Skip to content

Changelog

Jump to bottom
lrg4you edited this page Aug 7, 2026 · 1 revision

Changelog

This page mirrors patchnotes.md from the repository. For the canonical source, see patchnotes.md.

Version 1.4.1 — Current

🛡️ Data Integrity — Full Metadata Preservation

LRGEX Compress now preserves your files exactly across a compress → extract round-trip. Whatever goes into the archive comes out identical — no more "every extracted file shows today's date."

  • LastWriteTime (mtime) preserved on every file and folder — sort-by-date, build systems, sync tools, and incremental backups all keep working correctly.
  • CreationTime (ctime) preserved — the original creation date is restored, not lost.
  • Windows attributes preserved — Hidden, Read-only, System, and Archive flags survive the round-trip. A hidden file extracts hidden; a read-only file extracts read-only.
  • Symbolic links / junctions preserved — links recreate as links. On Windows, creating a link needs Administrator or Developer Mode; LRGEX asks once ("allow admin?") and, if yes, re-launches the extraction elevated to recreate every link exactly. If you decline, links degrade gracefully (never aborts the whole extraction).
  • Directory timestamps restored in a final pass — so writing children into a folder doesn't overwrite the folder's original mtime.

NTFS alternate data streams are not supported (rarely used; not planned).

⚡ Performance — Extract & Compress Wins

  • Metadata sidecar (.lrgex/meta.bin): one small packed blob at the front of the archive replaces 11k+ per-entry metadata headers on big folders. Both compress and extract handle roughly half as many tar entries now. Old v1.4.0 archives still read correctly via the per-entry fallback.
  • One-syscall metadata restore (SetFileInformationByHandle): mtime + ctime + attributes set in a single call on the already-open file handle — no per-file re-open.
  • Direct 4 MB chunked writes on large-file extract (no redundant BufWriter copy).
  • Cache-served metadata on compress walk (DirEntry::metadata() instead of a fresh CreateFile per file) — saves thousands of syscalls on big folders.
  • Removed the redundant 4 MB BufWriter on compress output — zstd already emits good-sized blocks; the buffer was a pure memcpy of every compressed byte.
  • Tuned zstd multithreading — JobSize(16MB) + OverlapSizeLog(0) at level 1, fewer thread handoffs.
  • Dedicated oversubscribed write pool (3× core count) for extraction writes — correct sizing for workers that block in CreateFile and Defender scans.

🐞 Fixes

  • Multi-select coordinator hardened: handles WAIT_ABANDONED (previous instance crashed holding the mutex) and stale coordinator-PID lockfiles — a second rapid launch can no longer silently no-op.
  • Round-trip verified lossless: mtime/ctime/Hidden/ReadOnly/System, Unicode names, deeply nested paths, and a 3000-file stress batch (SHA256-checked).

Version 1.4.0

🚀 New Features

  • Extract Here: right-click .zgx/.zip/.rar → LRGEX → Extract Here (extracts to current folder, no subfolder)

🔧 Improvements & Changes

  • Removed AppliesTo from installer (was hiding the right-click menu on all files)
  • Added Extract Here to installer .iss for all three archive extensions

🐞 Bug Fixes

  • Fixed: right-click menu disappeared on all files due to malformed AppliesTo value

Version 1.2.1

No changes (placeholder).

Version 1.2.0

🚀 New Features

  • Installer-based auto-update: updates now download and run the full installer (not just the bare exe), ensuring registry entries, icons, and file associations are always updated correctly.

🔧 Improvements & Changes

  • Installer filename now includes full version number (e.g. v1.2.0, not v1.2).
  • CloseApplications=force and RestartApplications=no added to installer for silent updates.
  • App silently exits when launched with no arguments (fixes post-update restart error).

Version 1.1.0

No changes (placeholder).

Version 1.0.0

🚀 New Features

  • Multi-format extraction: .zip and .rar extraction added alongside .zgx. Magic-byte detection (zstd/zip/rar) with extension fallback.
  • Real .rar progress: vendored unrar 0.5.8 with UCM_PROCESSDATA counter patch for byte-level extraction progress.
  • Extract Here: right-click .zgx/.zip/.rar → Extract Here dumps contents into the current folder.
  • Multi-select compress: selecting multiple files/folders and right-clicking Compress produces ONE .zgx containing all items.
  • Signed auto-updates: Ed25519 signature verification, deferred until after operation completes.
  • 8-byte uncompressed total header in .zgx format for accurate extraction progress.
  • Batch-parallel extraction: rayon batch writer (2048 entries/64MB) with directory cache for many-small-files workloads.
  • Version badge in lower right of progress window.
  • Shimmer animation on progress bar fill.
  • Auto-close 2s after successful operation.
  • Cancel kills process immediately — no orphan threads.
  • Window icon from product logo.
  • op-label/op-detail split: action word in white bold, filename in silver.

🔧 Improvements & Changes

  • ZIP backend switched to deflate-flate2-zlib-ng (C-accelerated, 4× faster decompression).
  • [profile.dev.package."*"] opt-level = 3 — dependencies compiled at -O3 even in debug builds.
  • File preallocation (set_len) for entries >1MB during extraction.
  • BufReader 256KB between zstd decoder and tar (reduced from 4MB to avoid cache eviction).
  • Removed redundant BufReader before zstd decoder (it reads 128KB chunks natively).
  • Folder name stripping: x.md folder → x.zgx (not x.md.zgx).
  • Stable install path for right-click verbs (survives exe rebuilds).

🐞 Bug Fixes

  • Empty directories now preserved through compress → extract round-trip.
  • Atomic writes: name.zgx.part → name.zgx only on full success.
  • Cancel during compress deletes the partial archive.
  • Zip-slip protection (path traversal guard for both .zip and .zgx extraction).
  • Failed extraction no longer hangs the UI (shows "Failed" instead of freezing).

⚠️ Known Issues

  • Many-small-files extraction is limited by Windows Defender real-time scanning (~5-15 MB/s). This is a documented Windows limitation, not a code issue.

Version 0.1.0

🚀 New Features

  • Right-click compression and extraction: one LRGEX Explorer entry that opens a hover submenu (Compress on folders/files, Extract on .zgx archives). Per-user, HKCU only, no UAC.
  • .zgx archive format: Double-clicking a .zgx extracts it.
  • Progress window: colored progress bar, live percent, MB/s, and ETA. A 500 ms heartbeat + animated spinner keep it from ever looking frozen, even on a cold cache.
  • Byte-based progress: percentage is always bytes_done / bytes_total (never file-count), so a single huge file does not sit at a misleading percent.
  • Multi-threaded compression: parallel reads across all CPU cores, designed for multi-gigabyte folders.
  • Atomic archive writes: compress writes to <name>.zgx.part and atomically renames to <name>.zgx only on full success, so an interrupted or crashed compress can never leave a complete-looking partial archive.
  • Empty-directory preservation: directory entries are archived before their contents, so extraction restores the exact source tree, including empty folders.
  • Cancel cleans up: aborting a compress releases the .part file and deletes it — no orphaned half-files are left behind.
  • Skip-with-warning on locked files: a single unreadable file does not fail a multi-GB job. The archive still publishes, and a <name>.zgx.skipped.txt sidecar lists anything that could not be read (auto-deleted on the next clean run).
  • Signed auto-updates: Ed25519 signature verification on every update.
  • Per-user installer: Inno Setup, PrivilegesRequired=lowest (no UAC prompt).
  • Self-contained executable: a single ~14 MB .exe, zero runtime dependencies.

Developed by LRGEX

Clone this wiki locally