v6.73+dev1230 - December 30 2022

[vitalker]

v6.73+dev1230 - December 30 2022

• * Includes feature branch: routing matrix/track wiring input activity indicators
• * Includes feature branch: preview item selection for grouped tracks
• * Includes feature branch: VST3 per-channel silence reporting compatibility option
• * Includes feature branch: better .RfxChain media explorer and clipboard integration
• * Includes feature branch: ReaReaRea timestretch mode
• * Includes feature branch: GR metering as embedded UI for third-party VSTs
• * Includes feature branch: JSFX new features and EEL2 preprocessor
• * Includes feature branch: render dialog statistics display improvements
• * Includes feature branch: selected media item appearance changes
• * Includes feature branch: media item fixed lanes
• + Filtering: fix matching of accented characters at start of search filter
• + Filtering: support matching latin 1 extended A characters when searching for unaccented characters
• + IDE/gfx.getchar: more international keyboard support tweaks for Windows (will probably break things)
• + IDE: improve home-key behavior (thanks Giorgos Vougioukas)
• + Media item lanes: allow moving items from a single non-fixed-lane track to specific lanes on a fixed lane track [p=2628432]
• + Mousewheel: improve responsiveness of 'View: Zoom vertically' action (Ctrl+Mousewheel)
• + Preferences: add preference for maximum vertical zoom
• + ReaScript: FX_GetNamedConfigParm add support for ACS x2/y2 parameters [t=274236]
• + Render wildcards: support case conversion for latin-supplemental and latin-extended-A accented characters [t=274162]
• + Routing matrix/track wiring: show if signal is present on audio/MIDI inputs
• + Routing matrix: improve record input selection indication on multichannel record inputs
• + Track grouping manager: support non-ascii characters in group names
• + Windows: fix track manager gridline display when hiding columns [t=274194]
• + macOS: fix ampersand drawing in tooltips [p=2628462]
• # API: midi editor play cursor preview override works when project playback stopped
• # Localization: correct group manager column context menu section in template [t=273946]
• # Media item lanes: fix interaction between track selectedness and lane button click
• # Razor edits: fix interaction with automation items and beat timebase [p=2627949]

This thread is for pre-release features discussion. Use the Feature Requests forum for other requests.

Changelog - Pre-Releases

Generated by X-Raym's REAPER ChangeLog to BBCode

[vitalker]

Not sure when it started, but it seems horizontal item copying is no longer possible. Mouse modifiers were resetted just before.

[vitalker]

Originally Posted by vitalker

+ Preferences: add preference for maximum vertical zoom

Nice one, very handy, thanks!

[vitalker]

Originally Posted by vitalker

+ Routing matrix/track wiring: show if signal is present on audio/MIDI inputs

I don't see MIDI signal monitoring in track wiring.

[schwa]

Originally Posted by vitalker

Not sure when it started, but it seems horizontal item copying is no longer possible.

Oops, thanks, fixing.

[dsyrock]

Originally Posted by vitalker

• * Includes feature branch: JSFX new features and EEL2 preprocessor

Any detail about these?

[vitalker]

Originally Posted by dsyrock

Any detail about these?

https://forum.cockos.com/showthread.php?t=273907

[IrishRover79]

You're giving us the ability to delete lanes one by one. Please could we also have the option to delete all lanes but the one we want to keep, i.e. the equivalent of crop to active take? Thanks.

[Meo-Ada Mespotine]

Regarding the gfx.getchar fixes, what could break due to them?

[Tale]

Well, for starters I can't type quotes/accents again in the IDE on my Windows 10 US-International keyboard layout, just like with v6.73+dev1219.

[PhelixK]

+ Preferences: add preference for maximum vertical zoom

Nice, but if zoom<100, then the action:
View: Toggle track zoom to maximum height doesn't toggle anymore..

[PhelixK]

* Includes feature branch: media item fixed lanes

I notice quite a marked change in the coverage of the muted overlay. When comparing to 6.63+dev075 I don't see any changes in the theme itself in that release, so it would be very nice to know if this change has a reason or maybe occurred unintentionally?

Here's a direct comparison

[ovnis]

As there are now many actions available to be able to manage the lanes, would it be possible to add an option to not display the number/button of the fixed lanes?



The problem is that it takes space at the TCP level. And it's less space to display other elements such as the list of FX, volume, pan, etc. Thx you!

[Justin]

Originally Posted by Tale

Well, for starters I can't type quotes/accents again in the IDE on my Windows 10 US-International keyboard layout, just like with v6.73+dev1219.

Hmm how do I enable that layout? I can do English (various places), but not US-International...


edit: ahh I think I have a solution figured out, silly me

edit:

as another side note -- gfx.getchar() returning unicode characters is a bit not OK, because we already used various high bits for modifiers for latin characters. I shouldn't have done the unicode support for 6.72/6.73 or whatever it was. So instead, I'm going to change it so it will return 'u'<<24 + unicode_char, to distinguish between that and the other keys.


a big underlying issue is that gfx.getchar() also reports if you type ctrl+alt+E, as Ctrl+Alt+E, which is "261". The issue is, on a german keyboard, that gets converted into the Euro character. So we should send that too! So -- I think this is OK, it will be up to the script on how to handle potentially duplicate keys. example:

'a' (the key a)
256+1 (ctrl+alt+e) -- either it uses this as a hotkey, or not?
'u'<<24 + whatever that euro character is I forget - unicode character, could be generated by the same keypress

the script can just look for the values that make sense in its current context and act accordingly. If entering text, it'd look for normal ascii characters or (code>>24)=='u'... anything else >=256 or 1..26 can be assumed to be ctrl/alt/etc modified keys, or arrows/etc.

[ovnis]

Copy/past (drag&drop) with CTRL+left click doesn't work with vertical move.

[Embass]

[Embass]

[akademie]

Originally Posted by vitalker

v6.73+dev1230 - December 30 2022

• + Render wildcards: support case conversion for latin-supplemental and latin-extended-A accented characters [t=274162]

Very nice,
although there is still strangely wrong capitalization conversion for "ž" character:

ěščřžýáíé-xxx-ĚŠČŘŽÝÁÍÉ.wav
ĚŠČŘſÝÁÍÉ-XXX-ĚŠČŘŽÝÁÍÉ.wav

[vitalker]

Originally Posted by ovnis

Copy/past (drag&drop) with CTRL+left click doesn't work with vertical move.

Yeah, check my first post and Schwa's answer.

[Meo-Ada Mespotine]

Originally Posted by Justin

Hmm how do I enable that layout? I can do English (various places), but not US-International...


edit: ahh I think I have a solution figured out, silly me

edit:

as another side note -- gfx.getchar() returning unicode characters is a bit not OK, because we already used various high bits for modifiers for latin characters. I shouldn't have done the unicode support for 6.72/6.73 or whatever it was. So instead, I'm going to change it so it will return 'u'<<24 + unicode_char, to distinguish between that and the other keys.


a big underlying issue is that gfx.getchar() also reports if you type ctrl+alt+E, as Ctrl+Alt+E, which is "261". The issue is, on a german keyboard, that gets converted into the Euro character. So we should send that too! So -- I think this is OK, it will be up to the script on how to handle potentially duplicate keys. example:

'a' (the key a)
256+1 (ctrl+alt+e) -- either it uses this as a hotkey, or not?
'u'<<24 + whatever that euro character is I forget - unicode character, could be generated by the same keypress

the script can just look for the values that make sense in its current context and act accordingly. If entering text, it'd look for normal ascii characters or (code>>24)=='u'... anything else >=256 or 1..26 can be assumed to be ctrl/alt/etc modified keys, or arrows/etc.

Hmm, just a thought but, to prevent possible breaking of thing, why not reverting gfx.getchar to previous behavior and adding gfx.getchar_utf to support only utf-characters without the high bit modifiers.

So if I want to get focusstate, I use gfx.getchar while if I want to get utf8, I use gfx.getchar_utf instead.

This wouldn't break thing, imho, and still add support for utf8 without bitshifting-complications.

Don't know if both getchar-functions should share the same keyboard queue just interpreted differently or if both would need their own.

Means, if

A=gfx.getchar()
B=gfx.getchar_utf()

both return the first character in the queue or if A holds the first and B holds the second...

[Justin]

Originally Posted by Meo-Ada Mespotine

Hmm, just a thought but, to prevent possible breaking of thing, why not reverting gfx.getchar to previous behavior and adding gfx.getchar_utf to support only utf-characters without the high bit modifiers.

So if I want to get focusstate, I use gfx.getchar while if I want to get utf8, I use gfx.getchar_utf instead.

hmm perhaps we'll do something like:

a,unicode_char = gfx.getchar()
a == ('u'<<24 + unicode) -- unicode

unicode_char == unicode -- unicode

-- or if unicode_char is 0, usual extended syntax:
a == 261 -- ctrl+alt e

(maybe, hmm need to think about this more)

[Justin]

Originally Posted by akademie

Very nice,
although there is still strangely wrong capitalization conversion for "ž" character:

ěščřžýáíé-xxx-ĚŠČŘŽÝÁÍÉ.wav
ĚŠČŘſÝÁÍÉ-XXX-ĚŠČŘŽÝÁÍÉ.wav

oops yep a big range of them are wrong, fixing!

[akademie]

Originally Posted by Justin

oops yep a big range of them are wrong, fixing!

Thanks

[Meo-Ada Mespotine]

Originally Posted by Justin

hmm perhaps we'll do something like:

a,unicode_char = gfx.getchar()
a == ('u'<<24 + unicode) -- unicode

unicode_char == unicode -- unicode

-- or if unicode_char is 0, usual extended syntax:
a == 261 -- ctrl+alt e

(maybe, hmm need to think about this more)

The first 'two retvals approach' would be a real good idea. That should keep compatibility and have the benefit of both codes available.
+1 for this approach

[vitalker]

Originally Posted by Meo-Ada Mespotine

The first 'two retvals approach' would be a real good idea. That should keep compatibility and have the benefit of both codes available.
+1 for this approach

Nice idea, yeah.

[Edison]

Originally Posted by Justin

oops yep a big range of them are wrong, fixing!

Acute accent

Something happens in some words for the Brazilian Portuguese, as it creates a small space to the left of the accented vowel shown in the red arrows. Arrows in yellow accents are OK! See pic.

PS: Syllabic separation for Brazilian Portuguese. I don't know if it would help...


Examples:

Es-ta-tís-ti-cas
Ru-í-do


Words: English

Statistics
Noise


Best regards,