Method names with leading double underscore are reserved by PHP.

move IXR XML RPC to composer dependency #1970

This is a monky patched version of kissyfrot/php-ixr until my PR has
been accepted upstream.

Plugins using the XML RPC client may need adjustments.

:fire: fix the calculation of file permissons

Our config allows to set the values for `dmode` and `fmode` to allow
users to explicitly define which permissions directories and files
should have.

To

:fire: fix the calculation of file permissons

Our config allows to set the values for `dmode` and `fmode` to allow
users to explicitly define which permissions directories and files
should have.

To avoid unnessary chmod operations, we check the current umask to
compare what permissions files and directories would get witout our
intervention. If the result is already what the user wants, no chmods
will happen later on. Otherwise we set new configs called `dperm` and
`fperm` which will be used in chmod ops. This is done in
`init_creationmodes()`

When we created new directories, we used to pass the original `dmode`
config to `mkdir()`. The system will then apply the umask to that
`dmode`.

This means the resulting directory will *always* have different
permissions than `dmode`, *always* requiring a chmod operation.
That's silly.

**Breaking Change:** This patch removes the passing of `dmode` as
second parameter to all `mkdir` calls, making it default to `0700`
which is also what we test against in `init_creationmodes()`.

Plugins not relying on our `io_*` functions and do create their own
directories and which currenlty pass `dmode` to it need to be
adjusted to remove that second parameter.

Users may want to reapply the proper file permissions to their data
folder.

**Revert:** In 9fdcc8fcd87114ca59a1764a84d213a53c655c8c @movatica
introduced a change to `init_creationmodes()` that compared the umask
against `fmode` instead of `0666`. I merged it because it looked logical
when compared to the code for directories which compared against `dmode`
as described above. However we do not pass `fmode` to any file creation
methods (that's not possible).

The result is that all changes made in the `fmode` setting resulted
in the wrong permissions for newly created files as first reported in
https://forum.dokuwiki.org/d/19463-setting-fmode-not-working-as-intended

I'm unsure about the orginal motivation behind @movatica's change. The
"fix" however, is wrong.

**Tests:** This patch introduces an integration test that will check the
actual results of directory and file creations under various umask,
`dmode` and `fmode` settings.

show more ...

Add support for SHA256 encrypted passwords

Enable detection of SHA256 hashes (prefixed with $5$) and allow to
use this hashing schemes in passwords.

add unit test for namespace exclusion in ft_pageLookup()

Add tests on sectionID and fixes #3436

This commit fixes a bug on sectionID which could lead to having duplicated id.

Note that this commit changes the API a bit $check used to be a key-value array

Add tests on sectionID and fixes #3436

This commit fixes a bug on sectionID which could lead to having duplicated id.

Note that this commit changes the API a bit $check used to be a key-value array
(with key=the origin id, and value=the number of time this id has been generated)
and is now just an array of string. That's because the previous structure
couldn't work.

As far as Dokuwiki sources are concerned, this change should have no impact because:
- most call to sectionID are done with $check=false (and the behavior in this case
isn't changed
- only Doku_Renderer->_headerToLink passes an actual array to this method. But this
array is initialized empty, it is only filled and read by sectionID itself.
(this Doku_Renderer->headers array is protected but the classes which extends it
aren't referring this array).

It could still break usages of pluging that would call sectionID and rely on the
format of this array. But even if this commits kept $check as an array<string, bool>
(for instance setting the values to 1) the semantic wouldn't be the same.

To put it in a nutshell:
- this change of API has 0 impact on the core of Dokuwiki
- there is no evidence that it would impact other code
- there doesn't seem to be a clean way to avoid this change

This commit fixes #3436

show more ...

test: run test in separate process in case of error

If there is an error when executing the method under test, then this
test caused the downstream test \remoteapicore_test::test_getBacklinks
to une

test: run test in separate process in case of error

If there is an error when executing the method under test, then this
test caused the downstream test \remoteapicore_test::test_getBacklinks
to unexpectedly fail. Probably, because the global plugin controller
would not have been reset to its original value.

Running this test in its own process ensures that it won't affect any
other test, even if it messes with the global state and doesn't clean up.

show more ...

test: fix two tests on PHP8

On PHP8 trying to access a non-existing array key leads to a warning,
which we might treat as errors?

This incidentally also fixes another test downstream which broke be

test: fix two tests on PHP8

On PHP8 trying to access a non-existing array key leads to a warning,
which we might treat as errors?

This incidentally also fixes another test downstream which broke because
this rendering test changes the global plugin controller, but doesn't
clean it up that if the test errors.

-9000 is used as the position to make it obvious that this number is
meaningless for this test.

show more ...

style: fix test code style

This makes all the basic phpcs style fixes and adds visibility to the
class methods.

Improve remote_test::test_notExistingCall()

There are actually 2 distinct test cases of invalid methods:

1. Not following the definition in Api class's PHPDoc block
- core methods begin by a 'do

Improve remote_test::test_notExistingCall()

There are actually 2 distinct test cases of invalid methods:

1. Not following the definition in Api class's PHPDoc block
- core methods begin by a 'dokuwiki' or 'wiki' followed by a . and
the method name itself.
- plugin methods are formed like 'plugin.<plugin name>.<method name>'
2. Method "type" does not begin with 'dokuwiki', 'wiki' or 'plugin'

A new check was added so both are now covered.

Test was modified to make use of PHPUnit expectException() method,
instead of relying on a try/catch block.

show more ...

Fix incorrect assertion in input_test::test_set()

$INPUT is set to 'test2', but the $_POST checks for 'test'.

Fix various errors in PHPUnit tests on PHP 8

fix version comparison in phpunit downloader

fix deprecated phpunit assertions

remove old DokuCLI class

This has been deprecated in 2017. Removing it instead of fixing tests
and PHP8 compatibility

fix PHP8 errors in tests

These were problems within the tests itself. There are many more
problems in the actual code.

add needed type hints for phpunit8

This will break a lot of plugin tests, but can't be avoided

remove unnamespaced php unit wrapper

default to phpunit8 now

test translations for correctness

This inspired by https://www.dokuwiki.org/teams:i18n:translation-check
by @schplurtz Instead of relying on an external script that only checks
after a translation w

test translations for correctness

This inspired by https://www.dokuwiki.org/teams:i18n:translation-check
by @schplurtz Instead of relying on an external script that only checks
after a translation was merged, this test can be run in out CI
environment and help us to avaoid merging faulty translations.

show more ...

simplified the media_searchlist tests

PHP 8: rename match() to apply() in Lexer #3298

add new log dir to test data directory

Replace Appveyor and Travis with Github actions

Travis has recently introduced built limits which makes using them very
slow sometimes and I never really liked the setup for appveyor for our
Windows

Replace Appveyor and Travis with Github actions

Travis has recently introduced built limits which makes using them very
slow sometimes and I never really liked the setup for appveyor for our
Windows testing.

This makes use of github actions instead. Some notes:

* since we only support PHP 7.2+ anymore, nothing below is tested
* MySQL and Postgres auth backends are only tested on Linux (as before)
* I am not actually sure the problem matchers are working as intended
* There is currently no good way to run tests that are allowed to fail
so no PHP 8 tests are run yet
* I haven't reestablished the no-gzip test runs again, I'm not sure they
are still relevant on modern PHP systems
* PHP 7.4 currently fails because of https://github.com/MarcusSchwarz/lesserphp/pull/18
* I'm currently using @v2 tags for the used action. We had a discussion
before if we should actually pin the action to a fixed commit. For
sake of easier maintainance, I would argue to trust the setup-php
action.

show more ...

Fix groups match in auth_ismanager and auth_isadmin

Even if a user was passed to the check but no groups, current user's groups were used for the match