(to set lower bound).

--adaptive-maxbuffer option obviously should be expected to be a positive
value. backing this up is that it is read into an unsigned int.

confusingly these two options use the same short text label, which is
problematic for interface distinction. regardless though,
--adaptive-livedelay is also read into an unsigned int so seems to be
expected to be a positive value.

(to set lower bound).

current logic is expecting a positive value (uses unsigned var).

(to set lower bound).

(to set lower bound).

(to set lower bound).

(to set lower bounds).

(to set lower bound).

(to set lower bound).

negative values simply resulted in an error.

also, explain in longtext that a value of zero means that a preset will be
used.

(to enforce lower bound).

Benjamin Arnaud authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

fix #26453

Romain Vimont authored 3 years ago and Romain Vimont committed 3 years ago

Even if RenderPicture() returns an error, this is not a reason to
increase the deadline of the next rendering.

Lyndon Brown authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

just as done with qt advanced prefs tree.

Lyndon Brown authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

this is the macos version of the previous qt commit.

Lyndon Brown authored 5 years ago and Hugo Beauzée-Luyssen committed 3 years ago

up until now the entire tree has been simply sorted alphabetically. this is
not at all ideal because this means that the "advanced" category comes
first, when logically it should come last. this is particularly unfortunate
because upon switching to advanced/complete/"all" view, the panel shown is
one titled "advanced settings" with a warning underneath saying "use with
care". this category also happens to be a particularly messy one in terms
of its organisation. this mess and warning gives a bad impression and
potential scares off users.

so, let's allow the subcat and plugin nodes to remain alphabetical, but
let's sort the root category nodes into a preferred order. we can do this
simply by resorting to match the order of entries within the category
lookup table in `vlc_config_cat.h`, which was reorganised into a suitable
order in the previous commit.

note that the solution accounts for the possibility that not all cats will
be in use in the tree.

Lyndon Brown authored 5 years ago and Hugo Beauzée-Luyssen committed 3 years ago

this is prep work for the next commit. the intention here is to have the
order of items in the category table reflect the preferred order for the
primary category nodes to be displayed within the advanced preferences
trees.

the current alphabetical ordering used in qt i feel is far from ideal, as
explain in the next commit.

the order i have chosen is based upon my perception of what categories
might be of the most interest to the most users:

 1. Playlist
 2. Interface
 3. Audio
 4. Video
 5. Input / Codecs
 6. Sout
 7. Advanced

i also re-ordered the subcat table, since it only involved moving one
block. this was purely for maintaining consistency.

Lyndon Brown authored 5 years ago and Hugo Beauzée-Luyssen committed 3 years ago

if you switch back and forth between simple and advanced views, every time
you go back to the advanced view it resets the tree node selection back to
the first entry.

this is sooooooooooo annoying. it has bugged me for such a long time.

Lyndon Brown authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

and a couple of syntax issues.

the hidden subcat is used here rather than `SUBCAT_VIDEO_VFILTER` because
according to discussion in !1181 this option is just used as a hack, not
meant for use as a normal user config option.

lack of a subcat entry when there are options is normally because the
author forgot it, which causes a bug of the options not being visible in
GUI preferences. we should at least have a comment here to highlight that
it is deliberate in this special case.

however, a test for option set validity could be added in future, which
would only end up failing on lack of subcat entry here, so to avoid that
let's add one. using the hidden subcat will "properly" keep the option
hidden from GUI prefs (and also potentially help output if the help output
logic ends up gaining the ability to also skip such hidden options).

This is a partial revert of 785c6dfe. Some servers (notably PLEX, UMS,
and some TV sets) amend the exposed media type with additional subtypes
for further differentiation not relevant to VLC, so it is actually
correct to compare the beginning of the string only.

For instance "object.container" can turn into
"object.container.storageFolder" or "object.container.person.musicArtist"
so the plain strcmp will fail to recognize the container nature.

Fixes vlc-ios#1239

Lyndon Brown authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

(to fix builds on Debian stable, as requested by @tguillem in !1141)

Lyndon Brown authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

(matching the order of attributes in the struct)

the object is allocated with `calloc()` via `vlc_custom_create()` but
despite this explicit allocation is wanted.

Lyndon Brown authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

This reverts commit 830ea36e.

as requested in follow-up discussion in !1104.

(actually this is now a partial revert - the change to the return line has
been left in place per review request in !1193).

Lyndon Brown authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

This reverts commit b10f5a02.

as requested in follow-up discussion in !1104.

Lyndon Brown authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

> WARNING : test/Makefile.am: 325:  noinst_PROGRAMS was already defined in condition TRUE, which includes condition HAVE_DYNAMIC_PLUGINS ...
> test/Makefile.am:15: ... 'noinst_PROGRAMS' previously defined here

Thomas Guillem authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

Better than the value guessed by GetVaProfile().

Rémi Denis-Courmont authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

By the time we got to this piece of code, this had already been taken
care of by the decoder device.

Rémi Denis-Courmont authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

This associates the correct X11 screen and does not depend on GL for no
particular reasons.

Rémi Denis-Courmont authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

This was necessary for sharing and for vlc_hold_x11(), but they were
both removed.

Note that the list is still used to match a display to an instance, so
it cannot be removed quite yet.

Rémi Denis-Courmont authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

Originally, both the decoder and the video output would instantiate
VDPAU devices on their own. The process-global list was necessary to
force paired decoder and output to use the same VDPAU instance: they
were matched by server name and screen number.

Now that all components in a given pipeline obtain the VDPAU instance
from a shared decoder device, this trick is no longer necessary.

Incidentally, VDPAU instances would also harmlessly but nevertheless
unnecessarily be shared between independent pipelines within the same
process using the same X11 screen. We can ignore this.

Rémi Denis-Courmont authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

Frames (video surfaces) are always held by one of more fields. Fields
already hold the video context, which holds the decoder device, which
holds the instance.

So frames can rely on any one of their referencing fields to hold the
video context.