The browsers currently do not default to same-origin behaviour for
modules, so we need to be explicit in order for necessary
credentials to be passed along. This seems to be changing though,
but we need to wait for the browsers to actually roll out more
lenient defaults:
https://github.com/whatwg/fetch/pull/585
v1.0.0-testing.2 is the "make sure all of our autopublishing works
properly" release. Assuming everything goes well, it should shortly be
followed by v1.0.0 proper.
This updates the .npmignore file to ignore some new files that were
recently added, and to keep the `core` directory, so that noVNC can be
used as an ES6 module from NPM.
This also updates package.json to clean when generating lib files in the
`prepare` phase (which has replaced the `prepublish` phase).
A previous commit started (quasi-correctly) rewriting vendor import
paths on built files. The gist of it was correct, but it incorrectly
rewrote paths in vendor itself.
The babel plugin in use operated on canonical absolute paths. This mean
that it saw no difference between the import
`../vendor/pako/lib/utils/foo` and `../utils/foo`, where the later was
actually in the `vendor/pako/lib/bar` directory. This rewrote imports
in files in the vendor directory itself. However, since those files
were *already* in the correct relative location, the new import was
incorrect by a degree of `..`.
Now, we only rewrite vendor paths on things in the `core` directory.
This adds an option to `utils/use_require.js` which cleans the lib
and build directories before regenerating them. This will enable us
to make sure we always have a fresh copy of the lib dir before
publishing.
Converted version downloaded from sinonjs.org. Fixed version that
doesn't register itself on the global object. This forces all modules to
do a proper import.
This restores basic support for Intel AMT servers. They refuse clients
that request more than 16 bits per pixels, so implement a fallback
in just the "Raw" encoding.
The VNC protocol can't handle different deltas or speeds for a mouse
wheel event. When using a device that sends a lot of small mouse wheel
events, instead of fewer larger steps, the effect was that mouse wheel
scrolling was way to sensitive.
This patch looks at the delta of wheel events and doesn't send events
until the combined delta has passed a threshold. Single events that
doesn't pass the threshold get sent after a timeout in order to not
loose any events.
Fixes#577.
Samuel Mannehed
Dominik Csapak
Giannis Kosmas
Yuri van Oers
Mariusz Jamro
Pierre Ossman
Solly Ross
Solly Ross
Frederik Carlier