https://bidule.menf.in/api/v1/statuses/82725 is an example of a valid status even though it contains a null url field. Either my application should reject this, or the API should explicitely allow it.
Just tested running `vagrant up --provider virtualbox` and it did not download the Ubuntu 14.04 base image, it downloaded xenial64, which is 16.04. Hopefully you will edit this :)
* add fields_attributes in /api/v1/accounts/update_credentials
* add fields in Accounts and verify_credentials
* add fields in Accounts and verify_credentials
* add fields in Accounts and verify_credentials
* Adding copy config step
As discussed in #591, docker will refuse to run the `mastodon:setup` step without some configuration file present.
* The secret generation part of docker installation process
As noticed in #6883, you need to generate the keys manually.
* Setting correct file owner is important even in the prebuilt-image case
... as evidenced by issue #6911
* This step should not be necessary -- let's fix the installation instead
From Yarn 1.0 onwards, scripts don't require "--" for options to be forwarded. In a future version, any explicit "--" will be forwarded as-is to the scripts.
mastodon-jfx is a desktop client for Mastodon written in kotlin using TornadoFX. It is designed for power tooters, particularly ones with multiple accounts spanning multiple instances. It is currently in an Alpha state. It is not yet feature-complete and the Look & Feel is in progress.
* Resorted apps, clarify what's FOSS and proprietary, and ToC
Cause CLI apps and Bots overlapped, I mostly combined them under Command line clients.
* That's not needed anymore
* Not needed, is already clear from the name 'web tools'
* Translated FAQ to french from mastodon funding question to search box question.
* Revert "Translated FAQ to french from mastodon funding question to search box question."
This reverts commit db0a35802a.
* Updated french translation for FAQ
* update after review
* minor adjustment on hastags question
As Mastodon temporaliry saves uploaded content to memory and disk (if
/tmp is a disk), unlimiting client max body size makes the server
vulnerable to DoS attack.
Without headless a lot of (for a server) unneeded dependencies needs be installed (on Debian stretch), like icon sets and x11-utils. Elasticsearch works without problems when you only install openjdk-8-jre-headless (that has much less dependencies).
Consolidated all basic configuration steps for the ideal case to top of document, moved all memory-related and "advanced" configuration to bottom. Removed remote Elasticsearch config from document entirely and replaced it with a link to ES documentation on the topic instead.