adobe-source-devel — ASL development topic discussion

Re: [Adobe-source-devel] Interest in ASL

[Sean P. <sp...@ad...>]

On May 7, 2009, at 10:42 AM, Robert Dailey wrote:

> Thanks for your detailed responses. I really appreciate the help.  
> I've provided some follow-up questions below:
>
> Correct - constant is immutable.  Input cells are set from outside -  
> input cells are used to initialize (or reinitialize) a sheet with  
> data from the application. A typical example is a command that  
> resizes a document might be seeded with the original document size  
> so you could constrain proportions. Input cells can be set using  
> sheet_t::set (either variant) <http://stlab.adobe.com/classadobe_1_1sheet__t.html#3cd494209eea1aa2b777c8da1585ba0d 
> >.
>
> So if I try to call sheet_t::get() on an Input Cell, I will get an  
> exception or some other type of failure? According to the  
> documentation, sheet_t::get() can be used on an Output Cell, but  
> this does not cause the logic of the output cell to be updated.  
> sheet_t::update() must be called before calling sheet_t:get() on an  
> Output Cell in order to get its most accurate and up-to-date value,  
> correct? Additionally, Logic Cells cannot have set() or get() called  
> on them, correct? Their scope is limited to the script file in  
> question?

sheet_t::get() was private until recently - it was made public as part  
of an effort to remove the dependency on the VM from the sheet_t. Soon  
you will be able to use sheet_t without the parser by providing  
function objects to define cells. Currently get() is called to do  
variable lookup by the VM during update(). If you call it after an  
update you will get the last cached value for the cell.

Generally, sheet_t is a push model - you monitor values, you don't get  
them.

If you try to call set() on an output, logic, or constant cell you  
will get an exception. If you try to monitor and input, logic, or  
constant cell you will get an exception.

>
> I'm just trying to make sure I have a grip on the concepts.
>
> Colon is used to denote an initializer. Normally an initializer is  
> only executed once (there is some support to reinitialize a sheet  
> that will cause some initializers to be evaluated again). The "<=="  
> is used to denote a defining expression. These are rules which are  
> evaluated anytime the sheet is updated.
>
> So if I have something like this:
>
> constant:
>     my_cell : 0 <== 4+5

Then you get a syntax error - constant and input cells can only have  
an initializer. It is only interface cells which play a duel role that  
can have both an initializer and a defining expression.
>
> Both a logic cell and an output cell have logical state (they are  
> like equation cells in a spreadsheet) so they have no initializer.  
> There value is always the value of the expression defining them. An  
> interface cell is the one cell type that has state _and_ a defining  
> expression (it can either be a contributing or derived depending on  
> the state of the sheet).
>        interface_cell : constant_cell <== input_cell;
>
> So if I do this:
>
> logic:
>     my_logic_cell : 0 <== 5+5
>
> This is malformed? Will this cause an error during parsing? What  
> would this behavior mean if it is not erroneous?

This is a syntax error.

> How are logic cells different from a regular cell delimited by a  
> colon? Furthermore, I am confused about 'logic' and 'output' as well  
> now.
>
> logic cells and ouput are just like equation cells in a spreadsheet  
> (where constant cells and input cells are like value cells in a  
> spreadsheet) - the major difference between an output cell and a  
> logic cell is that the output cell value can be monitored from  
> outside the sheet using sheet_t::monitor_value <http://stlab.adobe.com/classadobe_1_1sheet__t.html#48ec8465134b96deaee81c69be9084a2 
> >.
>
> But one can also use sheet_t::get() instead of monitoring the value,  
> correct? So since output cells can only be mutated by the expression  
> they are bound to, this means monitor_value() can only detect a  
> change when that "equation" is calculated with a different result.  
> Correct?

As mentioned above - you normally wouldn't use get() - you would  
monitor the cell. The monitor function is only called when the value  
of the cell is changed during an update.

>
> Keep the questions coming...
>
> Assuming I'm finally getting this, I think I've come a long way.  
> This is actually turning out to sound very intuitive! Thank you for  
> bearing with me through my questions. Let me know if I'm still  
> misunderstanding some things.
>
> My goal is to try to make an open source graphical tool for Adam &  
> Eve in the future. Right now it all seems to be hand-written, but it  
> would be nice to have a drag & drop approach to creating a GUI with  
> ASL. My first step, however, is to try to fully understand how to  
> use ASL in practice.

Nice - we have an internal tool for building these graphically -  
unfortunately we aren't able to open source it. I'm sure Foster can  
chime in with many suggestions. I've started working on yet another  
rewrite of sheet_t using the Boost Graph Library - this should make it  
fairly trivial to visualize a sheet_t with GraphViz which I think  
would really help with debugging sheets. I always draw them out by  
hand currently before coding them.

See slide 66 here for my set of "standard" symbols. <http://stlab.adobe.com/wiki/images/0/00/2008_04_03_accu.pdf 
 >

Sean

Re: [Adobe-source-devel] Interest in ASL

[Sean P. <sp...@ad...>]

On May 5, 2009, at 8:37 AM, Robert Dailey wrote:

> I also noticed that the CEL documentation does not discuss the '<=='  
> operator. Why?

Because they technically are not an operator.

<http://stlab.adobe.com/group__adam__reference.html>

Sean

Re: [Adobe-source-devel] Interest in ASL

[Sean P. <sp...@ad...>]

On May 4, 2009, at 9:56 PM, Robert Dailey wrote:

> Hi Sean,
>
> Sorry it's been a while, but I've just now got around to having time  
> to start learning ASL. I have a few questions about your last  
> response to me. I've interleaved my responses below.
>
> Also, the list doesn't seem very active at all. In the past few  
> months there has been a very limited amount of traffic. Is there a  
> more active place to post my questions?

This is a good list - I'm trying to encourage more traffic here. A lot  
of the use of ASL is internal at Adobe and there is a separate list  
for internal discussion - I'm cc'ing this message to that list. Non- 
confidential questions should go here.

>
> On Sat, Feb 14, 2009 at 3:55 AM, Sean Parent <sp...@ad...>  
> wrote:
>
> Absolutely - the terms are borrowed from spreadsheets. Cells are  
> named instead of having a coordinate system. Within a sheet, "value"  
> cells can be either constants or inputs (inputs are intended to be  
> set to from outside to initialize the sheet. A colon after the cell  
> name is followed by an expression to initialize the cell:
>
> sheet example {
>  constant:
>        constant_cell : "Initialized with a string";
>  input:
>        input_cell: 10; // initialized with a number
> }
>
> So far I think I get it. 'constant' here means that the variable  
> 'constant_cell' cannot be modified (it is immutable). This would be  
> like defining a variable as 'const' in C++, correct? And 'input' is  
> something that can only be set outside of the script? I'm not sure  
> how variables in Adam are set remotely or even internally other than  
> initializing them. Can you explain 'input' a little more? Perhaps  
> there is documentation for this already that I haven't found.

Correct - constant is immutable.  Input cells are set from outside -  
input cells are used to initialize (or reinitialize) a sheet with data  
from the application. A typical example is a command that resizes a  
document might be seeded with the original document size so you could  
constrain proportions. Input cells can be set using sheet_t::set  
(either variant) <http://stlab.adobe.com/classadobe_1_1sheet__t.html#3cd494209eea1aa2b777c8da1585ba0d 
 >.

>
> Corresponding to spreadsheet equation cells, you have logic cells  
> and output cells (output cells are logic cells that can be read  
> externally) - a defines operator "<==" after the cell name is  
> followed by the equation:
>
> sheet example {
>  constant:
>        constant_cell : "Initialized with a string";
>  input:
>        input_cell: 10; // initialized with a number
>  logic:
>        logic_cell <== input_cell + 10;
>  output:
>        output_cell <== { result: logic_cell }; // result is a  
> dictionary with the key/value pair "result" and the value of logic  
> cell
> }
>
> At this point I think I'm completely lost. So we use <== instead of  
> a colon here to designate a 'logic' cell?

Colon is used to denote an initializer. Normally an initializer is  
only executed once (there is some support to reinitialize a sheet that  
will cause some initializers to be evaluated again). The "<==" is used  
to denote a defining expression. These are rules which are evaluated  
anytime the sheet is updated.

Both a logic cell and an output cell have logical state (they are like  
equation cells in a spreadsheet) so they have no initializer. There  
value is always the value of the expression defining them. An  
interface cell is the one cell type that has state _and_ a defining  
expression (it can either be a contributing or derived depending on  
the state of the sheet).
	interface_cell : constant_cell <== input_cell;


> How are logic cells different from a regular cell delimited by a  
> colon? Furthermore, I am confused about 'logic' and 'output' as well  
> now.

logic cells and ouput are just like equation cells in a spreadsheet  
(where constant cells and input cells are like value cells in a  
spreadsheet) - the major difference between an output cell and a logic  
cell is that the output cell value can be monitored from outside the  
sheet using sheet_t::monitor_value <http://stlab.adobe.com/classadobe_1_1sheet__t.html#48ec8465134b96deaee81c69be9084a2 
 >.

>
> I can't really respond to any of the examples below this as they  
> just keep getting more and more confusing. I need to learn the  
> smaller examples you gave me first before I can understand the  
> harder ones.

Keep the questions coming...

[Adobe-source-devel] Replacing basic_sheet_t with sheet_t

[Sean P. <sp...@ad...>]

I've been doing a little ASL work on my sabbatical - todays project  
was replacing basic_sheet_t with sheet_t (the next step will be to  
extend the layout grammar to include more property model constructs  
(everything except input and output cells).

basic_sheet_t had become much more entrenched than I would have thought.

Everything seems to be building and running fine in my sandbox (all  
test cases and begin) - but if you were relying on any of the  
basic_sheet specific behaviors you may want to review the changes  
before I roll them to mainline:

http://stlab.adobe.com:8080/3879?ac=10

(If you didn't know - the web based diff tool is now working for our  
p4web client).

Sean

Re: [Adobe-source-devel] How to convert window's RC file to EVE

[erappy <er...@gm...>]

Hi Sean,Thanks for the reply.
>> Welcome! I do know of a couple of internal efforts at Adobe to write such
a tool, I don't believe they were particularly successful but >> I forwarded
your not to our internal list to see if anyone has something to share.
I am curious to know if there is any more useful information shared on the
other mailing lists.
Is there any major problem in solving this issue which caused the internal
efforts at Adobe less successful? I would like to understand the real
problems faced by the people tried to develop this kind of tool.
Because for my project with hundreds of RC files it is virtually impossible
to transform them manually into EVE layouts. It will definitely take few
man-months. I would rater try to find out or develop a tool which makes the
life easier.

>> I don't think it would take a lot to write a tool to at >>least transform
RC files into a list of widgets with the basic attributes (such as >> names)
carried over.
I agree that a list of widgets with the basic attributes by parsing the
dialog's controls can be done easily. A linked list of these resources can
be generated by parsing the Dialog's controls in the RC file. But the main
issue is the grouping of these controls into rows and columns. We should be
able to make a nested structure of controls wrapped into rows and columns
classes of EVE?

Looking forward for your help.

Thanks in Advance,
Apoorv


On Fri, Mar 6, 2009 at 12:16 AM, Sean Parent <sp...@ad...> wrote:

> Hi erappy,
>
> Welcome! I do know of a couple of internal efforts at Adobe to write such a
> tool, I don't believe they were particularly successful but I forwarded your
> not to our internal list to see if anyone has something to share. I don't
> think it would take a lot to write a tool to at least transform RC files
> into a list of widgets with the basic attributes (such as names) carried
> over.
>
> The best way that I know to proceed is to simply go through each dialog
> manually, to construct a layout I usually first rough in the basic structure
> without using any layout attributes (placement, alignment, spacing, margins,
> guide masks, etc) - the defaults usually do a reasonable job. Then you can
> add attributes to make the final tweaks.
>
> The Begin app makes some effort to reload changed files if you are using
> another editor (I think this works best with BBEdit on a Mac, but you might
> be able to get it working (and please contribute any improvements!) with
> your favorite editor - working in a good editor and having the view display
> every time you save can make the work go much faster.
>
> Hope that helps,
> Sean
>
>
> On Mar 5, 2009, at 12:45 AM, erappy wrote:
>
>  Hi Everyone,
>> I have recently started using ASL1.0.40, I have explored most of the
>> Adam-EVE examples and Begin app.
>> I am working on a typical C++ project on Win32 platform, my project has
>> hundreds of RC files having the layout information for dialogs used in the
>> various parts of the project.
>>
>> I am interested to know
>> [1] if there is any kind of automated tool available to convert RC files
>> to EVE files.
>> [2] In case there is no such tool available, what is the best way (or
>> steps) to proceed in order to convert RC files to EVE layouts for such a big
>> project.
>>
>> Thanks in Advance.
>> erappy
>> <ATT00001.txt><ATT00002.txt>
>>
>
>

Re: [Adobe-source-devel] How to convert window's RC file to EVE

[Sean P. <sp...@ad...>]

Hi erappy,

Welcome! I do know of a couple of internal efforts at Adobe to write  
such a tool, I don't believe they were particularly successful but I  
forwarded your not to our internal list to see if anyone has something  
to share. I don't think it would take a lot to write a tool to at  
least transform RC files into a list of widgets with the basic  
attributes (such as names) carried over.

The best way that I know to proceed is to simply go through each  
dialog manually, to construct a layout I usually first rough in the  
basic structure without using any layout attributes (placement,  
alignment, spacing, margins, guide masks, etc) - the defaults usually  
do a reasonable job. Then you can add attributes to make the final  
tweaks.

The Begin app makes some effort to reload changed files if you are  
using another editor (I think this works best with BBEdit on a Mac,  
but you might be able to get it working (and please contribute any  
improvements!) with your favorite editor - working in a good editor  
and having the view display every time you save can make the work go  
much faster.

Hope that helps,
Sean

On Mar 5, 2009, at 12:45 AM, erappy wrote:

> Hi Everyone,
> I have recently started using ASL1.0.40, I have explored most of the  
> Adam-EVE examples and Begin app.
> I am working on a typical C++ project on Win32 platform, my project  
> has hundreds of RC files having the layout information for dialogs  
> used in the various parts of the project.
>
> I am interested to know
> [1] if there is any kind of automated tool available to convert RC  
> files to EVE files.
> [2] In case there is no such tool available, what is the best way  
> (or steps) to proceed in order to convert RC files to EVE layouts  
> for such a big project.
>
> Thanks in Advance.
> erappy
> <ATT00001.txt><ATT00002.txt>

[Adobe-source-devel] How to convert window's RC file to EVE

[erappy <er...@gm...>]

Hi Everyone, I have recently started using ASL1.0.40, I have explored most
of the Adam-EVE examples and Begin app.
I am working on a typical C++ project on Win32 platform, my project has
hundreds of RC files having the layout information for dialogs used in the
various parts of the project.

I am interested to know
[1] if there is any kind of automated tool available to convert RC files to
EVE files.
[2] In case there is no such tool available, what is the best way (or steps)
to proceed in order to convert RC files to EVE layouts for such a big
project.

Thanks in Advance.
erappy

[Adobe-source-devel] [Boost] Range.Ex review feedback

[Thorsten O. <ne...@cs...>]

Dear developers,

The review of Neil Groves' RangeEx library starts today and lasts until 
March 3, 2009.

I'm writing to you to encourage one or more of you to participate in
the review. I know several people are world-class experts in this area, 
so I hope you can find the time to help make the Boost library even better.

best regards

Thorsten, RangeEx review manager

[Adobe-source-devel] ASL 1.0.41 Released

[Sean P. <sp...@ad...>]

ASL 1.0.41 is now released.
Highlights of this release include:

ASL:

Property model library (Adam) - rewrote main solver
no longer uses exceptions as flow control.
fixed bug with initializers that refer to cells with  
define_expressions has causing the define_expressions to evaluate  
which could refer to cells which had not been declared.
exported get() member, it is now the client responsibility to wire  
sheet_t::get() to the variable lookup in the VM (this also allows the  
clients to hook the variable lookup for supporting custom variables).  
In the future the VM will be separated entirely from the property  
model library.
added some more error reporting including reporting duplicate cell  
names.
virtual machine library
removed the contributing buts which were tracked through calculations.  
The VM stack is now just a stack of any_regular_t.
removed the use of storing size_t in an any_regular_t in the VM (was  
breaking serialization among other things because size_t is not a C++  
basic type on Win64).
any_regular_t
improved serialization by removing one heap alloc/dealloc per item  
serialized, also uses logN search for type instead of linear search.
removed the hack to handle size_t serialization on Win64.
serialization will now throw an exception when it encounters a type  
other than one of the 7 CEL types (it used to assert).
in debug builds, the version field of the any_regular_t vtable  
contains a short type name for debugging purposes only. See the wiki  
documentation.

Re: [Adobe-source-devel] Interest in ASL

[Sean P. <sp...@ad...>]

Hi Robert,
Welcome!

>
> I've been very interested in using ASL for an open source project of  
> mine, however I am having an extreme difficulty learning how to use  
> it. I have read as much documentation as possible on the ASL website  
> but it seems very incomplete to me.

It is incomplete - there is still much writing to do. Here are some  
suggestions to fill in the holes:

Check out the papers and presentations on the wiki:

<http://stlab.adobe.com/wiki/index.php/Papers_and_Presentations>

The ACCU talk goes into a fair amount of detail on both Adam and Eve,  
the Possible Future Talk has some discussion of Adam in the later  
half. The Property Model paper, though technical, is also a good  
source of background information.

Download the sources and the Begin application (binaries available for  
both Mac and Windows on Source Forge. There is a directory full of  
examples of both Adam and Eve files for Begin - most cover test cases  
and techniques but there is also an implementation of the Photoshop  
Image Size dialog as an example.

Also, if you haven't found the main TOC yet, it's here <http://stlab.adobe.com/modules.html 
 >

> The Eve grammar (in the scripting language) is very cryptic and  
> inconsistent to me and very hard to learn/follow. This page is all  
> I've been able to find in regards to documentation for the scripting  
> language. Is there any detailed documentation outlining syntax (What  
> each operator does, etc)? What exactly is the best approach to  
> learning Adam & Eve?

The best approach to learning is to start with the examples and ask  
lots of questions on the list. The documentation for the expression  
syntax (shared by both Adam and Eve) is here <http://stlab.adobe.com/group__expression__reference.html 
 >. A short summary is the basic syntax is very similar to C (except  
dynamically typed).
>
>
> The concepts "Sheets" and "Cells" are also very confusing. When I  
> hear these 2 words my first thought is a spreadsheet, and I have no  
> idea how they relate.

Absolutely - the terms are borrowed from spreadsheets. Cells are named  
instead of having a coordinate system. Within a sheet, "value" cells  
can be either constants or inputs (inputs are intended to be set to  
from outside to initialize the sheet. A colon after the cell name is  
followed by an expression to initialize the cell:

sheet example {
  constant:
	constant_cell : "Initialized with a string";
  input:
	input_cell: 10; // initialized with a number
}

Corresponding to spreadsheet equation cells, you have logic cells and  
output cells (output cells are logic cells that can be read  
externally) - a defines operator "<==" after the cell name is followed  
by the equation:

sheet example {
  constant:
	constant_cell : "Initialized with a string";
  input:
	input_cell: 10; // initialized with a number
  logic:
	logic_cell <== input_cell + 10;
  output:
	output_cell <== { result: logic_cell }; // result is a dictionary  
with the key/value pair "result" and the value of logic cell
}

Unlike a spreadsheet, an Adam sheet can have cells that behave as both  
input and output cells. These are known as "interface" cells and can  
be bound to externally, typically by a UI widget.

sheet example {
  constant:
	constant_cell : "Initialized with a string";
  input:
	input_cell: 10; // initialized with a number
  interface:
	interface_cell : input_cell;
  logic:
	logic_cell <== input_cell + 10;
  output:
	output_cell <== { result: logic_cell }; // result is a dictionary  
with the key/value pair "result" and the value of logic cell
}

By default, if you don't provide an defines operator on an interface  
cell, it is defined to be the identity - the above is equivalent to:

interface_cell : input_cell <== interface_cell;

Interface cells can participate in a relationship with one or more  
other cells using a relate clause - the idea with a relate clause is  
that for any state of the system, one of the defines holds:

sheet example {
  constant:
	constant_cell : "Initialized with a string";
  input:
	input_cell: 10; // initialized with a number
  interface:
	interface_cell : input_cell;
	another_interface_cell;
  logic:
	logic_cell <== input_cell + 10;

	relate {
		interface_cell <== another_interface_cell + 3;
		another_interface_cell <== interface_cell - 3;
	}

  output:
	output_cell <== { result: logic_cell }; // result is a dictionary  
with the key/value pair "result" and the value of logic cell
}


An interface cell can appear in more than one relate clause, and  
relates can have an optional predicate - for example if you want width  
and height to be the same when some predicate p is true it would be  
expressed as:

when (p) relate {
	width <== height;
	height <== width;
}

If you look at the Image Size example you can see how powerful the  
relate clause is.

That's my 2 minute tutorial - hopefully it helps.


> The documentation itself is very technical and isn't a good warmup  
> for beginners to the library. I need something "dummified" that  
> makes it easy to understand how ASL works (Analogies/Examples).

Play with the examples and with the Begin app and ask lots of  
questions - that's currently the best way to learn. Please also point  
out specific holes in the docs (I know there are _many_), knowing what  
people are struggling with helps me to prioritize the work.

> My head is spinning right now, so I hope someone can help point me  
> in the right direction.
>

I hope that help!

Sean

[Adobe-source-devel] Interest in ASL

[Robert D. <rcd...@gm...>]

I've been very interested in using ASL for an open source project of mine,
however I am having an extreme difficulty learning how to use it. I have
read as much documentation as possible on the ASL website but it seems very
incomplete to me. The Eve grammar (in the scripting language) is very
cryptic and inconsistent to me and very hard to learn/follow. This
page<http://stlab.adobe.com/group__eve__reference.html>is all I've
been able to find in regards to documentation for the scripting
language. Is there any detailed documentation outlining syntax (What each
operator does, etc)? What exactly is the best approach to learning Adam &
Eve?

The concepts "Sheets" and "Cells" are also very confusing. When I hear these
2 words my first thought is a spreadsheet, and I have no idea how they
relate. The documentation itself is very technical and isn't a good warmup
for beginners to the library. I need something "dummified" that makes it
easy to understand how ASL works (Analogies/Examples).

My head is spinning right now, so I hope someone can help point me in the
right direction.

[Adobe-source-devel] ASL 1.0.39 released

[Mat M. <mm...@ad...>]

Version 1.0.39 of the Adobe Source Libraries has been released. 

Highlights of this release include:

    ASL:
        - Added sheet_t::reinitialize(). "Input cells are
          re-initialized, in sheet order, and interface cell
          initializers are re-evaluated. Priorities are updated, but
          no callbacks are triggered. Calls to reinitialize should be
          followed by calls to update must be called, just as if the
          cells were updated by set operations on sorted ranges."
        - Adam documentaion improvements
        - Updated to (patched) boost 1.37.0
        - Added a command line property model evaluator tool, and
          programmer-oriented tutorial at:
          http://easyurl.net/pm_tutorial
        - New (theoretical) paper on Property Models from ACM GPCE
          2008 added to: http://easyurl.net/stlab_publications
        - Regressions now also run under Vista 64-bit
        - Regressions now also run under Mac OS X 64-bit
        - Deprecated pin library. Replaced with clamp library. Also
	  added median(of 3) and some select algorithms.

    User supported libs (APL): 
        - Formatter improvements.

Detailed change information can be found here:
http://stlab.adobe.com/asl_release_notes.html

Documentation to get started with the release is here:
http://stlab.adobe.com/asl_readme.html

Distribution files can be downloaded from here:
http://sourceforge.net/project/showfiles.php?group_id=132417

 - Mat

[Adobe-source-devel] ASL 1.0.38 released

[Mat M. <mm...@ad...>]

Version 1.0.37 of the Adobe Source Libraries has been released. 

Highlights of this release include:

    ASL:
        - New papers and presnetations at 
          http://stlab.adobe.com/wiki/index.php/Papers_and_Presentations
        - Major improvements to documentation structure
        - Fix 64-bit poly_test issue
        - Fixes for identity<> and pointer_to<> and their use.
        - Eliminated concept macros
        - Minor fix for static_table.hpp to include stdexcept
        - Fix for lexer to allow keywords as names, i.e. @empty.
        - Fix for MSVC compiler warning in memory.hpp (from Jon
           Clauson) 
        - Predicates for conditional relate declarations are now
          calculated up-front and cannot be dependent themselves on a
          relate declaration. 
        - Raising limit on number of interface cells to 256.
        - Added parameter to final to avoid misuse.
        - Added -L to curl commands in net setup scripts to protect
          against redirects
        - Made selection_t equality_comparable    
        - Added bool has_input(name) and bool has_output(name) to the
          property model library. 
        - removed outdated expression_filter routines; updated
          adam_tutorial output format; deleting some unnecessary
          sources 
        - Cleaned out static intializers
        - Collapsed entity lookup to a single place -- removed the
          once stuff from xml_parser
        - Fixed issue with implicit casts of enum node state in
          closed_hash.hpp
        - Fixed bug in property model library where an active monitor
          wasn't called on connect. 
        - Added code to layout library to reset metrics prior to
          calling client measure function to avoid errors where client
          code assumes defaulted value. 
        - Fixed bug with active monitor not being called on connection
          in Adam. 

    User supported libs (APL): 
        - Improvements to the expression, property model, and layout
          formatters. they print are much more readably now,
          especially with arrays and dictionaries with many items. 
        - Added shift as a modifier key to the new modifier key
          functionality 
        - Move controller and view concepts (compile and runtime) to
          APL 
        - BBedit language modules
        - xml_to_forest test app
        - Added xml_element_forest, a utility header that round-trips
          an XML file to a forest of dictionaries; also gives the user
          the ability to hand-generate a forest of dictionaries with a
          specific format and stream out a resulting XML file. 
        - Moved functions_as_controller to widget_proxies.hpp in APL
        - Renamed to_utf* to value_to_utf* in the singleton code point
        - Removed subview factory
        - Gave behaviors the ability to have their verbs pointed to
          another callback once they've been established. This gives
          clients the ability to do more than just resize the eve layout
          at the appropriate time (e.g., to resize a parent container
          not managed by eve.) 
        - sequence_model and associated view and controller concepts
          and implementations. Multiple changes and to improve the
          overall mvc architecture of the system. Also added a
          multiplexer/demultiplexer system to allow for the routing of
          commands between the sequence_model and any attached
          sequence views and controllers.
        - dictionary_set migration from obscure widgets lib location
          to APL 
        - Added breakpoint-compatible formatting to some of the
          casting type checks 
        - Logic fix for one of the selection algorithms
        - Fixed error reporting code in Begin.
        - Added formatter_tokens to eliminate duplicate symbols when
          both formatters are included in a single project 
        - Cleaned up a lot of the serialization code, pulled out
          dynamic memory allocations, etc 
        - Added expression_filter, for the escaping and unescaping of
          xml entities in a string; useful for serialization work 
        - Fixed some of the image_t leaks under win32. Problems run
          deeper though, since the architecture fails to store
          platform resources for cleanup when bitmaps updated or when
          window closes. Deeper fixes will be required to plug the
          rest of the leaks
        - Added the ability to add 'raw' functions to the function
          pack, bypassing all the function pack magic in case you'd like
          to do the work yourself. 
        - Update example code. inlined some function_pack routines,
          and added an nbitpath helper routine to the bitpath structure 

Detailed change information can be found here:
http://stlab.adobe.com/asl_release_notes.html

Documentation to get started with the release is here:
http://stlab.adobe.com/asl_readme.html

Distribution files can be downloaded from here:
http://sourceforge.net/project/showfiles.php?group_id=132417

 - Mat

[Adobe-source-devel] ASL 1.0.37 released

[Mat M. <mm...@ad...>]

Version 1.0.37 of the Adobe Source Libraries has been released. 

Highlights of this release include:

    ASL:
    - Update ASL to depend on boost 1.35.0
    - Patch boost 1.35.0 to fix bugs in boost threads and boost
    filesystem
    - Patch boost build to add a link-incompatible feature
    allowing _SECURE_SCL to be disabled, thus avoiding
    unjustifiable performance penalty on release builds.
    - Multiple net setup and build improvements.
    - Workaround the bjam MkDir bug that was causing build errors
    and the need for multiple invocations of bjam.
    - Tested against gcc 4.3.0.
    - Update ASL to depend on Intel Thread Building Blocks
    2008.04.08 stable release, now supporting atomics under ppc 32bit.
    - Patch TBB to support windows/cygwin-based gcc 4.1.1 and
    later.
    - Fix bug on 64 bit little-endian machines in closed_hash
    - Documentation improvements and corrections for any_regular_t
    and some algorithms
    - Fixed a potential memory leak in any_regular_t
    - Fixed thread safety issue with forest<>::size()
    - New binary_search (unlike std::binary_search, this is useful)
    - Added gather algorithm (thanks to Marshall Clow)
    - Started to break-up functional -
    functional/operator.hpp
    functional/is_member.hpp
    - Deleted find_first_of_set.hpp - use find_if with is_member
    instead
    - Change sheet and basic sheet API to remove
    attach_view/detach_view in favor of signals
    mechanism. Remove function_as_view since function/bind
    results can once again be used directly. poly_view is now
    optional for clients who find it more convenient.
    - Change adam API to remove
    attach_controller/detach_controller in favor of signals
    mechanism. poly_controller is now optional for clients who
    find it convenient.
    - Remove dependence of adam_tutorial on APL and move to
    ASL. Add visual project for adam_tutorial.
    - Rewrite copy_on_write to use a well-defined binary
    representation, add custom allocator support, move to
    version1 namespace.
    - Added script to update a directory in p4 based on "offline"
    changes, and other scripts to automate documentation update
    process. 
    - Fixes to vector and string for windows min/max problems,
    thanks to Jesper.
    - Add check ensuring operator== interoperability of forest
    iters and const iters.

    APL: 
    - Add property_model_formatter.hpp and layout_formatter.hpp
    for parsing layout and property model descriptions into a
    data structure instead of to the widget factory.
    - Add layout_tidy and property_model_tidy which check layout
    and property model definitions validity.


Change list information can be found here:
http://stlab.adobe.com/asl_release_notes.html

Documentation to get started with the release is here:
http://stlab.adobe.com/asl_readme.html

Distribution files can be downloaded from here:
http://sourceforge.net/project/showfiles.php?group_id=132417

 - Mat

Re: [Adobe-source-devel] Adobe Source Libraries 1.0.36 Released (corrected: not 1.0.46)

[Daniel J. <dan...@fm...>]

I think you have to include the full path with 'glob':

  [ glob ./examples/* : ./examples/.svn ]

In 'glob-tree' the patterns match any file in the current tree, in
'glob' they're relative to the current directory.

P.S. Sorry Mat for sending you this off-list just now. I'm used to the
boost lists which set 'reply-to'.

On 12/04/2008, Mat Marcus <mm...@ad...> wrote:
> Ok, sorry to hear that didn't work. From what I could gather about how
>  globs worked it seemed like it might. See for example
>  http://easyurl.net/globbing. In any case, I made the change that you
>  requested. It will ship as part of 1.0.37.
>
>
>  "Jacob Foshee" <ja...@gm...> writes:
>
>  > Sorry, I tried that change and it did not work.
>  >
>  >     error: Unable to find file or target named
>  >     error: '/C:/.../adobe_root/adobe_platform_libraries/test/begin/examples/.svn'
>  >     error: referred from project at
>  >     error: 'test/begin'
>  >
>  > On 3/27/08, Mat Marcus <mm...@ad...> wrote:
>  >
>  >     "Jacob Foshee" <ja...@gm...> writes:
>  >
>  >     Let me know whether [ glob ./examples/* : .svn ] performs the correct
>  >     .svn exclusion for you. If so I'll see about updating the
>  >     Jamfiles. Thanks for the suggestion.
>  >
>  >       - Mat

Re: [Adobe-source-devel] Adobe Source Libraries 1.0.36 Released (corrected: not 1.0.46)

[Mat M. <mm...@ad...>]

I spent a little time rummaging around in the boost build manual for
this, but didn't find an easy change to make this possible. As a
semi-manual approach, you could try:

Change the value of the variable $(TOP) in
adobe_source_libraries/project-root.jam and in
adobe_platform_libraries/project_root.jam to a directory to which
you'd like built_artifacts to be a peer. Also, when building, you can
invoke bjam with --build-directory=/some/path to alter the location of
the boost artifacts. 

I don't have more cycles to spend on the build system for a while, but
if you can figure out a way toset things up so that --build-directory
also affects the ASL artifact location, while keeping the default
location where it is now, I'd be happy to update ASL accordingly.

 - Mat



"Jacob Foshee" <ja...@gm...> writes:

> Great, yes, that did the trick!
>
> So, my next step is to incorporate this into our build.  In general we try to keep build artifacts out of the source tree.  Is there a way to
> specify a different destination for built_artifacts?  If not, I can make do w/out it, but might I make this a feature request?

Re: [Adobe-source-devel] Adobe Source Libraries 1.0.36 Released (corrected: not 1.0.46)

[Mat M. <mm...@ad...>]

FYI I spent a number of cycles improving the net setup scripts for
1.0.37. I've replaced untgz with a version of gnuwin32 tar and gzip
that seem to be able to handle the intel distribution. And I've worked
around the bjam MkDir bugs so that one should only have to build once
now to create an exectuable Adobe Begin.

 - Mat



"Jacob Foshee" <ja...@gm...> writes:

> Great, yes, that did the trick!
>
> So, my next step is to incorporate this into our build.  In general we try to keep build artifacts out of the source tree.  Is there a way to
> specify a different destination for built_artifacts?  If not, I can make do w/out it, but might I make this a feature request?
>
> Thanks, again. 
> I'll let you know as I have more questions.
> Jacob
>
> On Wed, Mar 26, 2008 at 9:41 PM, Mat Marcus <mm...@ad...> wrote:
>
>     "Jacob Foshee" <ja...@gm...> writes:
>    
>     Hi Jacob,
>    
>     Ok, this is good. You're almost there with your manual build. The
>     basic release  configuration and test suite is built by invoking:
>    
>      bjam release
>    
>     Now, in order to run, begin.exe requires certain other files to be
>     present in its directory. To execute the "install" step to make an
>     double clickable version of begin that won't immediately exit, try the
>     following from the asl_command_prompt:
>    
>      bjam test/begin//install_program
>    
>     This should result in an "installed" begin directory at:
>    
>      adobe_root/built_artifacts/install/NT/debug/Adobe Begin/
>    
>     You should be able to run the begin.exe in that directory, testing it
>     upon the examples in the Examples directory contained therein.
>    
>     Or, if you would prefer a release version:
>    
>      bjam release test/begin//install_program
>    
>     Or you could try to build both versions using 4 cores with:
>    
>       bjam -j4 debug release test/begin//install_program
>    
>     I will try to replace the buggy untgz with a separate tar/gzip soon to
>     restore the functionality of the automated build process. Thanks again
>     for your interest and efforts. It will be good to know that you can
>     successfully build and execute from the asl_command_prompt in the end.
>    
>      - Mat
>
>     > Hey, Mat,
>     > I tried a fresh download and followed your instructions.  Here's the workflow I arrived at:
>     > 1. Download and run asl_net_setup
>     > 2. untgz crashes, and click through it ("Don't Send")
>     > 3. asl_net_setup finishes  (Nothing built, TBB only partially extracted)
>     > 4. run asl_command_prompt
>     > 5. bjam (almost everything builds)
>     > 6. bjam (remaining ~150 targets build)
>     > However, only debug targets were built for ASL.  None of the release targets were built.  (How can I specify release
>     > configuration?)
>     >
>     > So, the conclusion is that something in the environment is different in the asl_net_setup script than with the
>     > asl_command_prompt.  I can try your suggested bjam parameters to see if I can isolate it.  (Also when I was using
>     > asl_net_setup both debug and release targets built)
>     >
>     > Finally, begin.exe still does not work.  I guess this is a separate issue.  Is this a known issue?
>     >
>     > thanks,
>     > jacob
>     >
>     > On Wed, Mar 26, 2008 at 12:40 PM, Mat Marcus <mm...@ad...> wrote:
>     >
>     >     Strange that we are getting different results. While it is true that
>     >     you could try turning off warnings-as-errors, that wouldn't address
>     >     the root problem. The provided jamfiles turn on or relax this setting
>     >     as needed when building different libraries. In particular, when I
>     >     build using the asl_net_setup scripts, I don't seem to get the error
>     >     that you get. This means that something deeper is wrong. However, if
>     >     you want to experiment, you can edit the line:
>     >
>     >       <toolset>msvc-8.0:<cxxflags>"/W3 /WX /Wp64 /Zc:wchar_t
>     >       /Zc:forScope"
>     >
>     >     in adobe_root/adobe_source_libraries/Jamfile.v2 to remove the /WX.
>     >
>     >     Here are the next ideas I can think of for getting to the bottom of
>     >     things. First, lets switch from using the full net setup script to a
>     >     faster/more usual way of building. Steps:
>     >
>     >     1) open asl_command_prompt from adobe_platform_libraries/tools
>     >
>     >     2) the (non-debug) build command to enter at the prompt there is
>     >     simply: bjam
>     >
>     >     Assuming that you are able to build in that manner, I recommend using
>     >     that method for the remainder of our discussion. The idea is to only
>     >     build, rather than trying to re-fetch boost each time, etc. In fact,
>     >     this will give you the opportunity to manually unpack the tbb 014
>     >     commercially aligned release and put it under adobe_root.
>     >
>     >     Now, if you get the same error we can add options to the bjam command
>     >     to get a better look at what is going on. In particular, the command
>     >     bjam -q -d+2 will tell us what actual commands were passed to the
>     >     compiler. It will also stop after the first error. I can then compare
>     >     the commands that are being passed to your compiler with those that
>     >     are being passed to mine.
>     >
>     >     Thanks for your patience,
>     >     Mat
>     >
>     >     "Jacob Foshee" <ja...@gm...> writes:
>     >
>     >     > The intel_tbb_libraries folder is getting moved whether untgz crashes or not.
>     >     > So my adobe_root contains the following:
>     >     > adobe_platform_libraries
>     >     > adobe_source_libraries
>     >     > boost_libraries
>     >     > built_artifacts
>     >     > intel_tbb_libraries
>     >     >
>     >     > After untgz and move, intel_tbb_libraries looks like this:
>     >     > +---examples
>     >     > |   +---concurrent_hash_map
>     >     > |   |   \---count_strings
>     >     > |   |       \---xcode
>     >     > |   |           \---count_strings.xcodeproj
>     >     > |   \---parallel_for
>     >     > |       +---seismic
>     >     > |       |   +---vc7.1
>     >     > |       |   +---vc8
>     >     > |       |   \---xcode
>     >     > |       |       +---English.lproj
>     >     > |       |       |   \---main.nib
>     >     > |       |       \---SeismicSimulation.xcodeproj
>     >     > |       \---tacheon
>     >     > |           +---dat
>     >     > |           +---src
>     >     > |           +---vc7.1
>     >     > |           +---vc8
>     >     > |           \---xcode
>     >     > |               +---English.lproj
>     >     > |               |   \---main.nib
>     >     > |               \---tacheon.xcodeproj
>     >     > \---include
>     >     >     \---tbb
>     >     >         \---machine
>     >     > So, it's definitely missing some files.
>     >     >
>     >     > I just tried another build, manually decompressing and moving TBB, but it failed as well.  I still suspect the
>     >     problem is all the warnings (which are
>     >     > being treated as errors) in intel_tbb_libraries\include\tbb\atomic.h.  For example:
>     >     > ..\intel_tbb_libraries\include\tbb/atomic.h(158) : warning C4244: 'argument' : conversion from
>     >     'tbb::internal::atomic_traits<Size,M>::word' to 'int',
>     >     > possible loss of data
>     >     >
>     >     > Thanks again for devoting time to this.
>     >     > Jacob
>     >     >
>     >     > On Tue, Mar 25, 2008 at 4:47 PM, Mat Marcus <mm...@ad...> wrote:
>     >     >
>     >     >     "Jacob Foshee" <ja...@gm...> writes:
>     >     >
>     >     >     Ok, thanks for the info. Based on the information you gave, I have a
>     >     >     theory as to what might be going on. If the untgz tool is crashing, it
>     >     >     is possible that the directory "intel_tbb_libraries" is not getting
>     >     >     moved to live inside the adobe_root directory. That is, the tool
>     >     >     failure may be resulting in the following directory structure:
>     >     >
>     >     >     C:\Development\3rdParty\Adobe\asl_1.0.36\
>     >     >       ...
>     >     >       gnuwin32
>     >     >       adobe_root
>     >     >          adobe_platform_libraries
>     >     >          adobe_source_libraries
>     >     >          boost_libraries
>     >     >          built_artifacts
>     >     >       intel_tbb_libraries
>     >     >       ...
>     >     >
>     >     >     When in fact it should be
>     >     >       ...
>     >     >       gnuwin32
>     >     >       adobe_root
>     >     >          adobe_platform_libraries
>     >     >          adobe_source_libraries
>     >     >          boost_libraries
>     >     >          built_artifacts
>     >     >          intel_tbb_libraries
>     >     >       ...
>     >     >
>     >     >     For the build to succeed, all of the directories whose name ends in
>     >     >     _libraries are supposed to be peers under adobe_root. If this is not
>     >     >     the case, try moving intel_tbb_libraries down into the adobe_root
>     >     >     directory and then execute the build script once more. If this works,
>     >     >     then I will know what happened and I will try to look for an
>     >     >     alternative to untgz for the next ASL release. Thanks for your
>     >     >     patience as we get to the bottom of this.
>     >     >
>     >     >      - Mat
>     >     >
>     >     >     > 1) asl_1.0.36_net_setup.bat
>     >     >     > 2) Something like: C:\Development\3rdParty\Adobe\asl_1.0.36\
>     >     >     > 3) I have an unpatched version of boost on my machine installed at C:\Boost.  I did have that version of
>     >     bjam
>     >     >     > on my path, but I removed it (to no avail).  asl net_setup is downloading and patching its own copy of
>     >     boost.
>     >     >     > It says:
>     >     >     >
>     >     >     >     ...found 5031 targets...
>     >     >     >     ...updating 671 targets...
>     >     >     >
>     >     >     > I am hoping the 5031 targets it found are not those already in my C:\Boost directory.
>     >     >     > 4) cmd
>     >     >     >
>     >     >     > Thanks!
>     >     >     > Jacob
>     >     >     >
>     >     >     > On Tue, Mar 25, 2008 at 3:27 PM, Mat Marcus <mm...@ad...> wrote:
>     >     >     >
>     >     >     >     "Jacob Foshee" <ja...@gm...> writes:
>     >     >     >     Hi Jacob,
>     >     >     >
>     >     >     >     What I'm looking for is:
>     >     >     >
>     >     >     >     1) what command line invocation did you use to initiate the build?
>     >     >     >     2) which directory did you invoke it from?
>     >     >     >     3) are you running with a patched or unpatched version of boost?
>     >     >     >     4) are your running under cmd or cygwin?
>     >     >     >
>     >     >     >     If I knew the answer to these questions I might be able to reason
>     >     >     >     further about whether there's a bug in the script, or how to help you
>     >     >     >     get things going on your machine. If everything is in place, another
>     >     >     >     experiment to try is to open the asl_command_prompt shell (found in
>     >     >     >     adobe_platform_libraries/tools) and invoke the command:  bjam
>     >     >     >
>     >     >     >     Sorry for the difficulties that you are experiencing,
>     >     >     >      - Mat
>     >     >     >
>     >     >     >     > Hi, Mat,
>     >     >     >     > Thanks for the response.
>     >     >     >     >
>     >     >     >     > I did continue building after the UNTGZ errors.  I also tried tried using a different tool to get a
>     >     >     >     complete
>     >     >     >     > decompression.  And I even tried d/loading the latest Intel TBB (which seemed to have the same
>     >     problem).
>     >     >     >      If I recall
>     >     >     >     > correctly, all the results were the same.
>     >     >     >     >
>     >     >     >     > I have a lot of .obj files, but no exe's (e.g. no begin.exe).  The only libs I have are from boost
>     >     >     >     (filesystem, signals,
>     >     >     >     > thread, and unit_test_framework).
>     >     >     >     >
>     >     >     >     > The final output of my build is as follows:
>     >     >     >     >
>     >     >     >     >     ...failed compile-c-c++ ..\built_artifacts\test\begin\msvc-8.0\release\asl-dev\atomics-tbb\
>     >     >     >     link-static\threading-multi\
>     >     >     >     >     user-interface-gui\main.obj...
>     >     >     >     >     ...skipped <p..\built_artifacts\test\begin\msvc-8.0\release\asl-dev\atomics-tbb\link-static\
>     >     >     >     threading-multi\
>     >     >     >     >     user-interface-gui>begin.exe for lack of <p..\built_artifacts\test\begin\msvc-8.0\release\
>     >     asl-dev\
>     >     >     >     atomics-tbb\
>     >     >     >     >     link-static\threading-multi\user
>     >     >     >     >     -interface-gui>express_viewer.obj...
>     >     >     >     >     ...skipped <p..\built_artifacts\test\selection_ops\main.test\msvc-8.0\release\asl-dev\
>     >     atomics-tbb\
>     >     >     >     link-static\
>     >     >     >     >     threading-multi>main.exe for lack of <pC:\Development\3rdParty\Adobe\asl_1.0.36_c\adobe_root\
>     >     >     >     built_artifacts\msvc-8.0\
>     >     >     >     >     release\asl-dev\a
>     >     >     >     >     tomics-tbb\link-static\threading-multi>libasl_dev.lib...
>     >     >     >     >     ...skipped <p..\built_artifacts\test\selection_ops\main.test\msvc-8.0\release\asl-dev\
>     >     atomics-tbb\
>     >     >     >     link-static\
>     >     >     >     >     threading-multi>main.run for lack of <p..\built_artifacts\test\selection_ops\main.test\msvc-8.0
>     >     \
>     >     >     >     release\asl-dev\
>     >     >     >     >     atomics-tbb\link-static\thre
>     >     >     >     >     ading-multi>main.exe...
>     >     >     >     >     ...failed updating 11 targets...
>     >     >     >     >     ...skipped 98 targets...
>     >     >     >     >     ...updated 454 targets...
>     >     >     >     >     ### Error : 1
>     >     >     >     >
>     >     >     >     > If I had to guess, I would suspect the source of the problem is treating warnings as errors w/
>     >     Intel's
>     >     >     >     TBB.  That is the
>     >     >     >     > first and most frequent error I see.  Is there an easy way to turn that off?
>     >     >     >     >
>     >     >     >     > thanks again,
>     >     >     >     > Jacob
>     >     >     >     >
>     >     >     >     > On Mon, Mar 24, 2008 at 7:39 PM, Mat Marcus <mm...@ad...> wrote:
>     >     >     >     >
>     >     >     >     >     Hi Jacob,
>     >     >     >     >
>     >     >     >     >     Sorry to hear that you're having troubles with the net-setup script. I
>     >     >     >     >     too encountered some crashes with UNTGZ. I kept hitting ok until the
>     >     >     >     >     crashes stopped, then eventually the net-setup script
>     >     >     >     >     succeeded. Unfortunately, I haven't been able to find an alternative
>     >     >     >     >     public tool for windows that can unpack the intel tbb tgz
>     >     >     >     >     files. Perhaps its stumbling on the long suffix ".xcodeproj".
>     >     >     >     >
>     >     >     >     >     It's not clear to me how you continued the build after the untgz
>     >     >     >     >     crashes. The net setup script actually executes the build command
>     >     >     >     >     multiple times to workaround apparent bjam bugs (mkdir issues). So it
>     >     >     >     >     is not unusual for some intermediate errors to show up, but in the end
>     >     >     >     >     the 1.0.36 build succeeded here. Is the same true for you if you "ok"
>     >     >     >     >     through the tgz crashes? If not, please give a few more details about
>     >     >     >     >     exactly how you are building if you would like further assistance.
>     >     >     >     >
>     >     >     >     >      - Mat
>     >     >     >     >
>     >     >     >     >     "Jacob Foshee" <ja...@gm...> writes:
>     >     >     >     >
>     >     >     >     >     > Hi,
>     >     >     >     >     > I just tried to d/load 1.0.36.  UNTGZ crashes while expanding Intel TBB.  Just
>     >     >     >     >     > curious if you had any ideas?  (I'm on WinXP SP2 w/ VC 8.0)
>     >     >     >     >     >
>     >     >     >     >     > I know this is not your archive, but, fwiw, it happens to crash after:
>     >     >     >     >     >
>     >     >     >     >     >     Creating directory: tbb20_014oss_src/examples/concurrent_hash_map/
>     >     >     >     >     >     count_strings/xcode/count_strings.xcodeproj OK
>     >     >     >     >     >
>     >     >     >     >     > I expanded the archive using another tool, and that seemed to work fine.
>     >     >     >     >     > However, the build breaks at this point:
>     >     >     >     >     >
>     >     >     >     >     >     compile-c-c++ C:\Development\3rdParty\Adobe\asl_1.0.36\adobe_root\
>     >     >     >     >     >     built_artifacts\msvc-8.0\debug\asl-dev\atomics-tbb\link-static\
>     >     >     >     >     >     threading-multi\
>     >     >     >     >     >     xstring.cpp
>     >     >     >     >     >     ..\intel_tbb_libraries\include\tbb/atomic.h(158) : error C2220: warning
>     >     >     >     >     >     treated as error - no 'object' file generated
>     >     >     >     >     >             ..\intel_tbb_libraries\include\tbb/atomic.h(158) : while compiling
>     >     >     >     >     >     class template member function 'tbb::internal::atomic_traits<Size,M>::word
>     >     >     >     >     >     Size,M>::word,tbb::internal::atomic_traits<Size,M>::word)'
>     >     >     >     >     >             with
>     >     >     >     >     >             [
>     >     >     >     >     >                 Size=4,
>     >     >     >     >     >                 M=__TBB_full_fence
>     >     >     >     >     >             ]
>     >     >     >     >     >             ..\intel_tbb_libraries\include\tbb/atomic.h(344) : see reference
>     >     >     >     >     >     to class template instantiation 'tbb::internal::atomic_traits<Size,M>'
>     >     >     >     >     >     being
>     >     >     >     >     >             with
>     >     >     >     >     >             [
>     >     >     >     >     >                 Size=4,
>     >     >     >     >     >                 M=__TBB_full_fence
>     >     >     >     >     >             ]
>     >     >     >     >     >             ..\intel_tbb_libraries\include\tbb/atomic.h(348) : see reference
>     >     >     >     >     >     to function template instantiation 'tbb::atomic<void*>::value_type
>     >     >     >     >     >     tbb::atomi
>     >     >     >     >     >     e_type)' being compiled
>     >     >     >     >     >     ..\intel_tbb_libraries\include\tbb/atomic.h(158) : warning C4244:
>     >     >     >     >     >     'argument' : conversion from 'tbb::internal::atomic_traits<Size,M>::word'
>     >     >     >     >     >     to 'int',
>     >     >     >     >     >             with
>     >     >     >     >     >             [
>     >     >     >     >     >                 Size=4,
>     >     >     >     >     >                 M=__TBB_full_fence
>     >     >     >     >     >             ]
>     >     >     >     >     >     ..\intel_tbb_libraries\include\tbb/atomic.h(158) : warning C4244:
>     >     >     >     >     >     'argument' : conversion from 'tbb::internal::atomic_traits<Size,M>::word'
>     >     >     >     >     >     to 'int',
>     >     >     >     >     >             with
>     >     >     >     >     >             [
>     >     >     >     >     >                 Size=4,
>     >     >     >     >     >                 M=__TBB_full_fence
>     >     >     >     >     >             ]
>     >     >     >     >     >     ..\intel_tbb_libraries\include\tbb/atomic.h(158) : warning C4244:
>     >     >     >     >     >     'argument' : conversion from 'tbb::internal::atomic_traits<Size,M>::word'
>     >     >     >     >     >     to 'int',
>     >     >     >     >     >             with
>     >     >     >     >     >             [
>     >     >     >     >     >                 Size=4,
>     >     >     >     >     >                 M=__TBB_full_fence
>     >     >     >     >     >             ]
>     >     >     >     >     >             ..\intel_tbb_libraries\include\tbb/atomic.h(158) : while compiling
>     >     >     >     >     >     class template member function 'tbb::internal::atomic_traits<Size,M>::word
>     >     >     >     >     >     e,M>::word)'
>     >     >     >     >     >             with
>     >     >     >     >     >             [
>     >     >     >     >     >                 Size=4,
>     >     >     >     >     >                 M=__TBB_full_fence
>     >     >     >     >     >             ]
>     >     >     >     >     >     ..\intel_tbb_libraries\include\tbb/atomic.h(158) : warning C4244:
>     >     >     >     >     >     'argument' : conversion from 'tbb::internal::atomic_traits<Size,M>::word'
>     >     >     >     >     >     to 'int',
>     >     >     >     >     >             with
>     >     >     >     >     >             [
>     >     >     >     >     >                 Size=4,
>     >     >     >     >     >                 M=__TBB_full_fence
>     >     >     >     >     >             ]
>     >     >     >     >     >             ..\intel_tbb_libraries\include\tbb/atomic.h(158) : while compiling
>     >     >     >     >     >     class template member function 'tbb::internal::atomic_traits<Size,M>::word
>     >     >     >     >     >     ize,M>::word)'
>     >     >     >     >     >             with
>     >     >     >     >     >             [
>     >     >     >     >     >                 Size=4,
>     >     >     >     >     >                 M=__TBB_full_fence
>     >     >     >     >     >             ]
>     >     >     >     >     >
>     >     >     >     >     > Please let me know if you have any ideas, or if I can provide more
>     >     >     >     >     > information.
>     >     >     >     >     > Thanks,
>     >     >     >     >     > Jacob
>     >     >     >     >     >
>     >     >     >     >     > On Fri, Mar 21, 2008 at 8:06 PM, Mat Marcus <mm...@ad...> wrote:
>     >     >     >     >     >
>     >     >     >     >     >     Corrected subject line: 1.0.36 has been released, not 1.0.46.
>     >     >     >     >     >
>     >     >     >     >     >     Mat Marcus <mm...@ad...> writes:
>     >     >     >     >     >
>     >     >     >     >     >     > Version 1.0.36 of the Adobe Source Libraries has been released.
>     >     >     >     >     >     > Highlights of this release include: ASL:
>     >     >     >     >     >     >
>     >     >     >     >     >     >     * version_1 types (replace version_0 types) - better, defined binary
>     >     >     >     >     >     layout and carry allocator for applications with replaced new/delete.
>     >     >     >     >     >     >     * capture_allocator, an std::allocator compatible allocator which
>     >     >     >     >     >     carries a paired new/delete.
>     >     >     >     >     >     >     * Allocator support in closed_hash_set<>, closed_hash_map<>, and
>     >     >     >     >     >     vector<>.
>     >     >     >     >     >     >     * Vastly improved move library (Thanks to Dave Abrahams!).
>     >     >     >     >     >     >     * Vastly improved copy-on-write library (now uses move library to
>     >     >     >     >     >     reduce number of increment/decrements).
>     >     >     >     >     >     >     * Bug fix in layout library that caused containers with the width
>     >     >     >     >     >     driven by the frame width to layout children incorrectly.
>     >     >     >     >     >     >     * Bug fix in forest::erase where the beginning of the erased range
>     >     >     >     >     >     has a lower depth then the end of the range.
>     >     >     >     >     >     >     * Forest test cases and minor fixes to forest types.
>     >     >     >     >     >     >     * Misc bug fixes.
>     >     >     >     >     >     >     * Fixes to build and distribution tools.
>     >     >     >     >     >     >     * Fixed search on documentation site to point to stlab and not
>     >     >     >     >     >     >     * opensource domain.
>     >     >     >     >     >     >
>     >     >     >     >     >     > Change list information can be found here:
>     >     >     >     >     >     > http://stlab.adobe.com/asl_release_notes.html
>     >     >     >     >     >     >
>     >     >     >     >     >     > Documentation to get started with the release is here:
>     >     >     >     >     >     > http://stlab.adobe.com/asl_readme.html
>     >     >     >     >     >     >
>     >     >     >     >     >     > Distribution files can be downloaded from here:
>     >     >     >     >     >     > http://sourceforge.net/project/showfiles.php?group_id=132417
>     >     >     >     >     >     >
>     >     >     >     >     >     >  - Mat
>     >     >     >     >     >
>     >     >     >     >     >     -------------------------------------------------------------------------
>     >     >     >     >     >     This SF.net email is sponsored by: Microsoft
>     >     >     >     >     >     Defy all challenges. Microsoft(R) Visual Studio 2008.
>     >     >     >     >     >     http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>     >     >     >     >     >     _______________________________________________
>     >     >     >     >     >     Adobe-source-devel mailing list
>     >     >     >     >     >     Ado...@li...
>     >     >     >     >     >     https://lists.sourceforge.net/lists/listinfo/adobe-source-devel

Re: [Adobe-source-devel] Adobe Source Libraries 1.0.36 Released (corrected: not 1.0.46)

[Mat M. <mm...@ad...>]

Ok, sorry to hear that didn't work. From what I could gather about how
globs worked it seemed like it might. See for example
http://easyurl.net/globbing. In any case, I made the change that you
requested. It will ship as part of 1.0.37.

"Jacob Foshee" <ja...@gm...> writes:

> Sorry, I tried that change and it did not work.
>
>     error: Unable to find file or target named
>     error: '/C:/.../adobe_root/adobe_platform_libraries/test/begin/examples/.svn'
>     error: referred from project at
>     error: 'test/begin'
>
> On 3/27/08, Mat Marcus <mm...@ad...> wrote:
>
>     "Jacob Foshee" <ja...@gm...> writes:
>
>     Let me know whether [ glob ./examples/* : .svn ] performs the correct
>     .svn exclusion for you. If so I'll see about updating the
>     Jamfiles. Thanks for the suggestion.
>
>       - Mat
>
>     P.S. This list is fine for these kinds of discussion. Thanks for your
>     interest.
>
>     > Hopefully one final issue about the build:
>     > I added ASL to our build and our source repository.  We use subversion.  Consequently every folder has hidden .svn folders.
>     >  This broke a glob statement in the install block of Begin's Jamfile.  My workaround is to change each line like the
>     > following:
>     >
>     >     : [ glob ./examples/* ]
>     >
>     > to:
>     >
>     >     : [ glob ./examples/*.adm ./examples/*.eve ]
>     >
>     > If there is an exclude mechanism for glob, we could use that instead, but I couldn't find one.
>     > I hope you will consider a change like this.
>     >
>     > btw, if this list is not for recommending source code changes, please let me know the proper avenue for that.
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace_______________________________________________
> Adobe-source-devel mailing list
> Ado...@li...
> https://lists.sourceforge.net/lists/listinfo/adobe-source-devel

Re: [Adobe-source-devel] Move documentation

[Sean P. <sp...@ad...>]

I read through you doc changes while traveling - all good. I've  
updated my local copy - I'll submit as soon as I have a chance.

Thanks!
Sean

On Apr 1, 2008, at 7:25 PM, Mat Marcus wrote:
> Hi Daniel,
>
> Great, thanks for your efforts. One of the ASL authors is traveling to
> a conference. We'll review your suggested in a week or so. Sorry for
> the delay.
>
> - Mat
>
>
> "Daniel James" <dan...@fm...> writes:
>
>> Hi,
>>
>> David Abrahams suggested 'boostifying' the adobe move library during
>> the Boost.Unordered review - so I'm currently working on doing that.
>> Other than changing the namespaces, I'm planning to supply a fallback
>> for compilers without SFINAE and add support for Boost.Thread's move
>> semantics. And possibly add some support for objects that aren't
>> regular.
>>
>> I've attached a patch for a few small changes to the documentation.
>> Hopefully, it should be self-explanatory. Does it look okay?
>>
>> thanks,
>>
>> Daniel
>>
>> --- adobe/move.hpp.orig	2008-03-31 00:52:50.000000000 +0100
>> +++ adobe/move.hpp	2008-03-31 01:22:30.000000000 +0100
>> @@ -42,7 +42,7 @@
>>
>> \par Implementing a Movable Type
>>
>> -A movable type models \ref concept_movable. There are four  
>> components of a movable type:
>> +A movable type models \ref concept_movable. There are three  
>> components of a movable type:
>> 	- Satisfy the requirements of concept \ref concept_regular_type.
>> 	- Implement a move-ctor using move_from<>.
>> 	- Modify the assignment operator to take the operand by value and  
>> consume it.
>> @@ -70,7 +70,7 @@
>>
>> struct implementation : boost::equality_comparable<implementation>
>> {
>> -    implementation(int x = 0) : member(x) { }
>> +    explicit implementation(int x = 0) : member(x) { }
>>
>> implementation(const implementation& x) : member(x.member)
>> { std::cout << "copy remote part: " << member << std::endl; }
>> @@ -93,7 +93,7 @@
>> public:
>> // model concept Regular
>>
>> -    movable(int x = 0) : member(new implementation(x)) { }
>> +    explicit movable(int x = 0) : member(new implementation(x)) { }
>> ~movable() { delete member; }
>> movable(const movable& x) : member(new implementation(*x.member)) { }
>> // operator=() implemented below
>> @@ -107,8 +107,8 @@
>> // model concept Movable
>>
>> // move-ctor assumes ownership of remote part
>> -    movable(move_from<movable> x) : member(x.source.member)
>> -    { x.member = 0; }
>> +    movable(adobe::move_from<movable> x) : member(x.source.member)
>> +    { x.source.member = 0; }
>>
>> // operator=() on a movable type takes parameter by value and  
>> consumes it
>> movable& operator=(movable x)
>> @@ -212,13 +212,12 @@
>>
>> <table>
>> <tr><td><b>Move</b></td><td><b>Copy</b></td><td><b>Comment</b></ 
>> td></tr>
>> -<tr><td>adobe::move()<td><td>std::copy</td><td>Not to be confused  
>> with the single argument adobe::move()</td></tr>
>> -<tr><td>adobe::move_backward()<td><td>std::copy_backward</td></tr>
>> -<tr><td>adobe::move_backward()<td><td>std::copy_backward</td></tr>
>> - 
>> < 
>> tr 
>> ><td>adobe::back_move_iterator()<td><td>std::back_insert_iterator</ 
>> td></tr>
>> -<tr><td>adobe::back_mover()<td><td>std::back_inserter</td></tr>
>> -<tr><td>adobe::move_construct()<td><td>std::construct</td></tr>
>> - 
>> <tr><td>adobe::uninitialized_move()<td><td>std::uninitialized_copy</ 
>> td></tr>
>> +<tr><td>adobe::move()</td><td>std::copy</td><td>Not to be confused  
>> with the single argument adobe::move()</td></tr>
>> +<tr><td>adobe::move_backward()</td><td>std::copy_backward</td></tr>
>> +<tr><td>adobe::back_move_iterator()</ 
>> td><td>std::back_insert_iterator</td></tr>
>> +<tr><td>adobe::back_mover()</td><td>std::back_inserter</td></tr>
>> +<tr><td>adobe::move_construct()</td><td>std::construct</td></tr>
>> +<tr><td>adobe::uninitialized_move()</ 
>> td><td>std::uninitialized_copy</td></tr>
>> </table>
>>
>> \par Advanced Topics
>> -------------------------------------------------------------------------
>> Check out the new SourceForge.net Marketplace.
>> It's the best place to buy or sell services for
>> just about anything Open Source.
>> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace_______________________________________________
>> Adobe-source-devel mailing list
>> Ado...@li...
>> https://lists.sourceforge.net/lists/listinfo/adobe-source-devel
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
> _______________________________________________
> Adobe-source-devel mailing list
> Ado...@li...
> https://lists.sourceforge.net/lists/listinfo/adobe-source-devel

Re: [Adobe-source-devel] Move documentation

[Mat M. <mm...@ad...>]

Forwarded message from Sean Parent:

---

I haven't had a chance to take a look at the patch yet (will soon) -

One more suggestion - if you can accurately detect swap() [there is
code from Dave for has_swap() in our implementation directory - I
don't recall the status of it.] then move() could fall back to
swap() - which is likely more efficient then copy if provided.
Alternatively it would be good to overload move for the std
containers and std::string in terms of swap and specialize
is_movable accordingly (there is an issue with that in that some
functions currently key off is_movable to directly invoke the
move_ctor, but we can detect the move_ctor and so could fix this
code.

Sean

---

"Daniel James" <dan...@fm...> writes:

> Hi,
>
> David Abrahams suggested 'boostifying' the adobe move library during
> the Boost.Unordered review - so I'm currently working on doing that.
> Other than changing the namespaces, I'm planning to supply a fallback
> for compilers without SFINAE and add support for Boost.Thread's move
> semantics. And possibly add some support for objects that aren't
> regular.
>
> I've attached a patch for a few small changes to the documentation.
> Hopefully, it should be self-explanatory. Does it look okay?
>
> thanks,
>
> Daniel
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace_______________________________________________
> Adobe-source-devel mailing list
> Ado...@li...
> https://lists.sourceforge.net/lists/listinfo/adobe-source-devel

Re: [Adobe-source-devel] Move documentation

[Mat M. <mm...@ad...>]

Hi Daniel,

Great, thanks for your efforts. One of the ASL authors is traveling to
a conference. We'll review your suggested in a week or so. Sorry for
the delay.

 - Mat


"Daniel James" <dan...@fm...> writes:

> Hi,
>
> David Abrahams suggested 'boostifying' the adobe move library during
> the Boost.Unordered review - so I'm currently working on doing that.
> Other than changing the namespaces, I'm planning to supply a fallback
> for compilers without SFINAE and add support for Boost.Thread's move
> semantics. And possibly add some support for objects that aren't
> regular.
>
> I've attached a patch for a few small changes to the documentation.
> Hopefully, it should be self-explanatory. Does it look okay?
>
> thanks,
>
> Daniel
>
> --- adobe/move.hpp.orig	2008-03-31 00:52:50.000000000 +0100
> +++ adobe/move.hpp	2008-03-31 01:22:30.000000000 +0100
> @@ -42,7 +42,7 @@
>  
>  \par Implementing a Movable Type
>  
> -A movable type models \ref concept_movable. There are four components of a movable type:
> +A movable type models \ref concept_movable. There are three components of a movable type:
>  	- Satisfy the requirements of concept \ref concept_regular_type.
>  	- Implement a move-ctor using move_from<>.
>  	- Modify the assignment operator to take the operand by value and consume it.
> @@ -70,7 +70,7 @@
>  
>  struct implementation : boost::equality_comparable<implementation>
>  {
> -    implementation(int x = 0) : member(x) { }
> +    explicit implementation(int x = 0) : member(x) { }
>      
>      implementation(const implementation& x) : member(x.member)
>      { std::cout << "copy remote part: " << member << std::endl; }
> @@ -93,7 +93,7 @@
>   public:
>  // model concept Regular
>  
> -    movable(int x = 0) : member(new implementation(x)) { }
> +    explicit movable(int x = 0) : member(new implementation(x)) { }
>      ~movable() { delete member; }
>      movable(const movable& x) : member(new implementation(*x.member)) { }
>      // operator=() implemented below
> @@ -107,8 +107,8 @@
>  // model concept Movable
>      
>      // move-ctor assumes ownership of remote part
> -    movable(move_from<movable> x) : member(x.source.member)
> -    { x.member = 0; }
> +    movable(adobe::move_from<movable> x) : member(x.source.member)
> +    { x.source.member = 0; }
>      
>      // operator=() on a movable type takes parameter by value and consumes it
>      movable& operator=(movable x)
> @@ -212,13 +212,12 @@
>  
>  <table>
>  <tr><td><b>Move</b></td><td><b>Copy</b></td><td><b>Comment</b></td></tr>
> -<tr><td>adobe::move()<td><td>std::copy</td><td>Not to be confused with the single argument adobe::move()</td></tr>
> -<tr><td>adobe::move_backward()<td><td>std::copy_backward</td></tr>
> -<tr><td>adobe::move_backward()<td><td>std::copy_backward</td></tr>
> -<tr><td>adobe::back_move_iterator()<td><td>std::back_insert_iterator</td></tr>
> -<tr><td>adobe::back_mover()<td><td>std::back_inserter</td></tr>
> -<tr><td>adobe::move_construct()<td><td>std::construct</td></tr>
> -<tr><td>adobe::uninitialized_move()<td><td>std::uninitialized_copy</td></tr>
> +<tr><td>adobe::move()</td><td>std::copy</td><td>Not to be confused with the single argument adobe::move()</td></tr>
> +<tr><td>adobe::move_backward()</td><td>std::copy_backward</td></tr>
> +<tr><td>adobe::back_move_iterator()</td><td>std::back_insert_iterator</td></tr>
> +<tr><td>adobe::back_mover()</td><td>std::back_inserter</td></tr>
> +<tr><td>adobe::move_construct()</td><td>std::construct</td></tr>
> +<tr><td>adobe::uninitialized_move()</td><td>std::uninitialized_copy</td></tr>
>  </table>
>  
>  \par Advanced Topics
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace_______________________________________________
> Adobe-source-devel mailing list
> Ado...@li...
> https://lists.sourceforge.net/lists/listinfo/adobe-source-devel

Re: [Adobe-source-devel] Adobe Source Libraries 1.0.36 Released (corrected: not 1.0.46)

[Jacob F. <ja...@gm...>]

Sorry, I tried that change and it did not work.

error: Unable to find file or target named
error:
'/C:/.../adobe_root/adobe_platform_libraries/test/begin/examples/.svn'
error: referred from project at
error: 'test/begin'


On 3/27/08, Mat Marcus <mm...@ad...> wrote:
>
> "Jacob Foshee" <ja...@gm...> writes:
>
>
> Let me know whether [ glob ./examples/* : .svn ] performs the correct
> .svn exclusion for you. If so I'll see about updating the
> Jamfiles. Thanks for the suggestion.
>
>
>   - Mat
>
>
> P.S. This list is fine for these kinds of discussion. Thanks for your
> interest.
>
>
> > Hopefully one final issue about the build:
> > I added ASL to our build and our source repository.  We use
> subversion.  Consequently every folder has hidden .svn folders.
> >  This broke a glob statement in the install block of Begin's
> Jamfile.  My workaround is to change each line like the
> > following:
> >
> >     : [ glob ./examples/* ]
> >
> > to:
> >
> >     : [ glob ./examples/*.adm ./examples/*.eve ]
> >
> > If there is an exclude mechanism for glob, we could use that instead,
> but I couldn't find one.
> > I hope you will consider a change like this.
> >
> > btw, if this list is not for recommending source code changes, please
> let me know the proper avenue for that.
>

[Adobe-source-devel] Move documentation

[Daniel J. <dan...@fm...>]

Hi,

David Abrahams suggested 'boostifying' the adobe move library during
the Boost.Unordered review - so I'm currently working on doing that.
Other than changing the namespaces, I'm planning to supply a fallback
for compilers without SFINAE and add support for Boost.Thread's move
semantics. And possibly add some support for objects that aren't
regular.

I've attached a patch for a few small changes to the documentation.
Hopefully, it should be self-explanatory. Does it look okay?

thanks,

Daniel

Re: [Adobe-source-devel] Adobe Source Libraries 1.0.36 Released (corrected: not 1.0.46)

[Mat M. <mm...@ad...>]

"Jacob Foshee" <ja...@gm...> writes:

Let me know whether [ glob ./examples/* : .svn ] performs the correct
.svn exclusion for you. If so I'll see about updating the
Jamfiles. Thanks for the suggestion.

 - Mat

P.S. This list is fine for these kinds of discussion. Thanks for your
interest. 

> Hopefully one final issue about the build:
> I added ASL to our build and our source repository.  We use subversion.  Consequently every folder has hidden .svn folders.
>  This broke a glob statement in the install block of Begin's Jamfile.  My workaround is to change each line like the
> following:
>
>     : [ glob ./examples/* ]
>
> to:
>
>     : [ glob ./examples/*.adm ./examples/*.eve ]
>
> If there is an exclude mechanism for glob, we could use that instead, but I couldn't find one.
> I hope you will consider a change like this.  
>
> btw, if this list is not for recommending source code changes, please let me know the proper avenue for that.

Re: [Adobe-source-devel] Adobe Source Libraries 1.0.36 Released (corrected: not 1.0.46)

[Jacob F. <ja...@gm...>]

Hopefully one final issue about the build:
I added ASL to our build and our source repository.  We use subversion.
 Consequently every folder has hidden .svn folders.  This broke a glob
statement in the install block of Begin's Jamfile.  My workaround is to
change each line like the following:

: [ glob ./examples/* ]

to:

: [ glob ./examples/*.adm ./examples/*.eve ]

If there is an exclude mechanism for glob, we could use that instead, but I
couldn't find one.
I hope you will consider a change like this.

btw, if this list is not for recommending source code changes, please let me
know the proper avenue for that.

Thanks,
Jacob

On 3/27/08, Jacob Foshee <ja...@gm...> wrote:
>
> Great, yes, that did the trick!
>
> So, my next step is to incorporate this into our build.  In general we try
> to keep build artifacts out of the source tree.  Is there a way to specify a
> different destination for built_artifacts?  If not, I can make do w/out it,
> but might I make this a feature request?
>
> Thanks, again.
> I'll let you know as I have more questions.
> Jacob
>
> On Wed, Mar 26, 2008 at 9:41 PM, Mat Marcus <mm...@ad...> wrote:
>
> > "Jacob Foshee" <ja...@gm...> writes:
> >
> > Hi Jacob,
> >
> > Ok, this is good. You're almost there with your manual build. The
> > basic release  configuration and test suite is built by invoking:
> >
> >  bjam release
> >
> > Now, in order to run, begin.exe requires certain other files to be
> > present in its directory. To execute the "install" step to make an
> > double clickable version of begin that won't immediately exit, try the
> > following from the asl_command_prompt:
> >
> >  bjam test/begin//install_program
> >
> > This should result in an "installed" begin directory at:
> >
> >  adobe_root/built_artifacts/install/NT/debug/Adobe Begin/
> >
> > You should be able to run the begin.exe in that directory, testing it
> > upon the examples in the Examples directory contained therein.
> >
> > Or, if you would prefer a release version:
> >
> >  bjam release test/begin//install_program
> >
> > Or you could try to build both versions using 4 cores with:
> >
> >   bjam -j4 debug release test/begin//install_program
> >
> > I will try to replace the buggy untgz with a separate tar/gzip soon to
> > restore the functionality of the automated build process. Thanks again
> > for your interest and efforts. It will be good to know that you can
> > successfully build and execute from the asl_command_prompt in the end.
> >
> >  - Mat
> >
> >
> >
> > > Hey, Mat,
> > > I tried a fresh download and followed your instructions.  Here's the
> > workflow I arrived at:
> > > 1. Download and run asl_net_setup
> > > 2. untgz crashes, and click through it ("Don't Send")
> > > 3. asl_net_setup finishes  (Nothing built, TBB only partially
> > extracted)
> > > 4. run asl_command_prompt
> > > 5. bjam (almost everything builds)
> > > 6. bjam (remaining ~150 targets build)
> > > However, only debug targets were built for ASL.  None of the release
> > targets were built.  (How can I specify release
> > > configuration?)
> > >
> > > So, the conclusion is that something in the environment is different
> > in the asl_net_setup script than with the
> > > asl_command_prompt.  I can try your suggested bjam parameters to see
> > if I can isolate it.  (Also when I was using
> > > asl_net_setup both debug and release targets built)
> > >
> > > Finally, begin.exe still does not work.  I guess this is a separate
> > issue.  Is this a known issue?
> > >
> > > thanks,
> > > jacob
> > >
> > > On Wed, Mar 26, 2008 at 12:40 PM, Mat Marcus <mm...@ad...>
> > wrote:
> > >
> > >     Strange that we are getting different results. While it is true
> > that
> > >     you could try turning off warnings-as-errors, that wouldn't
> > address
> > >     the root problem. The provided jamfiles turn on or relax this
> > setting
> > >     as needed when building different libraries. In particular, when I
> > >     build using the asl_net_setup scripts, I don't seem to get the
> > error
> > >     that you get. This means that something deeper is wrong. However,
> > if
> > >     you want to experiment, you can edit the line:
> > >
> > >       <toolset>msvc-8.0:<cxxflags>"/W3 /WX /Wp64 /Zc:wchar_t
> > >       /Zc:forScope"
> > >
> > >     in adobe_root/adobe_source_libraries/Jamfile.v2 to remove the /WX.
> > >
> > >     Here are the next ideas I can think of for getting to the bottom
> > of
> > >     things. First, lets switch from using the full net setup script to
> > a
> > >     faster/more usual way of building. Steps:
> > >
> > >     1) open asl_command_prompt from adobe_platform_libraries/tools
> > >
> > >     2) the (non-debug) build command to enter at the prompt there is
> > >     simply: bjam
> > >
> > >     Assuming that you are able to build in that manner, I recommend
> > using
> > >     that method for the remainder of our discussion. The idea is to
> > only
> > >     build, rather than trying to re-fetch boost each time, etc. In
> > fact,
> > >     this will give you the opportunity to manually unpack the tbb 014
> > >     commercially aligned release and put it under adobe_root.
> > >
> > >     Now, if you get the same error we can add options to the bjam
> > command
> > >     to get a better look at what is going on. In particular, the
> > command
> > >     bjam -q -d+2 will tell us what actual commands were passed to the
> > >     compiler. It will also stop after the first error. I can then
> > compare
> > >     the commands that are being passed to your compiler with those
> > that
> > >     are being passed to mine.
> > >
> > >     Thanks for your patience,
> > >     Mat
> > >
> > >     "Jacob Foshee" <ja...@gm...> writes:
> > >
> > >     > The intel_tbb_libraries folder is getting moved whether untgz
> > crashes or not.
> > >     > So my adobe_root contains the following:
> > >     > adobe_platform_libraries
> > >     > adobe_source_libraries
> > >     > boost_libraries
> > >     > built_artifacts
> > >     > intel_tbb_libraries
> > >     >
> > >     > After untgz and move, intel_tbb_libraries looks like this:
> > >     > +---examples
> > >     > |   +---concurrent_hash_map
> > >     > |   |   \---count_strings
> > >     > |   |       \---xcode
> > >     > |   |           \---count_strings.xcodeproj
> > >     > |   \---parallel_for
> > >     > |       +---seismic
> > >     > |       |   +---vc7.1
> > >     > |       |   +---vc8
> > >     > |       |   \---xcode
> > >     > |       |       +---English.lproj
> > >     > |       |       |   \---main.nib
> > >     > |       |       \---SeismicSimulation.xcodeproj
> > >     > |       \---tacheon
> > >     > |           +---dat
> > >     > |           +---src
> > >     > |           +---vc7.1
> > >     > |           +---vc8
> > >     > |           \---xcode
> > >     > |               +---English.lproj
> > >     > |               |   \---main.nib
> > >     > |               \---tacheon.xcodeproj
> > >     > \---include
> > >     >     \---tbb
> > >     >         \---machine
> > >     > So, it's definitely missing some files.
> > >     >
> > >     > I just tried another build, manually decompressing and moving
> > TBB, but it failed as well.  I still suspect the
> > >     problem is all the warnings (which are
> > >     > being treated as errors) in
> > intel_tbb_libraries\include\tbb\atomic.h.  For example:
> > >     > ..\intel_tbb_libraries\include\tbb/atomic.h(158) : warning
> > C4244: 'argument' : conversion from
> > >     'tbb::internal::atomic_traits<Size,M>::word' to 'int',
> > >     > possible loss of data
> > >     >
> > >     > Thanks again for devoting time to this.
> > >     > Jacob
> > >     >
> > >     > On Tue, Mar 25, 2008 at 4:47 PM, Mat Marcus <mm...@ad...>
> > wrote:
> > >     >
> > >     >     "Jacob Foshee" <ja...@gm...> writes:
> > >     >
> > >     >     Ok, thanks for the info. Based on the information you gave,
> > I have a
> > >     >     theory as to what might be going on. If the untgz tool is
> > crashing, it
> > >     >     is possible that the directory "intel_tbb_libraries" is not
> > getting
> > >     >     moved to live inside the adobe_root directory. That is, the
> > tool
> > >     >     failure may be resulting in the following directory
> > structure:
> > >     >
> > >     >     C:\Development\3rdParty\Adobe\asl_1.0.36\
> > >     >       ...
> > >     >       gnuwin32
> > >     >       adobe_root
> > >     >          adobe_platform_libraries
> > >     >          adobe_source_libraries
> > >     >          boost_libraries
> > >     >          built_artifacts
> > >     >       intel_tbb_libraries
> > >     >       ...
> > >     >
> > >     >     When in fact it should be
> > >     >       ...
> > >     >       gnuwin32
> > >     >       adobe_root
> > >     >          adobe_platform_libraries
> > >     >          adobe_source_libraries
> > >     >          boost_libraries
> > >     >          built_artifacts
> > >     >          intel_tbb_libraries
> > >     >       ...
> > >     >
> > >     >     For the build to succeed, all of the directories whose name
> > ends in
> > >     >     _libraries are supposed to be peers under adobe_root. If
> > this is not
> > >     >     the case, try moving intel_tbb_libraries down into the
> > adobe_root
> > >     >     directory and then execute the build script once more. If
> > this works,
> > >     >     then I will know what happened and I will try to look for an
> > >     >     alternative to untgz for the next ASL release. Thanks for
> > your
> > >     >     patience as we get to the bottom of this.
> > >     >
> > >     >      - Mat
> > >     >
> > >     >     > 1) asl_1.0.36_net_setup.bat
> > >     >     > 2) Something like:
> > C:\Development\3rdParty\Adobe\asl_1.0.36\
> > >     >     > 3) I have an unpatched version of boost on my machine
> > installed at C:\Boost.  I did have that version of
> > >     bjam
> > >     >     > on my path, but I removed it (to no avail).  asl net_setup
> > is downloading and patching its own copy of
> > >     boost.
> > >     >     > It says:
> > >     >     >
> > >     >     >     ...found 5031 targets...
> > >     >     >     ...updating 671 targets...
> > >     >     >
> > >     >     > I am hoping the 5031 targets it found are not those
> > already in my C:\Boost directory.
> > >     >     > 4) cmd
> > >     >     >
> > >     >     > Thanks!
> > >     >     > Jacob
> > >     >     >
> > >     >     > On Tue, Mar 25, 2008 at 3:27 PM, Mat Marcus <
> > mm...@ad...> wrote:
> > >     >     >
> > >     >     >     "Jacob Foshee" <ja...@gm...> writes:
> > >     >     >     Hi Jacob,
> > >     >     >
> > >     >     >     What I'm looking for is:
> > >     >     >
> > >     >     >     1) what command line invocation did you use to
> > initiate the build?
> > >     >     >     2) which directory did you invoke it from?
> > >     >     >     3) are you running with a patched or unpatched version
> > of boost?
> > >     >     >     4) are your running under cmd or cygwin?
> > >     >     >
> > >     >     >     If I knew the answer to these questions I might be
> > able to reason
> > >     >     >     further about whether there's a bug in the script, or
> > how to help you
> > >     >     >     get things going on your machine. If everything is in
> > place, another
> > >     >     >     experiment to try is to open the asl_command_prompt
> > shell (found in
> > >     >     >     adobe_platform_libraries/tools) and invoke the
> > command:  bjam
> > >     >     >
> > >     >     >     Sorry for the difficulties that you are experiencing,
> > >     >     >      - Mat
> > >     >     >
> > >     >     >     > Hi, Mat,
> > >     >     >     > Thanks for the response.
> > >     >     >     >
> > >     >     >     > I did continue building after the UNTGZ errors.  I
> > also tried tried using a different tool to get a
> > >     >     >     complete
> > >     >     >     > decompression.  And I even tried d/loading the
> > latest Intel TBB (which seemed to have the same
> > >     problem).
> > >     >     >      If I recall
> > >     >     >     > correctly, all the results were the same.
> > >     >     >     >
> > >     >     >     > I have a lot of .obj files, but no exe's (e.g. no
> > begin.exe).  The only libs I have are from boost
> > >     >     >     (filesystem, signals,
> > >     >     >     > thread, and unit_test_framework).
> > >     >     >     >
> > >     >     >     > The final output of my build is as follows:
> > >     >     >     >
> > >     >     >     >     ...failed compile-c-c++
> > ..\built_artifacts\test\begin\msvc-8.0\release\asl-dev\atomics-tbb\
> > >     >     >     link-static\threading-multi\
> > >     >     >     >     user-interface-gui\main.obj...
> > >     >     >     >     ...skipped <p..\built_artifacts\test\begin\msvc-
> > 8.0\release\asl-dev\atomics-tbb\link-static\
> > >     >     >     threading-multi\
> > >     >     >     >     user-interface-gui>begin.exe for lack of
> > <p..\built_artifacts\test\begin\msvc-8.0\release\
> > >     asl-dev\
> > >     >     >     atomics-tbb\
> > >     >     >     >     link-static\threading-multi\user
> > >     >     >     >     -interface-gui>express_viewer.obj...
> > >     >     >     >     ...skipped
> > <p..\built_artifacts\test\selection_ops\main.test\msvc-
> > 8.0\release\asl-dev\
> > >     atomics-tbb\
> > >     >     >     link-static\
> > >     >     >     >     threading-multi>main.exe for lack of
> > <pC:\Development\3rdParty\Adobe\asl_1.0.36_c\adobe_root\
> > >     >     >     built_artifacts\msvc-8.0\
> > >     >     >     >     release\asl-dev\a
> > >     >     >     >
> > tomics-tbb\link-static\threading-multi>libasl_dev.lib...
> > >     >     >     >     ...skipped
> > <p..\built_artifacts\test\selection_ops\main.test\msvc-
> > 8.0\release\asl-dev\
> > >     atomics-tbb\
> > >     >     >     link-static\
> > >     >     >     >     threading-multi>main.run for lack of
> > <p..\built_artifacts\test\selection_ops\main.test\msvc-8.0
> > >     \
> > >     >     >     release\asl-dev\
> > >     >     >     >     atomics-tbb\link-static\thre
> > >     >     >     >     ading-multi>main.exe...
> > >     >     >     >     ...failed updating 11 targets...
> > >     >     >     >     ...skipped 98 targets...
> > >     >     >     >     ...updated 454 targets...
> > >     >     >     >     ### Error : 1
> > >     >     >     >
> > >     >     >     > If I had to guess, I would suspect the source of the
> > problem is treating warnings as errors w/
> > >     Intel's
> > >     >     >     TBB.  That is the
> > >     >     >     > first and most frequent error I see.  Is there an
> > easy way to turn that off?
> > >     >     >     >
> > >     >     >     > thanks again,
> > >     >     >     > Jacob
> > >     >     >     >
> > >     >     >     > On Mon, Mar 24, 2008 at 7:39 PM, Mat Marcus <
> > mm...@ad...> wrote:
> > >     >     >     >
> > >     >     >     >     Hi Jacob,
> > >     >     >     >
> > >     >     >     >     Sorry to hear that you're having troubles with
> > the net-setup script. I
> > >     >     >     >     too encountered some crashes with UNTGZ. I kept
> > hitting ok until the
> > >     >     >     >     crashes stopped, then eventually the net-setup
> > script
> > >     >     >     >     succeeded. Unfortunately, I haven't been able to
> > find an alternative
> > >     >     >     >     public tool for windows that can unpack the
> > intel tbb tgz
> > >     >     >     >     files. Perhaps its stumbling on the long suffix
> > ".xcodeproj".
> > >     >     >     >
> > >     >     >     >     It's not clear to me how you continued the build
> > after the untgz
> > >     >     >     >     crashes. The net setup script actually executes
> > the build command
> > >     >     >     >     multiple times to workaround apparent bjam bugs
> > (mkdir issues). So it
> > >     >     >     >     is not unusual for some intermediate errors to
> > show up, but in the end
> > >     >     >     >     the 1.0.36 build succeeded here. Is the same
> > true for you if you "ok"
> > >     >     >     >     through the tgz crashes? If not, please give a
> > few more details about
> > >     >     >     >     exactly how you are building if you would like
> > further assistance.
> > >     >     >     >
> > >     >     >     >      - Mat
> > >     >     >     >
> > >     >     >     >     "Jacob Foshee" <ja...@gm...> writes:
> > >     >     >     >
> > >     >     >     >     > Hi,
> > >     >     >     >     > I just tried to d/load 1.0.36.  UNTGZ crashes
> > while expanding Intel TBB.  Just
> > >     >     >     >     > curious if you had any ideas?  (I'm on WinXP
> > SP2 w/ VC 8.0)
> > >     >     >     >     >
> > >     >     >     >     > I know this is not your archive, but, fwiw, it
> > happens to crash after:
> > >     >     >     >     >
> > >     >     >     >     >     Creating directory:
> > tbb20_014oss_src/examples/concurrent_hash_map/
> > >     >     >     >     >
> > count_strings/xcode/count_strings.xcodeproj OK
> > >     >     >     >     >
> > >     >     >     >     > I expanded the archive using another tool, and
> > that seemed to work fine.
> > >     >     >     >     > However, the build breaks at this point:
> > >     >     >     >     >
> > >     >     >     >     >     compile-c-c++
> > C:\Development\3rdParty\Adobe\asl_1.0.36\adobe_root\
> > >     >     >     >     >     built_artifacts\msvc-
> > 8.0\debug\asl-dev\atomics-tbb\link-static\
> > >     >     >     >     >     threading-multi\
> > >     >     >     >     >     xstring.cpp
> > >     >     >     >     >
> > ..\intel_tbb_libraries\include\tbb/atomic.h(158) : error C2220: warning
> > >     >     >     >     >     treated as error - no 'object' file
> > generated
> > >     >     >     >     >
> > ..\intel_tbb_libraries\include\tbb/atomic.h(158) : while compiling
> > >     >     >     >     >     class template member function
> > 'tbb::internal::atomic_traits<Size,M>::word
> > >     >     >     >     >
> > Size,M>::word,tbb::internal::atomic_traits<Size,M>::word)'
> > >     >     >     >     >             with
> > >     >     >     >     >             [
> > >     >     >     >     >                 Size=4,
> > >     >     >     >     >                 M=__TBB_full_fence
> > >     >     >     >     >             ]
> > >     >     >     >     >
> > ..\intel_tbb_libraries\include\tbb/atomic.h(344) : see reference
> > >     >     >     >     >     to class template instantiation
> > 'tbb::internal::atomic_traits<Size,M>'
> > >     >     >     >     >     being
> > >     >     >     >     >             with
> > >     >     >     >     >             [
> > >     >     >     >     >                 Size=4,
> > >     >     >     >     >                 M=__TBB_full_fence
> > >     >     >     >     >             ]
> > >     >     >     >     >
> > ..\intel_tbb_libraries\include\tbb/atomic.h(348) : see reference
> > >     >     >     >     >     to function template instantiation
> > 'tbb::atomic<void*>::value_type
> > >     >     >     >     >     tbb::atomi
> > >     >     >     >     >     e_type)' being compiled
> > >     >     >     >     >
> > ..\intel_tbb_libraries\include\tbb/atomic.h(158) : warning C4244:
> > >     >     >     >     >     'argument' : conversion from
> > 'tbb::internal::atomic_traits<Size,M>::word'
> > >     >     >     >     >     to 'int',
> > >     >     >     >     >             with
> > >     >     >     >     >             [
> > >     >     >     >     >                 Size=4,
> > >     >     >     >     >                 M=__TBB_full_fence
> > >     >     >     >     >             ]
> > >     >     >     >     >
> > ..\intel_tbb_libraries\include\tbb/atomic.h(158) : warning C4244:
> > >     >     >     >     >     'argument' : conversion from
> > 'tbb::internal::atomic_traits<Size,M>::word'
> > >     >     >     >     >     to 'int',
> > >     >     >     >     >             with
> > >     >     >     >     >             [
> > >     >     >     >     >                 Size=4,
> > >     >     >     >     >                 M=__TBB_full_fence
> > >     >     >     >     >             ]
> > >     >     >     >     >
> > ..\intel_tbb_libraries\include\tbb/atomic.h(158) : warning C4244:
> > >     >     >     >     >     'argument' : conversion from
> > 'tbb::internal::atomic_traits<Size,M>::word'
> > >     >     >     >     >     to 'int',
> > >     >     >     >     >             with
> > >     >     >     >     >             [
> > >     >     >     >     >                 Size=4,
> > >     >     >     >     >                 M=__TBB_full_fence
> > >     >     >     >     >             ]
> > >     >     >     >     >
> > ..\intel_tbb_libraries\include\tbb/atomic.h(158) : while compiling
> > >     >     >     >     >     class template member function
> > 'tbb::internal::atomic_traits<Size,M>::word
> > >     >     >     >     >     e,M>::word)'
> > >     >     >     >     >             with
> > >     >     >     >     >             [
> > >     >     >     >     >                 Size=4,
> > >     >     >     >     >                 M=__TBB_full_fence
> > >     >     >     >     >             ]
> > >     >     >     >     >
> > ..\intel_tbb_libraries\include\tbb/atomic.h(158) : warning C4244:
> > >     >     >     >     >     'argument' : conversion from
> > 'tbb::internal::atomic_traits<Size,M>::word'
> > >     >     >     >     >     to 'int',
> > >     >     >     >     >             with
> > >     >     >     >     >             [
> > >     >     >     >     >                 Size=4,
> > >     >     >     >     >                 M=__TBB_full_fence
> > >     >     >     >     >             ]
> > >     >     >     >     >
> > ..\intel_tbb_libraries\include\tbb/atomic.h(158) : while compiling
> > >     >     >     >     >     class template member function
> > 'tbb::internal::atomic_traits<Size,M>::word
> > >     >     >     >     >     ize,M>::word)'
> > >     >     >     >     >             with
> > >     >     >     >     >             [
> > >     >     >     >     >                 Size=4,
> > >     >     >     >     >                 M=__TBB_full_fence
> > >     >     >     >     >             ]
> > >     >     >     >     >
> > >     >     >     >     > Please let me know if you have any ideas, or
> > if I can provide more
> > >     >     >     >     > information.
> > >     >     >     >     > Thanks,
> > >     >     >     >     > Jacob
> > >     >     >     >     >
> > >     >     >     >     > On Fri, Mar 21, 2008 at 8:06 PM, Mat Marcus <
> > mm...@ad...> wrote:
> > >     >     >     >     >
> > >     >     >     >     >     Corrected subject line: 1.0.36 has been
> > released, not 1.0.46.
> > >     >     >     >     >
> > >     >     >     >     >     Mat Marcus <mm...@ad...> writes:
> > >     >     >     >     >
> > >     >     >     >     >     > Version 1.0.36 of the Adobe Source
> > Libraries has been released.
> > >     >     >     >     >     > Highlights of this release include: ASL:
> > >     >     >     >     >     >
> > >     >     >     >     >     >     * version_1 types (replace version_0
> > types) - better, defined binary
> > >     >     >     >     >     layout and carry allocator for
> > applications with replaced new/delete.
> > >     >     >     >     >     >     * capture_allocator, an
> > std::allocator compatible allocator which
> > >     >     >     >     >     carries a paired new/delete.
> > >     >     >     >     >     >     * Allocator support in
> > closed_hash_set<>, closed_hash_map<>, and
> > >     >     >     >     >     vector<>.
> > >     >     >     >     >     >     * Vastly improved move library
> > (Thanks to Dave Abrahams!).
> > >     >     >     >     >     >     * Vastly improved copy-on-write
> > library (now uses move library to
> > >     >     >     >     >     reduce number of increment/decrements).
> > >     >     >     >     >     >     * Bug fix in layout library that
> > caused containers with the width
> > >     >     >     >     >     driven by the frame width to layout
> > children incorrectly.
> > >     >     >     >     >     >     * Bug fix in forest::erase where the
> > beginning of the erased range
> > >     >     >     >     >     has a lower depth then the end of the
> > range.
> > >     >     >     >     >     >     * Forest test cases and minor fixes
> > to forest types.
> > >     >     >     >     >     >     * Misc bug fixes.
> > >     >     >     >     >     >     * Fixes to build and distribution
> > tools.
> > >     >     >     >     >     >     * Fixed search on documentation site
> > to point to stlab and not
> > >     >     >     >     >     >     * opensource domain.
> > >     >     >     >     >     >
> > >     >     >     >     >     > Change list information can be found
> > here:
> > >     >     >     >     >     >
> > http://stlab.adobe.com/asl_release_notes.html
> > >     >     >     >     >     >
> > >     >     >     >     >     > Documentation to get started with the
> > release is here:
> > >     >     >     >     >     > http://stlab.adobe.com/asl_readme.html
> > >     >     >     >     >     >
> > >     >     >     >     >     > Distribution files can be downloaded
> > from here:
> > >     >     >     >     >     >
> > http://sourceforge.net/project/showfiles.php?group_id=132417
> > >     >     >     >     >     >
> > >     >     >     >     >     >  - Mat
> > >     >     >     >     >
> > >     >     >     >     >
> > -------------------------------------------------------------------------
> > >     >     >     >     >     This SF.net email is sponsored by:
> > Microsoft
> > >     >     >     >     >     Defy all challenges. Microsoft(R) Visual
> > Studio 2008.
> > >     >     >     >     >
> > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> > >     >     >     >     >
> > _______________________________________________
> > >     >     >     >     >     Adobe-source-devel mailing list
> > >     >     >     >     >     Ado...@li...
> > >     >     >     >     >
> > https://lists.sourceforge.net/lists/listinfo/adobe-source-devel
> >
>
>