Support CEF 3.3202.1686.gd665578 / Chromium 62 #2200
Comments
|
Hiya, any ideas on when this is likely to be ready? I've found a bug with the version of cef that v57 is using and am looking forward to that being resolved. |
ASAP. 😄 Hopefully rather soon, but because of the nature of the project it's difficult to make promises at this stage. We'll let you and everyone else eagerly waiting know as soon as it's ready, for sure. |
|
CEF 62 probably fixes #1916 too. |
|
Thanks @SilverDragon135, added it to the list. |
|
Chrome 63 was recently released. We should probably re-target this to go with CEF 63 instead. If someone wants to do the initial work of bringing it into cef-binary, by all means, be my guest! |
|
thus.. no cefsharp for 62? |
Not yet, no. We could make a release but it still has the #1203 bug present which will prevent some people from using it. Please vote on this comment: 👎 if you want to have a 62 release asap, or 👍 if you'd rather wait some extra time but get a 63 release instead. Nothing is free in this world, everything takes a bit of time and it would be nice to hear your feedback before spending time on this. I will base my decision on your feedback. I personally would prefer the 63 approach (since it seems silly to release a version which is already outdated at the time of release), but I'm very open to your opinions also. |
|
Current Cefsharp release version (57) have stability issue (probably #2170) so we got many stabily issues on our latest product version. It may leads to security issue for ours systems, so I realy hope you can release new stable version soon. |
|
question is: is version 63 really uptodate when the new CEFVersion gets released ;-) |
Stability issues is one thing, I know about #2170, but security issues? Are we not mixing things up a bit here? I understand if people have problems with 57, that's fine. I think bumping the release to use 63 doesn't really have to be that much work to be honest though, so I am leaning towards that route since everyone will thank me for doing so in the long run... 😉 |
As I said, stabilty issue leads to some security issues in my product. Uncatchable exceptions lead to a app crash. And my app need to be running continuously (to ensure kiosk mode). So, I probably have to reingeneer my app architecture, but this is not a mixing... I agree, in the long run, 63 is better. You know the time cost of this choice. But you ask, so I give you my anser. ;) |
That's fine. 😄 I understand your concerns, thanks for using CefSharp. I noticed that CEF 63 isn't yet available at http://opensource.spotify.com/cefbuilds/index.html, so let's go with a CefSharp 62 (pre)release instead. Will try to get it done within the next few days. |
Maybe could you give me a trick about a way to catch this uncatchable exception ? |
|
@Cansii Let's move that discussion to Gitter or a separate issue. If it's exceptions being raised on the C++ side, I think the unfortunate answer is that they cannot be caught. But, as stated, let's move that part of the discussion elsewhere. |
|
@Cansii and others - a first 62.0.0 prerelease is now available: https://github.com/cefsharp/CefSharp/releases/tag/v62.0.0-pre01 Please test it and give us your feedback here in this issue. |
|
It seems to work really well for me. Solves two of the issues I had with v57 |
|
It's fine on our side, too! Thanks a lot! |
|
Any plans for another build with proprietary codecs? I have "62.0.0-CI2368" installed, if I try to update CefSharp.Common and CefSharp.Wpf to "62.0.0-pre01" I get this message: Unable to resolve dependencies. 'cef.redist.x86 3.3202.1690-proprietary-codecs' is not compatible with 'CefSharp.Common 62.0.0-pre01 constraint: cef.redist.x86 (= 3.3202.1686)'. I can roll with "62.0.0-CI2368" as it's working great, I was just hoping to test "62.0.0-pre01" for you as well while keeping H.264 support. Greatly appreciated as always! |
|
@vwmberry95 there are potential licensing issues with proprietary codecs so it will likely not be in the official tree. You can compile from scratch yourself these features in although it is tedious. As for the errors if you run into it on the current nuget packages it may be due to installing the former 62 test which had those dependencies and left elements in your project file. Remove cefsharp from your project and save the project file. Open the project file in a text editor and remove any cef references (you may find older versions as well). Then installing 62.0.0-pre01 should work. |
|
@mitchcapper Thanks, mitch. Yeah, I should have targeted that comment towards @perlun as he did the build with codecs under the myget tree due to licensing as you said. Being so tedious, having it done once in a fashion that correlates with the CefSharp release would be greatly appreciated by many, I'm sure. That being said, I did try building from scratch in the past but had issues, perhaps because CefSharp was still targeting v57 at the time. Even building cef at the same version had issues. |
Yeah, like @mitchcapper mentioned I think it will be hard to include that code in the official tree (unfortunately!). What we could perhaps do is to make the official packages be runnable towards either version of the (In other words: I don't have a clear plan for this at the moment. Right now #1203 kind of blocks us from doing a stable CefSharp 62 release anyway, so you can choose between "two different prereleases" at the moment - one "official" and one "unofficial".) |
|
Because we anticipate so many people would upgrade to 62.0.0 and have problems because of it, since people (unfortunately) tend to upgrade without reading release notes carefully enough, etc... So the expectancy is that it will cause more problems for people than it would solve.
The long story:
If all goes well and @amaitland gets enough funding, #1203 will be sorted out in the quite near future: https://gist.github.com/amaitland/91dae9ecfe8c47d0da08e000685eeef2 We can always have a vote: use the 👍 on this comment if you want to see us release a 62.0.0 asap with the bug present (and perhaps a setting flag like Alex suggested to make people be aware of it and "opt in" to "JS bindings with defect"). Use the 👎 if you rather prefer to wait on a 62.0.0 or 63.0.0 until the bug is fixed. We do take our user feedback very seriously, and if we get a reasonable enough amount of feedback on this, we will base the decision on this feedback. Thanks! |
|
63 is available now |
So I made a test with pre-release nuget package :
and catching UnhandledException with UnhandledExceptionEventHandler to restart my app. It seems really better for last 12 hours. But I will wait some times before putting it on most critical install. And I've got an issue on a particular install (it does not happen on mine) on a particular page with a video. The webm video plays well but the rest of the page is'nt displayed (maybe hidden). I've enought time to get more details at this time. I will investigate this as soon as possible. |
|
Closing this issue as |
|
Thanks @amaitland. Just to make things clear on everyone, the JS Bindings issue is still not resolved in the published packages - we're tracking that one in #1203. |
This is an umbrella release to support the Chromium 62 version in CefSharp, which I think a lot of you are waiting for.
Some of the known issues thus far:
Upgrade to 3.2924.1575.g97389a9 / Add support for Visual Studio 2017 cef-binary#46 - publish cef.redist and cef.sdk packages. This work has been started by @peters but is not fully completed yet. (and his PR is using an old CEF version as well.)I think we take the simpler approach in Upgrade to CEF 62 cef-binary#53 for now.cef.sdkdependency in CefSharp, making required changes as needed - merged in Upgraded to CEF 62 #2203.Fixes we get for free because of the upgrade:
Update: Prerelease packages are available on nuget.org, please ensure to check the "Include prereleases" checkbox in the NuGet settings in Visual Studio.
The text was updated successfully, but these errors were encountered: