有人知道Tachiyomi的记录导出为json的方法吗？备份文件是 .tachibk，尝试直接更改扩展名为 .json，乱码；直接更改为 .zip，居然能看到部分中文，但是打开会卡住只能强行退出浏览app
无标题无名氏No.62080645
2024-04-18(四)20:02:28
ID: 83YvGLK (PO主)
目前最大的疑问是解压缩 gzip，明明扩展名不是 gzip文件，是不是手动修改扩展名这一步出了问题？
无标题无名氏No.62083994
2024-04-19(五)00:33:42
ID: CzzTJVo
我电脑没在身边，帮你问了下ai
Here's a Python implementation of a backup file decoder based on the provided Kotlin files. This implementation assumes that the backup file is in Protocol Buffers format and uses the `protobuf` library to decode it.
```python
import gzip
import protobuf
# Define the protobuf message classes
BackupManga = protobuf.Descriptors.BackupManga
BackupChapter = protobuf.Descriptors.BackupChapter
BackupCategory = protobuf.Descriptors.BackupCategory
BackupTracking = protobuf.Descriptors.BackupTracking
BackupHistory = protobuf.Descriptors.BackupHistory
BackupSource = protobuf.Descriptors.BackupSource
BackupPreference = protobuf.Descriptors.BackupPreference
BackupSourcePreferences = protobuf.Descriptors.BackupSourcePreferences
Backup = protobuf.Descriptors.Backup
def decode_backup(backup_file_path):
with open(backup_file_path, 'rb') as f:
content = f.read()
# Check if the file is gzipped
if content[:2] == b'\x1f\x8b':
content = gzip.decompress(content)
# Parse the protobuf message
backup = Backup.FromString(content)
# Process the backup data
print(f"Number of manga: {len(backup.backupManga)}")
for manga in backup.backupManga:
print(f"Manga: {manga.title}")
for chapter in manga.chapters:
print(f" Chapter: {chapter.name}")
print(f"Number of categories: {len(backup.backupCategories)}")
for category in backup.backupCategories:
print(f"Category: {category.name}")
print(f"Number of tracking entries: {sum(len(manga.tracking) for manga in backup.backupManga)}")
for manga in backup.backupManga:
for tracking in manga.tracking:
print(f"Tracking: {tracking.title}")
print(f"Number of history entries: {sum(len(manga.history) for manga in backup.backupManga)}")
for manga in backup.backupManga:
for history in manga.history:
print(f"History: {history.url}")
print(f"Number of sources: {len(backup.backupSources)}")
for source in backup.backupSources:
print(f"Source: {source.name}")
print(f"Number of preferences: {len(backup.backupPreferences)}")
for pref in backup.backupPreferences:
print(f"Preference: {pref.key} = {pref.value}")
print(f"Number of source preferences: {len(backup.backupSourcePreferences)}")
for source_prefs in backup.backupSourcePreferences:
print(f"Source preferences for {source_prefs.sourceKey}:")
for pref in source_prefs.prefs:
print(f" {pref.key} = {pref.value}")
if __name__ == "__main__":
backup_file_path = input("Enter the path to the backup file: ")
decode_backup(backup_file_path)
```
This program defines the Protobuf message classes based on the Kotlin classes provided. The `decode_backup` function takes the path to the backup file as input, opens the file, decompresses it if it's gzipped, and parses the Protobuf message using the `protobuf` library.
After parsing the backup data, the function processes and prints the various components of the backup, such as manga, chapters, categories, tracking entries, history entries, sources, preferences, and source preferences.
Note that you'll need to have the `protobuf` library installed to run this program. You can install it using `pip install protobuf`.
无标题无名氏No.62084003
2024-04-19(五)00:34:39
ID: CzzTJVo
> Give me an example of the descriptor file.
Unfortunately, I don't have access to the actual Protobuf descriptor file for the data structures defined in the provided Kotlin files. However, I can provide an example of what the descriptor file might look like based on the structure of the data classes.
```protobuf
syntax = "proto3";
package eu.kanade.tachiyomi.data.backup.models;
message BackupManga {
int64 source = 1;
string url = 2;
string title = 3;
string artist = 4;
string author = 5;
string description = 6;
repeated string genre = 7;
int32 status = 8;
string thumbnailUrl = 9;
int64 dateAdded = 13;
int32 viewer = 14;
repeated BackupChapter chapters = 16;
repeated int64 categories = 17;
repeated BackupTracking tracking = 18;
bool favorite = 100;
int32 chapterFlags = 101;
repeated BrokenBackupHistory brokenHistory = 102;
int32 viewerFlags = 103;
repeated BackupHistory history = 104;
int32 updateStrategy = 105;
int64 lastModifiedAt = 106;
int64 favoriteModifiedAt = 107;
repeated string excludedScanlators = 108;
int64 version = 109;
}
message BackupChapter {
string url = 1;
string name = 2;
string scanlator = 3;
bool read = 4;
bool bookmark = 5;
int64 lastPageRead = 6;
int64 dateFetch = 7;
int64 dateUpload = 8;
float chapterNumber = 9;
int64 sourceOrder = 10;
int64 lastModifiedAt = 11;
int64 version = 12;
}
// ... other message definitions for BackupCategory, BackupTracking, etc.
message Backup {
repeated BackupManga backupManga = 1;
repeated BackupCategory backupCategories = 2;
repeated BrokenBackupSource backupBrokenSources = 100;
repeated BackupSource backupSources = 101;
repeated BackupPreference backupPreferences = 104;
repeated BackupSourcePreferences backupSourcePreferences = 105;
}
```
This is a simplified example of what the Protobuf descriptor file might look like based on the data structures in the provided Kotlin files. Note that the field numbers and message names are derived from the Kotlin code, but the actual structure and naming conventions may differ in the real descriptor file.
In this example, each data class is represented as a Protobuf message, and the fields within each class are defined with their corresponding data types and field numbers. Nested classes are represented as nested messages within their parent message.
无标题无名氏No.62084009
2024-04-19(五)00:35:34
ID: CzzTJVo
照着把剩下的定义文件补充完应该就行
无标题无名氏No.62084843
2024-04-19(五)02:37:58
ID: 83YvGLK (PO主)
>>No.62083994
感谢！完美解决了我的问题！现在终于能够正常读取到里面的字段没有乱码了！盲生啊，你让我发现了华点！！！原来问题出在 protobuff 的 schema 找错了！！！
不过就是你向 AI 问的这段 Python 代码是一定跑不起来的，因为 Python 里要使用 protobuf 要先通过protoc编译器生成对应的 proto 文件的处理模块然后用这个生成的处理模块才能进行解析！
