Re: Sometimes it feels that nobody cares... [WAS: Re: IntelliJ IDEA vs Netbeans]

[Christian Bourque]

Will do. Thanks!

On Sun, Apr 22, 2018 at 6:29 PM, Geertjan Wielenga <
geertjan.wiele...@googlemail.com> wrote:

> Spend some time exploring apache.org to see how Apache projects work.
>
> For example, this page: https://community.apache.org/
>
> Thanks,
>
> Gj
>
>
> On Mon, Apr 23, 2018 at 12:20 AM, Christian Bourque <
> christian.bour...@gmail.com> wrote:
>
> > Hi Emilian,
> >
> > Well, as I'm not yet very familiar with the NetBeans development
> > intricacies, I thought you were "the" reviewer because when I received
> the
> > first automated email regarding my PR it contained comments from you
> saying
> > that some changes were requested! But if, like you said: you were just
> "a"
> > reviewer, then who's in charge of making the final decision as whether or
> > not a pull request will be accepted?
> >
> > The only reason I said that it seemed nobody cared is simply because at
> > some point you stopped exchanging with me! My last question to you was
> > about the NetBeans module and how I could determine if it was OSGi or NBM
> > (like you were asking in your previous comment).
> >
> > Again, I'm not very familiar with all this so maybe I overreacted and I
> > apologize! Next time, I'll send a message to the mailing list instead...
> >
> > Thanks again for your message and comments, they were very constructive!
> >
> > Christian
> >
> > On Sun, Apr 22, 2018 at 7:04 AM, Emilian Bold <
> emilian.b...@protonmail.ch>
> > wrote:
> >
> > > Hello Christian,
> > >
> > > I believe I might have been the 'reviewer' on your PR. If so, I
> apologise
> > > for making you feel unwelcome.
> > >
> > > It would be nice if you would explain what your expectation in this
> case
> > > were. (There are some hints in your email, particularly the idea that
> > there
> > > should be continuous feedback, a completion, etc.)
> > >
> > > From my perspective, I was not "the" reviewer, I was just "a"
> reviewer. I
> > > noticed your PR, thought about it a bit and gave my feedback. Clearly
> the
> > > initial PR was too big and there were some problems with those
> > dependencies.
> > >
> > > How cool that you also found a bug in JGit, right?
> > >
> > > Anyhow, don't take a PR personal or so serious. Some "reviewer" might
> be
> > > more interested in a given PR and follow it to the end. Other
> "reviewers"
> > > might just be passing through, notice some problems and point them out.
> > >
> > > I will certainly think more about a review creating a commitment,
> > > especially for first-time contributors.
> > >
> > > Obviously, there's always the option of sending an email on dev@.
> > Usually
> > > GitHub shows me this notification when I get replies, I don't remember
> > > seeing anything new from that JGit PR...
> > >
> > > PS: Don't believe I'm undervaluing your time spent or your
> contributing.
> > I
> > > have multiple contributions that went nowhere, so I totally know the
> > > feeling. I guess for some things you also need good timing...
> > >
> > > --emi
> > >
> > > ‐‐‐ Original Message ‐‐‐
> > >
> > > On 21 April 2018 1:03 AM, Christian Bourque <
> christian.bour...@gmail.com
> > >
> > > wrote:
> > >
> > > > I have to agree with Gili...
> > > >
> > > > Sometimes it feels that nobody cares... I created a ticket a month
> ago
> > > >
> > > > about an improvement in the NetBeans Git integration and decided to
> > code
> > > it
> > > >
> > > > myself.
> > > >
> > > > Then I submitted a PR and eventually the reviewing process started.
> But
> > > the
> > > >
> > > > reviewer has had many objections to my request (which is perfectly
> > > fine). I
> > > >
> > > > told him that the Eclipse JGit library was OSGi based and that it had
> > > >
> > > > external dependencies that were triggered during the bootstrapping of
> > > >
> > > > NetBeans. I even contacted the Eclipse team to tell them that one of
> > > their
> > > >
> > > > OSGi dependencies wasn't even used in their JGit library, the lead
> > > >
> > > > developer realized that it was a mistake and he fixed it!
> > > >
> > > > After that, I sent more questions and comments to the reviewer to
> > > >
> > > > understand how I could resolve the remaining issues in an acceptable
> > way,
> > > >
> > > > but to no avail. The reviewer just stopped answering my questions!
> Now
> > > it's
> > > >
> > > > been 25 days and nothing... So, I put a lot of time into this for
> what
> > at
> > > >
> > > > the end?
> > > >
> > > > I don't think that this is how you're going to attract new developers
> > to
> > > >
> > > > contribute to this project...
> > > >
> > > > Christian
> > > >
> > > > On Fri, Apr 20, 2018 at 1:36 PM, cowwoc cow...@bbs.darktech.org
> wrote:
> > > >
> > > > > I did. Multiple times.
> > > > >
> > > > > One of the issues I brought up was the copy/paste bug I linked to
> > > below.
> > > > >
> > > > > When I brought it up in the mailing list in the past I got no
> > replies.
> > > > >
> > > > > That issue (and its linked counterparts) are very annoying (for

Re: Sometimes it feels that nobody cares... [WAS: Re: IntelliJ IDEA vs Netbeans]

[Geertjan Wielenga]

Spend some time exploring apache.org to see how Apache projects work.

For example, this page: https://community.apache.org/

Thanks,

Gj


On Mon, Apr 23, 2018 at 12:20 AM, Christian Bourque <
christian.bour...@gmail.com> wrote:

> Hi Emilian,
>
> Well, as I'm not yet very familiar with the NetBeans development
> intricacies, I thought you were "the" reviewer because when I received the
> first automated email regarding my PR it contained comments from you saying
> that some changes were requested! But if, like you said: you were just "a"
> reviewer, then who's in charge of making the final decision as whether or
> not a pull request will be accepted?
>
> The only reason I said that it seemed nobody cared is simply because at
> some point you stopped exchanging with me! My last question to you was
> about the NetBeans module and how I could determine if it was OSGi or NBM
> (like you were asking in your previous comment).
>
> Again, I'm not very familiar with all this so maybe I overreacted and I
> apologize! Next time, I'll send a message to the mailing list instead...
>
> Thanks again for your message and comments, they were very constructive!
>
> Christian
>
> On Sun, Apr 22, 2018 at 7:04 AM, Emilian Bold 
> wrote:
>
> > Hello Christian,
> >
> > I believe I might have been the 'reviewer' on your PR. If so, I apologise
> > for making you feel unwelcome.
> >
> > It would be nice if you would explain what your expectation in this case
> > were. (There are some hints in your email, particularly the idea that
> there
> > should be continuous feedback, a completion, etc.)
> >
> > From my perspective, I was not "the" reviewer, I was just "a" reviewer. I
> > noticed your PR, thought about it a bit and gave my feedback. Clearly the
> > initial PR was too big and there were some problems with those
> dependencies.
> >
> > How cool that you also found a bug in JGit, right?
> >
> > Anyhow, don't take a PR personal or so serious. Some "reviewer" might be
> > more interested in a given PR and follow it to the end. Other "reviewers"
> > might just be passing through, notice some problems and point them out.
> >
> > I will certainly think more about a review creating a commitment,
> > especially for first-time contributors.
> >
> > Obviously, there's always the option of sending an email on dev@.
> Usually
> > GitHub shows me this notification when I get replies, I don't remember
> > seeing anything new from that JGit PR...
> >
> > PS: Don't believe I'm undervaluing your time spent or your contributing.
> I
> > have multiple contributions that went nowhere, so I totally know the
> > feeling. I guess for some things you also need good timing...
> >
> > --emi
> >
> > ‐‐‐ Original Message ‐‐‐
> >
> > On 21 April 2018 1:03 AM, Christian Bourque  >
> > wrote:
> >
> > > I have to agree with Gili...
> > >
> > > Sometimes it feels that nobody cares... I created a ticket a month ago
> > >
> > > about an improvement in the NetBeans Git integration and decided to
> code
> > it
> > >
> > > myself.
> > >
> > > Then I submitted a PR and eventually the reviewing process started. But
> > the
> > >
> > > reviewer has had many objections to my request (which is perfectly
> > fine). I
> > >
> > > told him that the Eclipse JGit library was OSGi based and that it had
> > >
> > > external dependencies that were triggered during the bootstrapping of
> > >
> > > NetBeans. I even contacted the Eclipse team to tell them that one of
> > their
> > >
> > > OSGi dependencies wasn't even used in their JGit library, the lead
> > >
> > > developer realized that it was a mistake and he fixed it!
> > >
> > > After that, I sent more questions and comments to the reviewer to
> > >
> > > understand how I could resolve the remaining issues in an acceptable
> way,
> > >
> > > but to no avail. The reviewer just stopped answering my questions! Now
> > it's
> > >
> > > been 25 days and nothing... So, I put a lot of time into this for what
> at
> > >
> > > the end?
> > >
> > > I don't think that this is how you're going to attract new developers
> to
> > >
> > > contribute to this project...
> > >
> > > Christian
> > >
> > > On Fri, Apr 20, 2018 at 1:36 PM, cowwoc cow...@bbs.darktech.org wrote:
> > >
> > > > I did. Multiple times.
> > > >
> > > > One of the issues I brought up was the copy/paste bug I linked to
> > below.
> > > >
> > > > When I brought it up in the mailing list in the past I got no
> replies.
> > > >
> > > > That issue (and its linked counterparts) are very annoying (for
> obvious
> > > >
> > > > reasons). I suspect no one else is experiencing this problem because
> it
> > > >
> > > > seems too major to ignore.
> > > >
> > > > Gili
> > > >
> > > > On 2018-04-20 1:21 PM, Geertjan Wielenga wrote:
> > > >
> > > > > Rather than filing hundreds of issues, I recommend you pick one,
> > just one,
> > > > >
> > > > > and open a discussion thread around that. Once people have seen and
> 

Re: dpi on Windows Java 9

[Eirik Bakke]

JDK now has a standard way of loading HiDPI icons on both MacOS, Linux,
and Windows: An icon named "foo.png" can be paired with an icon named
"f...@2x.png" of double resolution. Swing will then pick the best icon to
use at any given point, even when a window is dragged from a HiDPI screen
to a non-HiDPI screen in a multi-monitor setup. This kind of @2x
resolution loading works with standard Image APIs such as "new
ImageIcon(URL)". Relevant resolved JDK issues:

https://bugs.openjdk.java.net/browse/JDK-8011059 (MacOS support)

https://bugs.openjdk.java.net/browse/JDK-8073320 (Windows support)

https://bugs.openjdk.java.net/browse/JDK-8044189 (on making MacOS and
Windows HiDPI work the same)
https://bugs.openjdk.java.net/browse/JDK-8055212 (Windows & Linux)


I think NetBeans should follow the same convention. Generating PNG files
from SVG files is something that should be done at icon design time--often
the icon designer needs to do pixel-level tweaks to the small version of
the icon, or even remove some details for the icon to remain legible. This
is especially the case for tiny 16x16 icons, or for the super-tiny 8x8
icon "badges" that netbeans superimposes on project files to indicate
errors, repo changes, and such.

As far as NetBeans is concerned, the first step in better HiDPI support
would be to get ImageUtilities to support the @2x.png loading convention
on all platforms. I think Emilian has already some some of this work:
https://github.com/emilianbold/nextbeans/commit/0f99dba0c1b3e8e0bc4e7cec407
b53d30e85ead1 .

Questions for Emilian about the work you already did on ImageUtilities:
(1) Did you ever try this on Windows? (I know you're a Mac guy, like me :-)
(2) Are you taking advantage of the new interfaces that were added to
Swing to handle HiDPI rendering? If so, that will likely make it easier to
get your code working on Windows and Linux as well. I see you have a class
that extends from MultiResolutionImage, which bodes well.
(3) Does your code work when a window is dragged from a HiDPI screen to a
non-HiDPI screen in a multi-monitor setup?

Once ImageUtilities is made to support @2x loading, the next step is to
actually update some of the icons. Some icons are much more important than
others. In particular, just getting 2x versions of the Windows and Mac
icons in openide.awt/src/org/openide/awt/resources would go a long way.
Those are all the little UI buttons that are part of the core window
system (maximize/close tab etc.). If I knew that ImageUtilities would
support it, and if I have some free time one day, I might be tempted to do
some of these myself. I'd just make them look exactly as they currently
do, but at a double resolution (to avoid long design discussions).

My next laptop will be a HiDPI Windows one, so I might have a chance to do
more testing on that platform eventually.

-- Eirik

On 3/11/18, 5:13 PM, "Tim Boudreau"  wrote:

>IMO, if we wanted to do this and be future-proof, the thing to do would be
>to convert the icons to SVG or some similar vector format and update the
>icon loading code in ImageUtilities to use it.
>
>There are some tools - particularly potrace - that can assist in initial
>conversion, but deal in tracing lines and shapes and give 2- or 3- color
>output (potrace lets you set a single interior color for closed shapes),
>but which could be helpful as a start.
>
>Given that SVG is XML-based, I could imagine that just using something
>like
>Batik to load SVG would be unacceptable;  but SVG could be used as a
>designer-friendly input stage, and then be compiled by the build process
>into something more performant (either literal Java code that paints the
>contents of the svg, or some binary representation of drawing
>instructions), much the same way resource bundle message annotations are
>turned into static methods.
>
>I've written code before that implements a Graphics2D that simply stores a
>list of everything it was told to do - wouldn't be that hard to take the
>list of drawing instructions from there and generate Java code to
>reproduce
>those steps.
>
>Straw man example:
>
> - Code that uses an SVG icon is annotated with
>@Icon("org/netbeans/modules/x/myIcon.svg")
> - At build time:
>   - Annotation processor looks up that file, reads it with Batik, paints
>it into a code-generating Graphics2D which outputs a generated static
>method myIcon() on a generated class in the same package
>   - Code that wants to use the icon calls the static method, the same way
>bundle messages are loaded in modern NetBeans code
> - At run time:
>   - If using the generated static method, the output is just an image or
>an icon, no surprises
>   - For the case where icons are shared across modules (this happens),
>ImageUtilities could be tweaked to, in the case of a .svg extension, try
>to
>look up the generated class / method (and optionally fall back to Batik if
>nobody minded a dependency on it, but that could be pluggable via Lookup
>if
>it's 

Re: Sometimes it feels that nobody cares... [WAS: Re: IntelliJ IDEA vs Netbeans]

[Christian Bourque]

Hi Emilian,

Well, as I'm not yet very familiar with the NetBeans development
intricacies, I thought you were "the" reviewer because when I received the
first automated email regarding my PR it contained comments from you saying
that some changes were requested! But if, like you said: you were just "a"
reviewer, then who's in charge of making the final decision as whether or
not a pull request will be accepted?

The only reason I said that it seemed nobody cared is simply because at
some point you stopped exchanging with me! My last question to you was
about the NetBeans module and how I could determine if it was OSGi or NBM
(like you were asking in your previous comment).

Again, I'm not very familiar with all this so maybe I overreacted and I
apologize! Next time, I'll send a message to the mailing list instead...

Thanks again for your message and comments, they were very constructive!

Christian

On Sun, Apr 22, 2018 at 7:04 AM, Emilian Bold 
wrote:

> Hello Christian,
>
> I believe I might have been the 'reviewer' on your PR. If so, I apologise
> for making you feel unwelcome.
>
> It would be nice if you would explain what your expectation in this case
> were. (There are some hints in your email, particularly the idea that there
> should be continuous feedback, a completion, etc.)
>
> From my perspective, I was not "the" reviewer, I was just "a" reviewer. I
> noticed your PR, thought about it a bit and gave my feedback. Clearly the
> initial PR was too big and there were some problems with those dependencies.
>
> How cool that you also found a bug in JGit, right?
>
> Anyhow, don't take a PR personal or so serious. Some "reviewer" might be
> more interested in a given PR and follow it to the end. Other "reviewers"
> might just be passing through, notice some problems and point them out.
>
> I will certainly think more about a review creating a commitment,
> especially for first-time contributors.
>
> Obviously, there's always the option of sending an email on dev@. Usually
> GitHub shows me this notification when I get replies, I don't remember
> seeing anything new from that JGit PR...
>
> PS: Don't believe I'm undervaluing your time spent or your contributing. I
> have multiple contributions that went nowhere, so I totally know the
> feeling. I guess for some things you also need good timing...
>
> --emi
>
> ‐‐‐ Original Message ‐‐‐
>
> On 21 April 2018 1:03 AM, Christian Bourque 
> wrote:
>
> > I have to agree with Gili...
> >
> > Sometimes it feels that nobody cares... I created a ticket a month ago
> >
> > about an improvement in the NetBeans Git integration and decided to code
> it
> >
> > myself.
> >
> > Then I submitted a PR and eventually the reviewing process started. But
> the
> >
> > reviewer has had many objections to my request (which is perfectly
> fine). I
> >
> > told him that the Eclipse JGit library was OSGi based and that it had
> >
> > external dependencies that were triggered during the bootstrapping of
> >
> > NetBeans. I even contacted the Eclipse team to tell them that one of
> their
> >
> > OSGi dependencies wasn't even used in their JGit library, the lead
> >
> > developer realized that it was a mistake and he fixed it!
> >
> > After that, I sent more questions and comments to the reviewer to
> >
> > understand how I could resolve the remaining issues in an acceptable way,
> >
> > but to no avail. The reviewer just stopped answering my questions! Now
> it's
> >
> > been 25 days and nothing... So, I put a lot of time into this for what at
> >
> > the end?
> >
> > I don't think that this is how you're going to attract new developers to
> >
> > contribute to this project...
> >
> > Christian
> >
> > On Fri, Apr 20, 2018 at 1:36 PM, cowwoc cow...@bbs.darktech.org wrote:
> >
> > > I did. Multiple times.
> > >
> > > One of the issues I brought up was the copy/paste bug I linked to
> below.
> > >
> > > When I brought it up in the mailing list in the past I got no replies.
> > >
> > > That issue (and its linked counterparts) are very annoying (for obvious
> > >
> > > reasons). I suspect no one else is experiencing this problem because it
> > >
> > > seems too major to ignore.
> > >
> > > Gili
> > >
> > > On 2018-04-20 1:21 PM, Geertjan Wielenga wrote:
> > >
> > > > Rather than filing hundreds of issues, I recommend you pick one,
> just one,
> > > >
> > > > and open a discussion thread around that. Once people have seen and
> > > >
> > > > discussed your issue on the mailing list, probably one of those will
> > > >
> > > > volunteer to tix it.
> > > >
> > > > Gj
> > > >
> > > > On Friday, April 20, 2018, cowwoc cow...@bbs.darktech.org wrote:
> > > >
> > > > This is precisely the kind of fixes I'm talking about. I understand
> that
> > > >
> > > > > IntelliJ has many features that Netbeans does not, but I'm
> primarily
> > > > >
> > > > > focused on improving the core development functionality (this is
> > > > >
> > > > 

Re: IntelliJ IDEA vs Netbeans

[Валера Солдатов]

22.04.2018 22:39, William L. Thomson Jr. пишет:

On Fri, 20 Apr 2018 22:01:48 +0200
Kai Uwe Pel  wrote:


Just wondering...  for example, if you need to coding with C++, the
only way to get decent C/C++ support within the IntelliJ IDEA is to
use/buy JetBrains CLion.
Anyway NetBeans IDE supports everything in one, and you no need to
have all possible features like IntelliJ. Simple said!

Lately I do more C in Netbeans than Java. It has good support for
CMake, my preferred C build system. Not really on the Meson bandwagon.
Only bits I am eager to see Oracle release :)

Oracle/Sun always used Netbeans for much more than Java. It was the
foundation for their other products for C/C++ and Fortran/
It has had a stupid amount of names/rebranding...
https://en.wikipedia.org/wiki/Oracle_Developer_Studio

Netbeans + Commercial stuff

Now almost all IDE-related Developer Studio features can be found in 
https://hg.netbeans.org/ sources.



-
To unsubscribe, e-mail: dev-unsubscr...@netbeans.incubator.apache.org
For additional commands, e-mail: dev-h...@netbeans.incubator.apache.org

For further information about the NetBeans mailing lists, visit:
https://cwiki.apache.org/confluence/display/NETBEANS/Mailing+lists

Re: IntelliJ IDEA vs Netbeans

[William L. Thomson Jr.]

On Fri, 20 Apr 2018 22:01:48 +0200
Kai Uwe Pel  wrote:

> Just wondering...  for example, if you need to coding with C++, the
> only way to get decent C/C++ support within the IntelliJ IDEA is to
> use/buy JetBrains CLion.
> Anyway NetBeans IDE supports everything in one, and you no need to
> have all possible features like IntelliJ. Simple said!

Lately I do more C in Netbeans than Java. It has good support for
CMake, my preferred C build system. Not really on the Meson bandwagon.
Only bits I am eager to see Oracle release :)

Oracle/Sun always used Netbeans for much more than Java. It was the
foundation for their other products for C/C++ and Fortran/
It has had a stupid amount of names/rebranding...
https://en.wikipedia.org/wiki/Oracle_Developer_Studio

Netbeans + Commercial stuff

-- 
William L. Thomson Jr.


pgp44UCAGKqtq.pgp
Description: OpenPGP digital signature

Re: Snap Packaging of NetBeans

[Matthias Bläsing]

Hi Neil,

Am Sonntag, den 22.04.2018, 12:10 + schrieb Neil C Smith:
> On Sun, 22 Apr 2018, 09:35 Matthias Bläsing,  
> wrote:
> This sounds like something I would be interested in - is there
> > indication, that Oracle wants to switch away from GPLv2-CP to
> > something
> > easier distributable?
> > 
> 
> No, but that's kind of opposite to what I was thinking. And GPLv2-CPE is
> generally easy to distribute, except by us!
> 
> I meant that we're allowed a JDK dependency because it's considered a
> platform dependency in Apache. But with a general move away from a system
> installed JDK and towards self-contained app bundles like this, we may find
> it harder to ship convenience binaries in future in ways that align with
> what people expect on each OS.

I see two options: Either we find a supplier, that ships Java with a
compatible license (yeah I doubt that, that even exists), or the apache
foundation revises its policy regarding GPLv2-CPE for this class of
redistribution.

And there is a third option: Create a secondary project outside the
apache foundation, that takes vanilla netbeans source, builds the IDE
and bundles it with the JRE and distributes the resulting binary.

The JDK can't be bundled in any case, as to my knowledge GPLv2 is
incompatible with the ALv2.

Greetings
Matthias


-
To unsubscribe, e-mail: dev-unsubscr...@netbeans.incubator.apache.org
For additional commands, e-mail: dev-h...@netbeans.incubator.apache.org

For further information about the NetBeans mailing lists, visit:
https://cwiki.apache.org/confluence/display/NETBEANS/Mailing+lists

Re: Tracking Issues, Versions

[Laszlo Kishalmi]

Well that one could work for me as well. Anyone else has any other idea 
or supporting this one?



On 04/21/2018 10:04 PM, cowwoc wrote:

The way I've seen JIRA used in the past is:

Setting a "Fix Version" of 9.0 on an unresolved issue implies a desire 
to fix it in that release. When that issue is subsequently marked as 
resolved, the Fix Version indicates what version it was fixed in.


Gili

On 2018-04-22 12:53 AM, Laszlo Kishalmi wrote:

Dear all,

Some of you might noticed that I spent some time digging JIRA up and 
down during the last few days.


I try to identify those issues which we really need to solve before 
9.0 release. Right now it is hard to say. We are working with two 
version field at the moment: Affects version(s) and Fix Version(s).


1. Affects Versions: is quite straight forward: It is used to define
   the version where the issue is found. For me it is not necessary the
   version it shall be fixed.
2. Fix Versions: It can be set when an issue is resolved, marking the
   version where the fix for the issue would be delivered.
3. There is no clear field which would specify if an issue shall be in
   a specific release. Well, I know this is not really agile, but we
   are still kind of waterfall, having infrequent major releases.

To see my problem, check: NETBEANS-656 Help is Broken (Offline and 
On-line) 


Affect Versions: 9.0 (well, yes we do not deliver JavaHelp)

Fix Version: Though there is an improvement PR applied from Geertjan, 
I still can't say it is fixed in 9.0. I set it to Next, though the 
issue is still unresolved and I do not feel myself comfortable with 
that setting either.


We can haggle on the Priority of the issue. It does not really 
affects me, but if we really would like to provide a decent IDE we 
might need to come up with a JavaHelp replacement. So critical would 
be fine for me either. Still this issue would not be delivered in NB 
9.0.


I'd propose label those issues which we would really like to deliver 
in 9.0 with "NB9.0", though I'm open any other 
clarifications/suggestions etc. So let's discuss this!






-
To unsubscribe, e-mail: dev-unsubscr...@netbeans.incubator.apache.org
For additional commands, e-mail: dev-h...@netbeans.incubator.apache.org

For further information about the NetBeans mailing lists, visit:
https://cwiki.apache.org/confluence/display/NETBEANS/Mailing+lists






-
To unsubscribe, e-mail: dev-unsubscr...@netbeans.incubator.apache.org
For additional commands, e-mail: dev-h...@netbeans.incubator.apache.org

For further information about the NetBeans mailing lists, visit:
https://cwiki.apache.org/confluence/display/NETBEANS/Mailing+lists

Re: Snap Packaging of NetBeans

[Laszlo Kishalmi]

Well then,

There are two major kind of snaps: Classic and Sandboxed

Classic containers has full access to the OS stuff as of filesystems, 
networking, devices


Sandboxed apps sees their own world.

As the IDE has many plugins and many external tools Sandboxed mode is 
not suitable. That's why I build classic snaps (you need to specify 
--classic on install)


Regarding classic snaps it can be considered as a read only tar.gz 
extract somewhere on the filesystem. What makes it appealing is the wide 
and hassle free distribution that snap offers, having different channels 
like experimental, beta, stable, etc. and the auto update feature.


So no JDK is distributed with this package, no external libraries, it is 
basically the result of the Ant based build with some Snapcraft 
generated wrappers and a desktop file.


If someone would like to try to build it it is really easy. If you are 
on Ubuntu 16.04+


You just need snapcraft, and the my snap branch (until the PR gets merged)

snap install snapcraft --classic
git clone g...@github.com:lkishalmi/incubator-netbeans.git
cd incubator-netbeans
git checkout snap
cd nbbuild/packaging
snapcraft

On 04/22/2018 05:10 AM, Neil C Smith wrote:

On Sun, 22 Apr 2018, 09:35 Matthias Bläsing, 
wrote:


Laszlo comment on this here:


https://github.com/apache/incubator-netbeans/pull/404#issuecomment-382096515


Yes, I'd seen that and your comment about bundled OpenJDK (which would
equally apply to Oracle JDK). Still wasn't sure, and have seen snaps
bundling.



This sounds like something I would be interested in - is there
indication, that Oracle wants to switch away from GPLv2-CP to something
easier distributable?


No, but that's kind of opposite to what I was thinking. And GPLv2-CPE is
generally easy to distribute, except by us!

I meant that we're allowed a JDK dependency because it's considered a
platform dependency in Apache. But with a general move away from a system
installed JDK and towards self-contained app bundles like this, we may find
it harder to ship convenience binaries in future in ways that align with
what people expect on each OS.

Best wishes,

Neil


--

Neil C Smith
Artist & Technologist
www.neilcsmith.net

Praxis LIVE - hybrid visual IDE for creative coding - www.praxislive.org




-
To unsubscribe, e-mail: dev-unsubscr...@netbeans.incubator.apache.org
For additional commands, e-mail: dev-h...@netbeans.incubator.apache.org

For further information about the NetBeans mailing lists, visit:
https://cwiki.apache.org/confluence/display/NETBEANS/Mailing+lists

Re: Snap Packaging of NetBeans

[Neil C Smith]

On Sun, 22 Apr 2018, 09:35 Matthias Bläsing, 
wrote:

> Laszlo comment on this here:
>
>
> https://github.com/apache/incubator-netbeans/pull/404#issuecomment-382096515


Yes, I'd seen that and your comment about bundled OpenJDK (which would
equally apply to Oracle JDK). Still wasn't sure, and have seen snaps
bundling.


> This sounds like something I would be interested in - is there
> indication, that Oracle wants to switch away from GPLv2-CP to something
> easier distributable?
>

No, but that's kind of opposite to what I was thinking. And GPLv2-CPE is
generally easy to distribute, except by us!

I meant that we're allowed a JDK dependency because it's considered a
platform dependency in Apache. But with a general move away from a system
installed JDK and towards self-contained app bundles like this, we may find
it harder to ship convenience binaries in future in ways that align with
what people expect on each OS.

Best wishes,

Neil

> --
Neil C Smith
Artist & Technologist
www.neilcsmith.net

Praxis LIVE - hybrid visual IDE for creative coding - www.praxislive.org

Sometimes it feels that nobody cares... [WAS: Re: IntelliJ IDEA vs Netbeans]

[Emilian Bold]

Hello Christian,

I believe I might have been the 'reviewer' on your PR. If so, I apologise for 
making you feel unwelcome.

It would be nice if you would explain what your expectation in this case were. 
(There are some hints in your email, particularly the idea that there should be 
continuous feedback, a completion, etc.)

>From my perspective, I was not "the" reviewer, I was just "a" reviewer. I 
>noticed your PR, thought about it a bit and gave my feedback. Clearly the 
>initial PR was too big and there were some problems with those dependencies.

How cool that you also found a bug in JGit, right?

Anyhow, don't take a PR personal or so serious. Some "reviewer" might be more 
interested in a given PR and follow it to the end. Other "reviewers" might just 
be passing through, notice some problems and point them out.

I will certainly think more about a review creating a commitment, especially 
for first-time contributors.

Obviously, there's always the option of sending an email on dev@. Usually 
GitHub shows me this notification when I get replies, I don't remember seeing 
anything new from that JGit PR...

PS: Don't believe I'm undervaluing your time spent or your contributing. I have 
multiple contributions that went nowhere, so I totally know the feeling. I 
guess for some things you also need good timing...

--emi

‐‐‐ Original Message ‐‐‐

On 21 April 2018 1:03 AM, Christian Bourque  wrote:

> I have to agree with Gili...
> 
> Sometimes it feels that nobody cares... I created a ticket a month ago
> 
> about an improvement in the NetBeans Git integration and decided to code it
> 
> myself.
> 
> Then I submitted a PR and eventually the reviewing process started. But the
> 
> reviewer has had many objections to my request (which is perfectly fine). I
> 
> told him that the Eclipse JGit library was OSGi based and that it had
> 
> external dependencies that were triggered during the bootstrapping of
> 
> NetBeans. I even contacted the Eclipse team to tell them that one of their
> 
> OSGi dependencies wasn't even used in their JGit library, the lead
> 
> developer realized that it was a mistake and he fixed it!
> 
> After that, I sent more questions and comments to the reviewer to
> 
> understand how I could resolve the remaining issues in an acceptable way,
> 
> but to no avail. The reviewer just stopped answering my questions! Now it's
> 
> been 25 days and nothing... So, I put a lot of time into this for what at
> 
> the end?
> 
> I don't think that this is how you're going to attract new developers to
> 
> contribute to this project...
> 
> Christian
> 
> On Fri, Apr 20, 2018 at 1:36 PM, cowwoc cow...@bbs.darktech.org wrote:
> 
> > I did. Multiple times.
> > 
> > One of the issues I brought up was the copy/paste bug I linked to below.
> > 
> > When I brought it up in the mailing list in the past I got no replies.
> > 
> > That issue (and its linked counterparts) are very annoying (for obvious
> > 
> > reasons). I suspect no one else is experiencing this problem because it
> > 
> > seems too major to ignore.
> > 
> > Gili
> > 
> > On 2018-04-20 1:21 PM, Geertjan Wielenga wrote:
> > 
> > > Rather than filing hundreds of issues, I recommend you pick one, just one,
> > > 
> > > and open a discussion thread around that. Once people have seen and
> > > 
> > > discussed your issue on the mailing list, probably one of those will
> > > 
> > > volunteer to tix it.
> > > 
> > > Gj
> > > 
> > > On Friday, April 20, 2018, cowwoc cow...@bbs.darktech.org wrote:
> > > 
> > > This is precisely the kind of fixes I'm talking about. I understand that
> > > 
> > > > IntelliJ has many features that Netbeans does not, but I'm primarily
> > > > 
> > > > focused on improving the core development functionality (this is
> > > > 
> > > > because
> > > > 
> > > > it impacts all other project types). We've got many of the pieces in
> > > > 
> > > > place,
> > > > 
> > > > but UI problems make that functionality annoying/slow to use.
> > > > 
> > > > Specifically
> > > > 
> > > > on the point of 4k monitors, I keep on saying that it is impossible to
> > > > 
> > > > copy/paste correctly: https://issues.apache.org/jira/browse/NETBEANS-235
> > > > 
> > > > (see linked issues as well)
> > > > 
> > > > I have already filed thousands of issues before (literally), but those
> > > > 
> > > > got
> > > > 
> > > > flushed down the drain when we migrated to Apache. I've already filed
> > > > 
> > > > some
> > > > 
> > > > issues in Apache JIRA but for now there hasn't been much movement
> > > > 
> > > > (committers are understandably working on their own issues before 
> > > > looking
> > > > 
> > > > at other people's issues). Nothing personal against you guys but I don't
> > > > 
> > > > have the time/energy to re-file all those bug reports. Family 
> > > > emergencies
> > > > 
> > > > are keeping me super busy nowadays and frankly I don't see the point of
> > > > 
> > > > doing all this work until 

Re: Snap Packaging of NetBeans

[Matthias Bläsing]

Hi Neil,

Am Sonntag, den 22.04.2018, 08:27 + schrieb Neil C Smith:
> On Sun, 22 Apr 2018, 07:05 Laszlo Kishalmi, 
> wrote:
> 
> > Once we can provide our official signed versions the '--dangerous'
> > flag
> > would not be required.
> > 
> 
> I've been reading this thread, but not looked at the actual snap yet.
> Does it bundle a JDK or rely on another snap? AFAIK we could never
> distribute a bundled JDK via Apache

Laszlo comment on this here:

https://github.com/apache/incubator-netbeans/pull/404#issuecomment-382096515

I had a quick look at the contents of the snap and indeed it is mostly
the copy of the netbeans directory.

>  - a situation I hope might change for platform
> dependencies in future given the evolving nature of Java deployment!

This sounds like something I would be interested in - is there
indication, that Oracle wants to switch away from GPLv2-CP to something
easier distributable?

Greetings,
Matthias

-
To unsubscribe, e-mail: dev-unsubscr...@netbeans.incubator.apache.org
For additional commands, e-mail: dev-h...@netbeans.incubator.apache.org

For further information about the NetBeans mailing lists, visit:
https://cwiki.apache.org/confluence/display/NETBEANS/Mailing+lists

Re: Snap Packaging of NetBeans

[Neil C Smith]

On Sun, 22 Apr 2018, 07:05 Laszlo Kishalmi, 
wrote:

> Once we can provide our official signed versions the '--dangerous' flag
> would not be required.
>

I've been reading this thread, but not looked at the actual snap yet. Does
it bundle a JDK or rely on another snap? AFAIK we could never distribute a
bundled JDK via Apache - a situation I hope might change for platform
dependencies in future given the evolving nature of Java deployment!

Best wishes,

Neil

> --
Neil C Smith
Artist & Technologist
www.neilcsmith.net

Praxis LIVE - hybrid visual IDE for creative coding - www.praxislive.org

Re: Netbeans encoding

[Tim Boudreau]

Okay, legacy web application - what appserver?

Seems like the bug here is that, if NetBeans is starting the application
server, that it should detect (from configuration files or wherever) the
encoding that application server will use for logging (or know what it
is).  That would be an actual fix without interfering with anything else.

Failing to detect or know output encoding from an application server is the
actual bug here.

In the meantime, you could either keep your netbeans.conf setting, *or* set
the system property file.encoding to your system default in your
appserver's configuration to be consistent with NetBeans' expectations.

-Tim

Re: HTML/Java API goes to Gradle

[Jaroslav Tulach]

Hello Emilian,
thanks for taking a look.

Dne neděle 22. dubna 2018 7:42:08 CEST, Emilian Bold napsal(a):
> I can't help with the patch. Just 2 remarks:
> 
> * didn't notice HTML4J is Maven-based.

I tried to avoid any dependencies on NetBeans Platform in the HTML4J. Thus I 
started from scratch. When starting from scratch I have chosen standard build 
system over NetBeans Ant Harness.

> * I suppose the Platform does not depend on html4j-mavan-plugin so this new
> Gradle dependency doesn't matter.

Right. The platform consumes just the binaries of the libraries. Instead of 
the Maven plugin it uses own special Ant task:
https://github.com/emilianbold/netbeans-releases/commit/a7b3e0732c078

-jt


-
To unsubscribe, e-mail: dev-unsubscr...@netbeans.incubator.apache.org
For additional commands, e-mail: dev-h...@netbeans.incubator.apache.org

For further information about the NetBeans mailing lists, visit:
https://cwiki.apache.org/confluence/display/NETBEANS/Mailing+lists

Re: Snap Packaging of NetBeans

[Laszlo Kishalmi]

Yes. It is my own build just being distributed as a file on an S3 bucket.

Once we can provide our official signed versions the '--dangerous' flag 
would not be required.


Snap help on dangerous:

--dangerous    Install the given snap file even if there are no 
pre-acknowledged signatures for it, meaning it was not verified and 
could be dangerous (--devmode implies this)



On 04/21/2018 10:43 PM, Emilian Bold wrote:

This being an unofficial package made by you Laszlo, correct?

The '--dangerous' flag doesn't look reassuring.

--emi

‐‐‐ Original Message ‐‐‐

On 19 April 2018 11:13 PM, Laszlo Kishalmi  wrote:


The (undecided) 9.0 release candidate is available as snap:

|wget

https://s3-us-west-2.amazonaws.com/lkishalmi-us-west-2-public/snaps/netbeans_9.0-rc_amd64.snap

sudo snap install netbeans_9.0-rc_amd64.snap --dangerous --classic|

On 02/20/2018 05:08 AM, Neil C Smith wrote:


On Tue, 20 Feb 2018 at 12:47 constantin drabo drconstan...@hotmail.com

wrote:


I think it will good . It will be an opportunity to bring back NetBeans to

Fedora Community after several years of mistrust .

+1 to providing the possibility of a Snap package, but I'm slightly

intrigued what this mistrust was about, and amused that Snap is a way to

address it!

There is the question of where Flatpak or AppImage sit with this?

Best wishes,

Neil



-
To unsubscribe, e-mail: dev-unsubscr...@netbeans.incubator.apache.org
For additional commands, e-mail: dev-h...@netbeans.incubator.apache.org

For further information about the NetBeans mailing lists, visit:
https://cwiki.apache.org/confluence/display/NETBEANS/Mailing+lists






-
To unsubscribe, e-mail: dev-unsubscr...@netbeans.incubator.apache.org
For additional commands, e-mail: dev-h...@netbeans.incubator.apache.org

For further information about the NetBeans mailing lists, visit:
https://cwiki.apache.org/confluence/display/NETBEANS/Mailing+lists