Activity
From 02/16/2015 to 03/17/2015
03/09/2015
-
03:11 PM
Bug #2560 (Closed): mvn integration-test doesn't seem to run the tests
- Great, thanks. I merged to master and will make a Maven release.
-
12:18 PM Bug #2560: mvn integration-test doesn't seem to run the tests
- Jeff, I believe that was an auto-refactor artifact and not intentional; I removed them in the latest commit.
-
11:43 AM
Bug #2560 (In Progress): mvn integration-test doesn't seem to run the tests
- Hi Andrew. Sounds good and thanks for using the convention for the branch name. It works for me after updating a few ...
-
- Jeff, check out the branch I published. I had to move the integration-tests back under the global tests folder for fa...
03/06/2015
-
02:39 PM Feature #2565: Implement async operations for Face.processEvents
- On my list it is near the top. Infinite loops with poll and sleep operations, specifically in android case would unn...
03/05/2015
-
11:56 AM
Task #2619 (Closed): Data.setName, etc. do not copy the object
-
10:59 AM
Task #2619 (Closed): Data.setName, etc. do not copy the object
- The API doc for Data.setName, etc. say that the method copies the object: http://named-data.net/doc/ndn-ccl-api/data....
03/04/2015
-
11:04 PM Feature #2565: Implement async operations for Face.processEvents
- +1. Any thoughts on priority relative to other issues?
-
- Makes sense to consider this. How much of a priority do you think it is?
02/27/2015
-
03:59 PM
Task #2545 (Closed): Blob should support hashCode()
- The change for Blob.hashCode to use ByteBuffer.hashCode is merged into the master branch.
02/26/2015
-
- No worries, I understand the concerns.
-
- Oh. ok, I'm got mislead by the question and like this approach. Please ignore my notes 28, 29.
-
- Look down to the answer to that question: the OP was not filling the buffer, therefore the hash was always the same.
-
- Andrew Brown wrote:
> I don't understand: in #28 you say ByteBuffer.hashCode() is not a real hash but it actually ... -
- I don't understand: in #28 you say ByteBuffer.hashCode() is not a real hash but it actually is (see #25, just not a c...
-
- Just forgot to add that we would still need to implement real (as it note-25) hashCode() for name and name components...
-
- ByteBuffer.hashCode() doesn't do any real hashing (value depends on the number of remaining bytes). I don't mind usi...
-
- Nice, I like it. Alex?
-
09:04 AM
Task #2545: Blob should support hashCode()
- The ByteBuffer position and limit will not change, nor will its contents. I hadn't noticed that ByteBuffer implement...
-
- I understand and that makes sense, just note from previous comments that I was advocating using the same hash as Stri...
02/25/2015
-
- Not sure about java-way, I'm kind of against reimplementing the wheel. Implementing hashing of the data buffer ourse...
-
- I still think hashCode() should be implemented in Blob (especially since it doesn't hurt to have it there) but... I d...
-
- I read the definitions and still think that it should not be generally defined for Blob (the use of Blob in hash tabl...
-
- It's more a Java-ism than anything else; see discussion above about why but basically the equals() contract assumes t...
-
03:14 PM
Task #2545 (In Progress): Blob should support hashCode()
- Re-open to address Alex's questions.
-
03:13 PM
Task #2545 (Closed): Blob should support hashCode()
-
- Why "Blob" abstract needs to be hashed? I can understand that name and name component should be hashable as they cou...
-
- I vote yes.
-
02:54 PM
Task #2545: Blob should support hashCode()
- Hi Andrew, I added basic unit tests.
https://github.com/named-data/jndn/commit/1ae2cb96380295c93dd91f8e375ae79177d07... -
- +1
-
02:17 PM
Task #2545: Blob should support hashCode()
- OK, I reverted Blob.equals to not use hashCode().
Let me know what you think of Name.hashCode() here:<br>
https:/... -
- #11, got it. #12, yes, that is more complete; I would still say leave it out because we can't guess how Blob will be ...
-
12:05 PM
Task #2545: Blob should support hashCode()
- In your argument to remove hashCode() from equals(), you say "If the Blobs are unequal, we will have to scan the enti...
-
12:02 PM
Task #2545: Blob should support hashCode()
- > Why do we have a type-specific equals?
To avoid an unnecessary call to instanceof if possible. Also, the CCL AP... -
- Jeff Thompson wrote:
>
> Do you mean equals(Object) which is used by HashMap, etc.? It will return false for !(... -
11:35 AM
Task #2545: Blob should support hashCode()
- > 1. When we add tests, can we add a test for equals(null); I "think" the implementation would return false but I'm n...
-
- This is good stuff. A few comments:
1. When we add tests, can we add a test for equals(null); I "think" the implem... -
10:13 AM
Task #2545: Blob should support hashCode()
- I made a branch issue/2545-Blob-hashCode. Please see the hashCode() method here:<br>
https://github.com/named-data/j... -
- Well, I'm not using it for equals() and it's a minor optimization, but I want to use a HashMap<Name, ...> to avoid lo...
-
08:09 AM
Task #2545: Blob should support hashCode()
- Sure we can. But I'm curious where do you use Name.equals in your application that a hash code can provide a signific...
-
- Callbacks such as OnData, OnInterest, ... and many others should not be exposed directly under net.named_data.jndn na...
-
- Right now, one is required to have an infinite loop with with polling of processEvents.
I believe, this can be don... -
12:07 PM
Bug #2560: mvn integration-test doesn't seem to run the tests
- Can you jump in with adding Failsafe to the global POM?
-
- Sounds good; why don't you add the failsafe XML to the global POM (let me know if you want me to jump in on that) and...
-
11:40 AM
Bug #2560: mvn integration-test doesn't seem to run the tests
- I think we should use the Failsafe plugin so that "mvn verify" at the top level runs all the integration tests. We c...
-
- In PR https://github.com/named-data/jndn/pull/6, I did add a sub-POM that closely mirrors the one in /examples; basic...
-
- Good call. I thought we put integration-test stuff in there... but I'm looking now and... nope. I will look into this...
-
11:18 AM
Bug #2560 (Closed): mvn integration-test doesn't seem to run the tests
- When I run "mvn integration-test" it doesn't seem to run the tests. I can even delete the integration-tests folder an...
02/24/2015
-
- Are we also considering adding hashCode() to Name? That would be helpful.
-
11:03 AM
Task #2545: Blob should support hashCode()
- I agree that it doesn't have to be a cryptographic hash. From what I've seen, hashCode() is only meant for hash maps...
-
- From what I can tell, the simplest and quickest hashCode() is the Java String version:
for (int i = 0; i < val... -
01:15 PM
Bug #2546 (Closed): Incorrect SCM url in pom.xml
- As requested, changed scm:git:git@github.com to the public scm:git:https://github.com according to http://maven.apach...
-
10:56 AM
Feature #2558 (Closed): Allow MemoryIdentityStorage to set/get defaults
- Your use of SecurityExceptions looks OK to me. I merged the pull request.
-
- (Alex, still no jndn project in Gerrit or I would have posted this there.) Here is a PR (https://github.com/named-dat...
-
- I like to test with MemoryIdentityStorage and MemoryPrivateKeyStorage so that I don't break the (I haven't identified...
02/21/2015
02/20/2015
-
- Yes, I just used it today and it worked as expected. I think we should close this for jndn.
02/19/2015
-
- You can change it to the https:// version according to http://maven.apache.org/scm/git.html but I also have no idea "...
-
- pom.xml currently defines scm block as
<scm>
<connection>scm:git:git@github.com:named-data/jndn.git</connectio... -
11:51 AM
Task #2545: Blob should support hashCode()
- We should use an efficient hash function like City/Farm Hash, but one that the Java library natively supports. What ...
-
11:51 AM
Task #2545 (Closed): Blob should support hashCode()
- The Blob class should support hashCode() with an efficient hash function of the content. This way, equals can use it.
-
08:47 AM
Task #2374 (Closed): Add Maven Support
- Version 0.3 is tagged in GitHub and is deployed to the Maven Central Repository.
02/18/2015
-
12:45 PM
Task #2374: Add Maven Support
- Hi Andrew. I think I fixed it. I added the protobuf dependency section to the ossrh profile. Is that right? https://...
-
11:56 AM
Task #2374: Add Maven Support
- ... I should ask you to first pull the latest from the main jndn branch so that it is the same pom.xml with the chang...
-
11:55 AM
Task #2374: Add Maven Support
- That pom.xml for building examples works great. I merged the changes into the main branch so that I could tag releas...
-
- It makes some assumptions that they are all in the same package, etc., but that simplifies the POM a bit.
-
- Yes, I think so. See this attachment; I dispensed with the profiles all together and just went with a command-line pr...
-
09:13 AM
Task #2374 (In Progress): Add Maven Support
-
09:12 AM
Task #2374: Add Maven Support
- When I removed version, Maven complained. So I simply set the version to 'test'. Also, to avoid confusion I changed ...
-
- I think you can remove it because it should inherit from the parent POM (according to http://maven.apache.org/pom.htm...
-
08:49 AM
Task #2374: Add Maven Support
- Great! The pom.xml in examples works now. Since we only need to run examples and not deploy, can we remove the 'vers...
02/17/2015
-
04:11 PM
Task #2374: Add Maven Support
- Thanks for the quick feedback. Sounds like a good solution. Can you attach a revised pom.xml?
-
- Makes sense; I wouldn't want to install it either. The problem is that the OSSRH deployment requirements use the GPG ...
-
02:54 PM
Task #2374: Add Maven Support
- Hi Andrew. Thanks for the pom.xml but I couldn't get it to work on OS X. 'mvn install' failed because gpg was not ins...
-
- Jeff, I attached one. I put this pom.xml in /examples and I ran it with 'mvn test -P encode-decode-fib'. As long as I...
Also available in: Atom