1. Simplify user creation to create a non system user (so the uid will usually
match the host user).
2. Set the user to redash & remove the need to change user in docker entrypoint.
Fix comment in bin/pre_compile.
Remove .gitattributes and .travis.yml in rd_ui/.
Remove bin/run from Procfile.heroku.
Update documentation:
-Add a note about upgrading from version to version.
-Remove commands for DATABASE_URL and REDIS_URL.
-Add importance to the cookie secret variable.
-Merge adding redis and postgres addons into 1 step.
Move .bowerrc, bower.json, gulpfile.js, package.json
down to root level.
Update paths in .bowerrc, gulpfile.js, Makefile
Add a heroku-postbuild step to package.json which
installs devDependencies and runs the build.
Add step in bin/pre_compile which adds the
requirements_all_ds.txt to requirements.txt to ensure that
cffi is installed. Also removes pymssql as this is
not supported on Heroku.
Add content from rd_ui/.gitignore to .gitignore and
remove rd_ui/.gitignore.
Add section in setup.rst about Heroku deployments.
Also reset circle.yml to be as in the main redash repo so our changes are not included in the changeset
Check earlier PR https://github.com/getredash/redash/pull/1027 to follow conversation.
* Web interface to add and delete data sources, without the need to ssh
into the server.
* Ability to safely delete datasources -- query results from this data sources
are deleted, while queries get assigned null datasource.
* Updated the BigQuery datasource to use the JSON key file from Google Developer
console. Also both BigQuery and the Google Spreadsheets datasource no longer store
their key on the filesystem, but rather in the DB.
* Minor updates to the Flask Admin.
