Thanks for the encouragement.
I would also like to discuss couple of related things.
I am now in the process of setting a development environment on a
64-bit Fedora Linux, and there are certain impediments that have to be
considered with regard to SDK.
- practically none of debugging runtimes wo
> Why should it need a DEB package
Because this is the common way to install programs on Debian-like
systems. SDK has tons of dependencies, resolving them all by hand is
difficult and time consuming.
It could be RPM too, or whatever there's the way to install on other
platforms. The more the merrie
Hi again,
After some thinking... this is my question, but I'm not yet mentally
ready to do it, only wondering about the consequences of it.
So... The SDK needs to be significantly changed in order for it to be
built and used on Linux. The changes aren't as much related to the
code of the SDK itself
OK then, I'll just put double quotes.
Best.
Oleg
In other places where equals sign is used the left hand side doesn't
have any decorations around it. But in other places single quotes
(apostrophes) are used. Like this:
*
* 0 = normal
* 1 = normal
* 2 = layer
* 3 = multiply
* 4 = screen
* 5 = lighten
* 6 = darken
* 7 = d
> You are welcome to file a bug and a patch.
But what should I put instead?
Best.
Oleg
If you are interested to know, the value of the "bad" characters is
224 (That's not a quotation mark character, certainly, or, at least
not in any one Latin, or Cyrillic, or Semitic language), not sure what
the writer wanted to write there. Most commonly it is the Greek
minuscule alpha character.
Hello,
I'm trying to write an alternative build using SCons and by accident I
discovered that
fxgutils/src/java/com/adobe/internal/fxg/dom/types/BlendMode.java uses
characters that cannot be mapped to Unicode somewhere in the comments:
* The following was introduced in FXG 2.0 which acts like
blen
Thanks for responses, Alex,
OK, let's say we (which is probably me...) will find a way to reduce
the Linux SDK to the compiler and the necessary utilities, and then,
perhaps try to port some / what will appear to be more feasible so it
works on Linux.
There are other things like the player version
Ah... sorry for the noise, I just realized that Linux is never
mentioned as a supported platform, so, my understanding it, it is not
possible by design? :/
Well, sorry if that's the case.
Best.
Oleg
Hello,
This is perhaps an old thread, but this is the only one that mentions
pbutil. So, I'll ask here.
I'm trying to build the SDK on Linux, and it wants me to have this
file in the PIXELBENDER_HOME - I don't have it, and can't find where
to download it from.
But secondly, seriously... why Pixel
Re' vector graphics:
This is one reason why most if not all vector graphics on Linux (GUI
including) are SVGs. SVG isn't really a good format from the performance or
structural point of view, but it appears to be easy to edit. There is also
a good (comparable to Freehand 9 at least) free oss vector
>
> Yes, if we have to, we can probably transform these FLAs into AS projects,
> but you lose the ability to edit the assets in a tool like Flash Pro later.
>
> Well, that's the disadvantage of proprietary tools... I can't edit those
files anyway.
Best.
Oleg
Alex, is the FLA File that you mention the one that has icons for
components / default image icon etc? I mean, it doesn't have any code,
right? If so, maybe we could redo it using SWFMill or similar? I kind of
suspected before this may be a problem.
Or, if not SWFMill, we could then have it as a SW
That could be a bit discouraging, but this approach is hardly any different
from the regular browser's cache. The "catch" in the Flex RSLs is that they
can be downloaded once from site X and then reused by site Y w/o ever
loading them from site Y. If you use the local storage ability, then it has
i
Ah, and, by the way, the practical meaning of -target-player argument for
projects that don't use the framework is simply the location of
playerglobals.swc. I.e. the file itself resides in a directory inside SDK
which has a version number as its name, something like
runtime/10.3/playerglobals.swc.
I can only give a definitive answer to #5. I remember that since FP 10
release there were several critical vulnerability fixes. I think I could
find a kb note on that one, if needed. However, 10.3 is a current official
download provided by Adobe, so, they must be fixed both there and in FP
11.2.
R
>
> localFonts.ser is currently mandatory if you want FlashBuilder to consider
> a directory a valid Flex SDK.
Oh well... you never know.
Best.
Oleg
Re'
unable to open 'C:\temp\flex\test123\frameworks/localFonts.ser'
error: Windows cannot understand paths with both correct and Windows
slashes - you should normalize it first, so that either all slashes are
correct, or Windows-style. The strategy would be to detect correct slashes
and replace the
Not sure, you might've tried it already, but saying it just in case, there
are two useful tasks if you want to canonicalize the path:
Or
I also think that the first one would reverse the slashes into "Windows
mode"... actually, from the time I built on Windows I can remember that
Windows
Hey, I was asking before about JS in Windows executables - and then I found
out that it is quite often disabled as part of the general policy of
disabling WSH. So it's sort of on par with PowerShell.
But I would agree that this may be a good option because compared to cmd
it's at leas a language yo
Aha, I see, thanks for the info!
Best.
Oleg
Hi,
Recently I had to look for a way to set up fdb so that several instances
can run and connect at the same time. Adobe off. docs say that you *cant
change the port* for FDB, but it appears to be a trivial thing to do, if
you read through the code in CLI wrapper, it actually expects -p=
argum
On Mon, May 7, 2012 at 9:09 PM, Carol Frampton wrote:
> The current issue is that ../../anything means one thing if you are in the
> flex root directory and another thing if you are in
> {flex-home}/frameworks. /abosolute/path/anything means the same thing in
> any of the flex subdirectories.
>
Sorry to be a bore (could it be worth than that?), but wouldn't fixing the
program such as to accept relative paths be a better option? I didn't have
the time to look into what Flex Ant task does exactly, but I wouldn't be
surprised that instead of some sort of Path.combine() it does simple string
OK, finally, it appears that
doesn't understand relative path. I.e.
if env.PLAYERGLOBAL_HOME=../../anything then it will simply strip leading
parent directories and combine the rest to the current directory. This
looks more like Flex Ant task's bug rather then c
At the time of posting I didn't know what to write in the readme. And,
Alex, I've just explained, knowing what AIR SDK is doesn't help knowing
what AIR_HOME is because the later can be an awful lot of things:
- the URL to fetch the binaries,
- the downloaded archive,
- the directory, where you extr
I get download results on the first page, same on the second and third, and
I also get a tutorial on installing it (defunct), but none mentions
AIR_HOME, and I can't infer from any other documentation what was the
intent behind that name. If I google for AIR_HOME, I get results related to
air condi
What is AIR SDK? Is it the archive you download, is it the directory where
you extract it or is it some directory inside that directory, or is it a
path to a file inside that directory?
(compare ANT_HOME is the bin directory inside the directory where you
extract Ant files, for example.)
JAVA_HOME
Readme doesn't say what it is.
Please, if you write readme, write it so to explain things. If it says
"Adobe AIR Integration Kit" this is for all intents and purposes equivalent
to "Three Random Words WITH Weird Capitalization" - how am I (or any user)
is supposed to know what it is?
Best.
Oleg
That doesn't really matter... it's not my question, whether player's API
are written in AS3 or not :S it could be COBOL for all I care - what
bothers me is why there is code in that library that shouldn't be there.
And yeah, I'm getting used to the sort of answer "everyone wears their
pants on the
I'm still puzzled - why does that library contains actual code? Maybe
asking Adobe for a *normal* library without that code (which we seems like
don't even need!) would sort out the patents / copyright issues?
On the other hand, I feel bad telling someone they have to download a piece
of *software
Regardless of how the legal side of this pans out, I'm terribly confused
about the content of playerglobal.swc - again, it is not "fakes". Or, if
you consider that analogy with C - it is not the headers only, it is both
the headers and the actual code / implementations. And implementations are
the
For flash.* classes there seem to be only their constructors / some
rather innocuous pieces of code, but the ES code is entirely embedded there.
There's something difficult for me to understand though - what pieces of AS
code from playerglobal actually go into resulting SWF, and why that code is
th
As I looked inside the playerglobal.swc, it has actual AS3 code compiled
into it, not just dummy definitions :S
All classes listed here:
http://hg.mozilla.org/tamarin-redux/file/fdf1416a3536/core are there (maybe
there are more of them, I didn't count).
Alex, why are you sure it violates anything?
Nick,
I don't understand why is it illegal to write an alternative
playerglobal.swc. This does not reproduce what Adobe has done, this is just
a program that serves the same purpose, which, as a consequence, exhibits
similar behavior. It is easy to show that you don't need to reverse
anything in o
>
> I'll ask legal about this, but it might fall under the reverse engineering
> restriction.
But there's published documentation with precise description of what the
API do - why copying that would be a reverse engineering? It's like if I
wanted to write a driver for NVidia adapter - there's no
>
> I'll try to rephrase to make sure I understand - IIUC someone needs to
> tell people to unpack the Apache Flex source distribution and mix that
> with other files which do not come from Apache, in order to use
> FlashBuilder which is an external tool that does not belong to us.
No, files like
>
> but your example applies to the way
> that mxmlc handles things, right?
>
> Yes, I don't know how Falcon does it.
To be honest, I feel very uncomfortable knowing that MXML will be parsed by
the compiler together with the AS code for the number of reasons I won't
list here because they might b
> Are you saying the compiler doesn't determine what the change events for
> each token in the chain? That would definitely be "very far from
reality".
That is some different sort of parsing, it is not based strictly on AS3
grammar, it doesn't contribute to the parse tree generation in the end. I
> In {a.b.c} the compiler has to understand what a, b, and c resolve to,
what [Bindable] metadata there is on each one, etc. It uses the syntax tree
and the symbol table to do this.
This is very far from reality. What compiler does is actually like this:
< ... {the expression that appears in betw
>
>
> Is there a bug# for that?
>
> I think I did file a bug for it about 2-3 years ago, but I can't promise
I'll find it. In any way, it was mentioned in another bug that I filed on
binding constants - which was resoled as "won't fix" or "cannot fix".
Besides, it's really difficult to create a re
> I consider the current binding system to be consistent.
No, it's not. It is possible to get the same code compiled with all the
same settings and get different behaviors. This is exactly what consistency
is not.
What I would be looking for is that the non-binding attributes were the
defaults, th
I think I can explain the undefined order.
The attributes are stored in the hash-map-like structure, which is
unordered, so you cannot build on some attributes' values being assigned
before the others. In languages which allow you to control how bindings
are executed (for example, in Lisp or Mira
>
>
> swfxprinter.java is in the donation and converts ABC code to text.
> Something like it could be used to do analysis.
>
> I believe Martin meant something like a tool that mechanically proves that
the generated bytecode is not erroneous. For example, if you have a
conditional branch you want t
This is going to be a very long term planning, and I don't expect any
immediate resolution, but wanted to hear your opinion anyway.
Reading through the framework code related to it's "developer facing" side
(i.e. the classes like SystemManager, Applicaiton and the company) I found
that:
- classes
Yes, it was it, the addingChild - it wouldn't be called because nothing
there is a UIComponent, so when children were added the styles weren't
initalized. Got it working.
Thanks.
>
> I assume the call to getStyleDeclaration returned something?
>
> Yes, it returns CSSStyleDeclaration
>
> Next, I believe component gets assigned a moduleFactory in addingChild
> handler. As long as the value is the systemManager that is used for the
> StyleManagerImpl, it should look in th
> XXX_Styles.as
Yes, I copied the code from there and from FlexInit. Combined them and this
code runs before the application starts, just as if SystemManager called
init().
> moduleFactory assigned and then the styles for the component are computed.
As you can see from my code above, I had to do t
> Registering the CSSStyleDeclaration in a SystemManagerImpl
I think you meant StyleManagerImpl (style, not system), right? If so, how
do I tell if it's registered? It does too many things you could call
"register".
Below are things that "register" something:
registerInheritingStyle
registerSizeIn
Hey, no problem, thanks for trying :)
I think I managed it once some time ago with Flex 3-something, but since
styles and skins changed, I don't follow really what's happening inside the
box.
But if you (not only Sanford) have any pointers, not necessary solutions,
things to check or things to try
Ah, OK, cool. I just bumped into one class (StyleManagerImpl) which is
relatively new... and it had plenty of these.
But your comments show it could be much worse :) It's actually only 17. Not
sure this was ever different in AS2, but I guess AS2-ers were more
superstitious then.
~/flex/SDK/4.6/frameworks/projects/framework/src/mx$ find . -type f -name
"*.as" -print0 | xargs -0 -e grep -nH -e " == true" | wc -l
Guess without trying, how many? :)
Sorry, I just had to share this.
On a more serious note - is this in the codestyle guides yet?
Best.
Oleg
> Seems like the expectation was that additional compiler args to mxmlc
might someday influence rsl dependencies.
it's hard to believe that those related to metadata may ever influence that.
> The idea is you can substitute swcdepends for mxmlc
mxmlc has twice the list of options (probably even mo
Hi and thanks for the links, this is somewhat helpful, but it leaves a lot
of questions unanswered. These articles only cover some small subset of all
options available to this utility, below is the full set:
-benchmark
-compiler.accessible
-compiler.actionscript-file-encoding
-compiler.compress
Hi,
I'm trying to write some conscious description of utilities inside SDK. I
never used swcdepends before. There is *some* command line documentation,
but no explanation of what to do with the program, and I can't understand
from the options what the program does. Can anyone provide an insight to
thoughtfulness -> thoughtless (sorry, I typoed).
As I re-read now my replies, I see how they are a little hectic, and the
important thing probably skipped your attention. I will first try to
explain what is the problem with vectors when they are serialized /
deserialized, then I will "reflect" on the current serialization mechanism
in the flash p
. . .
But I'm using the project only to edit the code, I don't use FB to compile
anything / it's too slow / inconvenient. I rather use it because of
autocompletion and F
svn:ignore is a Subversion documented property, whatever sets it is using
Subversion to support it, so it's fine to do that from any client. This
setting also gets transferred to other checkouts, so if you ignore a file
on your end, and I happen to create a file by the same name on my end, it
will
Hi Jeffry,
I replied in another thread, but in a nutshell: I'm used to different
project structure, so I won't be using the project files you are creating /
they won't fit my setup anyway (sorry if this sounds like a threat, I don't
intend it :). However, it does no harm either, so far any special
How I set up the project:
1. I didn't merge the SDKs, instead I imported from both locations the
libraries I needed.
2. My eclipse workspace contains a lot of settings (mostly syntax
highlighting), that I don't want to loose, and it's in an entirely
different place from where I check the source cod
>
> I have no idea what the "Apache" way is regarding this. But, checking in
>> the Flash Builder /Eclipse project files will make it insanely easy for
>> anyone--who uses Flash Builder--to share the project. They just "import
>> projects from another workspace" and it works. It's much easier t
I had this error happen to me before many times when using Surefire reports
(with Maven builds). It used to locks files, but it can be just any other
program not closing pipes to open files.
So, can you try to move all staged files elsewhere, just for the sake of
making sure they can be manipulated
Are you sure the error is not on your end? All files staged are accessible
(read permissions, not locked, not in the read-only fs)?
Best.
Oleg
On Fri, Apr 6, 2012 at 2:33 AM, Abdul Sattar wrote:
> Is there an ALL IN ONE package in your suggestion or it will be
> discontinued?
>
> Abdul Sattar
> (Director IT & Operations)
> 0321-6433805
> www.powersoft.com.pk
>
In a nutshell, packages declare dependencies. So, for example, if you have
b
>
>
> While I'm looking at the ASdocs code any requests for simple changes?
>
> I remember someone on support forums asked that there be an option to
exclude creation date from HTML pages - the argument towards it was that
they generated documentation during each build and committed it. The date
ch
Well, the whole thing about how serialization in Flash works is moot... but
it's not anything that we could help, even if we wanted to very much.
Probably, it should be made clear that this is not just serialization into
some amorphous format, especially because JavaScript, if we will ever get
the
There are several undocumented AMF types, Vector. is one of them,
however it is only partially implemented (it's also known as "type 13"
because 13 eventually is the value of the tag produced in AMF).
It is only half-baked because it doesn't preserve the runtime type of the
vector.
There are some
Alex,
I will. Although, this thread started as general discussion of how many and
of what kind bundles to make. So this one I'm very insistent about is
really only one of the longer list. It just got concentrated on this one
option. I'd really wanted to hear opinions (or rather reasoning for why i
Dough,
There is a great deal of a difference between those who need framework and
those who need the SDK (those who need framework still need the SDK, but
those who need the SDK, don't necessarily need framework).
This is why, for example, code editors for AS have "pure AS" projects. The
number of
Besides, we skipped few other important things:
- Eclipse-compatible install.
- RPM / DEB package(s).
I'm not convinced abut OS-particular bundles. It is only the matter of our
build scripts and intention, what they package. It's not going to divide
project between teams or anything like that. Tod
> I can't guarantee such a thing will make it into a future release.
>
> Why would I ask it otherwise?.. why would I "donate" it, if it is not
needed?
Best.
Oleg
Jeffry.
Well, then, if we do it for ourselves, then I have no interest in
UIComponent and related stuff. So, if I'm doing it for myself, I will be
more explicit - I want that there will be a separate small download of only
the compiler and basic utilities needed to compile AS3. I will be happy to
Omar,
The intent is not to ask about Adobe Flex. The intent is to ask about what
features the users of AS3 / Flash may possibly need / expect from this
project. It's good to have our own idea about that, but we after all, don't
do it exclusively for ourselves. Besides, I don't see Flex suddenly bei
Omar, you are confusing the target users with the developers. You can't
poll the developers about what do they think the users of their software
think. I mean, ok, you can, but what's the point if you can ask the users
firsthand?
I just offered to run more polls. I know the leader of UFUG / RFUG a
>
> And does anyone know what percentage of developers use FlashDevelop? I, for
> one, have never even seen their home page, let alone used or downloaded it.
> Nor do I believe that any polls on any such boards that you didn't mention
> really matter either. All of these boards will have only a per
>
> This is all conjecture, unless you have personally polled every single Flex
> and AS3 developer in the world these statements shouldn't be used to try
> and justify decisions. Unless you can back them up with hard data it just
> makes it sound like you are trying really hard to convince someone
> The current volume of the documentation is mostly given due to
unoptimized html output.
Documentation is known to be the major volume of just any software. Well,
good documentation :) Read on on what is minimal install, so you'll see why
/ how the actual size can be reduced below that of document
What is core and what is components?
There are, as of today, several developers groups involved. One working on
AIR SDK, another one works on the Flex SDK. These are quite distinct
projects, they both use the compiler and related things, but other than
that very few things in common. Compiler is a
>
> I was able to build it using msysgit console on windows:)
>
> Well, it's not possible to make it a requirement for everyone else :) You
could actually use less then the whole Git console (I suspect it installs
quite a lot of stuff), you could only use bash + tar (if you alias bash as
sh). But i
https://issues.apache.org/jira/browse/FLEX-38
I've attached patch, which should (hopefully) solve the issue for Windows.
Best.
Oleg
> Getting:
http://fpdownload.macromedia.com/pub/flashplayer/updaters/11/flashplayer_11_sa_debug.i386.tar.gz
Hi, I just realized that the build downloads flashplayer standalone
debugger for Linux. I have two questions about it:
- Wouldn't it be better to make it a separate _installation_. Well, in h
Marik Magar:
find these lines:
#!/bin/sh
tar -xjf "AIR Integration Kit.tbz2"
In the build could you try changing it to something like
I might not have it all right b/c I don't know if gzip caters for bz
archives, but you get the idea. Worst case: remove these lines altogether,
extract the a
Oh, sorry, I actually found some related info:
http://en.wikipedia.org/wiki/Windows_Script_Host
> By using *Software Restriction Policies* introduced with Windows XP, a
system may be configured to execute only those scripts which have been
digitally signed, thus preventing the execution of untruste
At the same time, could we ask the original authors to provide some
documentation to these programs? Some of them are well documented (such as
the compilers), but swfdump only shows usage options w/o any explanation,
for example.
I'm asking this because I wanted to add some autocompletion / better
>
> The patch you supplied does fix this issue but the contrib lib is also
> required for Mustella. Perhaps it's best to wait until that donated and
> then see how the build scripts can be improved?
>
> But Mustella isn't included in the SDK yet, it's only needed for
committers (who want to test th
It usually helps to post the error message you are getting. This way it is
easier to troubleshoot the exact problem you have.
Best.
Oleg
Oh, come on, seriously... there are people doing (or wanting to do) a Maven
build - let them staff their build with gigabytes of nonsense, please,
don't do it in the basic setup... Remember an old man who liked to shave so
much he was known for his razor? What's redundant is wrong. This particular
> Also mustella requires ant-contrib. I think ant-contrib
> should just be documented as part of the required software (which sorry I
> missed since I've had ant installed for so long).
>
> Also the formatting in the patch you submitted is messed up. Each indent
> should be 4 spaces. It looks l
https://issues.apache.org/jira/browse/FLEX-35 I posted the patch. It seem
to work for me.
Best.
Oleg
While we are at it, what is the convention for property names in Ant
scripts? I can find these: CAPS_LOCK_IS_HERE, names.with.dots and
sTuRdYcApS, which one should I use if I need to add a new property?
Best.
Oleg
>
>
> Oleg, if you want to see something like this get done, it is best to do it
> yourself and submit the patch.
>
> I absolutely will. Was just asking whether this was intended.
Hi Justin, I understand what it's used for. I am asking that because I
think that simplicity of compilation for others > simplicity of writing for
the author.
Besides, I know a lot of people who would, out of principle disallow ant
contrib libraries. And I quite share their viewpoint. Compilation s
Was this conrib library inherited or added on purpose? I'd rather we loose
it. It's not required really.
Best.
Oleg
Aha... so, as far as I could understand, CFF is different from TTF in that
it uses cubic curves for outlines instead of quadratic. Is there anything
else that is different?
Would this also mean that fonts, originally in OTF (CFF) format can be
compiled right away w/o needing to do some [more] compl
Yes Martin, I understood the legal problem. I was just wondering what is
particular about the code that it cannot be replaced by other (modified)
font managers. As you probably know, there are 4 managers currently in SDK
which can encode fonts using somewhat different techniques, but only one
can e
Hello,
more out of curiosity: what is the main difficulty behind using FontSwf? Is
it the generation of the proper SWF tag or is it the generation of the
outlines or the kinds of fonts it supports, or is it the parsing the source
font files? I'm asking this because there is another OSS code that m
As part of this effort, I was looking into documentation for swfdump
utility and I couldn't find any documentation in the developer manual /
user manual. It provides only the list of acceptable options on the command
line, but I only know what half of them does. I couldn't understand what
the rest
Hi,
I didn't file this as a bug or f/r because I don't know what were the
reasons for these design decisions. I'm talking about a whole bunch of
compiler options for warnings which follow the form of:
-compiler.warn-you-did-this-and-this
I was wondering why wouldn't it be possible to make it into
1 - 100 of 213 matches
Mail list logo
