java 从内存和硬盘驱动器的Zip是不同的，这导致下载后损坏的Zip

jaxagkaj  于 2024-01-05  发布在  Java
关注(0)|答案(1)|浏览(219)

问题已修复。请参见下文。

嘿，大家好，我想实现一个zip下载。我有一个Web应用程序，用户可以点击一个按钮：

  1. <template>
  2.   <button name="exportButton" id="exportButton" @click="convertConfigurationToZip" :class="['button']">{{
  3.       exportConfig
  4.     }}
  5.   </button>
  6. </template>
  8. methods: {
  9.     async convertConfigurationToZip() {
  10.       // Stores the current config in "store.config"
  11.       RestResource.convertConfigurationToZip();
  13.       const configData = store.config;
  14.       await new Promise((resolve) => setTimeout(resolve, 2000));
  16.       // A Blob represents raw binary data
  17.       const blob = new Blob([configData], {type: 'application/zip'});
  19.       // Use file-saver to trigger the download
  20.       saveAs(blob, 'config')

字符串
在RestResource中调用的函数如下：

  1. convertConfigurationToZip: () => {
  2.         axios
  3.             .get("http://localhost:8080/api/conversion/convertConfigurationToZip/")
  4.             .then((response: any) => {
  5.                 console.log(response.data);
  6.                 store.config = response.data;
  7.             })
  8.             .catch((error: any) => {
  9.                 console.log(error);
  10.             })
  11.     }

型
被调用的Controller看起来像这样：

  1. package controller;
  3. import io.swagger.annotations.*;
  4. import jakarta.annotation.Resource;
  5. import jakarta.ws.rs.GET;
  6. import jakarta.ws.rs.Path;
  7. import jakarta.ws.rs.container.AsyncResponse;
  8. import jakarta.ws.rs.container.Suspended;
  9. import jakarta.ws.rs.core.Response;
  10. import services.ConversionService;
  12. import java.util.concurrent.ExecutorService;
  13. import java.util.concurrent.Executors;
  15. // This Controller should be used for creating zip files that contains the ProCake-Configuration
  16. @Api(tags = "Conversion Controller", authorizations = { @Authorization(value = "basicAuth") })
  17. @Path("/conversion")
  18. public class ConversionController
  19. {
  20.     @Resource
  21.     ExecutorService executorService;
  23.     @ApiOperation(value = "Convert Config into Zip")
  24.     @ApiResponses({ @ApiResponse(code = 200, message = "Config as Zip"), @ApiResponse(code = 503, message = "ProCAKE not started/configured properly") })
  25.     @GET
  26.     @Path("/convertConfigurationToZip")
  27.     public void convertConfigurationToZip(@Suspended final AsyncResponse response)
  28.     {
  29.         executorService = Executors.newSingleThreadExecutor();
  30.         executorService.submit(() ->
  31.         {
  32.             try
  33.             {
  34.                 response.resume(Response.status(200).entity(new ConversionService().convertConfigurationToZip()).build());
  35.             }
  36.             catch (Exception e)
  37.             {
  38.                 response.resume(e);
  39.             }
  40.             executorService.shutdown();
  41.         });
  42.     }
  43. }

型
正在处理Zip文件的服务看起来像这样：

  1. package services;
  3. import de.uni_trier.wi2.procake.data.model.Model;
  4. import de.uni_trier.wi2.procake.data.model.ModelFactory;
  5. import de.uni_trier.wi2.procake.similarity.SimilarityModel;
  6. import de.uni_trier.wi2.procake.similarity.SimilarityModelFactory;
  7. import de.uni_trier.wi2.procake.utils.io.IOUtil;
  8. import org.apache.commons.compress.archivers.zip.ZipArchiveEntry;
  9. import org.apache.commons.compress.archivers.zip.ZipArchiveOutputStream;
  10. import org.apache.commons.io.IOUtils;
  11. import org.apache.jena.sparql.exec.RowSet;
  13. import java.io.*;
  14. import java.nio.file.Files;
  15. import java.nio.file.Paths;
  16. import java.util.ArrayList;
  17. import java.util.Arrays;
  18. import java.util.List;
  19. import java.util.zip.ZipInputStream;
  21. public class ConversionService
  22. {
  23.     public byte[] convertConfigurationToZip() throws IOException
  24.     {
  25.         try
  26.         {
  27.             // Create Zip on Harddrive
  28.             List<File> listOfFiles = new ArrayList<File>();
  29.             File hello = new File("pathToFile");
  30.             File world = new File("pathToFile");
  31.             listOfFiles.add(hello);
  32.             listOfFiles.add(world);
  33.             IOUtil.createZipFile("pathToZip", listOfFiles);
  34.             // ---------------------
  36.             // Create Zip File in Memory
  37.                ByteArrayOutputStream baos = IOUtil.createZipFileInMemory(listOfFiles);
  38. ;
  39.             // ---------------------
  41.             // Compare both Zip Files
  42.             byte[] zipContentHardDrive = Files.readAllBytes(Paths.get("pathToFile"));
  43.             byte[] zipContentMemory = baos.toByteArray();
  45.             int minLength = Math.min(zipContentMemory.length, zipContentHardDrive.length);
  47.             if (!Arrays.equals(
  48.                             Arrays.copyOf(zipContentMemory, minLength),
  49.                             Arrays.copyOf(zipContentHardDrive, minLength))) {
  50.                 throw new IOException("The Zip Files are different.");
  51.             }
  52.             // ---------------------
  54.             return zipContentMemory;
  55.         }catch (Exception e) {
  56.             System.out.println("Could not create Zip File");
  57.         }
  58.         return null;
  59.     }
  60. }

型
最后，在我给予更多细节之前，这里是IOUtil方法：

  1. public static ByteArrayOutputStream createZipFileInMemory(List<File> files) throws IOException {
  2.     ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream();
  3.     ZipArchiveOutputStream zipArchiveOutputStream = new ZipArchiveOutputStream(
  4.         byteArrayOutputStream);
  6.     for (File file : files) {
  7.       addZipEntry(zipArchiveOutputStream, file);
  8.     }
  10.     zipArchiveOutputStream.finish();
  11.     return byteArrayOutputStream;
  12.   }
  13. private static void addZipEntry(ArchiveOutputStream archiveOutputStream, File file)
  14.       throws IOException {
  15.     String entryName = file.getName();
  16.     ArchiveEntry archiveEntry = new ZipArchiveEntry(file, entryName);
  17.     archiveOutputStream.putArchiveEntry(archiveEntry);
  18.     if (file.isFile()) {
  19.       Files.copy(file.toPath(), archiveOutputStream);
  20.     }
  21.     archiveOutputStream.closeArchiveEntry();
  22.   }

型
Vim ZipFile运行得很好。我的问题是，我下载的zip文件比我直接在硬盘上创建的文件大了大约200字节。我也无法打开更大的zip文件。用vim打开更大的zip文件后，我可以看到内容与较小的zip文件相比完全不同：
更小的zip：“zip.Vim version v33“浏览zipfile pathToZip“用光标选择一个文件并按ENTER键
hello. txt world.txt
这是更大的一个：PK^C^D^T^@^H ^H^@�W^@^
我不知道我在内存中创建zip文件的方法是错误的，还是在传输到store.exe时出错了。还有一件值得注意的事情是zipContentHardDrive和zipContentMemory是不同的：zipContentHarddrive：[80，75，3，4，20，0，0，8，8，0，-72，-114，-103，87，32，48，58，54，8，0，0，0，6，0，0，0，9，0，53，0，104，101，108，108，111，46，116，120，116、85、84、13、0、7、-99、-77、-119、101、-42、-76、-119、101、-99、-77、-119、101、10、0、32、0、0、0、0、1、0、24、0、-106、85、-35、-18、82、55，-38，1，4，18，-88，-87，83，55，-38，1，-128，76，-114，-18，82，55，-38，1，-53，72，-51，-55，-55，-25，2，0，+322更多]
zipContentMemory：[80，75，3，4，20，0，8，8，0，-72，-114，-103，87，0，0，0，0，0，0，0，0，0，0，0，9，0，53，0，104，101，108，108，111，46，116，120，116，85、84、13、0、7、-99、-77、-119、101、-42、-76、-119、101、-99、-77、-119、101、10、0、32、0、0、0、0、0、1、0、24、0、-106、85、-35、-18、82、55、-38，1，4，18，-88，-87，83，55，-38，1，-128，76，-114，-18，82，55，-38，1，-53，72，-51，-55，-55，-25，2，0，+354更多]
我不确定这是否是因为zip文件在内存中和不在内存中时通常是不同的。
如果有人能帮助我理解并解决我的问题，我将非常感激。非常感谢您抽出时间！

更新1在以下测试之后，我发现由于未知的原因，ZipFileInMemory方法不能像CoonversionService中那样工作。

  1. @Test
  2.   public void testCreateZipFileInMemoryWithFiles() throws IOException {
  3.     IOUtil.writeFile(graph, PATH_GRAPH_TXT);
  4.     IOUtil.writeFile(graph, PATH_GRAPH_XML);
  6.     File fileText = new File(PATH_GRAPH_TXT);
  7.     File fileXML = new File(PATH_GRAPH_XML);
  9.     List<File> listOfFiles = new ArrayList<>();
  10.     listOfFiles.add(fileText);
  11.     listOfFiles.add(fileXML);
  13.     ByteArrayOutputStream zipFileInMemory = IOUtil.createZipFileInMemory(listOfFiles);
  14.     assertNotNull(zipFileInMemory);
  16.     // Now let's check if the data stored in the memory is actually the right Zip-file
  18.     // Therefore let's create an actual Zip-File on our hard drive
  19.     IOUtil.createZipFile(PATH_CREATED_ZIP, listOfFiles);
  20.     
  21.     // Read the content of the created zip file
  22.     byte[] actualZipContent = Files.readAllBytes(Paths.get(PATH_CREATED_ZIP));
  23.     
  24.     // Compare the contents of the files within the zip
  25.     ZipInputStream dataOfMemoryZip = new ZipInputStream(new ByteArrayInputStream(zipFileInMemory.toByteArray()));
  26.     ZipInputStream dataOfHardDriveZip = new ZipInputStream(new ByteArrayInputStream(actualZipContent));
  28.     ZipEntry memoryEntry;
  29.     ZipEntry hardDriveEntry;
  31.     while ((memoryEntry = dataOfMemoryZip.getNextEntry()) != null) {
  32.       hardDriveEntry = dataOfHardDriveZip.getNextEntry();
  34.       assertEquals(memoryEntry.getName(), hardDriveEntry.getName());
  36.       byte[] expectedFileContent = IOUtils.toByteArray(dataOfMemoryZip);
  37.       byte[] actualFileContent = IOUtils.toByteArray(dataOfHardDriveZip);
  39.       assertArrayEquals(expectedFileContent, actualFileContent);
  40.     }
  41.   }

型
这里的内容总是相同的。我不明白为什么ConversionService类中的内容不同，因为我使用相同的文件和相同的方法。

更新2这里是下载的zip文件的hexdump：这里是我从下载中得到的zip文件的hexdump：

504b0304140000808000efbfbdefbfbfbd5720303a36080000000600 00000900350068656c6c6f2e74787455540d0007efbfbdefbfbdefbfbd65 d6b4efbfbd65efbfbdefbfbdefbfbd650a002000000000001001800efbf bd55efbfbdefbfbd5237efbfbd010412efbfbdefbfbd5337efbfbd01efbf bd4cefbfbdefbfbd5237efbfbd01efbfbd48efbfbdefbfbdefbfbd0200504b0304140000080800efbfbdefbfbdefbfbd57efbfbd6138efbfbd 08000000060000009003500776f726c642e74787455540d000 7efbfbdef bfbdefbfbd65efbfbdefbfbdefbfbd65efbfbdefbfbdefbfbd650a002000 000@00000100180009efbfbd28efbfbd5237efbfbd01efbfbdefbfbdefbf bdefbfbd523 7efbfbd01efbfbd5aefbfbdefbfbd5237efbfbd012befbfbd 2fefbfbd49efbfbd0200504b0102140014000080800efbfbdefbfbdefbf bd5720303a36080000000060000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000686656c6fbdefbdefbfbdefbfbd650a002000 000000001001800efbfbd55efbfbdefbfbd5237efbfbd010412efbfbdef bfbd5337efbfbd01efbfbd4cefbfbdefbfbd5237efbfbd01504b01021400 140000080800efbfbdefbfbfbd57efbfbd6138efbfbd08000000600 000009002d000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 bd5aefbfbdefbfbd5237efbfbd01504b0506000000002000efbfbd00 00000efbfbd00000000
下面是工作zip文件的hexdump：
504b03041400000808000b88e995720303a3608000000060000009003500 68656c6c6f2e74787455540d00079db38965d6b489659db389650a002000 0000000010018009655ddee5237da010412a8a95337da01804c8eee5237 da01cb48cdc9c9e70200504b0304140000080800c08e9957a86138dd0800 00000600000009003500776f726c642e74787455540d0007a9b38965abb3 8965a9b389650a0020000000000000100180009ba28f65237da0183db6 5237da01805ab5f55237da012bcf2fca49e10200504b0102140014000008 0800b88e995720303a360800000000600000009002d000000000000000 0000000000068656c6c6f2e74787455540500079db389650a0020000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 504b01021400140000080800c08e9957a86138dd08000000060000000900 2d0000000000000000000640000000000776f726c642e7478745554050007 a9b389650a00200000000000100180009ba28f65237da0183da01805ab5f55237da01504b05060000000002000200c80000000c80000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000

**问题修复：**以下是更改：

  1. methods: {
  2.     async convertConfigurationToZip() {
  3.       if (this.conversionInProgress) {
  4.         return;
  5.       }
  7.       this.conversionInProgress = true;
  8.       // Stores the current config in "store.config"
  9.       const response = await RestResource.convertConfigurationToZip();
  11.       // A Blob represents raw binary data
  12.       const blob = new Blob([response.data], {type: 'application/zip'}); // #
  14.       saveAs(blob, 'config.zip');
  15.       this.conversionInProgress = false;
  16.     }

型
和

  1. convertConfigurationToZip: async () => {
  2.         return axios.get("http://localhost:8080/api/conversion/convertConfigurationToZip/", {
  3.             responseType: "blob", // Ensure the response is treated as a blob
  4.         })
  5.             .then((response) => {
  6.                 return response; // Return the entire Axios response object
  7.             })
  8.             .catch((error) => {
  9.                 console.error(error);
  10.                 throw error; // Rethrow the error to handle it appropriately in the calling context
  11.             });
  12.     }

型

我假设const configData = store.config是一个字符串而不是二进制数据。

Java

1条答案

按热度按时间
mm5n2pyu

mm5n2pyu1#

您创建的zip文件的部分十进制转储确实不同。
zipContentMemory创建了一个 streamed zip文件，而zipContentHarddrive没有。这足以解释为什么一个zip文件可以工作，而另一个不能。
注意事项：如果没有看到完整的zip文件，就不可能100%确定。可能是zip文件中的一些其他差异导致了您看到的问题。如果有机会，请发布完整zip的转储。
部分zipContentHarddrive zip文件如下所示（下面的数据是用zipdetails创建的

  1. 0000 LOCAL HEADER #1       04034B50 (67324752)
  2. 0004 Extract Zip Spec      14 (20) '2.0'
  3. 0005 Extract OS            00 (0) 'MS-DOS'
  4. 0006 General Purpose Flag  0800 (2048)
  5.      [Bits 1-2]            0 'Normal Compression'
  6.      [Bit 11]              1 'Language Encoding'
  7. 0008 Compression Method    0008 (8) 'Deflated'
  8. 000A Last Mod Date/Time    57E6F1C7 (1474752967) 'Invalid Date or Time'
  9. 000E CRC                   363A3020 (909783072)
  10. 0012 Compressed Size       00000008 (8)
  11. 0016 Uncompressed Size     00000006 (6)
  12. 001A Filename Length       0009 (9)
  13. 001C Extra Length          0035 (53)
  14. 001E Filename              'hello.txt'
  15. 0027 Extra ID #1           5455 (21589) 'Extended Timestamp [UT]'
  16. 0029   Length              000D (13)
  17. 002B   Flags               07 (7) 'mod access change'
  18. 002C   Mod Time            65F6CCE2 (1710673122) 'Sun Mar 17 10:58:42 2024'
  19. 0030   Access Time         65F6CBA9 (1710672809) 'Sun Mar 17 10:53:29 2024'
  20. 0034   Change Time         65F6CCE2 (1710673122) 'Sun Mar 17 10:58:42 2024'
  21. 0038 Extra ID #2           000A (10) 'NTFS FileTimes'
  22. 003A   Length              0020 (32)
  23. 003C   Reserved            00000000 (0)
  24. 0040   Tag1                0001 (1)
  25. 0042   Size1               0018 (24)
  26. 0044   Mtime               01A5375291A255E9 (118561792965367273) 'Thu Sep 16 07:08:16 1976 536727300ns'
  27. 004C   Atime               01A53753D6D71204 (118561798421418500) 'Thu Sep 16 07:17:22 1976 141850000ns'
  28. 0054   Ctime               01A5375291F14CFF (118561792970542335) 'Thu Sep 16 07:08:17 1976 54233500ns'

字符串
部分zipContentMemory zip文件如下所示

  1. 0000 LOCAL HEADER #1       04034B50 (67324752)
  2. 0004 Extract Zip Spec      14 (20) '2.0'
  3. 0005 Extract OS            00 (0) 'MS-DOS'
  4. 0006 General Purpose Flag  0808 (2056)
  5.      [Bits 1-2]            0 'Normal Compression'
  6.      [Bit  3]              1 'Streamed'
  7.      [Bit 11]              1 'Language Encoding'
  8. 0008 Compression Method    0008 (8) 'Deflated'
  9. 000A Last Mod Date/Time    57E6F1C7 (1474752967) 'Invalid Date or Time'
  10. 000E CRC                   00000000 (0)
  11. 0012 Compressed Size       00000000 (0)
  12. 0016 Uncompressed Size     00000000 (0)
  13. 001A Filename Length       0009 (9)
  14. 001C Extra Length          0035 (53)
  15. 001E Filename              'hello.txt'
  16. 0027 Extra ID #1           5455 (21589) 'Extended Timestamp [UT]'
  17. 0029   Length              000D (13)
  18. 002B   Flags               07 (7) 'mod access change'
  19. 002C   Mod Time            65F6CCE2 (1710673122) 'Sun Mar 17 10:58:42 2024'
  20. 0030   Access Time         65F6CBA9 (1710672809) 'Sun Mar 17 10:53:29 2024'
  21. 0034   Change Time         65F6CCE2 (1710673122) 'Sun Mar 17 10:58:42 2024'
  22. 0038 Extra ID #2           000A (10) 'NTFS FileTimes'
  23. 003A   Length              0020 (32)
  24. 003C   Reserved            00000000 (0)
  25. 0040   Tag1                0001 (1)
  26. 0042   Size1               0018 (24)
  27. 0044   Mtime               01A5375291A255E9 (118561792965367273) 'Thu Sep 16 07:08:16 1976 536727300ns'
  28. 004C   Atime               01A53753D6D71204 (118561798421418500) 'Thu Sep 16 07:17:22 1976 141850000ns'
  29. 0054   Ctime               01A5375291F14CFF (118561792970542335) 'Thu Sep 16 07:08:17 1976 54233500ns'

型

更新

复制到2.zip的原始文件没有问题

  1. $ unzip -t /tmp/2.zip
  2. Archive:  /tmp/2.zip
  3.     testing: hello.txt                OK
  4.     testing: world.txt                OK
  5. No errors detected in compressed data of /tmp/2.zip.

型
下载的文件，复制到/tmp/1.zip，是非常腐败的zip文件

  1. unzip -t /tmp/1.zip
  2. Archive:  /tmp/1.zip
  4. caution:  zipfile comment truncated
  5. error [/tmp/1.zip]:  missing 3232546215 bytes in zipfile
  6.   (attempting to process anyway)
  7. error [/tmp/1.zip]:  attempt to seek before beginning of zipfile
  8.   (please check that you have transferred or created the zipfile in the
  9.   appropriate BINARY mode and that you have compiled UnZip properly)

型
看看十六进制转储，有很多重复的EF BF BD序列。这三个字节恰好是Unicode替换字符的UTF8字节序列。
这表明根本原因是您的服务器代码或客户端代码将zip文件作为UTF8字符串而不是字节序列处理。

展开查看全部
赞(0）分享  回复(0）举报  2024-01-05
我来回答

相关问题

    查看更多

    热门标签

    更多
    JavaquerypythonNode开发语言requestUtil数据库Table后端算法LoggerMessageElementParser

    最新问答

    更多
    • 蜀ICP备13028337号-1 大数据知识库 https://www.saoniuhuo.com © All rights reserved
    • 本站内容来源互联网,如果侵犯您的权益请联系我们删除, 联系方式：448109455@qq.com