out of the substitution reference, since uppercase_file obviously works properly on any file path:
PROTO_JAVA_TARGETS := $(call uppercase_file,${PROTO_SPECS:$(SRCDIR)/%.proto=$(JAVAGEN)/%.java})
If $(PROTO_SPECS) resolves not to a single element, but rather to a list of elements, you can use foreach to call the function on every elements of a processed list:
PROTO_JAVA_TARGETS := $(foreach JAVA,${PROTO_SPECS:$(SRCDIR)/%.proto=$(JAVAGEN)/%.java},$(call uppercase_file,$(JAVA)))
The java files are generated with the following target: $(JAVAGEN)/%.java: $(SRCDIR)/%.proto
How can I apply the substitution also for targets?
Since Make matches targets first, and there is no way to run sed backwards, what you need here is either define an inverse function, or generate multiple explicit rules. I will show the latter approach.
define java_from_proto
$(call uppercase_file,$(1:$(SRCDIR)/%.proto=$(JAVAGEN)/%.java)): $1
# Whatever recipe you use.
# Use `$$#`, `$$<` and so on instead of `$#` or `$<`.
endef
$(foreach PROTO,$(PROTO_SPECS),$(eval $(call java_from_proto,$(PROTO))))
We basically generate one rule per file in $(PROTO_SPEC) using a multiline variable syntax, and then use eval to install that rule. There is also a very similar example on this documentation page that can be helpful.

perl quoting in ftp->ls with wildcard

contents of remote directory mydir :
blah.myname.1.txt
blah.myname.somethingelse.txt
blah.myname.randomcharacters.txt
blah.notmyname.1.txt
blah.notmyname.2.txt
...
in perl, I want to download all of this stuff with myname
I am failing really hard with the appropriate quoting. please help.
failed code
my #files;
#files = $ftp->ls( '*.myname.*.txt' ); # finds nothing
#files = $ftp->ls( '.*.myname.*.txt' ); # finds nothing
etc..
How do I put the wildcards so that they are interpreted by the ls, but not by perl? What is going wrong here?
I will assume that you are using the Net::FTP package. Then this part of the docs is interesting:
ls ( [ DIR ] )
Get a directory listing of DIR, or the current directory.
In an array context, returns a list of lines returned from the server. In a scalar context, returns a reference to a list.
This means that if you call this method with no arguments, you get a list of all files from the current directory, else from the directory specified.
There is no word about any patterns, which is not suprising: FTP is just a protocol to transfer files, and this module only a wrapper around that protocoll.
You can do the filtering easily with grep:
my #interesting = grep /pattern/, $ftp->ls();
To select all files that contain the character sequence myname, use grep /myname/, LIST.
To select all files that contain the character sequence .myname., use grep /\.myname\./, LIST.
To select all files that end with the character sequence .txt, use grep /\.txt$/, LIST.
The LIST is either the $ftp->ls or another grep, so you can easily chain multiple filtering steps.
Of course, Perl Regexes are more powerful than that, and we could do all the filtering in a single /\.myname\.[^.]+\.txt$/ or something, depending on your exact requirements. If you are desperate for a globbing syntax, there are tools available to convert glob patterns to regex objects, like Text::Glob, or even to do direct glob matching:
use Text::Glob qw(match_glob);
my #interesting = match_glob ".*.myname.*.txt", $ftp->ls;
However, that is inelegant, to say the least, as regexes are far more powerful and absolutely worth learning.

CMake macro across CMakeLists

I have a c++ project which directory structure like below:
server/
code/
BASE/
Thread/
Log/
Memory/
Net/
cmake/
CMakeList.txt
BASE/
CMakeList.txt
Net/
CMakeList.txt
here is part of /cmake/CMakeList.txt:
MACRO(SUBDIRLIST result curdir)
FILE(GLOB children RELATIVE ${curdir} ${curdir}/*)
SET(dirlist "")
FOREACH(child ${children})
IF(IS_DIRECTORY ${curdir}/${child})
SET(dirlist ${dirlist} ${child})
ENDIF()
ENDFOREACH()
SET(${result} ${dirlist})
ENDMACRO()
add_subdirectory(Base)
then use macro in /cmake/Base/CMakeList.txt:
SET(SUBDIR, "")
SUBDIRLIST(SUBDIRS, ${BASE_SRC_DIR})
message("SUBDIRS : " ${SUBDIRS})
output:
SUBDIRS :
I check ${dirlist} by output it's value in macro, I get directory list expected,but when message("result " ${result}) after SET(${result} ${dirlist}),I can not get output expected , what's wrong with my CMakeLists.txt?
There are a couple of minor issues here:
In your macro, SET(dirlist "") could be just SET(dirlist). Likewise, SET(SUBDIR, "") could be just SET(SUBDIRS) (I guess "SUBDIR" is a typo and should be "SUBDIRS". Also you don't want the comma in the set command - probably another typo?)
To output the contents of ${result} in the macro, use message("result: ${${result}}"), since you're not appending ${child} to result each time, but to ${result}. In your example ${result} is SUBDIRS, so ${${result}} is ${SUBDIRS}.
When you call SUBDIRLIST, don't use a comma between the arguments.
When you output the value of SUBDIRS, include ${SUBDIRS} in the quotes, i.e. message("SUBDIRS: ${SUBDIRS}") or you'll lose the semi-colon separators.
Other than those, your macro seems fine.

zsh filename globbling/substitution

I am trying to create my first zsh completion script, in this case for the command netcfg.
Lame as it may sound I have stuck on the first hurdle, disclaimer, I know how to do this crudely, however I seek the "ZSH WAY" to do this.
I need to list the files in /etc/networking but only the files, not the directory component, so I do the following.
echo $(ls /etc/network.d/*(.))
/etc/network.d/ethernet-dhcp /etc/network.d/wireless-wpa-config
What I wanted was:
ethernet-dhcp wireless-wpa-config
So I try (excuse my naivity) :
echo ${(s/*\/)$(ls /etc/network.d/*(.))}
/etc/network.d/ethernet-dhcp /etc/network.d/wireless-wpa-config
It seems that this doesn't work, I'm sure there must be some clever way of doing this by splitting into an array and getting the last part but as I say, I'm complete noob at this.
Any advice gratefully received.
General note: There is no need to use ls to generate the filenames. You might as well use echo some*glob. But if you want to protect the possible embedded newline characters even that is a bad idea. The first example below globs directly into an array to protect embedded newlines. The second one uses printf to generate NUL terminated data to accomplish the same thing without using a variable.
It is easy to do if you are willing to use a variable:
typeset -a entries
entries=(/etc/network.d/*(.)) # generate the list
echo ${entries#/etc/network.d/} # strip the prefix from each one
You can also do it without a variable, but the extra stuff to isolate individual entries is a bit ugly:
# From the inside, to the outside:
# * glob the entries
# * NUL terminate them into a single string
# * split at NUL
# * strip the prefix from each one
echo ${${(0)"$(printf '%s\0' /etc/network.d/*(.))"}#/etc/network.d/}
Or, if you are going to use a subshell anyway (i.e. the command substitution in the previous example), just cd to the directory so it is not part of the glob expansion (plus, you do not have to repeat the directory name):
echo ${(0)"$(cd /etc/network.d && printf '%s\0' *(.))"}
Chris Johnsen's answer is full of useful information about zsh, however it doesn't mention the much simpler solution that works in this particular case:
echo /etc/network.d/*(:t)
This is using the t history modifier as a glob qualifier.
Thanks for your suggestions guys, having done yet more reading of ZSH and coming back to the problem a couple of days later, I think I've got a very terse solution which I would like to share for your benefit.
echo ${$(print /etc/network.d/*(.)):t}
I'm used to seeing basename(1) stripping off directory components; also, you can use echo /etc/network/* to get the file listing without running the external ls program. (Running external programs can slow down completion more than you'd like; I didn't find a zsh-builtin for basename, but that doesn't mean that there isn't one.)
Here's something I hope will help:
haig% for f in /etc/network/* ; do basename $f ; done
if-down.d
if-post-down.d
if-pre-up.d
if-up.d
interfaces