* Rich Presence support using Discord Social SDK
Download DiscordSocialSdk-1.4.9649.zip from
https://discord.com/developers/applications/1394782217405862001/social-sdk/downloads
Add -DUSE_DISCORD:BOOL=ON to your cmake line.
The Discord app needs to be set to be a public client in the OAuth2 tab.
All Discord-related code are contained within one file, llstartup.cpp,
and other classes access it through some opaque layer, static functions,
otherwise we'd get these "duplicate symbol" linking errors.
* Move Discord-related code to llappviewer.cpp
The doFrame is the one called over and over again, so running the
Discord callbacks from there shouldn't have one extra function
overhead, while running the Discord initialisation is only once so
it's much more okay to have the extra function overhead there.
* panel_preferences_privacy tabs
Add tab and checkboxes for discord social SDK integration options to panel_preferences_privacy.xml
* Shorten Discord-related local variable names
* Connect to Discord now through privacy tab
Now the access token is saved the way passwords are saved, but
without a username, so we can have some persistence without having
to implement an OAuth2 backend server cause we would have to store
those tokens there anyway still, and it's just simpler to not go
that way. Discord Social SDK doesn't have a helper for sending code
to a custom server anyway, that we would have to have some
asynchronous HTTP requestor ready.
Show location check button gets enabled only when Discord
integration is enabled, though it's not functioning yet.
* Location for Discord Rich Presence Activity State
I was going to use LLAgentUI::buildLocationString but there's no
location format that shows only region and coords without having
to have the parcel name empty, so I copied buildLocationString
implementation in the case of LOCATION_FORMAT_NO_MATURITY but when
the parcel name is empty.
I had to make updateDiscordActivity check agent's ID and the
existence of agent avatar pointer first before trying to set
Activity Details or State, cause I like the "Show location" button
be checkable not only after online when both the ID & pointer will
have existed. I think this way is simpler than programmatically
enabling the "Show location" button after the user is logged in.
I put a trigger to Activity update somewhere after the user is
logged in for now, not yet after a TP.
The elapsed time gets reset whenever Activity is updated for now,
but I'll try to make elapsed time extended instead.
No Party for now, because I couldn't find a way to make a Party
shown without showing its CurrentSize (I could still get away not
showing its MaxSize by setting it to 0), so the State (location)
is shown above the elapsed time, not on the right of it.
I'll try to figure out to get some representative numbers for its
CurrentSize & MaxSize next.
Also no privacy on hiding the username for now, until the UI is
ready.
* Update Rich Presence location on region change
I had to find a spot in source code where it doesn't cause a crash
(it did in LLAgent::setRegion), but I'm not removing the one in
llstartup.cpp because on login, the one in llviewermessage.cpp
gets only the placeholder coords (10, 10, 10).
* Show display name too on Discord Rich Presence
Avatar name cache can be used right away upon login now after I
moved the update call to the end of PRECACHE section in llstartup.
* Show Discord Rich Presence Activity Party
By setting CurrentSize to the number of people within chat radius,
and MaxSize to the number of people within near range.
* Call updateDiscordActivity too in Discord init
so when the user enables the integration after being logged in,
the init can show the name and location right away.
* Discord Rich Presence: Hide name & connect to llappviewer.cpp
Add option to show/hide avatar name in privacy panel & connect rich presense directly to llappviewer.cpp
* Discord time elapsed not reset on region change
Time elapses right after viewer launch even before login.
Plus parameter name change in header to make it the same as in
implementation.
* Cache bool setting retrievals in updateDiscordActivity
As suggested by Andrey Kleschev.
getBOOL and getF32 are expensive, so using `static LLCachedControl<>`
is the way to do it in llappviewer.cpp.
* Check Discord creds existence before getting token
as suggested by Andrey Kleshchev, anticipating external factors
such as user moving settings from another PC.
* Tracy visibility for looped Discord function calls
As suggested by Andrey Kleshchev. They likely can get pricey so
they need to be visible in the profiler.
* Discord-related error handling/logging
plus delay saving Discord credentials to only after the access
token is successfully updated on Discord, and try to disconnect
from Discord when the integration gets disabled regardless whether
there are credentials to delete or not and whether there's an
access token to revoke or not.
* Use getAvatars already called for Discord Party numbers
so we don't have to make any extra getAvatars calls just for this,
as it's pricy in crowds, and we'll just be piggybacking
`updateSpeakerList` and `updateNearbyList`.
* Assemble Discord Activity Details only once
by saving it to a static global string for reuse.
* Remove updateDiscordActivity call in startup loop
The State field (region & coords) is updated well enough without it
now.
* Rename handleDiscordSocial to toggleDiscordIntegration
* Update Discord Activity only when integration is enabled
No need to check setting for the status change callback one,
because getting there would need to be connected to Discord first,
which in turn needs the integration to be enabled first.
---------
Co-authored-by: Secret Foxtail <remmy@megapahit.net>
With the emojis support, a new font was added, which not only provides emojis
but also fancy colorful replacements for UTF-8 characters that used to be
supported by our fallback (monochrome) fonts: this causes discrepancies and
unwanted/undesired changes in scripted objects menus (e.g. an empty circle or
square may render as a black, full one, a heart may render red instead of white),
not to mention the larger font size used by the emoji characters...
This patch restores the aspect of such menus/dialogs/UI elements with UTF-8
characters that *are* supported by the usual fallback fonts (fonts which may
also vary from one viewer to another, and from one OS to another), so that
everything keeps working/rendering as it always did so far, while not impairing
the use of new colorful emojis.
This second proposal ensures that:
- "genuine" emojis (in the 0x1f000-0x1ffff range), will *always* be rendered
using the new emojis font (this solves, for example, the monochrome "yellow
faces" issue seen with some characters in my first proposal).
- Special UTF-8 characters (in the 0x2000-0x32FF range) which have been used by
scripters so far, will render as they used to, using the monochrome fallback
fonts (this repairs scripted dialogs menus).
- Remaining special characters, that do not have a corresponding glyph in the
monochrome font, but do have one in the emojis font, will use the latter font
to render.
It also got the nice side-effect of removing the dependency on the ICU4C library.
Note however that the recent commit:
326055ba82
will need to be reverted to allow this patch to actually fix scripted dialogs.
Also, some cleanup might be needed in skins/default/xui/*/emoji_characters.xml to
remove from it the special UTF-8 characters that will no longer be rendered with
fanciful colors, but instead with the monochrome font glyphs.
Since now VLC is responsible for played parcel media (should be since
SL-19042) should be possible to switch remaining audio to OpenAL with
no loss of functionality
This commit adds the HBXX64 and HBXX128 classes for use as a drop-in
replacement for the slow LLMD5 hashing class, where speed matters and
backward compatibility (with standard hashing algorithms) and/or
cryptographic hashing qualities are not required.
It also replaces LLMD5 with HBXX* in a few existing hot (well, ok, just
"warm" for some) paths meeting the above requirements, while paving the way for
future use cases, such as in the DRTVWR-559 and sibling branches where the slow
LLMD5 is used (e.g. to hash materials and vertex buffer cache entries), and
could be use such a (way) faster algorithm with very significant benefits and
no negative impact.
Here is the comment I added in indra/llcommon/hbxx.h:
// HBXXH* classes are to be used where speed matters and cryptographic quality
// is not required (no "one-way" guarantee, though they are likely not worst in
// this respect than MD5 which got busted and is now considered too weak). The
// xxHash code they are built upon is vectorized and about 50 times faster than
// MD5. A 64 bits hash class is also provided for when 128 bits of entropy are
// not needed. The hashes collision rate is similar to MD5's.
// See https://github.com/Cyan4973/xxHash#readme for details.
This commit adds the HBXX64 and HBXX128 classes for use as a drop-in
replacement for the slow LLMD5 hashing class, where speed matters and
backward compatibility (with standard hashing algorithms) and/or
cryptographic hashing qualities are not required.
It also replaces LLMD5 with HBXX* in a few existing hot (well, ok, just
"warm" for some) paths meeting the above requirements, while paving the way for
future use cases, such as in the DRTVWR-559 and sibling branches where the slow
LLMD5 is used (e.g. to hash materials and vertex buffer cache entries), and
could be use such a (way) faster algorithm with very significant benefits and
no negative impact.
Here is the comment I added in indra/llcommon/hbxx.h:
// HBXXH* classes are to be used where speed matters and cryptographic quality
// is not required (no "one-way" guarantee, though they are likely not worst in
// this respect than MD5 which got busted and is now considered too weak). The
// xxHash code they are built upon is vectorized and about 50 times faster than
// MD5. A 64 bits hash class is also provided for when 128 bits of entropy are
// not needed. The hashes collision rate is similar to MD5's.
// See https://github.com/Cyan4973/xxHash#readme for details.
Ansariel