Repository-Candidates #1
Replies: 11 comments
-
|
From Stanford CoreNLP good ones: stanfordnlp/CoreNLP#1160 Less good ones stanfordnlp/CoreNLP#550 But they all have no related commits, so one must crawl through the affected pieces and changes. |
Beta Was this translation helpful? Give feedback.
-
|
There are 2 issues in DatumBox labelled bug They both are well documented and have a clear 'buggy version' as well as most of things for a test, but they do not have a clear tagged fix version. They are "just" closed in versions to be fixed, which is documented. It is Java 11 everyone get the party glasses on ! 😎 |
Beta Was this translation helpful? Give feedback.
-
|
JSAT has (many) bugs older than 2017, atleast from the issues. More recent bugs are There are also some PRs EdwardRaff/JSAT#89 They are from the same guy who is working on an analysis tool, reaching out to people https://github.com/Egor18/jdataflow JSAT is Java 8, and Java 9 is not on the ToDo list. |
Beta Was this translation helpful? Give feedback.
-
|
In Botsing I saw only one good candidate On a sidenote, Botsing could maybe be used to generate tests for some of the more complex issues e.g. in CoreNLP |
Beta Was this translation helpful? Give feedback.
-
|
Grobid looks huge, but most of the issues are requests or build related. These seem tested, but a bit crowded (very) hard to test: |
Beta Was this translation helpful? Give feedback.
-
|
Randoop is Java 8 and also talks about using Java 11 in randoop/randoop#523. Some (good) issues closed: Hard to test: This one has been fixed sideways with a newer version: randoop/randoop#716 But most of the Issues are marked as "not reproducible", which ... hmm. The project moved to Java 8 in 2018, so things before are Java 7 or likely related to it. The PRs are super flooded with Dependabot, maybe I should keep that in mind and not use that bot. This one overhauls all equals methods randoop/randoop#624 but no tests, so that is not truly a bug, but more a change of features. Potential ones are |
Beta Was this translation helpful? Give feedback.
-
|
Meka is a rather small project and a wrapper around the popular WEKA classifiers. One Nullpointer fixed sideways This Waikato/meka#43 is a perfect algorithmic example of an error, but it is from 2017 too One Issue from 2017 Waikato/meka#38 |
Beta Was this translation helpful? Give feedback.
-
|
As I do not speak Chinese, EasyML seems rather unfit. |
Beta Was this translation helpful? Give feedback.
-
|
Most of EvoSuite is about not being able to run it (this is the primary theme of everything around SE-tasks in Java). Good issues are other issues are Maybe hard to test are This one EvoSuite/evosuite#16 persisted from 2016, but was fixed 2020. So the related commits should be in range. |
Beta Was this translation helpful? Give feedback.
-
|
Smile is huge and also has a wild build tool (.sbt? should be scala). Good ones Ok ones are Hard ones are Most of these bugs are the kind of ML Bugs we are really going for. I checked up until 2017. |
Beta Was this translation helpful? Give feedback.
-
|
Jenetics is Java 11 and uses Gradle + TestNG, and is hence the first with something else than Junit. Good ones are Ok ones are Hard to test ones are In general it looks very nice and could in itself be a motivation to allow more than just JUnit for testing. Many tools will not like this (e.g. astor) but well, so be it? |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
A list of potential Repositories, in descending order of interest
Keywords searched (on Github, filter by java, sorted by stars) :
Beta Was this translation helpful? Give feedback.
All reactions