http://permalink.gmane.org/gmane.comp.java.openlaszlo.user hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5183 Here are some of the things I've tried. Note all the different effects for what appears to be equivalent code. <canvas> <script> System.security.loadPolicyFile("https://rest.mydomain.com/crossdomain.xml"); </script> <!-- NOTE: src is needed in order to get nsprefix to work --> <dataset name="dssend" proxied="false" nsprefix="true" src="https:"/> <node name="datactl"> <method name="send"> var ds = dssend; var x = '<![CDATA[<foo>bar</foo>]]>'; ds.setAttribute('src', 'https://rest.mydomain.com/foos'); ds.setQueryParams(null); // clear query params ds.setAttribute('querytype', 'POST'); // Sends "sid=abcdefg" as body, very odd. x is NOT sent. // ds.setQueryParam('sid', 'abcdefg'); // ds.setPostBody(x); // Sends encoded x, odd again. // BUT if the above ds.setQueryParams(null) is // removed it is sent as is (i.e. correctly). // ds.se Greg Denton 2008-12-03T23:42:07 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5182 Are you sure that query params cannot be set when there is a post body? I could have sworn we specifically fixed that to work. Can you send me a test case? Henry Minsky 2008-12-03T22:32:14 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5181 I just want to touch base on the SLOLO REST support for 4.2. This may be more of a swf question but if anyone could comment, it would be appreciated. (We are getting closer to a release and need to know whether/how to hack our backend service to do workarounds. I almost had a workaround for current 4.2b2 swf8 code but then discovered that query params could not be set when a body is given, argh!) Review of a completely transparent REST/HTTP interface: 1. All methods supported, PUT/DELETE in addition to GET/POST. 2. For all method types, independently set: (a) request headers, (b) query params, (c) body/content. 3. For all responses (all method types), whether status OK or not, get: (a) status, (b) response headers, (c) body/content (parse returned XML). (The odd thing is that this is the basic HTTP protocol and, in some ways, is simpler to implement than constraining things as is currently done.) I presume that this would only be available on swf9/DHTML, right? Thanks. On Mon, A Greg Denton 2008-12-03T22:30:09 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5180 I'd like to replace: v.setAttribute('x', ...); with: v.x = ...; someday. Having an alias `set` for `setAttribute` would be a fine interim step. On 2008-12-03, at 00:45EST, Max Carlson wrote: P T Withington 2008-12-03T14:05:40 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5179 If you have a simple test case, I can take a look. Like Tucker said, using Firebug can often show whatit thinks it is requesting. On Tue, Dec 2, 2008 at 7:14 AM, Brendan Fagan <suburbanantihero-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>wrote: Henry Minsky 2008-12-02T14:06:25 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5178 This was a config issue on my machine. I am running the OL war over Jetty. I have the project code in my workspace and under the exploded OL war folder via a symbolic link. This works fine for the OL war, but Jetty does not support symbolic links by default so it wasn't serving the image resources. To correct this, I modified $JETTY_HOME/etc/webdefault.xml to include the following for the "default" servlet: <init-param> <param-name>aliases</param-name> <param-value>true</param-value> </init-param> On Tue, Dec 2, 2008 at 7:54 AM, P T Withington <ptw-e+AXbWqSrlAAvxtiuMwx3w< at >public.gmane.org> wrote: Brendan Fagan 2008-12-02T14:00:26 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5177 This is supposed to 'just work'. How about looking at your browser's net traffic to see if it is fetching the .png's or maybe there is some path confusion? On 2008-12-02, at 07:14EST, Brendan Fagan wrote: P T Withington 2008-12-02T12:54:27 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5176 Do I need to do something special to use image resources in DHTML? I'm declaring PNG files in <resource tags and using those named resources in my views. Works fine in SWF8. No resources show up in DHTML. Brendan Fagan 2008-12-02T12:14:09 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5174 Hi, Thanks for fiiling this bug: LPP-7407. Amy Muntz 2008-11-28T15:16:10 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5173 Can you please file this as a bug on our JIRA bug tracker jira.openlazlo.org 2008/11/27 최영철 <nori11< at >volans.kr> Henry Minsky 2008-11-28T03:22:27 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5172 <canvas> <dataset name="cpu" request="false" type="http" src="cgi-bin/vlp_cgi.cgi" querystring="cpu_status"/> <method name="doRequest"> canvas.datasets.cpu.doRequest(); </method> </canvas> Compiling on window GET /cgi-bin/vlp_cgi.cgi?cpu_status= HTTP/1.1 Accept: */* Accept-Language: ko-KR Referer: http://192.168.0.39/main.lzx.lzr=swf8.swf x-flash-version: 10,0,12,36 UA-CPU: x86 Accept-Encoding: gzip, deflate User-Agent: Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0. 50727; .NET CLR 1.1.4322) Host: 192.168.0.39 Connection: Keep-Alive Authorization: Digest username="admin",realm="VoLANs Web Server Authentication",nonce="23433faedd0b1f27736aa90af44df1cd",uri="/cgi- bin/vlp_cgi.cgi?cpu_status=",cnonce="a868c05c95f21cf78488dcdf9bbb05bb",nc=00 000073,algorithm=MD5,response="e7c6cf9fc96aefb6dfcd94e2a8546374",qop="auth" Compiling on linux POST /main.lzr=swf8.swf HTTP/1.1 Accept: */* Accept-Language: ko-KR Referer: http://192.168.0.39/main.lzr=swf8.swf x-f 최영철 2008-11-28T02:46:26 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5171 The debug attribute has to be set at compile time -- it cannot be changed at run time. You don't have to set it in the canvas though. It can be set as a query argument on the application URL: ?debug=true. On 2008-11-25, at 06:29EST, Tim Dauer wrote: P T Withington 2008-11-25T12:05:50 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5170 Hi Laszlo-Users! I am using a configuration file which contains several constraints that are needed in my laszlo application - it's like an properties file. This file contains mostly attribute-TAGs and some resource-tags. What I would like to do is to manage the debug-attribute of the canvas in this configuration file. The problem is that, no matter what I try, this attribute is not read out correctly. I tried to bind the attribute in the canvas-TAG by using something like this <canvas debug="${canvas.isDebug}"... > and to add the attribute <attribute name="isDebug" type="boolean" value="true"/> in the configuration file which is included in the canvas-Class. No luck. And I tried to overwrite the debug-Attribute like <attribute name="debug" type="boolean" value="true"/> Again: No luck. Neither of these approaches worked out. Do you have any ideas on how to manage that issue in a configuration file? Or maybe this isn't possible because of initialization issues of the canvas? Greets Tim Tim Dauer 2008-11-25T11:29:58 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5169 If you're compiling with the command line compiler, lzc, you can set the flag like this -Dlzproxied=false On Mon, Nov 24, 2008 at 9:37 PM, Greg Denton <greg.denton-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote: Henry Minsky 2008-11-25T02:59:23 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5168 I'm a bit confused by the "3.3.2. lzproxied query parameter" section of the documentation: http://www.openlaszlo.org/lps4.2/docs/developers/proxied.html Is there a way to call the lzc compiler to compile for SOLO? I looked around some and haven't found any switch. It's more convenient in a build environment than hitting a lps server. I'm leaving "proxied=" off of <canvas> so I can debug in proxied mode. Thanks. Greg Denton 2008-11-25T02:37:25 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5167 http://www.openlaszlo.org/jira/browse/LPP-7380 On Fri, Nov 21, 2008 at 6:31 AM, P T Withington <ptw-e+AXbWqSrlAAvxtiuMwx3w< at >public.gmane.org> wrote: Greg Denton 2008-11-21T17:26:05 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5166 This bug is being tracked as: http://jira.openlaszlo.org/jira/browse/LPP-7272 if you would care to watch or comment. On 2008-11-20, at 16:50EST, Henry Minsky wrote: P T Withington 2008-11-21T15:05:42 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5165 We want drawview to work too, because we plan to use it more and more. It would be really helpful if you entered your test case and comments into Jira, so they can be tracked. Thanks! http://jira.openlaszlo.org/jira/secure/CreateIssue!default.jspa On 2008-11-20, at 18:45EST, Greg Denton wrote: P T Withington 2008-11-21T14:31:31 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5164 Here is a fix to the workaround (when drawing does not touch both axes and there is no subview), insert these lines instead of the arc, fillStyle, and fill calls: redlight.moveTo(0, 0); redlight.lineTo(0, 0); redlight.moveTo(0, r); redlight.lineTo(r, r); redlight.stroke(); Without the (0, 0) operations the height is just 1 as can be easily seen. Sorry for all the emails about this, but drawview is very powerful and would be great to get working consistently with view. It is, of course, very complex to maintain the width and height "manually". (Another example of drawview's power: it automatically detects context menu mouse downs on only the painted parts! though, I haven't tested this recently.) On Thu, Nov 20, 2008 at 2:42 PM, Greg Denton <greg.denton-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote: Greg Denton 2008-11-20T23:45:24 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5163 There was no response several months ago and now I'm just getting back to this part of my code and noticed different behavior in 4.2b2 and stumbled across a "workaround". The different behavior is that the drawview no longer resizes at all, neither expanding nor shrinking, whereas before it would expand. The workaround is to set the width and height to null before drawing. Insert these lines before "redlight.clear();": redlight.setAttribute('width', null); redlight.setAttribute('height', null); The drawview will both expand and shrink. There is a problem with this workaround, however, when drawing does not touch all the edges of the view, e.g. a 1 pixel line from (0,20) to (20,20) results in a width of 20 and height 1 (should be 20). This is "fixed", oddly enough, by including a subview to the drawview. On Wed, Sep 24, 2008 at 5:24 PM, Greg Denton <greg.denton-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote: Greg Denton 2008-11-20T22:42:02 http://permalink.gmane.org/gmane.comp.java.openlaszlo.user/5162 Sorry about that, we're still trying to decide how we want to expose this. Debug is now an instance of theLzDebugService, which is not a window object. It does have a pointer to the debugger window though, which currently can be found as "Debug.console" in swf8. This will probably change when we figure out what the best relationship is between the debugger service and the in-application debugger window component. On Thu, Nov 20, 2008 at 4:24 PM, Jesse Norell <jesse-xzVHdjf+epU< at >public.gmane.org> wrote: Henry Minsky 2008-11-20T21:50:45 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.comp.java.openlaszlo.user