[Laszlo-checkins] r6801 - openlaszlo/trunk
ben@openlaszlo.org
ben at openlaszlo.org
Wed Oct 10 16:46:08 PDT 2007
Author: ben
Date: 2007-10-10 16:46:07 -0700 (Wed, 10 Oct 2007)
New Revision: 6801
Modified:
openlaszlo/trunk/
Log:
Blocked revisions 6505,6575,6796-6797 via svnmerge
...............
r6505 | ben | 2007-09-16 14:18:38 -0700 (Sun, 16 Sep 2007) | 1 line
Creating paperpie branch for work on doctools
...............
r6575 | ben | 2007-09-24 17:04:34 -0700 (Mon, 24 Sep 2007) | 1 line
Changing version names and numbers, and turning up the maximum memory for the java doc build step, which speeds it up an awful lot.
...............
r6796 | ben | 2007-10-10 15:33:26 -0700 (Wed, 10 Oct 2007) | 3 lines
Initialized merge tracking via "svnmerge" with revisions "1-6504" from
http://svn.openlaszlo.org/openlaszlo/trunk
...............
r6797 | ben | 2007-10-10 16:17:04 -0700 (Wed, 10 Oct 2007) | 2154 lines
Merged revisions 6505-6796 via svnmerge from
http://svn.openlaszlo.org/openlaszlo/trunk
................
r6506 | ben | 2007-09-16 14:46:02 -0700 (Sun, 16 Sep 2007) | 1 line
Updating build.properties to 4.1.x.0
................
r6520 | ptw | 2007-09-19 09:56:48 -0700 (Wed, 19 Sep 2007) | 13 lines
Change 20070918-ptw-k by ptw at dueling-banjos.local on 2007-09-18 18:21:53 EDT
in /Users/ptw/OpenLaszlo/ringding
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Tweak output of Debug.bugReport
Details:
Use Debug.inspect, not inspectInternal so the 'hotlink' markup is hidden.
Tests:
IWFM
................
r6521 | lou | 2007-09-19 10:13:30 -0700 (Wed, 19 Sep 2007) | 24 lines
Change 20070919-lou-isO2ST by lou at lou-laptop on 2007-09-19 05:45:28 AST
in /home/lou/src/svn/openlaszlo/trunk/docs/src/developers
for http://svn.openlaszlo.org/openlaszlo/trunk/docs/src/developers
Summary: fix LPP-4529
New Features: spell-checked file
Bugs Fixed:
Technical Reviewer: Benjamin Shine
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details:
Tests:
................
r6522 | lou | 2007-09-19 10:20:19 -0700 (Wed, 19 Sep 2007) | 24 lines
Change 20070919-lou-idaGli by lou at lou-laptop on 2007-09-19 07:07:29 AST
in /home/lou/src/svn/openlaszlo/trunk/docs/src/developers
for http://svn.openlaszlo.org/openlaszlo/trunk/docs/src/developers
Summary: fix LPP-3826
New Features:
Bugs Fixed: LPP-3826 changes that made sense, and corrected many spelling errors
Technical Reviewer: Benjamine Shine
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details:
Tests:
................
r6529 | dda | 2007-09-19 13:26:31 -0700 (Wed, 19 Sep 2007) | 46 lines
Change 20070919-dda-f by dda at freddie.local on 2007-09-19 13:41:48 EDT
in /Users/dda/laszlo/src/svn/openlaszlo/trunk/WEB-INF/lps/server/src/org/openlaszlo/sc
for http://svn.openlaszlo.org/openlaszlo/trunk/WEB-INF/lps/server/src/org/openlaszlo/sc
Summary: generatePredictableTemps option for script compiler
New Features:
added new option generatePredictableTemps that allows for direct comparison
of compiler outputs.
Bugs Fixed:
Technical Reviewer: ptw
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
The new option, is specified to the compiler like this:
--option generatePredictableTemps=true
Note that this option is *not* the default, nor is it recommended for anyone
that is not doing side-by-side tests of the compiler. When used, it may
in fact *break* functionality. It provides some capabilities to assist
in internal testing, and specifically for testing the refactoring work.
When this is option is set, variables that normally use a UUID-like naming (for uniqueness between runs)
are changed to have predictable names, that is names that will be identical each time
the compiler is rerun. Such predictable names (like "$lzsc$1", "$lzsc$2", etc.) will
collide with the same names used in other compilations that appear in the same applications,
which may cause problems at runtime. This option uses a simple increment to generate
names. Seeding the random number generator with a fixed value, like 0, would also work
but the output will appear more obfuscated.
The intention is that this option can be used when making an architectural or 'cleanup'
change to the compiler that is not intended to change any behavior. When this change
is done, a suitably complete test input can be fed to the old and new compiler, both
with the generatePredictableTemps option, and comparisons of outputs can be made directly.
Tests:
................
r6531 | max | 2007-09-19 15:32:07 -0700 (Wed, 19 Sep 2007) | 28 lines
Change 20070919-maxcarlson-Z by maxcarlson at plastik on 2007-09-19 09:23:44 PDT
in /Users/maxcarlson/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Fix history for swf7
New Features:
Bugs Fixed: LPP-4746 - Javascript errors when using the history mechanism in swf7
Technical Reviewer: promanik
QA Reviewer: ben
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: LzHistory.as - Fix braino - use correct callJS() API.
embednew.js - Add 'loaded' property that is set to true when applications are loaded. Use loaded property to decide when apps are ready to receive events.
lzhistory.js - Add explicit scope for all methods, properties in Lz.history.set()
Tests: http://localhost:8080/trunk/test/history/history.lzx?lzr=swf7 no longer warns
................
r6533 | lou | 2007-09-19 17:11:42 -0700 (Wed, 19 Sep 2007) | 24 lines
Change 20070919-lou-If2BNY by lou at lou-laptop on 2007-09-19 15:27:06 AST
in /home/lou/src/svn/openlaszlo/trunk/docs/src/developers
for http://svn.openlaszlo.org/openlaszlo/trunk/docs/src/developers
Summary: spell check dguide
New Features:
Bugs Fixed: LPP-4754
Technical Reviewer: Benjamin Shine
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details: These are the files that actually had spelling errors
Tests:
................
r6539 | lou | 2007-09-20 03:17:33 -0700 (Thu, 20 Sep 2007) | 24 lines
Change 20070919-lou-Ccvkux by lou at lou-laptop on 2007-09-19 20:26:11 AST
in /home/lou/src/svn/openlaszlo/trunk/WEB-INF/lps/lfc/views
for http://svn.openlaszlo.org/openlaszlo/trunk/WEB-INF/lps/lfc/views
Summary: fix comment in source for addProxyPolicy
New Features:
Bugs Fixed: LPP-1552: doc for LzView.addProxyPolicy is wrong
Technical Reviewer: (pending)
QA Reviewer: Phil
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
Tests:
................
r6548 | dda | 2007-09-21 10:54:15 -0700 (Fri, 21 Sep 2007) | 67 lines
Change 20070919-dda-V by dda at freddie.local on 2007-09-19 17:35:41 EDT
in /Users/dda/laszlo/src/svn/openlaszlo/trunk/WEB-INF/lps/server/src/org/openlaszlo/sc
for http://svn.openlaszlo.org/openlaszlo/trunk/WEB-INF/lps/server/src/org/openlaszlo/sc
Summary: script compiler refactoring
New Features:
No externally visible changes. A rearrangement of source code to allow for better
sharing between modules, and separation of most of the platform independent parts of the compiler.
This more easily allows various future work (e.g. SWF9) to proceed in parallel with
general maintenance changes.
Bugs Fixed:
Technical Reviewer: ptw
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
Added a new class CommonGenerator that is the base class of
JavascriptGenerator and CodeGenerator. Methods from the two
subclasses that are textually the same, or can be easily coerced
to be the same are now only in CommonGenerator.
Data members and internal classes that are identical are
also now in the base class.
To get even greater sharing, many methods (like virtually all the
visit* methods) that had signature changes in JavascriptGenerator
(that is, they return 'SimpleNode' rather than void), have changed
across the board to return SimpleNode. That is, these functions return
the possibly modified node that is visited. In the CodeGenerator class, this
concept is generally not used, so even though this class plays by the
same rules, return values from such methods are ultimately ignored.
But this gives a lot more code sharing and many more functions could be moved
to CommonGenerator.
Likewise, methods that visit expressions that used to be declared
returning boolean in CodeGenerator now must return SimpleNode.
The backwards-seeming translation is that 'true' corresponds to null,
and false corresponds to the existing node. Doing it this way
gets the maximum code sharing, it clearly reflects the way that
the code was originally transformed after splitting.
CommonGenerator also implements a new interface: ASTVisitor, which is
the union of all the visit* methods (returning a SimpleNode, of
course). This allows methods in CommonGenerator (esp. visitStatement)
to freely call any other visit* method, whether or not it is actually
defined in CommonGenerator.
Most of these changes listed above are almost mechanical using the
indicated strategy; only occasionally were small (equivalent) coding
changes introduced, and then only if a method was otherwise identical.
More elaborate code sharing is possible, and was attempted, but
such changes tended to inhibit overall readability and maintainability.
Tests:
built all files in lps/includes/lfc with the previous compiler (without
these changes) using the flags "--option generatePredictableTemps=true".
built the same files with the new compiler (with these changes).
All *.js files compare exactly, all *.lzl files were disassembled,
and the output from that compares exactly.
................
r6550 | lou | 2007-09-21 12:11:51 -0700 (Fri, 21 Sep 2007) | 24 lines
Change 20070920-lou-pl9JDK by lou at lou-laptop on 2007-09-20 07:30:05 AST
in /home/lou/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: spell check tutorials
New Features:
Bugs Fixed: LPP-4761 spell check tutorial portion of the dguide
Technical Reviewer: Benjamin Shine
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details:
Tests:
................
r6559 | bargull | 2007-09-23 01:14:57 -0700 (Sun, 23 Sep 2007) | 28 lines
Change 20070922-bargull-0 by bargull at dell--p4--2-53 on 2007-09-22 22:15:29
in /home/Admin/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: basewindow must include basecomponent
New Features:
Bugs Fixed:
LPP-4758 "basewindow must include basecomponent"
Technical Reviewer: max
QA Reviewer: promanik
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
Applications which use solely basewindow do not compile without this change.
Just added the include-tag for basecomponent.lzx in basewindow.lzx
Tests:
see testcase at LPP-4758
................
r6568 | ben | 2007-09-24 11:59:55 -0700 (Mon, 24 Sep 2007) | 44 lines
Change 20070923-ben-l by ben at slim.local on 2007-09-23 15:15:48 PDT
in /Users/ben/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Add iPhone-specific hints to lzt=html wrapper to make iPhone apps look nice
New Features: apps are not zoomable or scalable with two-finger scaling on the iPhone
Bugs Fixed:
Technical Reviewer: ptw
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
As described in http://weblog.openlaszlo.org/archives/2007/07/our-first-iphone-app/
OpenLaszlo apps can look pretty decent on the iPhone just by adding a single
line to the html wrapper.
To see it in action, check out and run...
http://svn.openlaszlo.org/labs/newsmatch/
with and without this patch applied, with lzt=html. index.html,
in that directory, has links to the apps with the correct
query args, lzr=dhtml&lzt=html.
Without this patch, the app is scaled down to too small on the iPhone,
and the user can zoom and pan. With this patch, it appears at a good
size and is not zoomable. However, it is still pannable; I'm not sure
how to stop that.
This meta tag should be ignored by all other browsers. It is
a very small download cost.
Release Notes:
This change makes it possible to get apps to look right
on the iPhone without hand-editing wrappers.
Details:
Tests:
(see above, with newsmatch-labs)
................
r6582 | jcrowley | 2007-09-25 00:19:25 -0700 (Tue, 25 Sep 2007) | 44 lines
Change 20070920-jcrowley-8 by jcrowley at doctormanhattan.mshome.net on 2007-09-20 21:06:57 EDT
in /Users/jcrowley/src/svn/openlaszlo/trunk-two
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Update data tests to include dataloaders, remove
<datasource>, and more
New Features:
Bugs Fixed: LPP-4677 - Update data tests to include dataloaders,
remove <datasource>, and more
Technical Reviewer: henry
QA Reviewer: ben
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: This provides fixes for existing tests
and adds a couple other tests for new
dataprovider components. I'm working
on more tests for these, but this is
what I have so far.
You *might* get an error or failure with
clientcachebreaker, but it doesn't fail
consistently. I believe this is due to
the nature of the test. I think it fails
for me maybe once out of every five runs.
There were no tests of datasource that
I could find. (It was mentioned but
never instantiated anywhere.) So that
part of the bug is kind of N/A.
More extensive tests hopefully
forthcoming. I'll file a separate bug
for extending the testing.
Tests: Run all tests in test/lfc/data/
................
r6587 | ptw | 2007-09-25 09:36:15 -0700 (Tue, 25 Sep 2007) | 19 lines
Change 20070925-ptw-7 by ptw at dueling-banjos.local on 2007-09-25 09:38:45 EDT
in /Users/ptw/OpenLaszlo/ringding
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Fix merge error in r6548
Bugs Fixed:
LPP-4786 'Script compiler needs refactoring before SWF9 work can begin'
Technical Reviewer: dda at ddanderson.com (Message-Id: <C4801539-C2A3-403E-968B-1059AD82FA4C at ddanderson.com>)
QA Reviewer: jcrowley (pending)
Details:
Assertion left in common path that is supposed to only be in subclass.
Tests:
ant clean build will now build without assertion
................
r6589 | ptw | 2007-09-25 12:53:25 -0700 (Tue, 25 Sep 2007) | 26 lines
Change 20070925-ptw-2 by ptw at dueling-banjos.local on 2007-09-25 11:56:09 EDT
in /Users/ptw/OpenLaszlo/ringding-clean
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Fix SWF constraint lossage in r6548
Bugs Fixed:
LPP-4786 'Script compiler needs refactoring before SWF9 work can begin'
Technical Reviewer: dda at ddanderson.com (Message-Id: <24816C7A-777A-4188-9872-A5313BFD19F6 at ddanderson.com>)
QA Reviewer: hminsky (pending)
Doc Reviewer: (pending)
Details:
The remodularization tripped up the SWF Generator because it was
analyzing the constraint body _after_ it had been optimized to
inline the setAttribute call. Move the dependency analyzer to the
top of translateFunctionInternal, as is done in Javascript
Generator.
Tests:
Enabling ReferenceGenerator.debugConstraints shows that
constraints are actually being processed. smokecheck.lzx passes
in SWF now, and the SWF debugger window shows up again.
................
r6595 | max | 2007-09-25 15:04:25 -0700 (Tue, 25 Sep 2007) | 26 lines
Change 20070925-maxcarlson-7 by maxcarlson at plastik on 2007-09-25 13:55:11 PDT
in /Users/maxcarlson/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Fix xoffset and yoffset
New Features:
Bugs Fixed: LPP-4778 - "LzView#setX(..)" and "LzView#setY(..)" optimization breaks xoffset / yoffset
Technical Reviewer: promanik
QA Reviewer: bargull
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: lztest-view.lzx - add test for x/yoffset and one with rotation.
LaszloView.lzs - Add correct initial values for _x, _y, _width, _height. Add argument to setX/setY() to override cached values. setRotation/setXOffset/setYOffset() call setX/setY() with force set to true.
Tests: Run updated lztest-view.lzx with and without the changes to LaszloView.lzs. You should see the test fail without the LaszloView changes. Tested in swf7/8/dhtml.
................
r6602 | max | 2007-09-25 17:22:21 -0700 (Tue, 25 Sep 2007) | 24 lines
Change 20070925-maxcarlson-g by maxcarlson at plastik on 2007-09-25 15:42:07 PDT
in /Users/maxcarlson/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Fix stretches
New Features:
Bugs Fixed: LPP-4778 - "LzView#setX(..)" and "LzView#setY(..)" optimization breaks xoffset / yoffset
Technical Reviewer: promanik
QA Reviewer: andrebargull at aol.com
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: Shouldn't have set defaults for _x/y/width/height.
Tests: http://localhost:8080/trunk/examples/components/style_example.lzx runs correctly
................
r6604 | max | 2007-09-25 19:10:53 -0700 (Tue, 25 Sep 2007) | 24 lines
Change 20070925-maxcarlson-d by maxcarlson at plastik on 2007-09-25 16:15:32 PDT
in /Users/maxcarlson/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Fix explicit replicator
New Features:
Bugs Fixed: LPP-4664 - test/explicit-replicators/replicator.lzx visually deletes last two items when the second-to-last item is clicked and ...
Technical Reviewer: hminsky
QA Reviewer: andrebargull at aol.com
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: Per abargull and hminsky's suggestions, changed this._pointer.p to this.dataset for clarity, added destroy() to prevent memory leaks, _updateChildren() deals with a single node correctly, createChildren() calls super.createChildren(), __adjustVisibleClones() uses this.container instead of this.parent to lock/unlock layouts.
Tests: See LPP-4664.
................
r6608 | lou | 2007-09-26 02:38:21 -0700 (Wed, 26 Sep 2007) | 24 lines
Change 20070921-lou-l9Z4aa by lou at lou-laptop on 2007-09-21 14:17:20 AST
in /home/lou/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: correct minor doc issues
New Features:
Bugs Fixed: LPP-4762, fix minor issues (mostly typos) reported by user
Technical Reviewer: Ben Shine
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details:
Tests:
................
r6609 | lou | 2007-09-26 02:39:27 -0700 (Wed, 26 Sep 2007) | 25 lines
Change 20070925-lou-PdNB5U by lou at lou-laptop on 2007-09-25 10:31:35 AST
in /home/lou/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Update to performance chapter of dguide
New Features:
Bugs Fixed: LPP-1048: IDs vs names
Technical Reviewer: Ben Shine
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details:
Tests: Added the information from the JIRA bug. I pretty much
copied the text as is (I don't know JavaScript).
................
r6612 | lou | 2007-09-26 04:46:33 -0700 (Wed, 26 Sep 2007) | 25 lines
Change 20070926-lou-YHeW1d by lou at lou-laptop on 2007-09-26 05:49:05 AST
in /home/lou/src/svn/openlaszlo/trunk/docs/src/developers
for http://svn.openlaszlo.org/openlaszlo/trunk/docs/src/developers
Summary: LPP-4762: minor doc changes based on user input
New Features: This file should have gotten checked in with change
20070921-lou-l9Z4aa; I left it out by mistake.
Bugs Fixed:
Technical Reviewer: (pending)
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details:
Tests:
................
r6623 | lou | 2007-09-26 14:47:35 -0700 (Wed, 26 Sep 2007) | 24 lines
Change 20070926-lou-FCWTJO by lou at lou-laptop on 2007-09-26 12:53:34 AST
in /home/lou/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: check the dguide for correct descriptions of whitespace
New Features:
Bugs Fixed: LPP-984 check the data and solo chapters for correct descriptions of whitespace and trimwhitespace
Technical Reviewer: Ben Shine
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details:
Tests:
................
r6630 | max | 2007-09-26 18:41:06 -0700 (Wed, 26 Sep 2007) | 26 lines
Change 20070925-maxcarlson-3 by maxcarlson at plastik on 2007-09-25 17:03:02 PDT
in /Users/maxcarlson/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Make drawviews resize.
New Features:
Bugs Fixed: LPP-3905 - Changes on width/height must be propagated down to the HTML-canvas in DHTML
Technical Reviewer: promanik
QA Reviewer: andrebargull at aol.com
Doc Reviewer: (pending)
Documentation: Drawviews now resize when the width or height are changed. In DHTML, the drawview will be cleared so it must be redrawn whenever the oncanvas event is sent.
Release Notes:
Details: drawview.lzx - In _buildcanvas(), if the canvas exists it is resized and the oncontext event is sent because it will be cleared. For IE, the canvas must be completely destroyed and rebuilt. setWidth() and setHeight() call _buildcanvas() whenever a valid height and width are set.
roundrect.lzx - Clear cached color state when oncontext is sent to ensure canvas state is reset properly.
Tests: http://localhost:8080/trunk/lps/components/incubator/test/test-roundrectbutton.lzx?lzr=dhtml now resizes beyond 100 pixels in IE and firefox.
................
r6631 | max | 2007-09-26 18:43:06 -0700 (Wed, 26 Sep 2007) | 24 lines
Change 20070925-maxcarlson-m by maxcarlson at plastik on 2007-09-25 17:31:46 PDT
in /Users/maxcarlson/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: text inside a combobox should not resize
New Features:
Bugs Fixed: LPP-4787 - text inside a combobox should not resize
Technical Reviewer: promanik
QA Reviewer: jcrowley
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: Set text resize to false. False used to be the default in 3.x.
Tests: See LPP-4787.
................
r6637 | jcrowley | 2007-09-26 21:20:51 -0700 (Wed, 26 Sep 2007) | 59 lines
Change 20070926-jcrowley-r by jcrowley at doctormanhattan.mshome.net on 2007-09-26 05:45:59 EDT
in /Users/jcrowley/src/svn/openlaszlo/trunk-a
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: maxlength doesn't work for LzText (NOTE: also fixed
for inputtext)
New Features:
Bugs Fixed: LPP-2208 - maxlength doesn't work for LzText
Technical Reviewer: pbr
QA Reviewer: max
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: Added a check for maxlength to setText() functions
in both LzText and LzInputText. Also fixed it for
initial text construction. If maxlength is set,
text will not exceed the length, even with setText().
(Before this, it would catch it with inputtext,
but only when one entered text into the box.
setText() would ignore it.)
Tests: Run the following:
<canvas width="500" height="500">
<simplelayout axis="y"/>
<text id="tx1" text="This text shouldn't be truncated."/>
<text id="tx2" text="This text should be truncated after 12 characters." maxlength="12"/>
<inputtext id="tx3" text="Type something here fdafda." maxlength="20" width="300"
bgcolor="0xDDFFDD"/>
<button
onclick="tx1.setText('This text is pretty long, but it should be fine.')">
Click here to set the first text element using setText()
</button>
<button
onclick="tx2.setText('This text is much longer than the allowed limit of twelve characters.')">
Click here to set the second text element using setText()
</button>
<button
onclick="tx3.setText('This text is much longer than the allowed limit of twenty characters.')">
Click here to set the inputtext using setText()
</button>
</canvas>
This also passes smokecheck, and actually clears up
a failure in the lztest-text test.
................
r6640 | ptw | 2007-09-27 05:20:32 -0700 (Thu, 27 Sep 2007) | 16 lines
Change 20070926-ptw-p by ptw at dueling-banjos.local on 2007-09-26 17:30:12 EDT
in /Users/ptw/OpenLaszlo/ringding
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Declare clonenumber
Bugs Fixed:
LPP-4793 'Using clonenumber emits warning when using replicator'
Technical Reviewer: hminsky (pending)
QA Reviewer: pablo (Message-ID: <Pine.OSX.4.64.0709261631480.11512 at pkang-mac.local>)
Tests:
Bug test case no longer gives a warning
................
r6643 | lou | 2007-09-27 10:59:50 -0700 (Thu, 27 Sep 2007) | 24 lines
Change 20070927-lou-v6tk4a by lou at lou-laptop on 2007-09-27 13:59:34 AST
in /home/lou/src/svn/openlaszlo/trunk/docs/src/developers/programs
for http://svn.openlaszlo.org/openlaszlo/trunk/docs/src/developers/programs
Summary:
New Features:
Bugs Fixed:
Technical Reviewer: (pending)
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
Tests:
................
r6644 | lou | 2007-09-27 11:01:43 -0700 (Thu, 27 Sep 2007) | 24 lines
Change 20070927-lou-3dtDKq by lou at lou-laptop on 2007-09-27 14:01:01 AST
in /home/lou/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary:
New Features:
Bugs Fixed:
Technical Reviewer: (pending)
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
Tests:
................
r6645 | pbr | 2007-09-27 13:15:21 -0700 (Thu, 27 Sep 2007) | 30 lines
Change 20070926-Philip-4 by Philip at Philip-DC on 2007-09-26 20:26:00 EDT
in /cygdrive/f/laszlo/svn/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: DHTML: onkeyup/onkeydown sent twice
New Features:
Bugs Fixed: LPP-4267
Technical Reviewer: max
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
There are two code paths that are generating onkeyup/onkeydown events in dhtml (The details are in LPP-4267). I modified LzInputTextSprite.js to not generate onkeyup/onkeydown. It still generates onchange though. The path that generates onkeyup/onkeydown is in LzKeys, the same as swf.
Tests:
See LPP-4267. With this change, a single keyup/keydown is generated for each key click/release.
smokecheck.lzx
Files:
M WEB-INF/lps/lfc/kernel/dhtml/LzInputTextSprite.js
Changeset: http://svn.openlaszlo.org/openlaszlo/patches/20070926-Philip-4.tar
................
r6649 | jcrowley | 2007-09-27 15:30:49 -0700 (Thu, 27 Sep 2007) | 29 lines
Change 20070920-jcrowley-Y by jcrowley at doctormanhattan.mshome.net on 2007-09-20 21:28:58 EDT
in /Users/jcrowley/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Introduction->Form example does not work correctly
New Features:
Bugs Fixed: LPP-4763 - Introduction->Form example does not work correctly
Technical Reviewer: ben
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: A quick fix to prevent objects behind a view from
being clickable.
Tests: Run laszlo-explorer -> laszlo-in-ten -> introduction ->
form. Enter your name, select a new name, click
the submit button. Click around in the box. Note
that the objects behind this view that were still
clickable when the view was visible are no longer
so.
................
r6650 | jcrowley | 2007-09-27 15:34:07 -0700 (Thu, 27 Sep 2007) | 26 lines
Change 20070921-jcrowley-P by jcrowley at doctormanhattan.mshome.net on 2007-09-21 18:37:01 EDT
in /Users/jcrowley/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Calendar: Event info bar clipped at left of canvas
New Features:
Bugs Fixed: LPP-4692 - Calendar: Event info bar clipped at
left of canvas
Technical Reviewer: pbr
QA Reviewer: ben
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: Something was clipping when it shouldn't have been.
Tests: Run Calendar demo in DHTML and Flash; click on an
event on Sunday; notice that the event info bar is
no longer clipped on the left.
................
r6651 | jcrowley | 2007-09-27 15:52:59 -0700 (Thu, 27 Sep 2007) | 33 lines
Change 20070926-jcrowley-K by jcrowley at doctormanhattan.mshome.net on 2007-09-26 04:47:38 EDT
in /Users/jcrowley/src/svn/openlaszlo/trunk-a
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Local datasets do not work in datacomboboxes;
datacombobox getText() missing
New Features:
Bugs Fixed: LPP-3811 - Local datasets do not work in datacomboboxes
LPP-4468 - datacombobox getText() missing
Technical Reviewer: pbr
QA Reviewer: ben
Doc Reviewer:
Documentation:
Release Notes:
Details: Andre Bargull - LPP-3811. Implemented his change
to basedatacombobox. (This was still pending review;
ignore pending changeset 20070814-jcrowley-n.)
Chris Bartak - LPP-4468. Implemented his change
to add getText() to datacombobox.
Tests: For LPP-3811:
[Removed this code due to a complaint from SVN about line
ending styles being inconsistent.]
................
r6652 | dda | 2007-09-27 15:53:32 -0700 (Thu, 27 Sep 2007) | 201 lines
Change 20070921-dda-l by dda at freddie.local on 2007-09-21 17:52:38 EDT
in /Users/dda/laszlo/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Fix script compiler switch stack mixup in default label
New Features:
Bugs Fixed: LPP-3023
Technical Reviewer: ptw
QA Reviewer: henry
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
The problem is in the script compiler, specifically for the SW7/8 code generator.
When creating a jump table (a series of comparisons to jump to the case labels),
each jump action leaves a value at the top of the stack that must be popped as the
first action of the case label. However, default was handled differently.
The extra value was popped first and the jump to the default label is made,
so the default label did not pop as its first action.
Here's the problem. When case labels are joined with the default label like this:
case 10:
case 11:
default:
// do something...
the code generator attempts to share the code, but the stack levels are mismatched.
One solution is to force the default label to be the last and insert the extra pop
for the case labels only:
case 10:
case 11:
pop stack
// fall through
default:
I opted for a simpler solution, to guarantee that stack levels are always the same.
During the jump table, we avoid popping the stack before the default jump. We must
take care that if there is no default label, the stack is popped before the jump table
completes and we jump to the end of the switch code block.
With this approach, we now always emit a pop as the first action of a default
label, like every other case label. This allows case labels to be easily
joined whether or not they represent the default case.
I don't believe there is any performance impact: we are doing the same number of pops
at runtime (well, one extra that is needed to fix the buggy situation), and the
overall code size should be the same.
Tests:
Ran test listed in LPP-3023 and made sure it is now correct, both with debug and no debug.
Enhanced the app to test a number of other corner cases for default labels and verified
the output, again with/without debug. App included here:
<canvas>
<class name="switcher" extends="text">
<method name="testSwitch" args="val">
var result = "none";
switch (val) {
case 1:
result = "case 1";
break;
case 2:
case 3:
// if you uncomment the true below both 2 and 3 work
// true;
default:
// if you remove the default: above 2 and 3 also work
result = "default";
break;
}
this.setText(this.name + ": " + result);
</method>
<method name="testSwitchDefaultBreak" args="val">
var result = "none";
switch (val) {
case 1:
result = "case 1";
break;
case 2:
case 3:
// if you uncomment the true below both 2 and 3 work
// true;
default:
break;
}
this.setText(this.name + ": " + result);
</method>
<method name="testSwitchDefaultNoBreak" args="val">
var result = "none";
switch (val) {
case 1:
result = "case 1";
break;
case 2:
case 3:
// if you uncomment the true below both 2 and 3 work
// true;
default:
}
this.setText(this.name + ": " + result);
</method>
<method name="testSwitchNoDefault" args="val">
var result = "none";
switch (val) {
case 1:
result = "case 1";
break;
case 2:
case 3:
// if you uncomment the true below both 2 and 3 work
// true;
}
this.setText(this.name + ": " + result);
</method>
</class>
<simplelayout />
<switcher name="switch1" />
<switcher name="switch2" />
<switcher name="switch3" />
<switcher name="switch4" />
<switcher name="switch11" />
<switcher name="switch21" />
<switcher name="switch31" />
<switcher name="switch41" />
<switcher name="switch12" />
<switcher name="switch22" />
<switcher name="switch32" />
<switcher name="switch42" />
<switcher name="switch13" />
<switcher name="switch23" />
<switcher name="switch33" />
<switcher name="switch43" />
<text name="switch5">
<method name="otherSwitch" args="val">
var result = "none";
switch (val) {
case 1:
result = "case 1";
break;
case 2:
case 3:
// if you uncomment the true below both 2 and 3 work
// true;
default:
// if you remove the default: above 2 and 3 also work
result = "default";
break;
}
this.setText(this.name + ": " + result);
</method>
</text>
<text name="switch6">
<method name="nonSwitch" args="val">
var result = "none";
if (val == 1) {
result = "case 1";
} else {
result = "default";
}
this.setText(this.name + ": " + result);
</method>
</text>
<text name="switch7">
</text>
<script>
// the first case works
switch1.testSwitch(1);
// the default case works;
switch2.testSwitch(5);
// using 2 or 3 appears to work but breaks all method calls
switch3.testSwitch(2);
// none of these locally defined calls work
switch4.testSwitch(1);
switch11.testSwitchDefaultBreak(1);
switch21.testSwitchDefaultBreak(5);
switch31.testSwitchDefaultBreak(2);
switch41.testSwitchDefaultBreak(1);
switch12.testSwitchDefaultNoBreak(1);
switch22.testSwitchDefaultNoBreak(5);
switch32.testSwitchDefaultNoBreak(2);
switch42.testSwitchDefaultNoBreak(1);
switch13.testSwitchNoDefault(1);
switch23.testSwitchNoDefault(5);
switch33.testSwitchNoDefault(2);
switch43.testSwitchNoDefault(1);
switch5.otherSwitch(1);
switch6.nonSwitch(1);
switch7.setText(this.name + ": " + "test");
</script>
</canvas>
................
r6654 | jcrowley | 2007-09-27 16:06:22 -0700 (Thu, 27 Sep 2007) | 44 lines
Change 20070927-jcrowley-v by jcrowley at doctormanhattan.mshome.net on 2007-09-27 19:04:11 EDT
in /Users/jcrowley/src/svn/openlaszlo/trunk-a
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: [repackaging to be checked into trunk; this was sent out from legals as
20070814-jcrowley-S, and was approved by Ben] datagrid: If the columns are
not updatable, it's impossible to select a row.
New Features:
Bugs Fixed: LPP-4201 - datagrid: If the columns are not updatable, it's impossible to
select a row.
Technical Reviewer: bshine
QA Reviewer: max
Doc Reviewer:
Documentation:
Release Notes:
Details: Ben and I weren't _entirely_ sure what this one meant, but I'm
pretty sure I figured it out. I made it so that one can
select the text in a given row, without the row having to be
editable. All you have to do is specify selectable="true" on
the gridtext element, as with any other text element.
Tests: Run the following example:
<canvas width="800" height="400" bgcolor="0xEAEAEA">
<dataset name="contactsData" request="true" type="http" src="http:contactsdata.xml"/>
<grid name="foo" datapath="contactsData:/*" sizetoheader="false" selectable="true">
<gridtext editable="false" width="50" datapath="position()" sortable="false">
No.
</gridtext>
<gridtext editable="false" selectable="true" datapath="@firstname">First</gridtext>
<gridtext editable="false" selectable="true" datapath="@lastname">Last</gridtext>
<gridtext editable="false" datapath="@displayname">Full Name</gridtext>
</grid>
</canvas>
Note that you can now select the text in the row.
................
r6655 | pbr | 2007-09-27 16:14:14 -0700 (Thu, 27 Sep 2007) | 45 lines
Change 20070926-Philip-0 by Philip at Philip-DC on 2007-09-26 13:49:59 EDT
in /cygdrive/f/laszlo/svn/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Fix reference to hasNoDelegates in LzModeManager
New Features:
Bugs Fixed: LPP-4389, LPP-2564
Technical Reviewer: ptw
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
This bug is already fixed because hasNoDelegates has been removed.
However, it is still referenced in LzModeManager.lzs. I changed the reference to getDelegateCount() as recommended in LPP-4389.
Tests:
This test continues to work properly. Double clicking will produce a debug message.
<canvas height="500" debug="true" >
<view width="40" height="40" bgcolor="0xEAEAEA" clickable="true" >
<handler name="oninit" >
var del = new LzDelegate( this, "handleDblClick", this, "ondblclick" );
del.unregisterFrom( this.ondblclick );
del.register( this, "ondblclick" );
</handler>
<method name="handleDblClick" >
Debug.write( "handleDblClick" );
</method>
</view>
</canvas>
Files:
M WEB-INF/lps/lfc/services/LzModeManager.lzs
Changeset: http://svn.openlaszlo.org/openlaszlo/patches/20070926-Philip-0.tar
................
r6664 | hqm | 2007-09-28 07:37:40 -0700 (Fri, 28 Sep 2007) | 24 lines
Change 20070928-hqm-2 by hqm at IBM-2E06404CB67 on 2007-09-28 10:36:46 EDT
in /cygdrive/c/users/hqm/openlaszlo/trunk/demos/weather
for http://svn.openlaszlo.org/openlaszlo/trunk/demos/weather
Summary: uncomment weather canvas size
New Features:
Bugs Fixed:LPP-4782
Technical Reviewer: josh
QA Reviewer: amy
Doc Reviewer:
Documentation:
Release Notes:
Details:
Tests:
................
r6670 | lou | 2007-09-28 11:14:47 -0700 (Fri, 28 Sep 2007) | 24 lines
Change 20070928-lou-m1pLMm by lou at lou-laptop on 2007-09-28 14:10:42 AST
in /home/lou/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: change type for stretches from Boolean to String
New Features:
Bugs Fixed:LPP-4193
Technical Reviewer: ptw
QA Reviewer: (pending)
Doc Reviewer:
Documentation:
Release Notes:
Details: also enclosed in quotes width, height, and both
Tests:
................
r6676 | jcrowley | 2007-09-28 13:26:36 -0700 (Fri, 28 Sep 2007) | 36 lines
Change 20070928-jcrowley-u by jcrowley at doctormanhattan.mshome.net on 2007-09-28 16:24:50 EDT
in /Users/jcrowley/src/svn/openlaszlo/trunk-a
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: [repackaging for trunk - this was approved by Phil]
DHTML calendar demo wigs out on opening Event Info
New Features:
Bugs Fixed: LPP-4175 - DHTML calendar demo wigs out on opening
Event Info
Technical Reviewer: pbr
QA Reviewer: max
Doc Reviewer:
Documentation:
Release Notes:
Details: I found a larger bug with this one, which was auto-
assigned to Tucker. It's LPP-4499. The issue here is
really that calling LzFocus.setFocus() on something
off-canvas in DHTML will slide the canvas over so that
the focused item is visible. I worked around it in
this case by moving it into an onstop event for one
of the animators called by the same method, to ensure
it didn't get called until after the object was fully
on-screen.
Tests: Run modified Calendar versus unmodified in DHTML; note
the modified version doesn't slide over all crazy-like
when the info panel is opened by clicking the 'i' on
a selected event tag.
................
r6683 | max | 2007-09-28 15:02:38 -0700 (Fri, 28 Sep 2007) | 33 lines
Change 20070928-maxcarlson-r by maxcarlson at plastik on 2007-09-28 13:30:10 PDT
in /Users/maxcarlson/openlaszlo/trunk-clean
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: One more time: Hand marge r6661 from wafflecone to trunk
New Features:
Bugs Fixed:
Technical Reviewer: promanik
QA Reviewer: jcrowley
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: LzHistory.as - Use correct syntax for LzBrowser.callJS(). Clear out old __lzloading code. Initialize persistence and add explicit callback to Lz.history in receiveHistory().
LzHistory.js - Clear out old __lzloading code.
LaszloCanvas.lzs - Rely on browser history event callbacks to initialize history state instead of explicit call at canvas init time.
embednew.js - Clean up extra window.onload event, comments. Set Lz.loaded to true when load is finished. Use anonymous function in _callMethodSWF() to prevent overwrites.
lzhistory.js - Store and reset the document title in IE to keep the window title from changing. Make init code more specific to each browser. Use Lz.callMethod() to set the history state instead of calling dojo.flash.comm. Don't set top.location.hash to '#0' in IE - use '#' instead. Add __receivedhistory() callback to receive history from Flash.
flash.js - Ensure callback methods are executed in the proper scope for flash < 8.
Tests: http://localhost:8080/trunk/test/history/history.lzx in dhtml, swf7 and 8 in Firefox mac, IE 6/7 and Safari 2.
................
r6684 | max | 2007-09-28 15:25:28 -0700 (Fri, 28 Sep 2007) | 44 lines
Recorded merge of revisions 6661 via svnmerge from
http://svn.openlaszlo.org/openlaszlo/branches/wafflecone
........
r6661 | max | 2007-09-27 22:30:20 -0700 (Thu, 27 Sep 2007) | 38 lines
Change 20070927-maxcarlson-6 by maxcarlson at plastik on 2007-09-27 18:44:35 PDT
in /Users/maxcarlson/openlaszlo/wafflecone
for http://svn.openlaszlo.org/openlaszlo/branches/wafflecone
Summary: UPDATED: Fix history in swf7/8 and dhtml
New Features:
Bugs Fixed: LPP-4746 - Javascript errors when using the history mechanism in swf7, LPP-4731 - reloading browser page causes history to trigger "back" action
Technical Reviewer: promanik
QA Reviewer: lhenrywilkins at laszlosystems.com
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: LzSprite.js - Do callback to _ready when canvas initializes.
LzHistory.as - Use correct syntax for LzBrowser.callJS(). Clear out old __lzloading code.
LzBrowser.as - Reset __jscallback to null when there is no callback.
LzHistory.js - Clear out old __lzloading code.
LaszloCanvas.lzs - Rely on browser history event callbacks to initialize history state instead of explicit call at canvas init time.
embednew.js - Hand merge changes related to onloaded from trunk.
lzhistory.js - Store and reset the document title in IE to keep the window title from changing.
flash.js - Ensure callback methods are executed in the proper scope for flash < 8.
Tests: http://localhost:8080/trunk/test/history/history.lzx in dhtml, swf7 and 8 in Firefox mac, IE 6/7 and Safari 2. Also tested webtop and verified that both bugs are fixed in future/lzmail/main.lzx - I used r11 of uicontroller.lzx to bring back the history functionality.
........
................
r6687 | hqm | 2007-09-28 17:17:41 -0700 (Fri, 28 Sep 2007) | 34 lines
Change 20070928-hqm-4 by hqm at IBM-2E06404CB67 on 2007-09-28 20:15:33 EDT
in /cygdrive/c/users/hqm/openlaszlo/trunk2
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: disable compiler warning for duplicate resource name
New Features:
Bugs Fixed: LPP-4615
Technical Reviewer: ptw
QA Reviewer: pkang
Doc Reviewer:
Documentation:
Release Notes:
Details:
remove the warning for duplicated resource, for now.
There are a couple of proposals for adding a flag to resource
declaration which permits overrides explicitly.
This patch is just a temporary solution for now.
Tests:
test/compiler_errors/duplicate-resource.lzx
................
r6707 | hqm | 2007-10-01 14:14:24 -0700 (Mon, 01 Oct 2007) | 34 lines
Change 20071001-hqm-0 by hqm at IBM-2E06404CB67 on 2007-10-01 13:24:07 EDT
in /cygdrive/c/users/hqm/openlaszlo/trunk2
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: allow dhtml as default runtime
New Features:
Bugs Fixed: LPP-3705
Technical Reviewer: pbr
QA Reviewer: ptw
Doc Reviewer: jrs
Documentation:
Release Notes:
Details:
Assign default runtime from one place (LPS.java), and assume runtime default if no 'lzr'
query arg is supplied in RESPONDERObject.java for lzt=object requests
Tests:
restart tomcat, and see if apps come up as DHTML by default with no lzr arg, and with lzt=html queries
................
r6715 | max | 2007-10-02 16:03:03 -0700 (Tue, 02 Oct 2007) | 43 lines
ehange 20071002-maxcarlson-a by maxcarlson at plastik on 2007-10-02 11:15:45 PDT
in /Users/maxcarlson/openlaszlo/trunk-clean
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Fix html tag
New Features:
Bugs Fixed: LPP-4784 - OL4.0.5, <html> tag and SWF : http://www.openlaszlo.org/lps4/lps/components/extensions/test/html-swf.jsp NOT WOKING
Technical Reviewer: promanik
QA Reviewer: jcrowley
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: DojoExternalInterface.as - Better debugging info, make sure resultsChecker function is named, clear out _root._callbackID when done looking for results.
Library.lzs - Add default values for _root._currenthistory and _root._callbackID to prevent exceptions in IE.
LzHistory.as - Use _root._currenthistory to communicate current history state in swf7 to prevent concurrency issues calling into/out of flash simultaneously.
LaszloCanvas.lzs - Set canvas id passed in to init arguments.
test/html.lzx - Use w3.org as default URL to prevent runtime exceptions - openlaszlo.org fails to run some javascript when laoded in an iframe.
html-swf.js - Update to new wrapper format.
html.lzx - Pass in canvas id for callbacks. Add support for sendToBack(). Always update position when iframeid is set.
embednew.js - Pass id into flashargs to set canvas.id.
lzhistory.js - Watch _currenthistory for history state changes for swf7 in _checklocationhash(). __setFlash() only sets history state if _callbackID is null. Otherwise it waits 10ms.
iframemanager.js - Reorder init(). Add support for sendToBack().
dev-console.lzx.swf - Rebuild with default values for _root._currenthistory and _root. _callbackID to prevent exceptions in IE.
Tests: http://localhost:8080/trunk-clean/lps/components/extensions/test/html-swf.jsp and http://localhost:8080/trunk-clean/lps/components/extensions/test/html.zlx run in swf7, swf8 and dhtml. http://localhost:8080/trunk-clean/test/history/history.lzx passes in swf7, swf8 and dhtml. Tested under IE 6, 7 and Firefox 2 OS X.
................
r6726 | hqm | 2007-10-04 15:49:36 -0700 (Thu, 04 Oct 2007) | 41 lines
Change 20071004-hqm-2 by hqm at IBM-2E06404CB67 on 2007-10-04 18:43:58 EDT
in /cygdrive/c/users/hqm/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: replace jing validator with something better
New Features:
Bugs Fixed: LPP-4644
and other bugs too numerous to mention
Technical Reviewer: ptw
QA Reviewer: pbr
Doc Reviewer:
Documentation:
Release Notes:
Details:
removed JING RNG parser and validator, and all code in ViewSchema which uses it.
All validation is done from the lps/schema/lfc.lzx file now,
all validity checks done in updateschema or during compilation by
NodeModel and friends.
The "validate" flag is deprecated (we should warn for that I guess),
and can be removed at some point.
Tests:
compiled demos, examples/components, smokecheck, webtop lzmail app
ant runlzunit
tested test cases in test/compiler_errors
................
r6727 | hqm | 2007-10-04 15:59:29 -0700 (Thu, 04 Oct 2007) | 1 line
forgot this with 6726
................
r6729 | max | 2007-10-04 16:51:40 -0700 (Thu, 04 Oct 2007) | 34 lines
Change 20071004-maxcarlson-x by maxcarlson at plastik on 2007-10-04 12:49:38 PDT
in /Users/maxcarlson/openlaszlo/trunk-clean
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Fix Mode manager and LzText/Inputtext modality
New Features:
Bugs Fixed: LPP-3948 - LzModeManager and selectable LzText (real problem comes with edittext)
Technical Reviewer: promanik
QA Reviewer: jcrowley
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: LzTextSprite.js - Expose and set selectable attribute.
LzInputTextSprite.js - Don't become editable when selectable != true.
LzModeManager.lzs - Ensure mView.passModeEvent() exists before calling it.
LzInputText.lzs - Register for LzModeManager.onmode event. Ensure inputtext doesn't become selectable if it isn't inside the view that has modality applied.
LzText.lzs - Set selectable attribute in setSelectable().
edittext.lzx - Add support setSelection(), applyData(), updateData() and getValue().
Tests: See LPP-3948. Testcase passes in flash and DHTML.
................
r6731 | hqm | 2007-10-04 17:11:31 -0700 (Thu, 04 Oct 2007) | 27 lines
Change 20071004-hqm-8 by hqm at IBM-2E06404CB67 on 2007-10-04 19:22:03 EDT
in /cygdrive/c/users/hqm/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: fix a couple of warnings that now show up in dmoes
New Features:
Bugs Fixed:
Technical Reviewer: pbr
QA Reviewer: jsundman
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
These views should not have been named "frame", as it aliases an attribute of view.
Changed name of view to "myframe"
Tests:
................
r6736 | pbr | 2007-10-05 12:51:02 -0700 (Fri, 05 Oct 2007) | 48 lines
Also see LPP-1636
Change 20071005-Philip-6 by Philip at Philip-DC on 2007-10-05 13:42:37 EDT
in /cygdrive/f/laszlo/svn/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Pass arguments intact to Windows scripts
New Features:
Bugs Fixed: LPP-4843
Technical Reviewer: ptw
QA Reviewer: hqm
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
Arguments were not being transferred intact to java. The old scripts
used %1,%2,... to refer to arguments. However, Windows thinks that
--runtime=swf7 is two argument, "--runtime" and "swf7". The only fix
I found is to pass the arguments intact using %*.
Tests:
lzc,lzdc,lzmc pass arguments the same way. Tested on lzc. From the
windows command prompt:
- lzc --runtime=swf8 testfile.lzx
This command should product a file pbr113.lzr=swf8.swf
- lzc "--runtime=swf8" testfile.lzx
This should work the same as the first case. This makes sure that
people that are already escaping the argument can continue to do so.
Files:
M WEB-INF/lps/server/bin/lzc.bat
M WEB-INF/lps/server/bin/lzdc.bat
M WEB-INF/lps/server/bin/lzmc.bat
Changeset:
http://svn.openlaszlo.org/openlaszlo/patches/20071005-Philip-6.tar
................
r6740 | lou | 2007-10-06 07:38:01 -0700 (Sat, 06 Oct 2007) | 24 lines
Change 20071003-lou-c4MZCf by lou at lou-laptop on 2007-10-03 11:19:29 AST
in /home/lou/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Differential debugging SWF and DHTML
New Features:
Bugs Fixed: LPP-4824
Technical Reviewer: (pending)
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details: Added section 4.1 Differential debugging SWF and DHTML to the data tutorial.
Tests:
................
r6741 | lou | 2007-10-06 07:39:16 -0700 (Sat, 06 Oct 2007) | 20 lines
Change 20071003-lou-pXqghd by lou at lou-laptop on 2007-10-03 10:03:33 AST
in /home/lou/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Added Debug.bugReport() description
New Features:
Bugs Fixed: LPP-4798
Technical Reviewer: (pending)
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Details: Add section 1.6 Generating a bug report to the Debugging chapter.
Tests:
................
r6742 | lou | 2007-10-06 07:45:37 -0700 (Sat, 06 Oct 2007) | 24 lines
Change 20070928-lou-msVzZQ by lou at lou-laptop on 2007-09-28 13:31:10 AST
in /home/lou/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary:
New Features: replace references to Deployer's Guide to Administrator's Guide
Bugs Fixed: LPP-4819 dguide: references to Deployer's Guide shoud be Administrator's guide
Technical Reviewer: (pending)
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details:
Tests:
................
r6743 | lou | 2007-10-06 07:52:15 -0700 (Sat, 06 Oct 2007) | 24 lines
Change 20070928-lou-lvSbjV by lou at lou-laptop on 2007-09-28 14:05:16 AST
in /home/lou/src/svn/openlaszlo/trunk/docs/src
for http://svn.openlaszlo.org/openlaszlo/trunk/docs/src
Summary: Function.apply and Function.call now listed as supported
New Features: change table so that Function.apply and Function.call are now supported
Bugs Fixed: LPP-4338
Technical Reviewer: (pending)
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details:
Tests:
................
r6747 | lou | 2007-10-07 12:15:53 -0700 (Sun, 07 Oct 2007) | 25 lines
Change 20071007-lou-A by lou at loumac.local on 2007-10-07 14:56:08 AST
in /Users/lou/src/svn/openlaszlo/trunk/docs
for http://svn.openlaszlo.org/openlaszlo/trunk/docs
Summary: fix typos in language preliminaries chapter of dguide
New Features:
Bugs Fixed: LPP-4847
Technical Reviewer: (pending)
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: OpenLaszlo entirely case-sensitive. -> OpenLaszlo is entirely case-sensitive.
Fix 2 broken xrefs (these show up as ??? in the html file)
Tests:
................
r6749 | lou | 2007-10-07 15:53:03 -0700 (Sun, 07 Oct 2007) | 24 lines
Change 20071007-lou-8 by lou at loumac.local on 2007-10-07 17:54:39 AST
in /Users/lou/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Add Firebug array note to the debugging chapter of the dguide
New Features: Also changed references to FireFox and FireBug to Firefox and Firebug
Bugs Fixed: LPP-4846: explain arrays in Firebug debugger
Technical Reviewer: (pending)
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details:
Tests:
................
r6760 | jcrowley | 2007-10-08 13:45:40 -0700 (Mon, 08 Oct 2007) | 30 lines
Change 20071008-jcrowley-W by jcrowley at doctormanhattan.mshome.net on 2007-10-08 13:56:28 EDT
in /Users/jcrowley/src/svn/openlaszlo/trunk-b
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Components hierarchy browser should link to the right page in
the reference
New Features:
Bugs Fixed: LPP-4555 - Components hierarchy browser should link to the
right page in the reference
Technical Reviewer: pbr
QA Reviewer: ben
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: Fixed the links, many were outdated and the request
mechanism was kind of broken.
Tests: Run the components hierarchy; note that clicking all the
links in the modified version takes you to the appropriate
pages, whereas the links in an unmodified version will
result in 404 errors or simply taking you to the refguide
index.
................
r6761 | lou | 2007-10-08 14:47:08 -0700 (Mon, 08 Oct 2007) | 24 lines
Change 20071008-lou-j by lou at loumac.local on 2007-10-08 11:45:31 AST
in /Users/lou/src/svn/openlaszlo/trunk/docs/src
for http://svn.openlaszlo.org/openlaszlo/trunk/docs/src
Summary: Add section 4. Component auto-include to Designing Custom Components chapter of the dguide
New Features:
Bugs Fixed: LPP-1083: setting up component auto-include
Technical Reviewer: (pending)
QA Reviewer: (pending)
Doc Reviewer: John Sundman
Documentation:
Release Notes:
Details:
Tests:
................
r6769 | jcrowley | 2007-10-09 09:37:23 -0700 (Tue, 09 Oct 2007) | 153 lines
Change 20071009-jcrowley-z by jcrowley at doctormanhattan.mshome.net on 2007-10-09 12:34:27 EDT
in /Users/jcrowley/src/svn/openlaszlo/trunk-f
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: (Already approved by Phil; am integrating into a different
local branch before checking in.) long comobox does not
render properly on firefox 2.0
New Features:
Bugs Fixed: LPP-4192 - long comobox does not render properly on
firefox 2.0
Technical Reviewer: pbr
QA Reviewer: henry
Doc Reviewer:
Documentation:
Release Notes:
Details: This one wasn't really that much of a bug, since
you can set limits on the list size and give it
a scrollbar if you have to clip it, but I thought
there should still be some kind of check to make
sure it's not longer than the canvas. I added
attribute 'sizetocanvas', default true, to set
the number of visible elements in an expanded
list to fit the list on the canvas. In the likely
rare event one wants to let the list spill off the
canvas, they can always set sizetocanvas to false.
Tests: Run:
<canvas>
<combobox id="serviceCombo"
width="130"
defaulttext="choose one..." editable="false" enabled="true">
<textlistitem datapath="services:/services/service" text="$path{'@name'}"
value="$path{'@value'}"/>
</combobox>
<dataset name="services">
<services>
<service name="adamsboswe" value="adamsboswe"/>
<service name="323iactive" value="323iactive"/>
<service name="aecom" value="aecom"/>
<service name="1stomni" value="1stomni"/>
<service name="acisol" value="acisol"/>
<service name="ares" value="ares"/>
<service name="180solutio" value="180solutio"/>
<service name="ameris" value="ameris"/>
<service name="11below" value="11below"/>
<service name="advcommerc" value="advcommerc"/>
<service name="alloy" value="alloy"/>
<service name="abnerherrm" value="abnerherrm"/>
<service name="active" value="active"/>
<service name="activewind" value="activewind"/>
<service name="aei" value="aei"/>
<service name="adelante" value="adelante"/>
<service name="3spn" value="3spn"/>
<service name="ampere" value="ampere"/>
<service name="abundant" value="abundant"/>
<service name="advantage" value="advantage"/>
<service name="aethionsi" value="aethionsi"/>
<service name="altrec" value="altrec"/>
<service name="arin" value="arin"/>
<service name="aplia" value="aplia"/>
<service name="ailiant" value="ailiant"/>
<service name="about" value="about"/>
<service name="amerisec" value="amerisec"/>
<service name="alorica" value="alorica"/>
<service name="americasho" value="americasho"/>
<service name="americanap" value="americanap"/>
<service name="anomaly" value="anomaly"/>
<service name="anyware" value="anyware"/>
<service name="aloricaden" value="aloricaden"/>
<service name="appdiscove" value="appdiscove"/>
<service name="alpaul" value="alpaul"/>
<service name="advpersonn" value="advpersonn"/>
<service name="accesscomm" value="accesscomm"/>
<service name="absclothin" value="absclothin"/>
<service name="ainet" value="ainet"/>
<service name="aismedia" value="aismedia"/>
<service name="americom" value="americom"/>
<service name="accessline" value="accessline"/>
<service name="altntech" value="altntech"/>
<service name="apex" value="apex"/>
<service name="adinfuse" value="adinfuse"/>
<service name="actsgroup" value="actsgroup"/>
<service name="alliance" value="alliance"/>
<service name="alphazeta" value="alphazeta"/>
<service name="akamai" value="akamai"/>
<service name="aitcc" value="aitcc"/>
<service name="accellion" value="accellion"/>
<service name="accor" value="accor"/>
<service name="agile" value="agile"/>
<service name="alta" value="alta"/>
<service name="apptrix" value="apptrix"/>
<service name="arcap" value="arcap"/>
<service name="addante" value="addante"/>
<service name="3amlabs" value="3amlabs"/>
<service name="altadisusa" value="altadisusa"/>
<service name="amgen" value="amgen"/>
<service name="SJC-Lab" value="SJC-Lab"/>
<service name="accel" value="accel"/>
<service name="americanto" value="americanto"/>
<service name="apple" value="apple"/>
<service name="aci" value="aci"/>
<service name="accipiter" value="accipiter"/>
<service name="adpinvest" value="adpinvest"/>
<service name="acsonline" value="acsonline"/>
<service name="1ix" value="1ix"/>
<service name="aas" value="aas"/>
<service name="altoh" value="altoh"/>
<service name="adaptivepa" value="adaptivepa"/>
<service name="abs" value="abs"/>
<service name="aim2game" value="aim2game"/>
<service name="areatech" value="areatech"/>
<service name="appiq" value="appiq"/>
<service name="advancedac" value="advancedac"/>
<service name="1stfinanci" value="1stfinanci"/>
<service name="aig" value="aig"/>
<service name="800one" value="800one"/>
<service name="aptela" value="aptela"/>
<service name="activegrou" value="activegrou"/>
<service name="apress" value="apress"/>
<service name="3eco" value="3eco"/>
<service name="allrecipes" value="allrecipes"/>
<service name="amselect" value="amselect"/>
<service name="alpine" value="alpine"/>
<service name="argis" value="argis"/>
<service name="3dblinds" value="3dblinds"/>
<service name="ardennes" value="ardennes"/>
<service name="003realty" value="003realty"/>
<service name="amgenuk" value="amgenuk"/>
<service name="adaptec" value="adaptec"/>
<service name="aonwarrant" value="aonwarrant"/>
<service name="adpark" value="adpark"/>
<service name="apartment" value="apartment"/>
<service name="arenaone" value="arenaone"/>
<service name="adchemy" value="adchemy"/>
<service name="2wire" value="2wire"/>
<service name="advancecab" value="advancecab"/>
</services>
</dataset>
</canvas>
In both modified and unmodified branches. Note that in unmodified,
it spills off into forever. In modified, it stops just before, and
there's a scrollbar. (If one explicitly sets the number of visible
elements, it'll override this.)
................
r6779 | ptw | 2007-10-10 05:29:01 -0700 (Wed, 10 Oct 2007) | 27 lines
Change 20071009-ptw-M by ptw at dueling-banjos.local on 2007-10-09 18:28:50 EDT
in /Users/ptw/OpenLaszlo/ringding-2
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Re-enable Monitor messages
Bugs Fixed:
LPP-4855 'Debug.monitor does not appear to be working'
Technical Reviewer: max (pending)
QA Reviewer: sosullivan (pending)
Details:
LzDebug: Add MONITOR and TRACE as valid message levels
kernel.*: If message type does not have a valid level, always
print it.
LzMonitor, LzTrace: Give constructors a name.
Tests:
lzx> Debug.monitor(canvas, 'version')
lzx> canvas.version=888
MONITOR @simple.lzx#1: [845525.00] Debug.doEval: #canvas.version: '4.1.x.0' -> 888
888
................
r6782 | hqm | 2007-10-10 05:59:30 -0700 (Wed, 10 Oct 2007) | 26 lines
Change 20071010-hqm-4 by hqm at IBM-2E06404CB67 on 2007-10-10 08:57:32 EDT
in /cygdrive/c/users/hqm/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: remove debugging info that got included by mistake in r6726
New Features:
Bugs Fixed:
Technical Reviewer: hqm
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
Debug.info statements are adding noise to the output, should be
conditionalized to some high level
Tests:
................
r6783 | max | 2007-10-10 08:50:36 -0700 (Wed, 10 Oct 2007) | 24 lines
Change 20071009-maxcarlson-B by maxcarlson at plastik on 2007-10-09 15:20:56 PDT
in /Users/maxcarlson/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Add stub methods for view.getNextSelection() and getPrevSelection() for documentation
New Features:
Bugs Fixed: LPP-2041 - getNextSelection() and getPrevSelection() not documented in LzView reference page
Technical Reviewer: promanik
QA Reviewer: jcrowley
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details: Add stub methods and documentation for getNextSelection() and getPrevSelection().
Tests: Tabbing and selection in http://localhost:8080/trunk/examples/components/style_example.lzx?lzr=dhtml and http://localhost:8080/trunk/examples/components/style_example.lzx work as before.
................
r6785 | ptw | 2007-10-10 09:43:37 -0700 (Wed, 10 Oct 2007) | 20 lines
Change 20071001-ptw-m by ptw at dueling-banjos.local on 2007-10-01 19:55:05 EDT
in /Users/ptw/OpenLaszlo/ringding-2
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Remove old .make trampolines
Bugs Fixed:
LPP-4181 'Remove vestigial .make methods from built-ins'
Technical Reviewer: hminsky (Message-ID: <8c61fad60710100555y17b972e9j1d2fb8f0e74a2eee at mail.gmail.com>)
QA Reviewer: promanik (Message-Id: <20071010131315.AEC3497209 at hemicuda.laszlosystems.com>)
Details:
We decided we don't need to rewrite all new calls to .make calls,
so these old trampolines can be removed.
Tests:
smokecheck in SWF and DHTML, calendar, amazon
................
r6789 | hqm | 2007-10-10 11:20:57 -0700 (Wed, 10 Oct 2007) | 27 lines
Change 20071010-hqm-0 by hqm at IBM-2E06404CB67 on 2007-10-10 14:18:54 EDT
in /cygdrive/c/users/hqm/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: comment out diagnostic info in lzpix app
New Features:
Bugs Fixed: LPP-4779
Technical Reviewer: jcrowley
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
The inittimer and lps version widgets are commented out. if someone needs them
they should conditionally put instantiate them for internal releases, per tucker's suggestion
Tests:
................
...............
Property changes on: openlaszlo/trunk
___________________________________________________________________
Name: svnmerge-blocked
- /openlaszlo/branches/wafflecone:5808,5811,5813,6069,6206,6214-6215,6266,6369,6432,6452 /openlaszlo/trunk:3893,3953,4394,4462,4468,4472,5086,5172,5204,5210,5332
+ /openlaszlo/branches/paperpie:6505,6575,6796-6797 /openlaszlo/branches/wafflecone:5808,5811,5813,6069,6206,6214-6215,6266,6369,6432,6452 /openlaszlo/trunk:3893,3953,4394,4462,4468,4472,5086,5172,5204,5210,5332
More information about the Laszlo-checkins
mailing list