« Changelog History


FileDownloader v0.3.3 Release Notes

  • 2016-07-10

    🆕 New Interfaces

    • ➕ Add FileDownloadUtils#getTempPath: Get the temp path is used for storing the temporary file not completed downloading yet(filename.temp). Refs #172.
    • ➕ Add FileDownloader#getStatusIgnoreCompleted(id:int): Get the downloading status without cover the completed status(If completed you will receive INVALID).
    • ➕ Add FileDownloader#getStatus(id:int, path:String): Get the downloading status.
    • ➕ Add FileDownloader#getStatus(url:String, path:String): Get the downloading status.
    • ➕ Add FileDownloadUtils#isFilenameConverted(context:Context): Whether tasks from FileDownloader Database has converted all files' name from filename(in old architecture) to filename.temp, if it is not completed downloading yet.
    • ➕ Add FileDownloadUtils#generateId(url:String, path:String, pathAsDirectory:boolean): Generate a Download Id which can be recognized in FileDownloader.
    • ➕ Add BaseDownloadTask#setPath(path:String, pathAsDirectory:boolean): If pathAsDirectory is true, the path would be the absolute directory to store the downloading file, and the filename will be found in contentDisposition from the response#header as default.
    • ➕ Add BaseDownloadTask#isPathAsDirectory: Whether the result of BaseDownloadTask#getPath() is a directory path or directory/filename path.
    • ➕ Add BaseDownloadTask#getTargetFilePath: Get the target file path to store the downloading file.
    • ➕ Add FileDownloadQueueSet#setDirectory: Set the directory to store files in this queue.

    ✨ Enhancement

    • 👌 Improve Practicability: Support the path of the task as the directory to store the file, and in this case, the filename will be found in contentDisposition from the response#header as default. Refs #200.
    • 👌 Improve Practicability: Using the temp path to store the file not completed downloading yet(filename.temp). Refs #172.
    • 👌 Improve Performance: FileDownloader doesn't store completed tasks in Database anymore, and check whether the task has completed downloading with File#exists() directly. Refs #176, #172.
    • 👌 Improve Robust: Choosing the task which status is INVALID or progress to receive completed message preferentially, to ensure the callback of progress can be handled. Refs #123
    • 👌 Improve Robust: Expanding task-sync-lock to the outside of getting-same-id-downloading-task, to fix some messages can't be consumed because status changed during getting-same-id-downloading-task and waiting for task-sync-lock.

    🛠 Fix

    • 🛠 Fix(DB-maintain): Keeping models, whose status is pending and downloaded so far bytes is more than 0 because it can be used for resuming from the breakpoint. Closes #176.
    • 🛠 Fix(crash-NPE): FileDownloader might occur NPE when the download-listener was removed, but the task is still running in FileDownloader. Closes #171.