- Sep 16, 2019
-
-
Jaime Pérez Crespo authored
This resolves #1176
-
Jaime Pérez Crespo authored
Apparently, IE does not mark buttons as submit buttons when type is missing by default, making it impossible to submit a form. This resolves #908.
-
Jaime Pérez Crespo authored
This comment used to be true, but the behaviour was changed in 3c52b289 in order to purge old data more often.
-
Jaime Pérez Crespo authored
This is something the developer of a module is supposed to deal with. If you are just using SSP, it doesn't make much sense to have your log files flooded by errors about missing translation files or directories, since you cannot do anything about it and the original english text will be used anyway.
-
Jaime Pérez Crespo authored
If we reach this point, we're likely already logging the error somewhere else (i.e. the session handler in use), so there's no point on logging the exception's message again.
-
- Sep 13, 2019
-
-
Tim van Dijen authored
-
- Sep 12, 2019
-
-
Tim van Dijen authored
-
- Sep 11, 2019
-
-
Jørn Åne authored
-
Jaime Pérez Crespo authored
config of module.php cache headers
-
Jaime Pérez Crespo authored
Fix Admin UI glitch
-
- Sep 06, 2019
-
-
Tim van Dijen authored
-
Tim van Dijen authored
Resort to entityid
-
Jørn Åne authored
-
Jørn Åne authored
-
Jørn Åne authored
-
- Sep 04, 2019
-
-
Jaime Pérez Crespo authored
The SimpleSAML\Session::expireData() method did not mark the session as dirty when there was expired data on it, so if nothing else changed, the data was never actually purged. It was done like this by design, but in practice, it seems like sessions aren't modified as often, meaning they end up growing a lot with each state array that's stored on them, and expired data is never removed. We now check for expired data in the save() method (which is run every time a session is destroyed, if not manually) and if there is any, we mark the session as dirty, so that it is actually updated in the backend. Most of the time this will be transparent and have no visible performance hit, as it'll be run after the response is sent, during shutdown. This closes #1053
-
- Aug 31, 2019
-
-
Tim van Dijen authored
-
Tim van Dijen authored
* Rationalize SQL Store insertOrUpdate
-
- Aug 30, 2019
-
-
Galen Charlton authored
It wasn't immediately clear to me where to put the cert and private key when I went through these instructions recently, so I thought I'd propose a change to make the destination explicit.
-
- Aug 29, 2019
-
-
Tim van Dijen authored
-
Tim van Dijen authored
Closes #1103
-
Tim van Dijen authored
Integers for example are not uncommon.. This closes #1152
-
Tim van Dijen authored
-
Tim van Dijen authored
-
Tim van Dijen authored
-
Tim van Dijen authored
-
Tim van Dijen authored
Allow the new major version to be used as well (lock-file remains at 3.x)
-
Tim van Dijen authored
-
Tim van Dijen authored
-
Tim van Dijen authored
-
Jaime Pérez Crespo authored
This was inline in the script, and later migrated to a twig template. This made us force the use of twig, or otherwise the template would not work for the old user interface. We shouldn't do that, and we should have a proper PHP template in place.
-
Jaime Pérez Crespo authored
When we are logging something and the logger hasn't been initialized, the messages are deferred and flushed by a shutdown handler. However, when we initialize a session, we set the track ID in the logger, effectively initializing it with everything we need. When we log something and then load the session (or create it), any subsequent logs will be dumped, but the deferred logs will wait until shutdown, changing the order in which they were emitted. This patch makes sure we flush the deferred logs as soon as we have a track ID (can either be set manually or by, most typically, loading a session). That way, we get those logs out as soon as we can, and subsequent logs entries will be dumped afterwards, keeping their relative order. This closes #1107
-
- Aug 28, 2019
-
-
Tim van Dijen authored
* Force template to be rendered by Twig
-
Tim van Dijen authored
-
Tim van Dijen authored
-
Tim van Dijen authored
-
Tim van Dijen authored
-
- Aug 27, 2019
-
-
Tim van Dijen authored
* Fix column and index-size for MySQL only
-
- Aug 26, 2019
-
-
Tim van Dijen authored
-
Tim van Dijen authored
-