All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

• Support Google Drive file URLs ending in /preview or without a /view or /edit suffix. (#532)

• Optional cancellation events for download() and cached_download() interrupt stalled response reads and download waits without shortening network timeouts, raising DownloadCancelled while preserving partial-file cleanup. (#525)

• An invalid --speed value now reports the accepted units instead of an internal function name. (#527)

• importlib is no longer exported from the gdown package; it was listed in __all__ by accident. (#526)

• Add a secretstorage installation extra for importing Chromium cookies from GNOME Keyring with pip install 'gdown[secretstorage]' or uvx --from 'gdown[secretstorage]' gdown. (#505)
• download(), download_folder(), and cached_download() accept a requests-style timeout, exposed on the command line as --timeout SECONDS, so a stalled server no longer blocks forever. (#510)
• Add opt-in --retries N and Python retries=N for automatic recovery of interrupted file and folder downloads, with bounded backoff and validated byte-range resume; --continue still controls reuse of earlier downloads. (#517)

• cached_download() verifies hash while streaming instead of reading the finished file back from disk. (#512)
• The command line reports a hit --timeout as a timeout instead of asking to file an issue. (#513)

• Propagate progress callback exceptions unchanged instead of misreporting chunked-encoding errors as incomplete downloads. (#519)

• Add --cookies-from-browser BROWSER to download as your signed-in Google account using cookies read straight from a local browser, and --cookies FILE to use a cookies file other than ~/.cache/gdown/cookies.txt. The cookies file is now written owner-only and keeps session cookies. (#494)

• Continue downloading the remaining files in a folder after individual downloads fail, then report all failures together. (#495)
• Automatically detect Google Drive folder URLs, making --folder necessary only for ambiguous bare IDs. (#496)

• Close download sessions and streamed responses on early exits, failures, and response replacement, including single-file Listing and resumed downloads. (#499)
• Remove cached-download staging directories after success or interruption, and create the cache on demand instead of during import. (#500)
• Honor an explicitly supplied user agent for file downloads within folders while retaining the existing default user agents. (#501)
• Fix folder --json export paths and preserve export extensions for dotted Google-native filenames. (#503)
• Preserve the saved cookies file when browser cookie extraction reports an error instead of replacing working session cookies with invalid decrypted values. (#504)

• Fixed downloads leaking HTTP sessions, progress bars, or output files when an exception interrupted a file or folder download. (#478)
• Reject incomplete response bodies instead of saving them as completed downloads. (#481)
• Normalized archive member paths returned by extractall, so ZIP and tar results use native path separators consistently. (#487)

• Added --json output for folder downloads as an array of URL and path records. (#460)
• Extended --json to resolve single-file URLs and filenames without downloading the file body. (#463)

• Marked --json as beta and added a warning that can be suppressed with --quiet. (#465)

See the GitHub Releases page for changelogs of v6.0.0 and earlier.