java - 有人成功使用 JPype 加载 Netflix 光子类吗？

Question

我从 https://github.com/Netflix/photon 构建了 Netflix Photon docker 镜像它的工作原理就像我能够启动容器并运行提供的 java 应用程序一样。

除此之外，我安装了 Python3、gcc 和 JPype1-py3，但无法加载 Photon java 类。

我可以使用以下方法成功启动 JVM:

startJVM(getDefaultJVMPath(), "-ea", "-Djava.class.path=/source/build/libs") java.lang.System.out.println("hello world") Hello World

所以我知道 JVM 本身正在工作，但每当我尝试从 Photon 加载类时，我都会收到错误:

>>> s = JClass("com.netflix.imflibrary.st2067_2.ImpAsset")
File "<stdin>", line 1, in <module>
File "/usr/local/lib/python3.5/dist-packages/jpype/_jclass.py", line 55, in JClass
  raise _RUNTIMEEXCEPTION.PYEXEC("Class %s not found" % name)
jpype._jexception.ExceptionPyRaisable: java.lang.Exception: Class com.netf;ox/imflibrary.st2067_2.IMPAsset not found

有人能够使用 JPype 直接与 Netflix 光子类集成吗？我宁愿不限于通过子进程运行提供的应用程序。

Answer 1

蟋蟀!??

好吧，无论如何，我想通了，必须像这样加载 jar 文件:

try:
    from jpype import *
    classpath="/source/build/libs/Photon-0.1.0-SNAPSHOT.jar:/source/build/libs/log4j-1.2.17.jar:/source/build/libs/regxmllib-1.1.1.jar:/source/build/libs/slf4j-log4j12-1.7.25.jar:/source/build/libs/jsr305-3.0.1.jar:/source/build/libs/log4j.properties:/source/build/libs/slf4j-api-1.7.25.jar"
    startJVM(getDefaultJVMPath(), "-Djava.class.path=%s" % classpath)
    print("JVM started: ",getDefaultJVMPath())
    print("classpath=",classpath)
except Exception as e:
print("error launching JVM:", traceback.format_exc())
exit()

print("Starting scan of",scanfolder)
IMPAnalyzer = JClass("com.netflix.imflibrary.app.IMPAnalyzer")
IMPFile = java.io.File(scanfolder)
error_fatal = 0
error_nonfatal = 0
error_warning = 0
if IMPFile.isDirectory():
    result_dict = {}
    IMPresult_iterator = IMPAnalyzer.analyzePackage(IMPFile).entrySet().iterator()
    while IMPresult_iterator.hasNext():
        resultpair = IMPresult_iterator.next()
        # print(resultpair.getKey(),":")
        if len(resultpair.getValue()) > 0:
            error_list = []
            list_iterator = resultpair.getValue().iterator()
            while list_iterator.hasNext():
                errorobject = list_iterator.next()
                error_list.append(str(errorobject.errorLevel) + ":" + str(errorobject.errorCode) + ":" + errorobject.toString())
                if str(errorobject.errorLevel) == "WARNING":
                        error_warning += 1
                elif str(errorobject.errorLevel) == "NON FATAL":
                        error_nonfatal += 1
                elif str(errorobject.errorLevel) == "FATAL":
                        error_fatal += 1
            result_dict[resultpair.getKey()] = error_list
        else:
            result_dict[resultpair.getKey()]='Ok'
    print(scanfolder,"scanned with",error_fatal,"fatal error(s),",error_nonfatal,"nonfatal error(s), and",error_warning,"warning(s)")
    pprint.pprint(result_dict)
else:
    print(scanfolder,"does not appear to be a valid folder")