You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@abdera.apache.org by James M Snell <ja...@gmail.com> on 2006/09/29 23:41:04 UTC
IRI Code
Ok all, for those not monitoring the SVN commits, I just checked in a
large update that replaces our use of java.net.URI with the new IRI
implementation. This is an extensive change that needs to be well
tested. All of the existing test cases pass* and new test cases
covering the new IRI/IDNA/Unicode support have been added to the core
test suite.
I would like to ask anyone and everyone to please grab the latest from
SVN and test like crazy. I know for sure that we'll need to go in and
work on optimizing the IRI implementation -- that is, it's works but can
most definitely be improved.
Just a recap: the new IRI support gives us full support for
internationalized resource identifiers and internationalized domain
names. It does automatic conversion to and from US-ASCII, groks Unicode
normalization and implements the Punycode standard. And it seems to
actually work properly :-)
- James
* Btw, I will be looking at the client test cases that are causing our
build process to hang this weekend
Re: IRI Code
Posted by Garrett Rooney <ro...@electricjellyfish.net>.
On 9/29/06, James M Snell <ja...@gmail.com> wrote:
> Out of curiosity, are you testing on Windows or Linux?
>
> I'm not seeing these problems at all. I'm running them in Eclipse and
> via Ant. (on Linux).
This is on Linux, Sun JDK:
$ java -version
java version "1.5.0_05"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_05-b05)
Java HotSpot(TM) Client VM (build 1.5.0_05-b05, mixed mode, sharing)
Re: IRI Code
Posted by James M Snell <ja...@gmail.com>.
Out of curiosity, are you testing on Windows or Linux?
I'm not seeing these problems at all. I'm running them in Eclipse and
via Ant. (on Linux).
- James
Garrett Rooney wrote:
> On 9/29/06, James M Snell <ja...@gmail.com> wrote:
>> Ok all, for those not monitoring the SVN commits, I just checked in a
>> large update that replaces our use of java.net.URI with the new IRI
>> implementation. This is an extensive change that needs to be well
>> tested. All of the existing test cases pass* and new test cases
>> covering the new IRI/IDNA/Unicode support have been added to the core
>> test suite.
>>
>> I would like to ask anyone and everyone to please grab the latest from
>> SVN and test like crazy. I know for sure that we'll need to go in and
>> work on optimizing the IRI implementation -- that is, it's works but can
>> most definitely be improved.
>>
>> Just a recap: the new IRI support gives us full support for
>> internationalized resource identifiers and internationalized domain
>> names. It does automatic conversion to and from US-ASCII, groks Unicode
>> normalization and implements the Punycode standard. And it seems to
>> actually work properly :-)
>
> FWIW, I'm seeing two test failures in the IRI code's tests.
> Everything else tests cleanly (other than the hanging client stuff,
> which I've got commented out).
>
> [echo] Running Core Tests...
> [java] .........F.......F.....
> [java] Time: 0.86
> [java] There were 2 failures:
> [java] 1)
> testURItoIRI2(org.apache.abdera.test.iri.TestIRI)junit.framework.ComparisonFailure:
>
> expected:<...> but was:<...%FCrst>
> [java] at
> org.apache.abdera.test.iri.TestIRI.testURItoIRI2(TestIRI.java:56)
> [java] at sun.reflect.NativeMethodAccessorImpl.invoke0(Native
> Method)
> [java] at
> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
>
> [java] at
> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
>
> [java] at
> org.apache.abdera.test.core.TestSuite.main(TestSuite.java:30)
> [java] 2)
> testMixedEncodings(org.apache.abdera.test.iri.TestIRI)junit.framework.ComparisonFailure:
>
> expected:<...�sum�...> but was:<...%E9sum%E9...>
> [java] at
> org.apache.abdera.test.iri.TestIRI.testMixedEncodings(TestIRI.java:122)
> [java] at sun.reflect.NativeMethodAccessorImpl.invoke0(Native
> Method)
> [java] at
> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
>
> [java] at
> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
>
> [java] at
> org.apache.abdera.test.core.TestSuite.main(TestSuite.java:30)
>
>
>> * Btw, I will be looking at the client test cases that are causing our
>> build process to hang this weekend
>
> Thanks, I've been meaning to get around to that...
>
> -garrett
Re: IRI Code
Posted by Ugo Cei <ug...@gmail.com>.
On Sep 30, 2006, at 9:08 PM, James M Snell wrote:
> Can you post a Jira issue with the full details of the errors you are
> seeing?
See https://issues.apache.org/jira/browse/ABDERA-7
Ugo
Re: IRI Code
Posted by James M Snell <ja...@gmail.com>.
Hey Ugo,
Can you post a Jira issue with the full details of the errors you are
seeing?
- James
Ugo Cei wrote:
>
> On Sep 29, 2006, at 11:50 PM, Garrett Rooney wrote:
>
>> FWIW, I'm seeing two test failures in the IRI code's tests.
>> Everything else tests cleanly (other than the hanging client stuff,
>> which I've got commented out).
>>
>> [echo] Running Core Tests...
>> [java] .........F.......F.....
>
> I'm seeing many more failures on OSX:
>
> test:
> [echo] Running Core Tests...
> [java] .....F..F.F.F.F.F.....F...F.F.
> [java] Time: 1.169
> [java] There were 9 failures:
> [java] 1)
> testSimple(org.apache.abdera.test.iri.TestIRI)junit.framework.ComparisonFailure:
> expected:<...?sum?...> but was:<...ésumé...>
>
> <snip/>
>
> [java] FAILURES!!!
> [java] Tests run: 21, Failures: 9, Errors: 0
>
> Plus one more error in the Stax parser tests, which again seems related
> to character encoding:
>
> [java] There was 1 failure:
> [java] 1)
> testSection4275LinkTitleWithBadchars(org.apache.abdera.test.parser.stax.FeedValidatorTest)junit.framework.ComparisonFailure:
> expected:<...??test.?> but was:<...£test.>
>
> Ugo
>
>
> --Ugo Cei
> Blog: http://agylen.com/
> Open Source Zone: http://oszone.org/
> Company: http://www.sourcesense.com/
>
>
>
Re: IRI Code
Posted by Ugo Cei <ug...@gmail.com>.
On Sep 29, 2006, at 11:50 PM, Garrett Rooney wrote:
> FWIW, I'm seeing two test failures in the IRI code's tests.
> Everything else tests cleanly (other than the hanging client stuff,
> which I've got commented out).
>
> [echo] Running Core Tests...
> [java] .........F.......F.....
I'm seeing many more failures on OSX:
test:
[echo] Running Core Tests...
[java] .....F..F.F.F.F.F.....F...F.F.
[java] Time: 1.169
[java] There were 9 failures:
[java] 1) testSimple(org.apache.abdera.test.iri.TestIRI)
junit.framework.ComparisonFailure: expected:<...?sum?...> but
was:<...ésumé...>
<snip/>
[java] FAILURES!!!
[java] Tests run: 21, Failures: 9, Errors: 0
Plus one more error in the Stax parser tests, which again seems
related to character encoding:
[java] There was 1 failure:
[java] 1) testSection4275LinkTitleWithBadchars
(org.apache.abdera.test.parser.stax.FeedValidatorTest)
junit.framework.ComparisonFailure: expected:<...??test.?> but was:<...
£test.>
Ugo
--
Ugo Cei
Blog: http://agylen.com/
Open Source Zone: http://oszone.org/
Company: http://www.sourcesense.com/
Re: IRI Code
Posted by Garrett Rooney <ro...@electricjellyfish.net>.
On 9/29/06, James M Snell <ja...@gmail.com> wrote:
> Ok all, for those not monitoring the SVN commits, I just checked in a
> large update that replaces our use of java.net.URI with the new IRI
> implementation. This is an extensive change that needs to be well
> tested. All of the existing test cases pass* and new test cases
> covering the new IRI/IDNA/Unicode support have been added to the core
> test suite.
>
> I would like to ask anyone and everyone to please grab the latest from
> SVN and test like crazy. I know for sure that we'll need to go in and
> work on optimizing the IRI implementation -- that is, it's works but can
> most definitely be improved.
>
> Just a recap: the new IRI support gives us full support for
> internationalized resource identifiers and internationalized domain
> names. It does automatic conversion to and from US-ASCII, groks Unicode
> normalization and implements the Punycode standard. And it seems to
> actually work properly :-)
FWIW, I'm seeing two test failures in the IRI code's tests.
Everything else tests cleanly (other than the hanging client stuff,
which I've got commented out).
[echo] Running Core Tests...
[java] .........F.......F.....
[java] Time: 0.86
[java] There were 2 failures:
[java] 1) testURItoIRI2(org.apache.abdera.test.iri.TestIRI)junit.framework.ComparisonFailure:
expected:<...> but was:<...%FCrst>
[java] at
org.apache.abdera.test.iri.TestIRI.testURItoIRI2(TestIRI.java:56)
[java] at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
[java] at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
[java] at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
[java] at
org.apache.abdera.test.core.TestSuite.main(TestSuite.java:30)
[java] 2) testMixedEncodings(org.apache.abdera.test.iri.TestIRI)junit.framework.ComparisonFailure:
expected:<...�sum�...> but was:<...%E9sum%E9...>
[java] at
org.apache.abdera.test.iri.TestIRI.testMixedEncodings(TestIRI.java:122)
[java] at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
[java] at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
[java] at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
[java] at org.apache.abdera.test.core.TestSuite.main(TestSuite.java:30)
> * Btw, I will be looking at the client test cases that are causing our
> build process to hang this weekend
Thanks, I've been meaning to get around to that...
-garrett