| Age | Commit message (Collapse) | Author |
|---|
|
|
|
Fix #1067 Corrected exports links regexp
|
|
|
|
refact inInternationalComposition
|
|
Add `userColor` query param to set initial color
|
|
Somehow </iframe> was no more. Now it is back.. Oh boy.
|
|
|
|
|
|
|
|
4 was a bit stingy :) 12 is a bit more friendly from a UX persepctive.
|
|
The utility functions colorutils.js assume that background colors are in
CSS hex format, so require userColor to do the same, rather than
allowing inputs like "red" and "rgba(...)", to insure that inversion
checks will succeed.
|
|
Use 'transparent' instead of 'white' as a reference color for validating
CSS color values. Presumably, a user setting a userColor wants some
color other than 'transparent' if they are setting it (they could always
duplicate the background's color if not).
|
|
Fixes issue introduced in 9be69ef2582dfc2c1b050041d4586cbd90d20e2c.
|
|
Add a URL parameter which sets the initial color for a user, e.g.:
http://example.com/p/mypad?userColor=%2300ff00
Sanitize the given color value to ensure that it's a valid css value
(could be any supported CSS color format -- #fff, rgba(), "red", etc).
Shortly after rejoining a pad, the server responds with a USER_NEWINFO
message which may contain an old color value; however, this message
arrives after we have set and sent the new color value to the server.
To avoid this race condition, if the query parameter has been set,
ignore the color value in a USER_NEWINFO message which matches our user
ID.
|
|
Added hooks and made some ace functions available to editorInfo Object
|
|
Fixed international composition issues (e.g., Japanese Input method)
|
|
moved inInternationalComposition from Ace2Inner to top window
fix bindTheEventHandlers() because ie9 implement CompositionEvent
when inInternationalComposition, NEW_CHANGES msg and ACCEPT_COMMIT msg
are pushed msgQueue.
when handleUserChanges(), apply msgQueue.
|
|
Instead of hacking with the internals of require, make client_plugins aware
and capable of sharing behavior.
|
|
There is virtually no shared code for the client, extract it into its own
module and do away with the switches.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Windows server fixes
|
|
|
|
With loopback removed, embedding concatenated resources doesn't happen. It
was a marginal feature to begin with so it will not be missed.
|
|
|
|
|
|
|
|
Fixed ESC key handling issue of Firefox 15
|
|
Loopback avoidance
|
|
|
|
|
|
|
|
|
|
For some reason, the client was sending the server a Unicode-normalized
version of inserted strings. So if for example we inserted the string
'ä' (i.e. \x61\xCC\x88) into the document, what would be sent to the
server would be 'ä' (i.e. \xC3\xA4).
This wouldn't be a problem on its own. BUT JavaScript reports that the
length of the first string is 2, while the length of the second one is
1.
So the command that was being sent to the server was 'Z:1>2*0+1$ä', when
it should really be 'Z:1>1*0+1$ä'. When the `checkRep` method checks the
length of the inserted string, it finds an inconsistency, and
disconnects the client.
We now normalize the inserted string before the command is generated, so
the length is always correct.
|
|
|
|
The new hook will accept only the message named in the hook call.
It will be used primarily for adding new message handlers, but it
can alse be used to handle existing message types.
|
|
Nice popups instead of modals
|
|
|
|
After 6507614e459dac868d3c76355ca013d13132bc79 the contentcollector does not properly "exit" lists back to the `none` listType anymore.
The repro steps on #769 seem to pass correctly after this change.
|
