Linux Version

Showing 76-221 of 221 messages
Linux Version Stramel 6/29/12 11:10 AM
I wasn't sure how to contact you as it says to in your README. But I would like to help out with creating a Linux version. The problem is that I am new to the Brackets-app/shell project and CEF1/3. I am studying computer science and have a fair knowledge and understanding but may need a little direction to get started. ie. outline of how building it for win/mac would normally go. 
Re: Linux Version Jason San Jose 6/29/12 11:29 AM
Thanks, Stramel. I was about to post something similar to get a discussion started. There's been a few requests spread across twitter, our github issues, Trello and IRC. Here's the user story for reference https://trello.com/c/GcNAJPye as well as the repo https://github.com/adobe/brackets-shell.

To be clear, no one on the core team is working on this yet. And as far as I know, there isn't any active work from external contributors...yet. There are a few interested contributors @sub889 on IRC has been pretty vocal about it.

The first step is to define some tasks for the user story. Here's the original user story for mac and win https://trello.com/c/8Vuom2dA.

I'm pretty sure one of the first tasks for getting brackets-shell running on linux is to build a binary distribution of CEF3 on linux. The CEF project has posted binary builds for mac and windows only. Find CEF here http://code.google.com/p/chromiumembedded/ and their support forum here http://www.magpcss.org/ceforum/index.php.

Figuring out the tasks and getting CEF3 to build is a good starting point. For reference, look at the brackets-shell README for build instructions on mac and win.


On Friday, June 29, 2012 11:10:33 AM UTC-7, Stramel wrote:
I wasn't sure how to contact you as it says to in your README. But I would like to help out with creating a Linux version. The problem is that I am new to the Brackets-app/shell project and CEF1/3. I am studying computer science and have a fair knowledge and understanding but may need a little direction to get started. ie. outline of how building it for win/mac would normally go. 
Re: Linux Version Stramel 6/29/12 11:43 AM
Thanks for the great foundation to start off of. I have checked out the brackets-shell, brackets-app, and brackets repos on github as well as the two posts that you had mentioned on Trello. I guess this weekend I will have to check the IRC since I don't have access from work. 

I had guessed we would probably need to compile a Linux version and I started working towards that last night. So I guess I will look more into the user story stuff and continue on to creating a binary distribution of CEF3 on Linux. Once I complete that I guess I will come back here and see if I can get some further direction for implementation. Thanks Jason!

If anyone is interested in getting this started or helping out just let me know, I have a pretty good start on compiling the CEF3 but if you would like to give it a shot go right ahead.
(unknown) 7/1/12 5:42 AM <This message has been deleted.>
Re: Linux Version sashless 7/1/12 5:45 AM
Hey,

i am interested in supporting Linux too. How is your progess and what are you working on, stramel ?

kind regards
Re: Linux Version Stramel 7/1/12 7:42 AM
Glad you are interested too! I have actually been tied up working on another project currently for work and haven't gotten to it this weekend like I had hoped. So far I am in the process of compiling the CEF3 on Linux and we still need to finish that and compile a list tasks for Trello. 
Re: [brackets-dev] Re: Linux Version Morteza Milani 7/1/12 7:48 AM
Do you really think CEF3 is ready for use? Last time I checked, there
were serious issues open on issue tracker.

Anyway, compiling CEF on linux is not a big task. I already have CEF1
binaries here if they help you to get start:

https://github.com/milani/appjs/downloads

If you manage to build it using ninja, you get better results.

$ cd chromium/src/cef/
$ GYP_GENERATORS=ninja ./cef_create_project.sh

Regards,
Re: [brackets-dev] Re: Linux Version Stramel 7/1/12 7:56 AM
I feel like the CEF3 is ready for use. They have released recompiled binaries for CEF3 for mac and windows, which is being worked on under brackets-shell on github. The CEF people have also stated that there are still some bugs in CEF3 but that its fairly stable now.

I know its not a big task, I just haven't gotten around to doing it between my two jobs and college. 

I will take a look at your binaries that you posted and perhaps working off that for an initial version would be a good idea?
Re: [brackets-dev] Re: Linux Version Narciso (nj) Jaramillo 7/8/12 2:41 PM
Hi folks--the Brackets team is on vacation this week, but I wanted to point at a few resources that discuss our shell migration plans, though none of them directly address Linux support yet. They're listed as links from this issue comment--please take a look:


nj
Re: Linux Version Adam Lehman 7/19/12 4:45 PM
It looks like we got our first attempt at a Linux port from Dmitry Petrov!

https://github.com/can3p/brackets-app 

We don't have much expertise in Linux at Adobe -- big surprise, right? So, if there is anyone willing to compile, package and test this, it would be awesome!

-Adam
Re: Linux Version Gerardo Cobo 7/21/12 5:35 PM
Hi, I managed to get it compiled, and got it to launch. But that's about it. It won't do anything else.
Here's the console output, I don't know if you could point me in the direction of the problem.

** Message: console message: file:///home/cobo/Programming/brackets/brackets/src/thirdparty/less-1.3.0.min.js @8: less: parsed file:///home/cobo/Programming/brackets/brackets/src/styles/brackets.less successfully.

** Message: console message: file:///home/cobo/Programming/brackets/brackets/src/thirdparty/less-1.3.0.min.js @8: less: css for file:///home/cobo/Programming/brackets/brackets/src/styles/brackets.less generated in 3206ms

** Message: console message: file:///home/cobo/Programming/brackets/brackets/src/thirdparty/less-1.3.0.min.js @8: less: css generated in 3207ms

** Message: console message: file:///home/cobo/Programming/brackets/brackets/src/thirdparty/require.js @26: Error: Load timeout for modules: thirdparty/path-utils/path-utils.min thirdparty/smart-auto-complete/jquery.smart_autocomplete thirdparty/jslint/jslint thirdparty/CodeMirror2/mode/xml/xml thirdparty/CodeMirror2/mode/javascript/javascript thirdparty/CodeMirror2/mode/css/css thirdparty/CodeMirror2/mode/less/less thirdparty/CodeMirror2/mode/htmlmixed/htmlmixed thirdparty/CodeMirror2/mode/clike/clike thirdparty/CodeMirror2/mode/php/php thirdparty/CodeMirror2/mode/coffeescript/coffeescript thirdparty/CodeMirror2/mode/clojure/clojure thirdparty/CodeMirror2/mode/perl/perl thirdparty/CodeMirror2/mode/ruby/ruby thirdparty/CodeMirror2/mode/mysql/mysql thirdparty/CodeMirror2/mode/diff/diff thirdparty/CodeMirror2/mode/markdown/markdown 


Thanks
Re: Linux Version Gerardo Cobo 7/21/12 5:43 PM
Oh, wow, i'm so dumb. Turns out I was missing the third party libraries.
(unknown) 7/22/12 1:05 PM <This message has been deleted.>
Re: Linux Version Dmitry Petrov 7/22/12 1:07 PM
Hi,

I've joined the group and can answer to any questions concerning the code for linux wrapper.
In order to build succesfully please check the readme and in order to run the app make sure
that full brackets code lies at brackets-app/brackets

- can3p

пятница, 20 июля 2012 г., 3:45:39 UTC+4 пользователь Adam Lehman написал:
Re: Linux Version wolfv 7/23/12 4:22 AM
Awesome. I was about to attempt the same port via WebkitGTK / Python.

So far it works for me!

Regards,

-wolfv
Re: Linux Version Peter Flynn 7/27/12 12:29 AM
Hi can3p, thanks for joining the group!  It's great to see so much progress on the Linux front already.

I see comments from Geraldo and wolfv... has anyone else here tried out can3p's port?  Any feedback?  Anyone interested in trying it but having trouble with the build?

From a technical standpoint, I think one concern the core team has is that this implementation is so different from the Windows & Mac app shells -- using WebKitGTK with Vala, as opposed to CEF with C++.  It will have a higher cost to maintain, going forward, when there's less shared between them.  Ideally we'd like to see a Linux version based on some build of CEF.  Of course, there are really two pieces to the native shell: the core browser engine, and the implementation of Brackets's custom APIs.  The Linux implementation of those APIs seems like an extremely valuable piece of can3p's work.  Hopefully that code could be used even if the browser engine piece is changed... what do you think?

On the CEF front, Morteza mentioned that CEF 1 works on Linux pretty well already.  Brackets is still in the process of transitioning from CEF 1 to CEF 3, so it seems reasonable to me to start out by experimenting with a CEF 1 build for now (if that's easiest).  Morteza and can3p, would you be interested in collaborating on a CEF-1-based Linux build?  It would be awesome to see Morteza's CEF build fused with can3p's implementation of the custom APIs.

Anyway, again -- I think this is awesome progress!  Please keep posting here, and we on the core team will try to support you as best we can!

- Peter
Re: [brackets-dev] Re: Linux Version Morteza Milani 7/27/12 1:25 AM
Hi,

> On the CEF front, Morteza mentioned that CEF 1 works on Linux pretty well
> already.  Brackets is still in the process of transitioning from CEF 1 to
> CEF 3, so it seems reasonable to me to start out by experimenting with a CEF
> 1 build for now (if that's easiest).  Morteza and can3p, would you be
> interested in collaborating on a CEF-1-based Linux build?  It would be
> awesome to see Morteza's CEF build fused with can3p's implementation of the
> custom APIs.

What CEF1 revision do you use in Brackets? I have CEF builds here[1]
for Mac/Win/Linux with some bug fixes. I will fork brackets-app and
give it a try. Just a side note, brackets-app folder structure seems
odd to me;) It has a src folder which includes Release and Debug
folders with binary files in it. and another bin folder. Do you mind
if I change folder structure a little?

[1] https://github.com/appjs/appjs/downloads
(unknown) 7/27/12 6:48 AM <This message has been deleted.>
Re: [brackets-dev] Re: Linux Version Randall Edmunds 7/27/12 9:42 AM
Morteza,


> brackets-app folder structure seems
> odd to me;) It has a src folder which includes Release and Debug
> folders with binary files in it. and another bin folder. Do you mind
> if I change folder structure a little?

All feedback is welcome. How would you like to change it?

We have already started a CEF3 repo in github.com/adobe/brackets-shell, and the directory structure is already different, so please take a look. Notice that there is no longer a bin folder. We plan to stop adding executable files to the repo (to keep it smaller) and instead deliver them using an installer.

Thanks for the feedback,
Randy

Re: [brackets-dev] Re: Linux Version Morteza Milani 7/27/12 10:32 AM
Randall,

The new repository is much cleaner than brackets-app. So I prefer to
work on it. Even though it can be managed better. You don't need to
add visual studio or xcode project files to the repo, gyp generates
them automatically.

if you agree, I'll build CEF3 on linux and send you a pull request
with changes to brackets-shell.
Re: [brackets-dev] Re: Linux Version Randall Edmunds 7/27/12 10:43 AM
> You don't need to add visual studio or xcode project files to the repo, gyp generates them automatically.

Good point. Please create an issue.


> if you agree, I'll build CEF3 on linux and send you a pull request with changes to brackets-shell

Yes, brackets-shell is the best starting point. That would be awesome.

Randy

Re: Linux Version Rado Rodopski 9/20/12 1:10 AM
Hi, Randy,

I am very fond of the idea of an editor that can edit itself (it is as natural as a C++ editor editing itself).
And what is more natural than an open-source software running on an open-source OS, right?

Since I see that CEF3 supports Linux (not sure exactly what this means, probably that it compiles fine and runs on Linux?) and I would like to see Brackets running on Linux, I thought I might contribute to the project and bootstrap a native Linux version. But as I don't have much time to look at the code and see what the architecture of Brackets is, could you point me to a resource where the architecture is explained (if such a resource exists) and maybe summarize what changes would need to be done (I'm aware that FS API will have to be reimplemented) ?

I suppose I could copy the Windows part of the code and tweak it to run in Linux?

If you manage to respond within a couple of days, maybe I could start working on it this weekend (that's the only free time I have right now).

Thanks,
Rado Rodopski
Re: Linux Version Randall Edmunds 9/20/12 10:16 AM

Rado,

Thanks for your interest in Brackets.

> Since I see that CEF3 supports Linux (not sure exactly what this
> means, probably that it compiles fine and runs on Linux?)

Yes, CEF3 runs on Linux, but the binaries are not yet delivered, so, for now, you’ll need to build those yourself. See the brackets-shell wiki for more info:

https://github.com/adobe/brackets-shell/wiki/Building-brackets-shell

 

> I suppose I could copy the Windows part of the code and tweak it to run in Linux?

That’s a reasonable approach. All of the OS-specific code for Windows is in files that end with “_win”, so start by copying each of those. I think that appending them with “_linux” is probably too generic, since there are many flavors, so you should use something more specific than that. Here is the usage:

 

- client_handler_win.cpp – this has all of the startup and shutdown code for each instance

- appshell_extensions_win.cpp – this is where the FS API is implemented

- client_app_win.cpp, resource_util_win.cpp, cef_types_win.h – various utilities

 

Let us know if you have more questions.

 

Randy

Re: Linux Version Rado Rodopski 9/22/12 10:58 AM
Hi, Randy,

Thanks for the info.

I read the CEF wiki pages and it seems that they have managed to successfully build CEF3 on Ubuntu, which is good news for me. But I read the build instructions for CEF3 and in the prerequisites it says that I need a machine with at least 4GB of RAM and 8GB swap area and also over 10GB to checkout and build the source tree. Since I don't have a machine that meets those requirements, I can't build CEF3.
Could someone from the Brackets team (who has access to a good Linux box) possibly build CEF3 and upload the binary someplace?

Thanks,
Rado
Re: Linux Version Andrei Zisu 9/23/12 4:29 AM
I would be happy to help and test out once that CEF3 ubuntu build is done.
Re: Linux Version Andrei Zisu 9/23/12 11:09 AM
Hello, 

Please read the following comment on a issue I posted today on CEF tracker: http://code.google.com/p/chromiumembedded/issues/detail?id=726

It seems a release build is definetely less resource intensive.

On Saturday, September 22, 2012 8:58:06 PM UTC+3, Rado Rodopski wrote:
Hi, Randy,

Thanks for the info.

I read the CEF wiki pages and it seems that they have managed to successfully build CEF3 on Ubuntu, which is good news for me. But I read the build instructions for CEF3 and in the prerequisites it says that I need a machine with at least 4GB of RAM and 8GB swap area and also over 10GB to checkout and build the source tree. Since I don't have a machine that meets those requirements, I can't build CEF3.
Could someone from the Brackets team (who has access to a good Linux box) possibly build CEF3 and upload the binary someplace?

Thanks,
Rado

20 септември 2012, четвъртък, 20:16:45 UTC+3, Randall Edmunds написа:

Rado,

Thanks for your interest in Brackets.

> Since I see that CEF3 supports Linux (not sure exactly what this
> means, probably that it compiles fine and runs on Linux?)

Yes, CEF3 runs on Linux, but the binaries are not yet delivered, so, for now, you’ll need to build those yourself. See the brackets-shell wiki for more info:

https://github.com/adobe/brackets-shell/wiki/Building-brackets-shell

 

> I suppose I could copy the Windows part of the code and tweak it to run in Linux?

That’s a reasonable approach. All of the OS-specific code for Windows is in files that end with “_win”, so start by copying each of those. I think that appending them with “_linux” is probably too generic, since there are many flavors, so you should use something

...
Re: Linux Version Rado Rodopski 9/24/12 4:02 AM
Andrei,

Thanks for the tip.
Did you manage to compile it? If so, could you share the binaries?

Thanks.


On Sunday, September 23, 2012 9:09:31 PM UTC+3, Andrei Zisu wrote:
Hello, 

Please read the following comment on a issue I posted today on CEF tracker: http://code.google.com/p/chromiumembedded/issues/detail?id=726

It seems a release build is definetely less resource intensive.
Re: Linux Version Andrei Zisu 9/24/12 5:52 AM
I am going to try later today. Will let you know as soon as I get anything consistent....
(unknown) 10/2/12 6:14 AM <This message has been deleted.>
Re: Linux Version Rado Rodopski 10/3/12 5:18 AM
Hi, Andrei,

It's been a week since your last post. Did you make any progress?

-- Rado


24 септември 2012, понеделник, 15:52:53 UTC+3, Andrei Zisu написа:
I am going to try later today. Will let you know as soon as I get anything consistent.

On Monday, September 24, 2012 2:02:39 PM UTC+3, Rado Rodopski wrote:
Andrei,

Thanks for the tip.
Did you manage to compile it? If so, could you share the binaries?

Thanks.

On Sunday, September 23, 2012 9:09:31 PM UTC+3, Andrei Zisu wrote:
Hello, 

Please read the following comment on a issue I posted today on CEF tracker: 
...
Re: Linux Version Ivan Fraixedes 10/4/12 5:53 AM
Hi Rado, 

Although this is not a good solution maybe, you may use Brackets in Wine like a workaround.
...
Re: Linux Version Jerzy 10/5/12 4:27 PM
Maybe somebody will bundle Brackets with Adobe AIR? Its API has a file system API and in works on Windows/Linux/Mac.
Re: Linux Version Aezop Shyft 10/6/12 1:21 PM
Yeah I looked at everything and thought it was great, but then I found out they haven't even started on a Linux version. Seeing as the team hasn't even started work on the programming OS, I don't think it'd be a wise use of time or resources to contribute to this. The team seems misguided. 
Re: [brackets-dev] Re: Linux Version Cheng Zhao 10/6/12 6:02 PM
I once migrated brackets to node-webkit, and it runs well on Linux, see
https://github.com/zcbenz/brackets

you can download node-webkit at
https://github.com/rogerwang/node-webkit and run it by:
./nw /path-to-brackets-directory

Some functions that make use of brackets-shell APIs did not work but
it's easy to change them to node-webkit's API, see what I have done:
https://github.com/zcbenz/brackets/commit/c76c208fb1e90e04e94653c77331217e1a0eac43

You still need some work to make all functions work but it should be
much easier than port brackets-shell to Linux.

On Sun, Oct 7, 2012 at 4:21 AM, Brackets-Dev on behalf of Aezop Shyft
<brackets-dev+noreply-APn2wQdRyuMZ15SCEiZpNRNz4xTFrHdDsNNbRLxJZ0F@googlegroups.com>
wrote:
> Yeah I looked at everything and thought it was great, but then I found out
> they haven't even started on a Linux version. Seeing as the team hasn't even
> started work on the programming OS, I don't think it'd be a wise use of time
> or resourc...
Re: Linux Version Marc Boisvert-Dupras 10/7/12 8:48 PM
Very interested in this as well. Progress update ?
Re: Linux Version Gabriel C 10/8/12 2:42 PM

I am interested in supporting Linux too. How is your progess?
Re: Linux Version Konstantin Zolotarev 10/9/12 3:10 AM
Today i compiled and created package with CEF1 binaries on Ubuntu 12.04 x64. From this repo : http://code.google.com/p/chromiumembedded/source/browse/#svn%2Fbranches%2F1271%2Fcef1
CEF3 sources contains broken includes and couldn't be compiled (I'll try to fix it tomorrow). 

If somebody need them I'll upload binaries in few hours. But I really have no enough time to finish creating of brackets packages for Linux.


On Tuesday, 9 October 2012 03:42:18 UTC+6, Gabriel C wrote:

I am interested in supporting Linux too. How is your progess?
Re: Linux Version Rado Rodopski 10/9/12 6:26 AM
Hi,

I'm interested in CEF3 binaries if it's not much of a trouble.
I'm also lacking some free time, but maybe some weekend I could take a look.

Thanks


09 октомври 2012, вторник, 13:10:32 UTC+3, Konstantin Zolotarev написа:
Today i compiled and created package with CEF1 binaries on Ubuntu 12.04 x64. From this repo : http://code.google.com/p/chromiumembedded/source/browse/#svn%2Fbranches%2F1271%2Fcef1
CEF3 sources contains broken includes and couldn't be compiled (I'll try to fix it tomorrow). 

If somebody need them I'll upload binaries in few hours. But I really have no enough time to finish creating of brackets packages for Linux.

On Tuesday, 9 October 2012 03:42:18 UTC+6, Gabriel C wrote:
...
Re: Linux Version Konstantin Zolotarev 10/15/12 3:11 AM
I'll try to compile them this week...
(unknown) 10/22/12 4:45 AM <This message has been deleted.>
Re: Linux Version Chris Franklin 10/22/12 7:59 PM
Hey,

Did you have any luck? If your short on time and you point me in the right direction I may be able to help.

Cheers,

Chris...
Re: Linux Version Pritam Baral 10/26/12 5:22 AM
Hey everyone, I have built the binary distribution of CEF3 (1271 branch, 1180 is on the way,) with a few caveats:
  • dynamically-linked to libudev.so.0 (my Arch Linux setup only has libudev.so.1. A symlink from .so.0 to .so.1 works fine, for now)
  • cef_unittests haven't been built (linker complained about a missing symbol, didn't have time to hunt that down myself)
  • some compiler flags (-Werror=c++0x-compat, -Werror=uninitialised, -Werror=deprecated-declarations) were disabled, but only in some targets respectively, not all.

I have also gone through the Building Brackets-shell wiki, but the crux of the compilation stuff seems to be done by XCode/Visual Studio, using the project files generated from .gyp files. The next obvious step for me seems to be to get/generate a Makefile and compile this stuff.

This is where I need a little guidance.

On Tuesday, October 23, 2012 8:29:45 AM UTC+5:30, Chris Franklin wrote:
Hey,

Did you have any luck? If your short on time and you point me in the right direction I may be able to help.

Cheers,

Chris

On Monday, October 15, 2012 11:11:48 AM UTC+1, Konstantin Zolotarev wrote:
I'll try to compile them this week

On Tuesday, 9 October 2012 19:26:28 UTC+6, Rado Rodopski wrote:
Hi,

I'm interested in CEF3 binaries if it's not much of a trouble.
I'm also lacking some free time, but maybe some weekend I could take a look.

Thanks

09 октомври 2012, вторник, 13:10:32 UTC+3, Konstantin Zolotarev написа:
Today i compiled and created package with CEF1 binaries on Ubuntu 12.04 x64. From this repo : http://code.google.com/p/chromiumembedded/source/browse/#svn%2Fbranches%2F1271%2Fcef1
CEF3 sources contains broken includes a
...
Re: [brackets-dev] Re: Linux Version Chris Franklin 10/26/12 12:54 PM
Hey,

Good work on all of this, thanks for your hard work. I am in the middle of launching something at work so I am pretty busy this week. 

Next week on the other hand I shall be chilling in Cornwall with a laptop and a good amount of free time. I will have a look at the build scripts then. 

I shall also be looking at Django and some sort of interactive Python shell integration as I see a lot of potential in that area but that is for another post.

Cheers,

Chris
...
Re: [brackets-dev] Re: Linux Version Chris Franklin 10/26/12 1:09 PM
Hey,

Had a quick look at the code to see what you were talking about, I was having a little google around to see if any of what we are trying to do could be automated and found this: http://wiki.gnustep.org/index.php/Pbxbuild do you think an approach like this could save us time or should we go the manual route?

Cheers,

Chris...
Re: [brackets-dev] Re: Linux Version Pritam Baral 10/26/12 2:02 PM
Hey,

I managed to generate the Makefile by installing and running gyp on the appropriate directory and .gyp file.

I also managed compile it all in, after manually adding a few include_dirs to the OS==linux section of the .gyp file. The linker, on the other hand, presented with trouble as I haven't yet ported any of the platform-specific code to linux/gtk. I'm currently working on it, but it would be encouraging if there was a way to skip the platform-specific code and just compile/link the brackets-shell and cefclient, libcef_dll_wrapper together.
 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 10/29/12 2:16 AM
Hi,

That's a very good progress you've made!

Can you point me to the platform-specific code? I will take a look and there's a big chance I can help you with porting them for Linux.
Also, any pointers on how to compile brackets-shell are welcome since it will save me a couple of hours trying on my own (and I have a very tight schedule).

Thanks,
Rado...
Re: [brackets-dev] Re: Linux Version Pritam Baral 10/29/12 3:52 AM
Latest update: https://twitter.com/_PritamBaral_/status/262666953741697024

Still needs a few basic native functions to be ported, and the entire LiveBrowserManagerLin class.

We just need to add a few include_dirs and some system libraries at link time. I have added the relevant include_dirs in the appshell.gyp, (probably not in the best way possible, as they are local to my system,) but I didn't have the time to look into GYP's syntax in order to find out how to add linker libraries.

Run make BUILDTYPE=Default and use the out/Default/ directory, I am yet to find out how to set the BUILDTYPE from GYP. This step WILL FAIL at the link-time. See below

For the linking step, I manually modified the command, and have included it at the bottom of the appshell.gyp file as a comment.


PS: Some compile-time/link-time issues may crop up, as my set-up is the furthest thing from portable at the moment. Do post them here, so we can modify it and make it more portable for the Linux community at large.
 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Pritam Baral 10/31/12 4:08 PM
The 32-bit binary CEF3 distribution (from a vanilla Ubuntu 10.04 lucid)  has been posted on the downloads link above.
Only one caveat this time: libudev.so.0

 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 10/31/12 5:12 PM
Hi,

It seems that CEF is linked to libudev.so.0 dynamically. I did a little digging and could think of 3 possible solutions
1) As you pointed - creating a soft symlink to libudev.so.1 is the easiest solution
2) Cahnging the .gyp file to make it link to libudev.so.1 insteda of so.0 (I'm not sure how the linking is done)
3) In Release/lib.target/ : sed -i 's/libudev.so.0/libudev.so.1/' libcef.so -- this should trick the library to look for so.1

I hope this helps.

I noticed that you managed to make the Live browsing work, do you need some help? I might have some free time this weekend so I will try to build brackets on Linux on my own. let me know if I can help with something.

Thanks,
Rado
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 10/31/12 6:13 PM
I'm not quite sure if forcing a particular version of libudev is such a good thing. It would be better if we could get it to link to just libudev.so. That may or may not be permitted by ld, needs some research..

Another avenue we could research is see what the Chromium devs do about this, as this is clearly a Chromium specific library requirement.

Although, the best solution would be to remove the requirement of libudev completely, as brackets doesn't seem to be needing any udev services. Might be difficult, and there's the possibility of breaking other projects that use our CEF build.

 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Pritam Baral 10/31/12 6:26 PM
As far as I know, the list of required libraries is generated at gyp time, and is dependent on the host system. But some flag/variable in the many Chromium gypi files might be explicitly setting libudev.

The major remaining part of the porting consists of the LiveBrowserMgrLin{} class. I'm not quite sure if it's functionality is entirely needed, and I have some thoughts about the similar logic used on other platforms. Some input from the original devs of the Windows and Mac versions would be helpful.

Tiny things to port: GTK extras, e.g. Window management, Icon and window title.

I have managed to modify the appshell.gyp file to be friendly enough for building straightaway. There are still some shims in there for architecture detection that aren't the best. Will be putting up on my linux branch on github after a little clean-up.


 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Narciso (nj) Jaramillo 10/31/12 6:45 PM
Wow, great progress on this--thanks for pushing forward!

I don't remember exactly who worked on the appshell stuff for live development, but I'll ask around on the team and have someone chime in.

Thanks,

nj
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 10/31/12 6:52 PM
Thanks, that would be great.
Also, this might be of some reference: https://github.com/adobe/brackets/issues/1317#issuecomment-9968553

 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Pritam Baral 10/31/12 8:26 PM
For anyone looking to build brackets-shell on linux from my branch:

The appshell.gyp code has been updated. With a nifty new resource compilation feature. More on that

It also supports compiling 32-bit binaries on a 64-bit system. Just run a

GYP_DEFINES="target_arch=ia32" gyp --depth .

to switch to 32-bit buiding.

(Use target_arch=x64, or just skip the GYP_DEFINES flag to switch back)

 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Glenn Ruehle 11/1/12 1:37 PM
I added details to this bug:

In summary, there are two ways that Brackets will close Chrome tabs/windows:
  1. When stopping a Live Preview session we use the remote debugging connection to call "window.close()" on the tab. This will only close the tab that was showing the Live Preview content, but will also close Chrome if that was the only tab that was showing.
  2. If Chrome is launched but we can't start a remote debugging session, the user is prompted to restart Chrome with remote debugging enabled. In this case Chrome is terminated and relaunched. This is only done on the Mac.
On Linux, do you know if you need to restart chrome to enable remote debugging? That will determine how much functionality will need to be ported.

Glenn
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/1/12 2:14 PM
No I don't think enabling remote debugging without a restart is possible, just checked.

That said, chrome may or may *not* restore a user's session upon relaunch, (i.e.: not under our control,) which is further reason to use a separate profile of our own....
Re: [brackets-dev] Re: Linux Version Dennis Kehrig 11/2/12 6:35 AM
On Thursday, November 1, 2012 10:14:55 PM UTC+1, Pritam Baral wrote:
That said, chrome may or may *not* restore a user's session upon relaunch, (i.e.: not under our control,) which is further reason to use a separate profile of our own.

There's "--restore-last-session" (http://peter.sh/experiments/chromium-command-line-switches/), can anybody think of reasons why not to set this flag when restarting Chrome?
Re: [brackets-dev] Re: Linux Version Dennis Kehrig 11/2/12 6:47 AM

There's "--restore-last-session" (http://peter.sh/experiments/chromium-command-line-switches/), can anybody think of reasons why not to set this flag when restarting Chrome?

Nevermind, couldn't get it to work with Brackets on the Mac. But maybe I did something wrong. 
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/2/12 7:22 AM
Yes, this flag does work on Linux Google Chrome. We can use this in the relaunching code.

Note that Brackets itself can't be expected to respond to all Chrome startup flags, as I've seen during my testing..

 

Regards,
Chhatoi Pritam Baral




On Fri, Nov 2, 2012 at 7:17 PM, Dennis Kehrig via Brackets-Dev <brackets-dev+noreply-APn2wQf9TeTZL5-4CT5EXKDHuosO7QDfcaG6de-e_ma@googlegroups.com> wrote:

There's "--restore-last-session" (http://p
...
Re: [brackets-dev] Re: Linux Version Peter Flynn 11/2/12 12:22 PM
We have a bug filed for using "--restore-last-session" on the Mac too: https://github.com/adobe/brackets/issues/1713
...if anyone would like to just fix both platforms in one fell swoop :-)

- Peter
...
Re: [brackets-dev] Linux Version Jonathan Diehl 11/2/12 3:07 PM
That was easy enough to fix. See my pull request :)

Have a great weekend!
- jonathan
...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/2/12 5:42 PM
Hi,

I generated a Makefile with gyp and run
make BUILDTYPE=Default

but I got the following output:
make: *** No rule to make target `out/Default/obj.target/libcef_dll_wrapper/libcef_dll/transfer_util.o', needed by `out/Default/obj.target/libcef_dll_wrapper.a'.  Stop.


What am I doing wrong?

Thanks,
Rado...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/2/12 5:50 PM
Sorry, I should've mentioned it:

I changed the appshell.gyp to include only two configurations (BUILDTYPEs); Release and Debug. Release is default, and a simple make should suffice

The Debug BUILDTYPE mightn't work, as the CEF binary isn't compiled in debug mode, yet.

Also, I'm looking into ways to drop some unneeded/troublesome dependencies, e.g.: udev. It's still a work-in-progress, so expect issues.

 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/2/12 6:01 PM
I got the same error.
How do you generate the Makefile?
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/2/12 6:05 PM
Also, on the downloads page I cannot access the 32-bit binary. It says the file doesn't exists. No problem with the 64-bit binary though.
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/2/12 6:20 PM
The Makefile needs to be regenerated, from appshell.gyp. Use 
gyp --depth .

The 32-bit binary must have become corrupted during my last upload. I have re-uploaded it again.

PS: You may have to git pull my changes to appshell.gyp regarding the libudev.so shim

 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/3/12 8:38 AM
Sorry for bothering you, I'm such an idiot :D
I forgot to run scripts/make_symlinks.sh, that's why I got that error from the Makefile.

I managed to get it compiling, but it failed at link stage with the following error:
 
 LINK(target) out/Release/Brackets
Release/lib.target/libcef.so: undefined reference to `udev_monitor_unref'
Release/lib.target/libcef.so: undefined reference to `
udev_enumerate_scan_devices'
.....
Release/lib.target/libcef.so: undefined reference to `udev_device_get_sysattr_value'

Release/lib.target/libcef.so: undefined reference to `udev_enumerate_add_match_subsystem'
collect2: ld returned 1 exit status
make: *** [out/Release/Brackets] Error 1


I wonder what I missed this time... Any ideas?

Thanks,
Rado...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/3/12 8:50 AM
BTW, I am using the 32-bit binaries.

On another thought, maybe we need libcef.so linked against the udev library. I tried to run "cefclient" inside the Release dir and it showed a similar error on launching:
symbol lookup error: /home/rado/Desktop/brackets/brackets-shell/deps/cef/Release/lib.target/libcef.so: undefined symbol: udev_new

What's your thoughts?

Thanks,
Rado...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/3/12 12:48 PM
Yes, we do need to link libudev.so with the final binary. libcef.so is the Uploaded CEF binary only has the linking the libudev.so removed, it is deferred to gyp time on the target build system.

I did mention the need to git pull to pull my libudev changes in appshell.gyp. I have temporarily added libudev to the link libraries of the final target. Just generate a new Makefile (with the latest copy of appshell.gyp) carry on.

PS: I have managed to fully remove libudev requiring functions from the Chromium source code, effectively allowing us to drop libudev altogether. The code, of course, needs to be cleaned up if it is to be maintainable later on.
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/3/12 1:41 PM
Hi,

Yes, I pulled the changes from you and still no effect,
Btw, the output of "pkg-config --libs-only-l udev" is " ", i.e. empty. However "pkg-config --libs-only-l libudev" returns "-ludev -lrt", so I changed the .gyp file accordingly and still got the same error during linking.

I checked /usr/lib/ and I don't have a file named "libudev.so.0/1" or anything like that. (I'm using Ubuntu 12.04).
I ran "locate libudev.so" and got the following:
/lib/i386-linux-gnu/libudev.so.0
/lib/i386-linux-gnu/libudev.so.0.13.0
(I think I got these after installing libudev-dev)

Maybe this is what the problem consists of, the location of the libudev.so file?

--Rado



On Saturday, November 3, 2012 9:48:08 PM UTC+2, Pritam Baral wrote:
Yes, we do need to link libudev.so with the final binary. libcef.
...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/3/12 4:49 PM
Hi,

I finally managed to build Brackets.
I experimented a little and managed to make it link correctly.

Here's how it worked for me:
in appshell.gyp [lines 228 - 231]

              '<!@(<(pkg-config) --libs-only-l udev)',
             
'<!@(<(pkg-config) --libs-only-l gtk+-2.0 gthread-2.0)',
             
'$(BUILDTYPE)/lib.target/libcef.so',

changed it to:

              '<!@(<(pkg-config) --libs-only-l gtk+-2.0 gthread-2.0)',
             
'$(BUILDTYPE)/lib.target/libcef.so',
             
'<!@(<(pkg-config) --libs-only-l libudev)',

and it linked correctly.

I will play with it a little, and then I will look at how to set the icon.

I noticed that it shows the address bar, is it intentional? Or is it some sort of a side effect?
Maybe I could try to think of a way to hide/remove it?

Regards,
Rado



On Saturday, November 3, 2012 10:41:16 PM UTC+2, Rado Rodopski wrote:
Hi,

Yes, I pulled the changes from you and still no effect,
Btw, the output of "pkg-config --libs-only-l udev" is " ", i.e. empty. However "pkg-config --libs-only-l libudev" returns "-ludev -lrt", so I changed the .gyp file accordingly and still got the same error during linking.

I checked /usr/lib/ and I don't have a file named "libudev.so.0/1" or anything like that. (I'm using Ubuntu 12.04).
I ran "locate libudev.so" and got the following:
/lib/i386-linux-gnu/libudev.so.0
/lib/i386-linux-gnu/libudev.so.0.13.0
(I think I got these after installing libudev-dev)

Maybe this i
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/3/12 6:34 PM
Hey,

Good work on the linking! My original gyp was indeed wrong. The linker only looks for symbols in the libraries supplied after the one needing it, in our case, after libcef.so

The address bar and all is how it was in the original example cefclient application. I found it handy during brackets development as I could quickly refresh it. It needs to be removed ultimately, of course.

Another thing: remote debugging of brackets doesn't work. I'm not sure if it's because of the newer, still in-development, 1271 version of CEF3, or some build flags of CEF3 or what. The older, 1180 version doesn't have libcrypto statically linked in, so it causes a libudev-like library version clash issue. Still working on this one.

 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/4/12 2:19 AM
Hi, Pritam,
 
The address bar and all is how it was in the original example cefclient application. I found it handy during brackets development as I could quickly refresh it.
Brackets has a way of refreshing itself - Debug --> Reload Brackets (F5). But then again, it really is handy to see what address we're looking at.


Another thing: remote debugging of brackets doesn't work. I'm not sure if it's because of the newer, still in-development, 1271 version of CEF3, or some build flags of CEF3 or what. The older, 1180 version doesn't have libcrypto statically linked in, so it causes a libudev-like library version clash issue. Still working on this one.
Yeah, I noticed that I can launch Live preview, but then Brackets somehow misbehaves and doesn't show we're in Live preview, neither does it refresh the page when editing CSS.
I can have a look at the library issue if you want me to, but I would need a 32-bit release build of the 1180 version of CEF. If you can upload that too, maybe I could help somehow.

I was thinking about something - brackets-shell looks in the user folder for the HTML files, is that a good idea? This way we should supply the HTML files for every user on the system, and updating will be a lot harder that if the files were in a central location like where the brackets-shell binary is installed. The Windows and Mac versions are implemented this way, could you tell me what was the idea behind this decision?

Thanks,
Rado...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/4/12 2:55 AM
I only once ever used Live Preview, and it did work then, the indicator went offline after a while though. Since the crux of Live Preview work uses direct communication between Brackets and Chrome, via websockets, I didn't think brackets-shell might have anything to do with.

The LiveBrowserMgrLin{} class is still incomplete though, if that has something to do with.

About the location of Brackets www files, I didn't put much thought into it and just appended "www/index.html" to the end of cache_dir. The cache_dir needed to be in a place the user had write access to, and Unix convention has been to use directories, sometimes called xdg-cache-dirs, like $HOME/.config/<app-name> or $HOME/.<app-name>

We could (should) use a central system location, perhaps the same dir as brackets, or a separate resources dir like libcef.so has lib? It really is a last-minute choice, easily changeable. Feel free to modify.

A 32-bit 1180 build has been uploaded. You will need to add -lcrypto to the list of libraries (pkg-config --libs-only-l openssl) for it to finish. The thing is, version 23 of Chromium, (the one used by CEF 1271) has it's own set of crypto libraries built-and-statically-linked into it. I think 21 doesn't.

 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/4/12 3:15 AM
The issue with LiveDevelopment may be related to this https://github.com/pritambaral/brackets/commit/bea752a6eff50e49379230b448e151027d7a5191

Basically, brackets adds an extra forward-slash to the live preview path on Windows. If brackets thinks we're on Windows, and since Chrome auto-corrects extra forward-slashes, Live Browser is gonna complain.

 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/5/12 7:03 PM
Status Quo:
  • Brackets is up and running on Linux.
  • We have binary distributions of both the stable(1180) and in-development(1271) branches of CEF3
  • 1271 is easier for distribution: cleaner library dependencies
  • Native functions are all implemented, only exception being CloseBrowser() (see below)
  • Some clean up is required before shipping

I have a few queries for the Brackets team, regarding distribution:
  1. I have tried to compile in trivial resources, like appshell_extensions.js and the application icon.
    Is that okay, or is it explicitly required that they be distributed as separate files along with the Brackets binary?

  2. I'm currently using a corrupted codebase of Chromium to build CEF3; corrupted in the sense I've dummified all udev functions. Not really necessary, but it relieves having to worry about something (libudev) which we don't need anyway. No maintenance required until CEF decides to update to the next version, two cycles away.
    Is it okay if we use this build of CEF?

  3. I prefer using the in-development branch, as it's a little easier to build with. It has caused no apparent troubles as of now, but then again, we aren't running the entire gamut of Chromium's features.
    Is it okay if we stick with 1271 for the Linux version?

 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Jason San Jose 11/6/12 9:16 AM
Nice work.
  1. Compiling in those resources is fine. They're compiled in the mac and win binaries as well.
  2. Ideally, we wouldn't want to fork CEF. If we can at least document the changes why the changes exist, it should be fine for now in this stage of development. 
  3. Again, you're early in development, so do what you have to do. Ideally, all 3 platforms would use the same CEF build.
I think there's a lot of good information on this thread. I was wondering if you could summarize the build instructions for us on our Github wiki? That will help more of us on the team and in the community get involved quickly and maybe find any issues. 
...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/7/12 5:30 AM
Hi, Pritam,

I managed to compile against both 1180 and 1271 builds of CEF3. I disn't see any difference while playing with brackets. As you mentioned, we need to install the libssl-dev package and add -lcrypto flag to linker when using build 1180 of CEF3. So from developer perspective, it is a little bit easier if we used CEF3 build 1271.

I also noticed that the Live Preview feature now works fine. Nice work!
Just out of curiosity, what was the issue with Live Preview? Was it JS-related?

One more question. When you talk about the icon and other resources, what do you mean by 'distribution' ? Do you mean as some kind of package (.deb for Debian / .rpm for Red Hat) or just binary distribution?
It would be nice if we could create such packages to ease the initial setup of brackets on a new machine. In fact, I would like to help create at least a .deb package. Any comments on this?

Thanks,
Rado
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/7/12 7:14 AM
Hi Rado,


The problem with Live Development was that Brackets thought it was running on Windows. It sets a javascript flag brackets.platform on start. Previously this was set to either "mac" or "win". And on Windows, the Live Development code used an extra forward-slash in the url when communicating with Chrome.

By distribution, I meant creating any sort of collection that would ease the installation on user's systems. I have two things in mind about this:
  1. A tarball with the following file-system configuration: 
    /usr/bin/brackets -> /opt/Brackets/Brackets
    /opt/Brackets/Brackets
    /opt/Brackets/lib/libcef.so
    /opt/Brackets/locales/
    .
    .
    .

  2. A tarball with no pseudo-filesystem, just a folder containing all of Brackets components.
Both are feasible, and good enough. While (1) is suitable for system-wide installation, (2) is a portable distribution. The advantage with a (2) is that the user need not install it system-wide (no sudo?) and can just try it right away after downloading.

Creating a .deb/.rpm/.pkg.tar.xz is relatively simple, because we're going to make it simple for individual distro packagers. In fact, (1) can be made into a .deb with little addition.

And, yes 1271 is indeed friendlier for us, as I previously noted. :)
 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/7/12 6:19 PM

Hi, Pritam,

I agree with the points you made in your last post.

I found a little time and removed the toolbar at the top, here's how it looks like:

(Without the native menu it looks even better, but the Windows and Mac versions have such a menu and I left it this way. I guess the Brackets team is planning on switching to only native menu at some point.)

Next I will set an icon set for the window, set the www location to be relative to Brackets main executable, and finally package it and upload it somewhere so that people can try it themselves and give some feedback.

Is there anything else left to be done?

Thanks,
Rado
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/7/12 7:04 PM
Hi Rado,

Nice work!

I've done some of that stuff too. I removed native menu though (yes it does look better!) Some extra bugfixes/stabilisation with the appshell_extensions functions too. (Check the latest commit!)

I think only CloseLiveBrowser() remains. (I also removed the LiveBrowserMgrLin{} class, as it wasn't being used anyway.)

I was just planning on building an automatic installer and wrapping this up.

Do commit it on your git repo and send a pull request, we'll see to it that we include both our contributions.
 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/8/12 10:19 AM
Hi, Pritam,

That's a very good approach with the icon, but I'm not sure how it works. What is the resulting .h file after running gdk-pixbuf-csource on the .ico file? The Windows icon implementation includes images with the most common sizes (i.e. 16x16, 32x32, 48x48, 256x256), is it the same case here?
Would we be able to see correct icons (icons of the correct size) in the main window and also if we, say, make a launcher on the Desktop?
I hope you understand what I'm asking about. (I was going to propose converting the icon to a .xpm file, but I guess there's not much difference.)

What's the CloseLiveBrowser() function used for (and when) ?

If we manage to complete all the tasks left before the Sprint 16 release of Brackets, maybe we would be able to upload a package (or installers for different linux distros). Do you know how the main Brackets team feels about this?

Thanks,
Rado...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/8/12 6:35 PM
Hi, Rado,

gdk-pixbuf-csource a picture file's binary data into a stream of textual representation. If you look into the .h file, you'll notice that this data gets assigned to a variable. We compile in this file, and access this variable directly, as opposed to loading the icon from a picture file at runtime.

Frankly, .xpm is just another picture format. So is .png, and so is .ico. Given that GDK can load a large number of picture formats (including .png, .ico and .xpm) anyway, I don't think it makes a difference.

GDK automatically handles scaling when displaying the icon in the title bar, taskbar etc. But, unlike Windows, this icon would be visible only when Brackets is running, and within Brackets only. We would have to supply an icon file anyway for Launchers. So, I'm going to load the icon from the relative path ./icons/scalable/bracekts.ico Compiling the icon into the binary itself was cool, but since we're gonna have to distribute a separate file anyway...

CloseLiveBrowser() is used when Live Preview fails to detect a working remote-debugging interface for the page, and the user chooses to Relaunch Chrome. There are discussions on how best to do this, and option that seems best as of now is to use a custom user-profile for Chrome, and close instances of only that profile.

PS: What are the tasks that are required to be completed?
 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/9/12 8:59 AM
Hi, Pritam,

XPM is indeed an image format, but it perhaps has a similar structure like the .h file you described. An .xpm (the same goes to .xbm files I guess) contains basically a C array of all the pixels of the image and this way you can include the .xpm so that you don't worry about distributing the icon.

What are the tasks that are required to be completed?
Well, what I meant was, everything that needs to be done before we (or probably the Brackets team) can distribute the application for testing.

And, btw, using a custom user profile for Chrome sounds like a really good idea.
Let me know if there's a way that I can help.

Thanks,
Rado
...
Re: [brackets-dev] Re: Linux Version Narciso (nj) Jaramillo 11/9/12 9:05 AM
Hi Rado & Pritam,

Thanks so much (again) for continuing to push this forward. It's really exciting to see this coming together.

Regarding making a package for distribution, that would be great. I don't think we'd be able to host it in our Downloads area until we'd had a chance to review and test it ourselves, but if you put it up somewhere public, we could certainly let people know about it to try it out. The next step would probably be for you to submit pull requests to the brackets-shell and brackets repos for the Linux changes you made so that we could get them integrated into master. After that, we'd want to look into making an automated build process for Linux binaries and packages, which would then lead to being able to post them in our Downloads area.

Thanks!

nj
...
Re: [brackets-dev] Re: Linux Version Rado Rodopski 11/9/12 11:00 PM
Hi, nj,

That's very nice to hear.

In order to build on linux you need to install some prerequisite libraries (only if you miss them on the machine) and run the following commands:
gyp --depth=.  --> this will generate the Makefile, Pritam has done an amazing job with the gyp file
make

Then we could create a script (or just an additional custom step in the gyp file) to generate some sort of a package (or 10, we'll discuss more on this). and we just have to wait for everythin to build and pack up.

Let me know if you have more questions and I'll wait for Pritam to add his comments.

Thanks,
Rado
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/10/12 4:46 AM
I've added build instructions to Building-brackets-shell

As I said before, in our case, packaging is very simple since all files are relatively together. I would give the list of files/folders if required, but I guess it's fairly obvious what needs to be packaged. (PS: It's all in the out/Release/ folder)

As about launch for testing. the current situation is fairly usable, with only few uncritical functions here and there remaining. I'm having some trouble with a particularly deceptive focus bug at the moment, and I hope someone can see to it.

Builds are posted on Github | Downloads.

NB: These are using slightly patched builds of CEF. Irrelevant to the udev thingy of before, which isn't that big an issue and can be overlooked. More on this later. The patch in question is related to a bug in CEF which was hindering the asynchronous window closing behaviour. Reported CEF issue upstream
 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Khrono 11/10/12 5:11 AM
the link http://github.com/pritambaral/downloads to download the CEF is broken, return a 404 page


2012/11/10 Brackets-Dev on behalf of Pritam Baral <bracke...@googlegroups.com>
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/10/12 5:14 AM
@ricardo. Thanks for pointing that out. Fixed.
 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Khrono 11/10/12 5:15 AM
no problem, i'm tryng to build it up on debian :P...
Re: [brackets-dev] Re: Linux Version Khrono 11/10/12 5:23 AM
a simple question, I have to use the master branch ?? if yes, there is no makefile in the root of the project...

ricca@ego:~/Apps/brackets-shell-master$ make
make: *** No targets specified and no makefile found.  Stop.

this is the output. where am I doing wrong?


2012/11/10 Khrono <galanos...@gmail.com>







nj
Hi Rado,

Nice work!

 

Regards,
Chhatoi Pritam Baral




Nice work.

 

Regards,
Chhatoi Pritam Baral





 

Regards,
Chhatoi Pritam Baral





 

Regards,
Chhatoi Pritam Baral





 

Regards,
Chhatoi Pritam Baral




              '<!@(<(pkg-config) --libs
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/10/12 5:30 AM
Sorry, I should've mentioned it. It's not in master yet, you need to use the linux branch. The Makefile isn't supplied by default, because the locations of headers and libraries vary from distribution to distribution. You'll need gyp (available on Debian repos) to generate the Makefile, (GYP is like cmake in this regard.)

Modified instructions on wiki accordingly. Thanks for pointing this out! :)
 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Khrono 11/10/12 5:34 AM
ok, i'll pull clone the linux branch and try it again...
Re: [brackets-dev] Re: Linux Version Khrono 11/10/12 5:53 AM
if u're on irc contact me plz i'm kronoo



2012/11/10 Khrono <galanos...@gmail.com>







2012/11/10 Khrono <galanos...@gmail.com>







nj
Hi Rado,

Nice work!

 

Regards,
Chhatoi Pritam Baral




Nice work.

 

Regards,
Chhatoi Pritam Baral





 

Regards,
Chhatoi Pritam Baral





 

Regards,
Chhatoi Pritam Baral





 

Regards,
Chhatoi Pritam Baral




              '<!@(<(pkg-config) --libs-only-l udev)'
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/10/12 7:21 PM
I notice that a major part of the platform-dependent code in brackets-shell are operations Chromium does anyway, like reading a file, or prompting user to select a file etc.. So I guess Chromium, hence CEF, must be able to handle these tasks on it's own. Instead of we implementing these operations in native code by ourselves, we might as well just call CEF functions to do the task.

A lot of the code is rendered redundant if this is possible.
 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Pritam Baral 11/12/12 12:27 AM
Some help from the Brackets team needed.

We're having trouble figuring out how the focus to the main editor is handled. Is it via CefBrowser::SetFocus(true) ? We've noticed that the editor div loses the class "CodeMirror-focused" when the user navigates to a different window on the OS's window manager. The same class is applied back on the div, when the Brackets window gains focus. So there's something JS-side going on here, and it's definitely getting it's input/signal from brackets-shell

The problem at hand is, brackets' main editor doesn't accept keystrokes on start-up. One has to manually click inside the editor, or press Tab 13 times (cycling through some HTML elements,) or press Shift+Tab once to correctly focus on the main editor. We've tried CefBrowser::SetFocus(true) but to no avail.

Any help would be appreciated
 

Thanks,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Jason San Jose 11/14/12 11:25 AM
Most, if not all, of the filesystem i/o (appshell.fs) platform-specific code will be replaced with platform-neutral NodeJS code. Sometime in the next few sprints, we're looking at starting our first NodeJS integration story.

For the file dialog operations, I'm not sure if there's anything in CEF that we can use that's not tied to browser operations (e.g. input type=file). Did you find something that's suited to our usage?

@jasonsanjose
...
Re: [brackets-dev] Re: Linux Version Jason San Jose 11/14/12 11:28 AM
I talked with Rado on IRC last night. We weren't able to narrow down why the linux shell CEF window isn't getting focus at startup. It doesn't seem to be a JavaScript-side issue on my initial investigation.

I think we could sanity check by trying a simple page with just a standard text input and try to set focus to it on $(document).ready().
...
Re: Linux Version codersofthedark 11/28/12 12:01 PM
Do we have brackets for linux/ubuntu now?

On Friday, June 29, 2012 11:40:33 PM UTC+5:30, Stramel wrote:
I wasn't sure how to contact you as it says to in your README. But I would like to help out with creating a Linux version. The problem is that I am new to the Brackets-app/shell project and CEF1/3. I am studying computer science and have a fair knowledge and understanding but may need a little direction to get started. ie. outline of how building it for win/mac would normally go. 
Re: [brackets-dev] Re: Linux Version Pritam Baral 12/4/12 5:17 PM
Hello everyone,

Sorry for the delay, I went offline as my university exams had begun. I have just pushed a workaround for the focus-on-start bug.

With reasonable confidence I can say Brackets for Linux is ready for testing. Two builds (32 and 64 bit respectively) have been posted. The core brackets code from Sprint 16 has also been packaged and posted on the same page. A tester may chose to build it for himself, or download the brackets-shell-[64,32] and brackets-sprint-16 builds and run it.
 

Regards,
Chhatoi Pritam Baral




On Thu, Nov 29, 2012 at 1:31 AM, codersofthedark via Brackets-Dev <brackets-dev+noreply-APn2wQfU_G4K17S-VBYCqwvL15Rxk6R1e-fxPnZ53NX@googlegroups.com> wrote:
Do we have brackets for linux/ubuntu now?

On Friday, June 29, 2012 11:40:33 PM UTC+5:30, Stramel wrote:
I wasn't sure how to contact you as it say
...
Re: [brackets-dev] Re: Linux Version Narciso (nj) Jaramillo 12/5/12 6:20 PM
Awesome! I have on my todo list to get a Linux vm up and running sometime soon, and will try it out.

I'm assuming we need both the "linux" branch from pritambaral/brackets as well as the shell download from pritambaral/brackets-shell? (Or does that download already have brackets packaged inside it?)

Thanks,

nj
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 12/5/12 6:29 PM
Both brackets-shell and brackets are packaged up as brackets-shell-[32,64]-bit.tar.bz2 and brackets-sprint-16.tar.bz2 respectively on the downloads page. Download both and extract to the proper directory. The folders www/ and samples/ being in the same folder as the Brackets binary is preferable, but you extract anywhere and manually choose the location of www/index.html on start-up.

Or, if you want to test the building process, the linux branches on both pritambaral/brackets and pritambaral/brackets-shell are the ones to use.
 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Randall Edmunds 12/7/12 12:55 PM
Pritam Baral,

I was able to get Brackets running on OpenSuse 12.1. This is awesome. Thanks for the hard work by you, and everyone else who has contributed.


> The folders www/ and samples/ being in the same folder as the Brackets binary is preferable,
> but you extract anywhere and manually choose the location of www/index.html on start-up.

FYI, it wouldn't work for me when I put these folders side-by-side, but it did work when I put the www/ and samples/ being in the same folder as the Brackets.

Thanks,
Randy

Re: [brackets-dev] Re: Linux Version Ilya Radchenko 12/10/12 6:14 PM
I'm having trouble building the linux version, I downloaded this CEF(cef_binary_3.1271.889_linux-32.zip), and the master branch. Installed "gyp", but get this message when run "make":

......$ make
  CXX(target) out/Common_Base/obj.target/libcef_dll_wrapper/libcef_dll/wrapper/cef_byte_read_handler.o
In file included from ./include/internal/cef_types.h:46:0,
                 from ./include/internal/cef_types_wrappers.h:36,
                 from ./include/cef_base.h:44,
                 from ./include/wrapper/cef_byte_read_handler.h:40,
                 from libcef_dll/wrapper/cef_byte_read_handler.cc:5:
./include/internal/cef_types_linux.h:38:21: fatal error: gtk/gtk.h: No such file or directory
compilation terminated.
make: *** [out/Common_Base/obj.target/libcef_dll_wrapper/libcef_dll/wrapper/cef_byte_read_handler.o] Error 1

When I try to install GTK2, I get a message that it already exists. I'm running on Ubuntu12.10. Thanks!



On Friday, December 7, 2012 3:55:39 PM UTC-5, Randall Edmunds wrote:
Pritam Baral,

I was able to get Brackets running on OpenSuse 12.1. This is awesome. Thanks for the hard work by you,
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 12/11/12 6:54 AM
@Ilya, can you mail me (on my personal mail id, as I think it might be considered spam if we discussed this on the group) the output of 

pkg-config gtk+-2.0 gthread-2.0

On another note, you need to use the "linux" branch of my repo, master won't do. Do try again with the linux branch.
 

Regards,
Chhatoi Pritam Baral




On Tue, Dec 11, 2012 at 7:44 AM, Ilya Radchenko via Brackets-Dev <brackets-dev+noreply-APn2wQcDNqdudYrzcSxTtS9zO_-SYNlLsmWCCloICJJ@googlegroups.com> wrote:
I'm having trouble building the linux version, I downloaded this CEF(cef_binary_3.1271.889_linux-32.zip), and the master branch. Installed "gyp", but get this message when run "make":

......$ make
  CXX(target) out/Common_Base/obj.target/libcef_dll_wrapper/libcef_dll/wrapper/cef_byte_read_handler.o
In file included from ./include/internal/cef_types.h:46:0,
                 from ./include/internal/cef_types_wrappers.h:36,
                 from ./include/cef_base.h:44,
                 from ./include/wrapper/cef_byte_read_handler.h:40,
                 from libcef_dll/wrapper/cef_byte_read_handler.cc:5:
./include/internal/cef_types_linux.h:38:21: fatal error: gtk/gtk.h: No such file or directory
compilation terminated.
make: *** [out/Common_Base/obj.target/libcef_dll_wrapper/libcef_dll/wrapper/cef_byte_read_handler.o] Error 1

When I try to install GTK2, I get a message that it already exists. I'm running on Ubuntu12.10. Thanks!



On Friday, December 7, 2012 3:55:39 PM UTC-5, Randall Edmu
...
Re: [brackets-dev] Re: Linux Version Randall Edmunds 12/11/12 8:17 AM
> can you mail me (on my personal mail id, as I think it might be considered spam if we discussed this on the group)...

Discussing the building of the Linux version of Brackets is very much welcome on this forum, but please start a new topic. This "Linux Version" thread is getting way too big to follow. :)

Thanks,
Randy

Re: [brackets-dev] Re: Linux Version Ilya Radchenko 12/12/12 5:04 PM
Thanks for the advice! I've started a new post here.
Re: [brackets-dev] Re: Linux Version Pablo Luis Botta 12/16/12 11:26 AM
It's Works! Thank you guys!
Re: [brackets-dev] Re: Linux Version Sebastian 12/20/12 7:26 AM
Really Awesome! Thank you very much. I love seeing brackets on Linux.

If anyone is interested (and it is allowed?), I can try and build debian packages...

Greetings,
Sebastian
Re: [brackets-dev] Re: Linux Version Jason San Jose 12/20/12 9:47 AM
Yes, that would be great! I don't think Pritam has a build script yet, but you're able to write a script for packaging that would be ideal.

Jason
...
Re: [brackets-dev] Re: Linux Version Ioannis Panteleakis 12/21/12 3:31 PM
Sebastian,
debian/ubuntu packages would certainly most welcome! Thank you! Lots of ppl are waiting for an easy way to install brackets on their linux pc.
Thank you again!
Re: [brackets-dev] Re: Linux Version Jasdeep Khalsa 12/29/12 3:56 PM
Thanks Pritam - This is working for me fine on Ubuntu too!

I did have some trouble finding the link to your downloads though, for anyone else looking for them:


Merry Christmas!

Jasdeep.
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 12/29/12 5:20 PM

Thanks Jasdeep.

About the downloads link: github is decommissioning the downloads service, and so has removed the explicit link from their pages. The url will keep working till 90 days after their announcement though.

I'll have to move them to a different location, once I'm back at my university.

Re: Linux Version János Pásztor 1/3/13 5:16 AM
Hi Rado,

I just discovered Brackets and I have plenty of resources to burn on one of my servers. I'm rather short on time, so I can't do this myself but if you are willing, I can give you a virtual machine with enough resources to build Brackets.

Please contact me privately via contact details at http://www.janoszen.com/contact/ if you're interrested.

Janos

On Saturday, September 22, 2012 7:58:06 PM UTC+2, Rado Rodopski wrote:
Hi, Randy,

Thanks for the info.

I read the CEF wiki pages and it seems that they have managed to successfully build CEF3 on Ubuntu, which is good news for me. But I read the build instructions for CEF3 and in the prerequisites it says that I need a machine with at least 4GB of RAM and 8GB swap area and also over 10GB to checkout and build the source tree. Since I don't have a machine that meets those requirements, I can't build CEF3.
Could someone from the Brackets team (who has access to a good Linux box) possibly build CEF3 and upload the binary someplace?

Thanks,
Rado

20 септември 2012, четвъртък, 20:16:45 UTC+3, Randall Edmunds написа:

Rado,

Thanks for your interest in Brackets.

> Since I see that CEF3 supports Linux (not sure exactly what this
> means, probably that it compiles fine and runs on Linux?)

Yes, CEF3 runs on Linux, but the binaries are not yet delivered, so, for now, you’ll need to build those yourself. See the brackets-shell wiki for more info:

https://github.com/adobe/brackets-shell/wiki/Building-brackets-shell

 

> I suppose I could copy the Windows part of the code and tweak it to run in Linux?

That’s a reasonable approach. All of the OS-specific code for Windows is in files that end with “_win”, so start by copying each of those. I think that appending them with “_linux” is probably too generic, since there are many

...
Re: Linux Version Brian Hinton 1/6/13 7:26 PM
Anyone know of any updates from Adobe on the Linux version? I did successfully get the sprint 16, linked above, running on elementary OS. I'm hoping we'll see some official support soon. Is it possible to build a updated version? It was a bit unclear on whether building it would build the latest sprint or just sprint 16.
Re: [brackets-dev] Re: Linux Version Eric Dykstra 1/25/13 8:07 AM
Hello Pritnam,

I am on Debian Squeeze. I followed the instructions here: https://github.com/adobe/brackets/wiki/Linux-Version

Now, when I start Brackets, it prompts me for the index.html file, and I browse to the /brackets/src/index.html file but I get the following error.

Failed to load URL data:text/html,chromewebdata with error (-3).


In the terminal where I started Brackets, I see the following:

No cache_path supplied by default
[0125/110158:ERROR:renderer_main.cc(214)] Running without renderer sandbox

Then I close the Brackets window and in the terminal I see:

Segmentation fault

Ho
...
Re: [brackets-dev] Re: Linux Version Pritam Baral 1/25/13 8:26 AM
HI Eric,

Make sure you try a sprint release of Brackets and not from a git working that is not a sprint release. We don't want to debug a temporary issue with a git snapshot.

Also, I personally haven't tested any of the sprint builds after sprint-16. A sprint-16 bzipped tarball is provided on github.com/pritambaral/brackets-shell/downloads
 

Regards,
Chhatoi Pritam Baral

...
Re: [brackets-dev] Re: Linux Version Eric Dykstra 1/25/13 8:38 AM
I just downloaded the Sprint 16 tarball from there. (Brackets (not brackets-shell) Sprint 16 release)

When starting Brackets, I then navigated to the /www/index.html file. Exact same results.

I'm sure I am doing something wrong. No idea what though.

On Friday, 25 January 2013 11:26:55 UTC-5, Pritam Baral wrote:
HI Eric,

Make sure you try a sprint release of Brackets and not from a git working that is not a sprint release. We don't want to debug a temporary issue with a git snapshot.

Also, I personally haven't tested any of the sprint builds after sprint-16. A sprint-16 bzipped tarball is provided on github.com/pritambaral/brackets-shell/downloads
 
...
Re: [brackets-dev] Re: Linux Version Randall Edmunds 1/25/13 8:59 AM
> When starting Brackets, I then navigated to the /www/index.html file. Exact same results.

I had a similar problem. Moving the /www folder to the same folder as the Brackets executable fixed it for me on OpenSUSE 12.1.

The downside of this fix is that the Debug > Run Tests (for running unit tests) menu item is disabled, but it allows me to run Brackets for now.

Randy

Re: [brackets-dev] Re: Linux Version Eric Dykstra 1/25/13 9:07 AM
Randy,

That helped move me forward, but presented a new error. It was more descriptive, so it prompted me to move the /samples folder there along with the /www folder per your suggestion and VIOLA! It works!

THANKS!

Eric
...
Re: [brackets-dev] Re: Linux Version Randall Edmunds 1/25/13 9:11 AM
Excellent!

FYI,

This thread has grown so large and hard to follow that I created this wiki page to summarize the info so far, and also as a collecting point for more info:

https://github.com/adobe/brackets/wiki/Testing-Pritam%27s-Linux-build

That wiki page has a table for logging what has been tested, and links to where to log issues and to find issues to fix. Feel free to update that page with any info that you think is relevant.

Thanks,
Randy

Re: [brackets-dev] Re: Linux Version Eric Dykstra 2/6/13 7:03 PM
Hey everyone,

I am not seeing any updates on the Linux build anywhere. Has it stalled?

Thanks,

Eric
...
Re: [brackets-dev] Re: Linux Version Randall Edmunds 2/7/13 9:27 AM
Hi Eric,

I just made an update to the [wiki page](https://github.com/adobe/brackets/wiki/Testing-Pritam%27s-Linux-build) with my results.

Have you tried the build? If so, please open any issues that you are seeing, and add your results to the wiki page.

Thanks,
Randy

Re: [brackets-dev] Re: Linux Version Eric Dykstra 2/7/13 1:26 PM
Randy,

Ugh - NOW 'Post Reply' works, but AFTER I did a 'Post Reply To Author'. Anyway - resubmitting here for all to benefit from.

I tried a few weeks ago, and from what I can remember, I was limited to Sprint 16. I got it working, but there were a lot of 'issues'. Since it is an older Sprint, I didn't know if they were 'bugs', or known missing functionality, or errors in my build. It didn't make sense to spend the effort to determine which category the issues are, when most of them could be resolved/completed in the latest Sprint. So, I couldn't really update the Wiki without possibly filling it with useless info.

I just went back to the docs to see if I can build for a newer Sprint. I must say - it is confusing. I am not finding the same information this time, and in fact, I can't figure out how to build the shell for Linux now. The links in the docs send me in circles or dead ends.

If I recall, the shell must be aligned with the brackets code, and there were file sets for both for Sprint 16, and nothing newer for the shell.

Have I got that right?

Someone needs to have ONE page detailing the exact steps for Linux users and all references to Linux point there.

I would be happy to play 'the confused newbie' and point out where the steps are lacking. It is pretty easy to miss those areas when you've done it a bunch of times and take things for granted. I'm not there yet. :)

Thanks!

Eric

On Thursday, 7 February 2013 12:27:54 UTC-5, Randall Edmunds wrote:
Hi Eric,

I just made an update to the [wiki pa
...
Re: [brackets-dev] Re: Linux Version Randall Edmunds 2/7/13 2:36 PM
Eric,

Thanks for following up.


> Someone needs to have ONE page detailing the exact steps for Linux users
> and all references to Linux point there

That is the goal of the wiki page I started here:
https://github.com/adobe/brackets/wiki/Testing-Pritam%27s-Linux-build

Feel free to update it with any information that you think is relevant.


> I tried a few weeks ago, and from what I can remember, I was limited to Sprint 16.
> I got it working, but there were a lot of 'issues'. Since it is an older Sprint, I didn't know
> if they were 'bugs', or known missing functionality, or errors in my build. It didn't make
> sense to spend the effort to determine which category the issues are, when most of them
> could be resolved/completed in the latest Sprint. So, I couldn't really update the Wiki
> without possibly filling it with useless info.

Yes, this is a Sprint 16 build. I am encouraged to see that you want to get to the latest build, but I think it makes the most sense for us to resolve as many basic issues from this build, so it will be easier to move forward. So, please compare against Sprint 16, for now.

At this point we're not concerned about exhaustively testing to find...
Re: [brackets-dev] Re: Linux Version Eric Dykstra 2/7/13 4:07 PM
Randy,

Re: "So, please compare against Sprint 16, for now."

Sorry. I can't compare it against Windows or Mac. I have Linux everywhere, because I refuse to work with closed source systems. That's what attracted me to Brackets in the first place.

Wait a minute ... isn't the whole point of Brackets is that it is built ON and FOR open source technologies? If so, why isn't Linux THE development platform, with the others lagging behind? Maybe the tag line should be; "Open-Source code editor, built with the web, for the web, but only if you paid for a Closed-Source OS." (Rant switch is off now.)

Having said that, I just got it working using the builds for Sprint 16. However, after opening a folder with many sub-folders, the file navigation tree on the left is full, and I can't clear it. Even after re-starting. That makes even high-level testing very difficult. So, Brackets is useless to me after 10 seconds. :(

Eric  

On Thursday, 7 February 2013 17:36:43 UTC-5, Randall Edmunds wrote:
Eric,

Thanks for following up.

> Someone needs to have ONE page detailing the exact steps for Linux users
> and all references to Linux point there

That is the goal of the wiki page I started here:
https://github.com/adobe/brackets/wiki/Testing-Pritam%27s-Linux-build

Feel free to update it with any information that you think is relevant.

> I tried a few weeks ago, and from what I can remember, I was limited to Sprint 16.
> I got it working, but there were a lot of 'issues'. Since it is an older Sprint, I didn't know
> if they were 'bugs', or known missing functionality, or errors in my build. It didn't make
> sense to spend the effort to determine which category the issues are, when most of them
> could be resolved/completed in the latest Sprint. So, I couldn't really update the Wiki
> without possibly filling it with useless info.

Yes, this is a Sprint 16 build. I am encouraged to see that you want to get to the latest build, but I think it makes the most sense for
...
Re: [brackets-dev] Re: Linux Version Randall Edmunds 2/7/13 4:22 PM
> Re: "So, please compare against Sprint 16, for now."
>
> Sorry. I can't compare it against Windows or Mac. I have Linux everywhere

Fair enough. The Release Notes for each Sprint are here: https://github.com/adobe/brackets/wiki/Release-Notes


> after opening a folder with many sub-folders, the file navigation tree on the left is full,
> and I can't clear it. Even after re-starting. That makes even high-level testing very difficult.

The sidebar (aka Project Panel) that contains the file navigation tree has scroll bars that only appears when you move the mouse over that panel. Are you not seeing those?

If not, please file an issue (https://github.com/pritambaral/brackets/issues) with info about your OS and version.

Thanks,
Randy

Re: [brackets-dev] Re: Linux Version Pritam Baral 2/7/13 4:44 PM
Hi Eric,

I'm sorry it lags way behind the Windows and Mac editions too.I guess the last update I did to it was back in November. The guys have Adobe, and many other contributors seem to have made a lot of progress on the "Closed-source OS" editions in the meanwhile.

I hate to see such a good open-source project fall behind on an open-source platform. But the good thing about Brackets being open-source is that it allows *anyone* to modify and improve. I doubt we'd have Linux version today if Brackets wasn't even open-source. (Response to rant is off now)

That being said, I have included a link to build instructions on the Testing wiki. Please give me some time to update Brackets (other contributors are welcome too.)

@Randy, I guess the file-nav-tree-too-long is a Brackets issue, and not -shell related. Do tell me if I'm wrong
 

Regards,
Chhatoi Pritam Baral



Yes, this is a Sprint 16 build. I am encouraged to see that you want to get to the latest build, but I think it makes the most sense for us to resolve as many basic issues from this build, so it will be easier to move forward. So, please compare agai
...
Re: [brackets-dev] Re: Linux Version Eric Dykstra 2/7/13 5:20 PM
Pritam,

Re: "Response to rant is off now." Well played and it gave me a good laugh. Thanks.

Perhaps you could give a high-level and concise description of what's involved with keeping the Linux shell upto date with the Brackets code, and why it needs to be in synch in the first place. Perhaps my problem is that I picture a 'shell' as a 'wrapper' that is separate from the Brackets code, in which case it should be easier to keep it in sync. Or, maybe I am correct, but the API is still changing dramatically? Something isn't making sense to me. If it is clearly described, maybe others in the Open-Source community can help out?

Thanks,

Eric


On Thursday, 7 February 2013 19:44:06 UTC-5, Pritam Baral wrote:
Hi Eric,

I'm sorry it lags way behind the Windows and Mac editions too.I guess the last update I did to it was back in November. The guys have Adobe, and many other contributors seem to have made a lot of progress on the "Closed-source OS" editions in the meanwhile.

I hate to see such a good open-source project fall behind on an open-source platform. But the good thing about Brackets being open-source is that it allows *anyone* to modify and improve. I doubt we'd have Linux version today if Brackets wasn't even open-source. (Response to rant is off now)

That being said, I have included a link to build instructions on the Testing wiki. Please give me some time to update Brackets (other contributors are welcome too.)

@Randy, I guess the file-nav-tree-too-long is a Brackets issue, and not -shell related. Do tell me if I'm wrong
 
...
Re: [brackets-dev] Re: Linux Version Randall Edmunds 2/7/13 5:22 PM
Pritam,


> I guess the file-nav-tree-too-long is a Brackets issue, and not -shell related.

I don't understand the issue. Here's how it's supposed to work:

If the height (or width) of the file tree exceeds the height (or width) of it's container, then you get a vertical (or horizontal) scroll bar. If scroll bars are required on the Brackets Sidebar, they are *only* shown if you move the mouse over the Sidebar.

Note that the scrollbars in the Sidebar are not standard in appearance, and they are different from the scrollbars that you see in the editor. It's a designer thing.

That's what I'm seeing on OpenSUSE 12.1, so let me know what is not right for you.

Thanks,
Randy

Re: [brackets-dev] Re: Linux Version Eric Dykstra 2/7/13 5:41 PM
Randy,

Re: " ....the file navigation tree has scroll bars that only appears when you move the mouse over that panel. Are you not seeing those?"

Sorry - forgot to answer that. Yes - I didn't realize it was there, because it is almost the same color as the background. Regardless, that is not the issue. What I really need is a way to clear the navigation tree, or it will become a mile long in no time.

Re: " The Release Notes for each Sprint are here ..."

Yes. I skimmed those before. But, expecting a person to keep the Release Notes for 4 sprints in their head (or review them each time) so that they don't waste everyone's time generating bug reports for bugs already resolved, is not realistic. That's why I want to setup an environment where I can keep current.

Thanks,

Eric


On Thursday, 7 February 2013 19:22:06 UTC-5, Randall Edmunds wrote:
> Re: "So, please compare against Sprint 16, for now."
>
> Sorry. I can't compare it against Windows or Mac. I have Linux everywhere

Fair enough. The Release Notes for each Sprint are here: https://github.com/adobe/brackets/wiki/Release-Notes

> after opening a folder with many sub-folders, the file navigation tree on the left is full,
> and I can't clear it. Even after re-starting. That makes even high-level testing very difficult.

The sidebar (aka Project Panel) that contains the file navigation tree h
...
Re: [brackets-dev] Re: Linux Version Randall Edmunds 2/7/13 6:23 PM
> Perhaps you could give a high-level and concise description of what's involved
> with keeping the Linux shell upto date with the Brackets code, and why it needs
> to be in synch in the first place. Perhaps my problem is that I picture a 'shell' as
> a 'wrapper' that is separate from the Brackets code, in which case it should be
> easier to keep it in sync. Or, maybe I am correct, but the API is still changing
> dramatically?

In Brackets, the shell is operating system specific, native code mostly for accessing the file system. Yes, Brackets is still pretty early, so the API is still changing. we don't anticipate a lot of other changes to the native shell, but I'm sure there will be some.

After Sprint 16, we implemented native menus and a few other minor changes. You can get around implementing native menus by always setting brackets.inBrowser to true on Linux.

Randy

Re: [brackets-dev] Re: Linux Version Randall Edmunds 2/7/13 6:34 PM
> What I really need is a way to clear the navigation tree, or it will become a mile long in no time.

The left Sidebar is also known as the Project Panel. It basically shows the contents of any folder on your file system, where the "folder" is your "project". You can't really "clear" it, just change it. Use File > Open Folder... to change the project folder. Also, click on the little triangle next to the root folder name to see a list of all of the previous projects so tyou can quickly switch between them.

Once you make an edit to a file, or double-click on it in the project tree, it gets added to the "working set" list above the project tree. You can remove files from that list using File > Close, Ctrl-W, or mouse over and click the (x) icon on the left.

Does that make sense?



> Re: " The Release Notes for each Sprint are here ..."
>
> Yes. I skimmed those before. But, expecting a person to keep the Release Notes for
> 4 sprints in their head (or review them each time) so that they don't waste everyone's
> time generating bug reports for bugs already resolved, is not realistic. That's why I want
> to setup an environment where I can keep current.

I don't see the differen
...
Re: [brackets-dev] Re: Linux Version Eric Dykstra 2/8/13 9:17 AM
Randy,

Re: "Does that make sense?"

Yes. Not really a 'standard' approach, but I think a better one. Glad to see Brackets is not just doing things the way they always have been just 'because'. What got me confused is that I was not able to open new folders. Turns out, it is because they are remote and mounted via SSHFS, and Brackets doesn't like that for some reason. Not sure how it can even know. I've never had an issue doing it that way with any other editor. 

Thanks,

Eric


On Thursday, 7 February 2013 21:34:53 UTC-5, Randall Edmunds wrote:
> What I really need is a way to clear the navigation tree, or it will become a mile long in no time.

The left Sidebar is also known as the Project Panel. It basically shows the contents of any folder on your file system, where the "folder" is your "project". You can't really "clear" it, just change it. Use File > Open Folder... to change the project folder. Also, click on the little triangle next to the root folder name to see a list of all of the previous projects so tyou can quickly switch between them.

Once you make an edit to a file, or double-
...
Re: [brackets-dev] Re: Linux Version Romuald TISSERAND 2/9/13 2:26 AM
Hi there,
New to Brackets and volunteer to help.

Brackets S16 runs pretty well on my 2 Ubuntu 12.10 64 systems. But I'm totally unable to run the tests. I don't really understand this sentence from the wiki:
"Note that Unit Tests are disabled if you use the www folder download." What are we supposed to do?

I've tried to move the www and samples folder outside my brackets, but in that case, I get the following error when running Brackets:

"Failed to load URL data:text/html,chromewebdata with error (-3)."


Any help would make me ready to help !

Thanks.


Le jeudi 7 février 2013 23:36:43 UTC+1, Randall Edmunds a écrit :
Eric,

Thanks for following up.

> Someone needs to have ONE page detailing the exact steps for Linux users
> and all references to Linux point there

That is the goal of the wiki page I started here:
https://github.com/adobe/brackets/wiki/Testing-Pritam%27s-Linux-build

Feel free to update it with any information that you think is relevant.

> I tried a few weeks ago, and from what I can remember, I was limited to Sprint 16.
> I got it working, but there were a lot of 'issues'. Since it is an older Sprint, I didn't know
> if they were 'bugs', or known missing functionality, or errors in my build. It didn't make
> sense to spend the effort to determine which category the issues are, when most of them
> could be resolved/completed in the latest Sprint. So, I couldn't really update the Wiki
> without possibly filling it with useless info.

Yes, this is a Sprint 16 build. I am encouraged to see that you want to get to the latest build, but I think it makes the most sense for us to resolve as m
...
Re: [brackets-dev] Re: Linux Version Romuald TISSERAND 2/9/13 2:48 AM
A little update as I've just seen the tests are not packaged into to the Pritam archives. What's the recommended procedure to get the tests sources?...
Re: [brackets-dev] Re: Linux Version Randall Edmunds 2/9/13 9:50 AM
Romuald,

The correct way to install Brackets for running unit tests is:

1. Get brackets-shell setup either by:
a. Downloading it (from here: https://github.com/pritambaral/brackets-shell/downloads).
b. Cloning it (from here: https://github.com/pritambaral/brackets-shell) and downloading cef libraries (from here: https://github.com/pritambaral/brackets-shell/downloads), and then building it.

2. Clone brackets (from here: https://github.com/pritambaral/brackets)

3. Start Brackets and when you are prompted for index.html file, choose brackets/src/index.html.

I get the same error that you noted (which is captured in: https://github.com/pritambaral/brackets/issues/2).


I am able to invoke Unit Test dialog using these hacks:

1. Download brackets and copy www/ and samples/ folders from the download to the same folder as Brackets executable.
2. Clone brackets repo and copy the src/ and test/ folders from brackets folder to the same folder as Brackets executable.
3. Start Brackets
4. Use Debug > Run Tests to invoke the Jasmine Spec Runner dialog...
Re: [brackets-dev] Re: Linux Version Romuald TISSERAND 2/10/13 11:59 AM
Thanks Randall,
Your hack works...almost.

I'm able to display the "Jasmine Spec Runner" window, with a menu inside "Brackets Tests, Unit, Performance, Extensions, Reload, Show Developer tools", but nothing more. If a click any menu item, the menu bar blinks one time and...that's all. The window main area remains totally empty.

Curious thing, from the Brackets window, Debug -> "Show Developer tools" doesn't work, it does nothing. Same on the Jasmine Spec Runner window, by from there, a right click on the menu bar raises a "Show DevTools" menu item which works, it shows the Developer Tools window.

Here I am, is there a way to run the tests?

Le samedi 9 février 2013 18:50:57 UTC+1, Randall Edmunds a écrit :
Romuald,

The correct way to install Brackets for running unit tests is:

1. Get brackets-shell setup either by:
a. Downloading it (from here: https://github.com/pritambaral/brackets-shell/downloads).
b. Cloning it (from here: https://github.com/pritambaral/brackets-shell) and downloading cef libraries (from here: https://github.com/pritambaral/brackets-shell/downloads), and then building it.

2. Clone brackets (from here: https://github.com/pritambaral/brackets)

3. Start Brackets and when you are prompted for index.html file, choose brackets/src/index.html.

I get the same error that you noted (which is captured in: https://github.com/pritambaral/brackets/issues/2).


I am able to invoke Unit Test dialog using these hacks:

1. Download brackets and copy www/ and samples/ folders from the download to the same folder as Brackets executable.
2. Clone b
...
Re: [brackets-dev] Re: Linux Version Romuald TISSERAND 2/11/13 2:15 AM
Hi,
Another question, I've seen that building CEF & the Brackets shell required GTK 2.0. But why using such a nearly deprecated API and not using GTK 3.x ?
Re: [brackets-dev] Re: Linux Version Romuald TISSERAND 2/11/13 5:45 AM
I've tried to build the whole project and it fails because it doens't find the gtk.h header. Here is the whole trace:


  CXX(target) out/Common_Base/obj.target/libcef_dll_wrapper/libcef_dll/wrapper/cef_byte_read_handler.o
In file included from ./include/internal/cef_types.h:46:0,
                 
from ./include/internal/cef_types_wrappers.h:36,
                 
from ./include/cef_base.h:44,
                 
from ./include/wrapper/cef_byte_read_handler.h:40,
                 
from libcef_dll/wrapper/cef_byte_read_handler.cc:5:
./include/internal/cef_types_linux.h:38:21: fatal error: gtk/gtk.h: No such file or directory
compilation terminated
.
make
: ***
...
(unknown) 2/11/13 10:17 PM <This message has been deleted.>
Re: [brackets-dev] Re: Linux Version Reuben Walker 2/11/13 10:19 PM
That seems to be the same problem found here. https://groups.google.com/forum/#!topic/brackets-dev/vGMA3wOMFDs

Are you using the Linux branch? It got me too.

git checkout linux

Or something like that.
...
Re: [brackets-dev] Re: Linux Version Romuald TISSERAND 2/12/13 12:52 AM
Obviously, this seems to be the same error, thanks a lot Reuben. The git command is "git clone -b linux http://xxxxxxxxxx" and now the shell builds!...
Re: [brackets-dev] Re: Linux Version Romuald TISSERAND 2/25/13 2:01 AM
Hi there,
This forum has a very low activity, is there any news about Brackets for Linux? Good or bad...
Re: [brackets-dev] Re: Linux Version Valéry Herlaud 6/7/13 8:22 AM
Hi,

I have an error a bit like this.

make: *** No rule to make target `Release/lib.target/libcef.so', needed by `out/Release/lib/libcef.so'. Stop.

Any idea ?

Thx !

Valéry


Le samedi 3 novembre 2012 01:50:12 UTC+1, Pritam Baral a écrit :
Sorry, I should've mentioned it:

I changed the appshell.gyp to include only two configurations (BUILDTYPEs); Release and Debug. Release is default, and a simple make should suffice

The Debug BUILDTYPE mightn't work, as the CEF binary isn't compiled in debug mode, yet.

Also, I'm looking into ways to drop some unneeded/troublesome dependencies, e.g.: udev. It's still a work-in-progress, so expect issues.

 

Regards,
Chhatoi Pritam Baral




On Sat, Nov 3, 2012 at 6:12 AM, Rado Rodopski via Brackets-Dev <brackets-dev+noreply-APn2wQcqmncJeKB1mQpsS5dMzcdH_ouy7lpHddynbFM@googlegroups.com> wrote:
Hi,

I generated a Makefile with gyp and run
make BUILDTYPE=Default

but I got the following output:
make: *** No rule to make target `out/Default/obj.target/libcef_dll_wrapper/libcef_dll/transfer_util.o', needed by `out/Default/obj.target/libcef_dll_wrapper.a'.  Stop.


What am I doing wrong?

Thanks,
Rado



On Monday, October 29, 2012 12:52:31 PM UTC+2, Pritam Baral wrote:
Latest update: https://twitter.com/_PritamBaral_/status/262666953741697024

Still needs a few basic native functions to be ported, and the entire LiveBrowserManagerLin class.

We just need to add a few include_dirs and some system libraries at link time. I have added the relevant include_dirs in the appshell.gyp, (probably not in the best way possible, as they are local to my system,) but I didn't have the time to look into GYP's syntax in order to find out how to add linker libraries.

Run make BUILDTYPE=Default and use the out/Default/ directory, I am yet to find out how to set the BUILDTYPE from GYP. This step WILL FAIL at the link-time. See below

For the linking step, I manually modified the command, and have included it at the bottom of the appshell.gyp file as a comment.


PS: Some compile-time/link-time issues may crop up, as my set-up is the furthest thing from portable at the moment. Do post them here, so we can modify it and make it more portable for the Linux community at large.
 

Regards,
Chhatoi Pritam Baral




On Mon, Oct 29, 2012 at 2:46 PM, Brackets-Dev on behalf of Rado Rodopski <brackets-dev+noreply-APn2wQcqmncJeKB1mQpsS5dMzcdH_ouy7lpHddynbFM@googlegroups.com> wrote:
Hi,

That's a very good progress you've made!

Can you point me to the platform-specific code? I will take a look and there's a big chance I can help you with porting them for Linux.
Also, any pointers on how to compile brackets-shell are welcome since it will save me a couple of hours trying on my own (and I have a very tight schedule).

Thanks,
Rado

27 октомври 2012, събота, 00:02:26 UTC+3, Pritam Baral написа:
Hey,

I managed to generate the Makefile by installing and running gyp on the appropriate directory and .gyp file.

I also managed compile it all in, after manually adding a few include_dirs to the OS==linux section of the .gyp file. The linker, on the other hand, presented with trouble as I haven't yet ported any of the platform-specific code to linux/gtk. I'm currently working on it, but it would be encouraging if there was a way to skip the platform-specific code and just compile/link the brackets-shell and cefclient, libcef_dll_wrapper together.
 

Regards,
Chhatoi Pritam Baral




On Sat, Oct 27, 2012 at 1:39 AM, Brackets-Dev on behalf of Chris Franklin <bracke...@googlegroups.com> wrote:
Hey,

Had a quick look at the code to see what you were talking about, I was having a little google around to see if any of what we are trying to do could be automated and found this: http://wiki.gnustep.org/index.php/Pbxbuild do you think an approach like this could save us time or should we go the manual route?

Cheers,

Chris


On Fri, Oct 26, 2012 at 8:54 PM, Chris Franklin <ch...@piemonster.me> wrote:
Hey,

Good work on all of this, thanks for your hard work. I am in the middle of launching something at work so I am pretty busy this week. 

Next week on the other hand I shall be chilling in Cornwall with a laptop and a good amount of free time. I will have a look at the build scripts then. 

I shall also be looking at Django and some sort of interactive Python shell integration as I see a lot of potential in that area but that is for another post.

Cheers,

Chris


On Fri, Oct 26, 2012 at 1:22 PM, Pritam Baral via Brackets-Dev <brackets-dev+noreply-APn2wQdDiymJwGOCgT_wcRBAeOVV4MXzDoo9fZX4MMX@googlegroups.com> wrote:
Hey everyone, I have built the binary distribution of CEF3 (1271 branch, 1180 is on the way,) with a few caveats:
  • dynamically-linked to libudev.so.0 (my Arch Linux setup only has libudev.so.1. A symlink from .so.0 to .so.1 works fine, for now)
  • cef_unittests haven't been built (linker complained about a missing symbol, didn't have time to hunt that down myself)
  • some compiler flags (-Werror=c++0x-compat, -Werror=uninitialised, -Werror=deprecated-declarations) were disabled, but only in some targets respectively, not all.

I have also gone through the Building Brackets-shell wiki, but the crux of the compilation stuff seems to be done by XCode/Visual Studio, using the project files generated from .gyp files. The next obvious step for me seems to be to get/generate a Makefile and compile this stuff.
...
Re: [brackets-dev] Re: Linux Version Jason San Jose 6/7/13 10:09 AM
Sounds like you need to download the patched CEF binaries from @pritambaral here https://github.com/pritambaral/brackets-shell/downloads.

However, I've been working on a simplified setup script. Please see this thread instead https://groups.google.com/d/msg/brackets-dev/C_4TJ3A8ppg/cUD3fJYvBKwJ.

Jason

 

Regards,
Chhatoi Pritam Baral








Th
...
More topics »