David,

I need to go double check, but I believe there is another error
in the system relating what may be a similar error. In addition,
there is a recent change which may be a cause. So, I'm hoping
there is enough info in your report in this case for me to make
progress.

I'll post info about issue numbers in due course.

-- Jon

On 04/10/2013 11:55 PM, Dawid Weiss wrote:
Thanks Jon. Let me know if you need a (small) reproducible example of
(2), I'll try to prepare it. This sequence of commands also reproduces
this issue but it's far from a small isolated example...

git clone git://github.com/carrot2/carrot2.git
# go grab coffee, takes a while...
cd carrot2
ant javadoc

This will result in plenty if warnings about missing tags but also the
issues I mentioned (and since they're errors it'll stop the
compilation process).

Let me know if I can be of any help,
Dawid

On Thu, Apr 11, 2013 at 2:30 AM, Jonathan Gibbons
<jonathan.gibb...@oracle.com> wrote:
Both seem odd.

I will investigate.  Thanks for the reports.

You can probably work around the issues by disabling doclint for now,
-Xdoclint:none.

-- Jon


On 04/10/2013 06:31 AM, Dawid Weiss wrote:
Hi there,

I've switched to the latest 1.8 snapshot (b84) and javadoc went nuts
about the following:

1)

/**
   * @see "http://www.google.com";
   */
public class Main {}

C:\carrot2\carrotsearch.lingo3g\carrot2\tmp>javadoc -quiet  Main.java
Main.java:3: error: unexpected content
   * @see "http://www.google.com";
     ^
1 error

This seems odd. Then there's this thing which I didn't try to reproduce
yet:

2)

    [javadoc]
C:\carrot2\carrotsearch.lingo3g\carrot2\core\carrot2-core\src\org\carrot2\core\Cluster.java:543:
error: missing method body, or declare abstract
    [javadoc]         .nullsFirst().onResultOf(new Function<Cluster,
Integer>(){
    [javadoc]
^

on a snippet of code that creates an inner class (and compiles cleanly).

      public static final Comparator<Cluster> BY_SIZE_COMPARATOR =
Ordering.natural()
          .nullsFirst().onResultOf(new Function<Cluster, Integer>(){
              public Integer apply(Cluster cluster)
              {
                  return cluster.size();
              }
          });

Like I said, I couldn't reproduce it on a small example but it's
definitely deterministic. Any clues? Are these known issues or should
I file a bug (and try to write a reproducible snippet)?

Dawid


Reply via email to