exist-development — eXist Developer's List

Re: [Exist-development] Some thoughts about betterforms @ exist

[Dannes W. <da...@ex...>]

Hi.

On Wed, Aug 18, 2010 at 10:58 PM, Joern Turner
<joe...@go...>wrote:

> Yes, i know Saxon is in endorsed and we rely on this now (not bundling
> our own any more)
>

Ok....



> I fear you mean saxon 9.2 don't you? I just spend 3 hours on Monday to
> make our code run with 9.1.0.8 which (due to my last info) is the
> current vesion used in eXist. We still used 9.0. If i'd known that you
> want to switch quickly to 9.2 i had not done that upgrade to 9.1.
>

Sorry, i do not have the intention to waste development time. The thing is
however that we already had dependencies with Saxon 9.2: the ant target
"download-saxon" downloads a 9.2 version for some time ago, and also the
xmlcalabash extension has a dependency there. That is why the latter does
not build anymore, due to the 9.0/9.1 version>

Bit unhappy about that but anyway - must be done one day anyway.
>
Sorry :-(


> Please let me know when the upgrade to 9.2 comes.
>

On my behalf... ASAP. The work around is to put the 9.2 jar into lib/user
for this moment.
So... if your code works with 9.2 .. the rest of exist is OK.




> And yes, we have deep dependency on Saxon as its the XPath engine we
> use and all of the XPath 2, XForms and custom functions are
> implemented using Saxon so changes here are the most critical issue in
> a tight integration of eXist and betterFORM. We should coordinate on
> version upgrades in time to avoid breaking code.
>

Ok, i was not aware of the dependency until the build failed.


cheers

Dannes

-- 
eXist-db Native XML Database - http://exist-db.org
Join us on linked-in: http://www.linkedin.com/groups?gid=35624

Re: [Exist-development] Some thoughts about betterforms @ exist

[Joern T. <joe...@go...>]

On Wed, Aug 18, 2010 at 8:46 PM, Dannes Wessels <da...@ex...> wrote:
> Hi,
> On 16 Aug 2010, at 13:28 , Joern Turner wrote:
>
> Think this is solved by Wolfgangs change to the build.xml
>
> yes, thnx wolf.
>
> (2) Remove jars from  lib/user
>
> Just to  clarify: we have agreed that libs that are only used by
> betterFORM should go to extensions/betterform lib directory. So, if
> commons-lang is not used elsewhere in eXist we would move it there.
>
> The same is true for the DWR lib.
>
> Ok. Well commons-lang is quite a powerfull library, but few ppl know it. So
> yeas we can put it under the betterforms directories, and move it later.
>
> WARNING: An API change between ehcache-1.1 and ehcache-1.2 results in
> the persistence path being set to
>
> /Users/wessels/Development/eXist-db/Contrib/Betterforms/tools/jetty/work/Jetty_0_0_0_0_8080_webapp__exist__.8xgt21/cocoon-files/cache-dir/ when
> the ehcache-1.1 constructor is used. Please change to the 1.2 constructor.
>
> This should go away when Cocoon is removed.
>
> make sense....
>
> At the  moment we still have our own log4j.xml in WEB-INF. This should
>
> be removed. Please direct us to the right location to add our
> debugging.
>
> Hmmm do you want to have your own log file? that makes sense. Probably we
> can use an XSLT script to update the exist-db log4j.xml, by adding another
> appender specific for the bf-packages
That's exactly what i asking. How should we integrate with eXist
logging. I think there is a logging dir and yes it makes some sense to
have a log just for xforms logging as it can be quite verbose when you
try to find issues in a complex form.

An xslt to merge our log4j into eXists' make sense - we did the same
for the web.xml already. But we also need to adapt our code as we
initialize our logging in XFormsFilter. Think that probably an issue
and must be suppressed? Shouldn't be hard to add a config option for
that.

>
> Kind regards
> Dannes
> --
> eXist-db Native XML Database - http://exist-db.org
> Join us on linked-in: http://www.linkedin.com/groups?gid=35624
>
>
>
>
>
>
>
>

Re: [Exist-development] Some thoughts about betterforms @ exist

[Joern T. <joe...@go...>]

On Wed, Aug 18, 2010 at 8:42 PM, Dannes Wessels <da...@ex...> wrote:
>> > (4) Saxon
>> > It became clear that we have some issues with the newest saxon jar
>> > files. In
>> > november I found out about a work around, i'd like to propose to follow
>> > the
>> > work around: install saxon in lib/user [for this we have already an
>> > excellent download task], remove all lib/endorsed/saxon*jar files.  If
>> > we
>> > decide to install saxon into the exist-db distribution [adam] this
>> > download
>> > code can be removed.
>>
>> Would be nice to have a decision here
>
> Saxon must be under endorsed.. for sure, but the current version has a bug.
> So in the mean time I;d like to propose to put it under lib/user , if mr kay
> fixes the issue we'll revert.
> I had problems with starting betterforms with the newest Saxon version,
> there is a deep dependacy between bf and saxon?
Yes, i know Saxon is in endorsed and we rely on this now (not bundling
our own any more)

I fear you mean saxon 9.2 don't you? I just spend 3 hours on Monday to
make our code run with 9.1.0.8 which (due to my last info) is the
current vesion used in eXist. We still used 9.0. If i'd known that you
want to switch quickly to 9.2 i had not done that upgrade to 9.1.

Bit unhappy about that but anyway - must be done one day anyway.
Please let me know when the upgrade to 9.2 comes.

And yes, we have deep dependency on Saxon as its the XPath engine we
use and all of the XPath 2, XForms and custom functions are
implemented using Saxon so changes here are the most critical issue in
a tight integration of eXist and betterFORM. We should coordinate on
version upgrades in time to avoid breaking code.


> D.
> --
> eXist-db Native XML Database - http://exist-db.org
> Join us on linked-in: http://www.linkedin.com/groups?gid=35624
>

Re: [Exist-development] Please review setup.xqm

[Joern T. <joe...@go...>]

On Wed, Aug 18, 2010 at 9:14 PM, Dannes Wessels <da...@ex...> wrote:
> Joern,
>
> On 18 Aug 2010, at 21:07 , Dannes Wessels wrote:
>
>> I am pretty sure that wolfgang fixed this.......... what has happened? Why is there a XSLT1 message?
>
> unfortunately you undid the changes of Wolfgang. I repaired it.
All i can say is that it worked on my machine when i checked in this
afternoon. And i ran all the tests and it did fine. So don't know how
that comes.

Best

Joern
>
> Kind regards
>
> Dannes
>
> --
> eXist-db Native XML Database - http://exist-db.org
> Join us on linked-in: http://www.linkedin.com/groups?gid=35624
>
>
>
>
>
>
>
>
>

Re: [Exist-development] Please review setup.xqm

[Dannes W. <da...@ex...>]

Joern,

On 18 Aug 2010, at 21:07 , Dannes Wessels wrote:

> I am pretty sure that wolfgang fixed this.......... what has happened? Why is there a XSLT1 message?

unfortunately you undid the changes of Wolfgang. I repaired it.

Kind regards

Dannes

--
eXist-db Native XML Database - http://exist-db.org
Join us on linked-in: http://www.linkedin.com/groups?gid=35624

Re: [Exist-development] Please review setup.xqm

[Dannes W. <da...@ex...>]

I tried BF tonight......

extension-betterform:

install:
     [echo] ************************************************************************************
     [echo] add betterFORM config parameters to eXist to /Users/wessels/Development/eXist-db/Contrib/Betterforms/extensions/betterform/../../webapp/WEB-INF/web.xml
     [echo] ************************************************************************************

patchWebXml:
     [xslt] Processing /Users/wessels/Development/eXist-db/Contrib/Betterforms/webapp/WEB-INF/web.xml to /Users/wessels/Development/eXist-db/Contrib/Betterforms/webapp/WEB-INF/web.xml.bf
     [xslt] Loading stylesheet /Users/wessels/Development/eXist-db/Contrib/Betterforms/extensions/betterform/src/main/etc/MergeWebXML.xsl
     [xslt] Unknown file:15:120: Fatal Error! Could not find function: exists
     [xslt] : Fatal Error! Fatal error during transformation Cause: Fatal error during transformation
     [xslt] Failed to process /Users/wessels/Development/eXist-db/Contrib/Betterforms/webapp/WEB-INF/web.xml

BUILD FAILED
/Users/wessels/Development/eXist-db/Contrib/Betterforms/build/scripts/build-impl.xml:658: The following error occurred while executing this line:
/Users/wessels/Development/eXist-db/Contrib/Betterforms/extensions/betterform/build.xml:53: The following error occurred while executing this line:
/Users/wessels/Development/eXist-db/Contrib/Betterforms/extensions/betterform/build.xml:91: Fatal error during transformation


I am pretty sure that wolfgang fixed this.......... what has happened? Why is there a XSLT1 message?

hmmmm

D.


On 18 Aug 2010, at 15:39 , Joern Turner wrote:

> Dannes,
> 
> could you (or someone else responsible for the examples) please review
> our changes to setup.xqm? We have added some code to install our
> sample app and it works as far as we can tell. The code is still
> commented out.
> 
> When uncommenting and running install examples from admin page the
> application should be installed and can be accessed via:
> http://localhost:8080/exist/rest/db/betterform/apps/timetracker/index.xql.
> 
> When you see a list of time tracking data everything should be fine.
> 
> Thanks,
> 
> Joern

Kind regards

Dannes

--
eXist-db Native XML Database - http://exist-db.org
Join us on linked-in: http://www.linkedin.com/groups?gid=35624

Re: [Exist-development] Some thoughts about betterforms @ exist

[Dannes W. <da...@ex...>]

Hi,
On 16 Aug 2010, at 13:28 , Joern Turner wrote:

> Think this is solved by Wolfgangs change to the build.xml

yes, thnx wolf.

> 
>> (2) Remove jars from  lib/user
>> 
> Just to  clarify: we have agreed that libs that are only used by
> betterFORM should go to extensions/betterform lib directory. So, if
> commons-lang is not used elsewhere in eXist we would move it there.
> 
> The same is true for the DWR lib.

Ok. Well commons-lang is quite a powerfull library, but few ppl know it. So yeas we can put it under the betterforms directories, and move it later.


>> WARNING: An API change between ehcache-1.1 and ehcache-1.2 results in the persistence path being set to
>> /Users/wessels/Development/eXist-db/Contrib/Betterforms/tools/jetty/work/Jetty_0_0_0_0_8080_webapp__exist__.8xgt21/cocoon-files/cache-dir/ when the ehcache-1.1 constructor is used. Please change to the 1.2 constructor.
> 
> This should go away when Cocoon is removed.

make sense....

>> At the  moment we still have our own log4j.xml in WEB-INF. This should
> be removed. Please direct us to the right location to add our
> debugging.

Hmmm do you want to have your own log file? that makes sense. Probably we can use an XSLT script to update the exist-db log4j.xml, by adding another appender specific for the bf-packages


Kind regards

Dannes

--
eXist-db Native XML Database - http://exist-db.org
Join us on linked-in: http://www.linkedin.com/groups?gid=35624

Re: [Exist-development] Some thoughts about betterforms @ exist

[Dannes W. <da...@ex...>]

>
> > (4) Saxon
> > It became clear that we have some issues with the newest saxon jar files.
> In
> > november I found out about a work around, i'd like to propose to follow
> the
> > work around: install saxon in lib/user [for this we have already an
> > excellent download task], remove all lib/endorsed/saxon*jar files.  If we
> > decide to install saxon into the exist-db distribution [adam] this
> download
> > code can be removed.
>
> Would be nice to have a decision here


Saxon must be under endorsed.. for sure, but the current version has a bug.
So in the mean time I;d like to propose to put it under lib/user , if mr kay
fixes the issue we'll revert.

I had problems with starting betterforms with the newest Saxon version,
there is a deep dependacy between bf and saxon?

D.

-- 
eXist-db Native XML Database - http://exist-db.org
Join us on linked-in: http://www.linkedin.com/groups?gid=35624

[Exist-development] bug? empty xforms model in admin.xql

[Joern T. <joe...@go...>]

Posted this before as part of another message - probably got lost.

We have encountered that there is an empty xf:model element in line106
of admin.xql. This does not seem to make any sense but maybe someone
can enlighten me. At least - if that element is output by admin.xql
the XFormsFilter of betterFORM picks that page up and tries to process
it as XForms document. This fails as there is no complete XForms in
there and this in turn leads to some of the admin pages break.

I'd like to know if it's save to remove this element. Otherwise we
need (a more complicated) way to suppress betterFORM processing for
the admin area.

[Exist-development] Please review setup.xqm

[Joern T. <joe...@go...>]

Dannes,

could you (or someone else responsible for the examples) please review
our changes to setup.xqm? We have added some code to install our
sample app and it works as far as we can tell. The code is still
commented out.

When uncommenting and running install examples from admin page the
application should be installed and can be accessed via:
http://localhost:8080/exist/rest/db/betterform/apps/timetracker/index.xql.

When you see a list of time tracking data everything should be fine.

Thanks,

Joern

Re: [Exist-development] test-compile fails

[Joern T. <joe...@go...>]

Ok, got it

a ant exist.rebuild did it.

Sorry for disturbing.

On Wed, Aug 18, 2010 at 10:36 AM, Joern Turner
<joe...@go...> wrote:
> -checked out fresh copy of trunk
> -ran ant exist.compile
> -ran ant test
>
> result:
> test-compile:
>    [mkdir] Created dir: /Users/dev/projects/exist-trunk/test/classes
>     [echo] Compiling with Java 1.6 from 1.6 source to 1.6 target,
> debug on, optimize on, deprecation off
>    [javac] Compiling 278 source files to
> /Users/dev/projects/exist-trunk/test/classes
>    [javac] /Users/dev/projects/exist-trunk/test/src/org/exist/xquery/functions/util/CounterTest.java:9:
> package org.exist.xquery.modules.counter does not exist
>    [javac] import org.exist.xquery.modules.counter.CounterModule;
>    [javac]                                        ^
>    [javac] /Users/dev/projects/exist-trunk/test/src/org/exist/xquery/functions/util/CounterTest.java:25:
> cannot find symbol
>    [javac] symbol  : variable CounterModule
>    [javac] location: class org.exist.xquery.functions.util.CounterTest
>    [javac]     private final static String IMPORT = "import module
> namespace counter=\"" + CounterModule.NAMESPACE_URI + "\" " +
>    [javac]
>                     ^
>    [javac] Note: Some input files use or override a deprecated API.
>    [javac] Note: Recompile with -Xlint:deprecation for details.
>    [javac] Note: Some input files use unchecked or unsafe operations.
>    [javac] Note: Recompile with -Xlint:unchecked for details.
>    [javac] 2 errors
>
> BUILD FAILED
> /Users/dev/projects/exist-trunk/build/scripts/junit.xml:44: Compile
> failed; see the compiler error output for details.
>
>
> On Wed, Aug 18, 2010 at 10:20 AM, Joern Turner
> <joe...@go...> wrote:
>> Sorry - i don't get it. Still get compile errors for the tests.
>>
>> What can be wrong?
>>
>>
>> On Mon, Aug 16, 2010 at 6:23 PM, Dannes Wessels <da...@ex...> wrote:
>>> Hi Joern,
>>> this is an ordering issue in the build. You need to build exist first, then
>>> the tests.
>>> cheers
>>> Dannes
>>> On 16 Aug 2010, at 17:55 , Joern Turner wrote:
>>>
>>> We have done our mods and tried to run tests but there is a compile
>>> error in test-compile:
>>>
>>> CounterTest.java:9: package org.exist.xquery.modules.counter does not
>>> exist and so forth
>>>
>>> Can someone confirm that? We have swtiched off all of our stuff so it
>>> does not seem to have to do with the betterFORM extension
>>>
>>> Kind regards
>>> Dannes
>>> --
>>> eXist-db Native XML Database - http://exist-db.org
>>> Join us on linked-in: http://www.linkedin.com/groups?gid=35624
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>

Re: [Exist-development] test-compile fails

[Joern T. <joe...@go...>]

-checked out fresh copy of trunk
-ran ant exist.compile
-ran ant test

result:
test-compile:
    [mkdir] Created dir: /Users/dev/projects/exist-trunk/test/classes
     [echo] Compiling with Java 1.6 from 1.6 source to 1.6 target,
debug on, optimize on, deprecation off
    [javac] Compiling 278 source files to
/Users/dev/projects/exist-trunk/test/classes
    [javac] /Users/dev/projects/exist-trunk/test/src/org/exist/xquery/functions/util/CounterTest.java:9:
package org.exist.xquery.modules.counter does not exist
    [javac] import org.exist.xquery.modules.counter.CounterModule;
    [javac]                                        ^
    [javac] /Users/dev/projects/exist-trunk/test/src/org/exist/xquery/functions/util/CounterTest.java:25:
cannot find symbol
    [javac] symbol  : variable CounterModule
    [javac] location: class org.exist.xquery.functions.util.CounterTest
    [javac]     private final static String IMPORT = "import module
namespace counter=\"" + CounterModule.NAMESPACE_URI + "\" " +
    [javac]
                     ^
    [javac] Note: Some input files use or override a deprecated API.
    [javac] Note: Recompile with -Xlint:deprecation for details.
    [javac] Note: Some input files use unchecked or unsafe operations.
    [javac] Note: Recompile with -Xlint:unchecked for details.
    [javac] 2 errors

BUILD FAILED
/Users/dev/projects/exist-trunk/build/scripts/junit.xml:44: Compile
failed; see the compiler error output for details.


On Wed, Aug 18, 2010 at 10:20 AM, Joern Turner
<joe...@go...> wrote:
> Sorry - i don't get it. Still get compile errors for the tests.
>
> What can be wrong?
>
>
> On Mon, Aug 16, 2010 at 6:23 PM, Dannes Wessels <da...@ex...> wrote:
>> Hi Joern,
>> this is an ordering issue in the build. You need to build exist first, then
>> the tests.
>> cheers
>> Dannes
>> On 16 Aug 2010, at 17:55 , Joern Turner wrote:
>>
>> We have done our mods and tried to run tests but there is a compile
>> error in test-compile:
>>
>> CounterTest.java:9: package org.exist.xquery.modules.counter does not
>> exist and so forth
>>
>> Can someone confirm that? We have swtiched off all of our stuff so it
>> does not seem to have to do with the betterFORM extension
>>
>> Kind regards
>> Dannes
>> --
>> eXist-db Native XML Database - http://exist-db.org
>> Join us on linked-in: http://www.linkedin.com/groups?gid=35624
>>
>>
>>
>>
>>
>>
>>
>>
>

Re: [Exist-development] test-compile fails

[Joern T. <joe...@go...>]

Sorry - i don't get it. Still get compile errors for the tests.

What can be wrong?


On Mon, Aug 16, 2010 at 6:23 PM, Dannes Wessels <da...@ex...> wrote:
> Hi Joern,
> this is an ordering issue in the build. You need to build exist first, then
> the tests.
> cheers
> Dannes
> On 16 Aug 2010, at 17:55 , Joern Turner wrote:
>
> We have done our mods and tried to run tests but there is a compile
> error in test-compile:
>
> CounterTest.java:9: package org.exist.xquery.modules.counter does not
> exist and so forth
>
> Can someone confirm that? We have swtiched off all of our stuff so it
> does not seem to have to do with the betterFORM extension
>
> Kind regards
> Dannes
> --
> eXist-db Native XML Database - http://exist-db.org
> Join us on linked-in: http://www.linkedin.com/groups?gid=35624
>
>
>
>
>
>
>
>

Re: [Exist-development] org.exist.xquery.regex

[Leif-Jöran O. <lj...@ex...>]

Den 2010-08-17 18:16, Dmitriy Shabanov skrev:
> Good day,
>
> As far as we use saxon as default xsl processor, should
> org.exist.xquery.regex be removed?

I see your point. But I am not sure it can be removed. I only see it 
possible if it is not an option to use eXist-db without saxon or make 
sure if it is shifted out that this functionality is provided by the 
replacement processor (or still by saxon classes).

I am currently reviewing the differences between your and mine take at 
this saxon borrow. So I will need to think some more about it.

Cheers,
Leif-Jöran

[Exist-development] org.exist.xquery.regex

[Dmitriy S. <sha...@gm...>]

Good day,

As far as we use saxon as default xsl processor, should
org.exist.xquery.regex be removed?

-- 
Cheers,

Dmitriy Shabanov

Re: [Exist-development] adhoc perf benchmark

[Leif-Jöran O. <lj...@ex...>]

Den 2010-08-02 16:25, Wolfgang Meier skrev:
>> e.g.
>>
>> http://demo.exist-db.org/jgoethe-tei.zip
>>
>> not there to download from this url
>
> Correct. I have a local copy of those files, so I did not recognize
> the URL is failing now. I'll try to move the stuff elsewhere.


Well, someone moved it ;)

existdb@serengeti:~$ find . -name "jgoethe-tei.zip"
./exist-db.org/eXist-trunk/webapp.bak/jgoethe-tei.zip

I assume I just can move it back to the webapp directory.

Leif-Jöran

Re: [Exist-development] httpclient issues

[Adam R. <ad...@ex...>]

Do you need a new module? Could you not build this as a .xqm using the
existing httpclient module?

On 16 August 2010 20:58, Adam Retter <ad...@ex...> wrote:
> webdavclient would seem appropriate
>
> On 16 August 2010 20:05, Loren Cahlander <lor...@gm...> wrote:
>> Hello Dannes,
>>
>> I will create a new function module in extensions.  Should I call it webdav or webdav-client?
>>
>> Loren
>>
>> On Aug 16, 2010, at 01:40 PM, Dannes Wessels wrote:
>>
>>> Hi Loren,
>>>
>>> On 16 Aug 2010, at 19:45 , Loren Cahlander wrote:
>>>
>>>> I could copy the code from the library and implement it myself, but I have a timeframe issue.  I can first implement webdav:proppatch() with the jar and then change over to our own code.  I just have to get something implemented this week.
>>>
>>> Ok, I was not aware of this constraint. Ok, let's do a phased implementation then. We need to consider, if only proppatch() would be in webdav:, that would not make sense. If we can put more functions (logically) into webdav: then that would have my preference. In the end.
>>>
>>> If we can isolate all webdav-functions into one function library, we can probably come away with adding another 3rd party library.....
>>>
>>> Kind regards
>>>
>>> Dannes
>>>
>>> --
>>> eXist-db Native XML Database - http://exist-db.org
>>> Join us on linked-in: http://www.linkedin.com/groups?gid=35624
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>>
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by
>>
>> Make an app they can't live without
>> Enter the BlackBerry Developer Challenge
>> http://p.sf.net/sfu/RIM-dev2dev
>> _______________________________________________
>> Exist-development mailing list
>> Exi...@li...
>> https://lists.sourceforge.net/lists/listinfo/exist-development
>>
>
>
>
> --
> Adam Retter
>
> eXist Developer
> { United Kingdom }
> ad...@ex...
> irc://irc.freenode.net/existdb
>



-- 
Adam Retter

eXist Developer
{ United Kingdom }
ad...@ex...
irc://irc.freenode.net/existdb

Re: [Exist-development] httpclient issues

[Adam R. <ad...@ex...>]

webdavclient would seem appropriate

On 16 August 2010 20:05, Loren Cahlander <lor...@gm...> wrote:
> Hello Dannes,
>
> I will create a new function module in extensions.  Should I call it webdav or webdav-client?
>
> Loren
>
> On Aug 16, 2010, at 01:40 PM, Dannes Wessels wrote:
>
>> Hi Loren,
>>
>> On 16 Aug 2010, at 19:45 , Loren Cahlander wrote:
>>
>>> I could copy the code from the library and implement it myself, but I have a timeframe issue.  I can first implement webdav:proppatch() with the jar and then change over to our own code.  I just have to get something implemented this week.
>>
>> Ok, I was not aware of this constraint. Ok, let's do a phased implementation then. We need to consider, if only proppatch() would be in webdav:, that would not make sense. If we can put more functions (logically) into webdav: then that would have my preference. In the end.
>>
>> If we can isolate all webdav-functions into one function library, we can probably come away with adding another 3rd party library.....
>>
>> Kind regards
>>
>> Dannes
>>
>> --
>> eXist-db Native XML Database - http://exist-db.org
>> Join us on linked-in: http://www.linkedin.com/groups?gid=35624
>>
>>
>>
>>
>>
>>
>>
>>
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> _______________________________________________
> Exist-development mailing list
> Exi...@li...
> https://lists.sourceforge.net/lists/listinfo/exist-development
>



-- 
Adam Retter

eXist Developer
{ United Kingdom }
ad...@ex...
irc://irc.freenode.net/existdb

Re: [Exist-development] httpclient issues

[Loren C. <lor...@gm...>]

Hello Dannes,

I will create a new function module in extensions.  Should I call it webdav or webdav-client?

Loren

On Aug 16, 2010, at 01:40 PM, Dannes Wessels wrote:

> Hi Loren,
> 
> On 16 Aug 2010, at 19:45 , Loren Cahlander wrote:
> 
>> I could copy the code from the library and implement it myself, but I have a timeframe issue.  I can first implement webdav:proppatch() with the jar and then change over to our own code.  I just have to get something implemented this week.
> 
> Ok, I was not aware of this constraint. Ok, let's do a phased implementation then. We need to consider, if only proppatch() would be in webdav:, that would not make sense. If we can put more functions (logically) into webdav: then that would have my preference. In the end. 
> 
> If we can isolate all webdav-functions into one function library, we can probably come away with adding another 3rd party library.....
> 
> Kind regards
> 
> Dannes
> 
> --
> eXist-db Native XML Database - http://exist-db.org
> Join us on linked-in: http://www.linkedin.com/groups?gid=35624
> 
> 
> 
> 
> 
> 
> 
> 

Re: [Exist-development] httpclient issues

[Dannes W. <da...@ex...>]

Hi Loren,

On 16 Aug 2010, at 19:45 , Loren Cahlander wrote:

> I could copy the code from the library and implement it myself, but I have a timeframe issue.  I can first implement webdav:proppatch() with the jar and then change over to our own code.  I just have to get something implemented this week.

Ok, I was not aware of this constraint. Ok, let's do a phased implementation then. We need to consider, if only proppatch() would be in webdav:, that would not make sense. If we can put more functions (logically) into webdav: then that would have my preference. In the end. 

If we can isolate all webdav-functions into one function library, we can probably come away with adding another 3rd party library.....

Kind regards

Dannes

--
eXist-db Native XML Database - http://exist-db.org
Join us on linked-in: http://www.linkedin.com/groups?gid=35624

Re: [Exist-development] httpclient issues

[Loren C. <lor...@gm...>]

Hello Dannes,

I do not have any problem with implementing a WebDAV client function module.  I am planning on implementing all of the HTTP methods for WebDAV client.  I can implement the PROPPATCH as webdav:proppatch().  I just need to get it implemented.

I could copy the code from the library and implement it myself, but I have a timeframe issue.  I can first implement webdav:proppatch() with the jar and then change over to our own code.  I just have to get something implemented this week.

Loren


On Aug 16, 2010, at 12:35 PM, Dannes Wessels wrote:

> Hi,
> 
> 
> Hmm the log() dependency, that is a strange one... what is happening there?
> 
> What is the idea of having a PROPPATCH method? What about the other webdav METHODS like PROPFIND? Is it a good idea to have webdav specific functionality in a generic HTTP client?  WebDAV is HTTP but HTTP is not  WebDAV.... are you sure both map on each other well enough?
> 
> If you want to implement a webdav client, maybe we need to have a different module, specific for Webdav? 
> 
> 
> On 16 Aug 2010, at 18:30 , Loren Cahlander wrote:
> 
>> There is also another issue.  The commons-httpclient-3.1.jar has been superseded with the HttpComponents project http://hc.apache.org/.
> 
> version 3.x has been the primary library for a long time. Until 6 months ago or so, the 4.x release was not good usable for production. Good that things change over time :-) Adam has experience with the 4.x release in our attempt to introduce jsr-225 aka xqj.
> 
> 
>> What should be our plan of attack?
>> I would like to add the downloading of the webdavlib-2.0.jar in the building of the httpclient extension module so that I can get the PROPPATCH method implemented.
> 
> Why? I estimate that the PROPPATCH layer is very thin, I would vote against another library with limited added value. I guess it can be coded ourself easily. Or ultimately copy some code from an other distribution?
> 
> 
>> We can then discuss HttpComponents and Axis2.
> 
> Sorry i don't see the relation with axis2; upgrading axis to axis2 does imply lot's of work [check of dependancies etc], and probably some risks. Not sure what the added value is compared to 'the costs'. And we have the SOAPserver which is a better powerful interface....
> 
> Kind regards
> 
> Dannes
> 
> --
> eXist-db Native XML Database - http://exist-db.org
> Join us on linked-in: http://www.linkedin.com/groups?gid=35624
> 
> 
> 
> 
> 
> 
> 
> 

Re: [Exist-development] httpclient issues

[Dannes W. <da...@ex...>]

Hi,


Hmm the log() dependency, that is a strange one... what is happening there?

What is the idea of having a PROPPATCH method? What about the other webdav METHODS like PROPFIND? Is it a good idea to have webdav specific functionality in a generic HTTP client?  WebDAV is HTTP but HTTP is not  WebDAV.... are you sure both map on each other well enough?

If you want to implement a webdav client, maybe we need to have a different module, specific for Webdav? 


On 16 Aug 2010, at 18:30 , Loren Cahlander wrote:

> There is also another issue.  The commons-httpclient-3.1.jar has been superseded with the HttpComponents project http://hc.apache.org/.

version 3.x has been the primary library for a long time. Until 6 months ago or so, the 4.x release was not good usable for production. Good that things change over time :-) Adam has experience with the 4.x release in our attempt to introduce jsr-225 aka xqj.


> What should be our plan of attack?
> I would like to add the downloading of the webdavlib-2.0.jar in the building of the httpclient extension module so that I can get the PROPPATCH method implemented.

Why? I estimate that the PROPPATCH layer is very thin, I would vote against another library with limited added value. I guess it can be coded ourself easily. Or ultimately copy some code from an other distribution?


> We can then discuss HttpComponents and Axis2.

Sorry i don't see the relation with axis2; upgrading axis to axis2 does imply lot's of work [check of dependancies etc], and probably some risks. Not sure what the added value is compared to 'the costs'. And we have the SOAPserver which is a better powerful interface....

Kind regards

Dannes

--
eXist-db Native XML Database - http://exist-db.org
Join us on linked-in: http://www.linkedin.com/groups?gid=35624

[Exist-development] httpclient issues

[Loren C. <lor...@gm...>]

Hello folks,

I am needing to add the HTTP method PROPPATCH to the httpclient function module.  That requires adding a WebDAV client jar file.  I originally was going to add webdavlib-2.0.jar, but Dannes informed me that it is orphaned.  I am looking at adding the jackrabbit-standalone-2.1.1.jar, but am encountering a compilation error.

In all of these files there appears to be a problem with the org.apache.tools.ant.Task log method:

src/org/exist/ant/UserTask.java
src/org/exist/ant/AddUserTask.java
src/org/exist/ant/BackupTask.java
src/org/exist/ant/ChmodTask.java
src/org/exist/ant/ChownTask.java
src/org/exist/ant/ListGroupsTask.java
src/org/exist/ant/ListUsersTask.java
src/org/exist/ant/LockResourceTask.java
src/org/exist/ant/RemoveUserTask.java
src/org/exist/ant/RestoreTask.java
src/org/exist/ant/UserPasswordTask.java
src/org/exist/ant/XMLDBCopyTask.java
src/org/exist/ant/XMLDBCreateTask.java
src/org/exist/ant/XMLDBExtractTask.java
src/org/exist/ant/XMLDBListTask.java
src/org/exist/ant/XMLDBMoveTask.java
src/org/exist/ant/XMLDBQueryTask.java
src/org/exist/ant/XMLDBRemoveTask.java
src/org/exist/ant/XMLDBShutdownTask.java
src/org/exist/ant/XMLDBStoreTask.java
src/org/exist/ant/XMLDBXPathTask.java
src/org/exist/ant/XMLDBXQueryTask.java
src/org/exist/ant/XMLDBXUpdateTask.java


    [javac] symbol  : method log(java.lang.String,org.xmldb.api.base.XMLDBException,int)
    [javac] location: class org.exist.ant.XMLDBXUpdateTask
    [javac]                 log(msg, e, Project.MSG_ERR);


I think that the simplest approach would be to use webdavlib-2.0.jar since we are also using orphaned axis code as well.

There is also another issue.  The commons-httpclient-3.1.jar has been superseded with the HttpComponents project http://hc.apache.org/.

What should be our plan of attack?
I would like to add the downloading of the webdavlib-2.0.jar in the building of the httpclient extension module so that I can get the PROPPATCH method implemented.

We can then discuss HttpComponents and Axis2.

Cheers,
Loren

Re: [Exist-development] test-compile fails

[Dannes W. <da...@ex...>]

Hi Joern,

this is an ordering issue in the build. You need to build exist first, then the tests.

cheers

Dannes

On 16 Aug 2010, at 17:55 , Joern Turner wrote:

> We have done our mods and tried to run tests but there is a compile
> error in test-compile:
> 
> CounterTest.java:9: package org.exist.xquery.modules.counter does not
> exist and so forth
> 
> Can someone confirm that? We have swtiched off all of our stuff so it
> does not seem to have to do with the betterFORM extension

Kind regards

Dannes

--
eXist-db Native XML Database - http://exist-db.org
Join us on linked-in: http://www.linkedin.com/groups?gid=35624

[Exist-development] test-compile fails

[Joern T. <joe...@go...>]

We have done our mods and tried to run tests but there is a compile
error in test-compile:

CounterTest.java:9: package org.exist.xquery.modules.counter does not
exist and so forth

Can someone confirm that? We have swtiched off all of our stuff so it
does not seem to have to do with the betterFORM extension