Change Logs

Changelog

v2.17.0 — 2026-06-10

Added

  • Add: word to webp and word to avif — two new browser-side Word converters.

v2.16.0 — 2026-06-09

Fixed

  • Fix: Word to JPG / PNG / PDF no longer hang on large, image-heavy .docx files (4 MB+, 50+ pages).
  • Fix: a stalled conversion now surfaces a clear error message instead of an infinite progress bar.

v2.14.0 — 2026-06-08

Added

  • Add: word to png — new browser-side Word to PNG converter.

v2.13.2 — 2026-06-06

Added

  • Add: word to pdf — new browser-side Word to PDF converter.

v2.13.0 — 2026-05-29

Added

  • Add: image compressor — new browser-side image compressor. Batch up to 20 JPG, PNG or WebP images. Quality slider (1–100%)

v2.12.9 — 2026-05-25

Added

  • Dot-separated value prop row to all tools

Fixed

  • All client-side downloads now use branded filenames (resizelab.app-XXXXXX-XXXXXX.ext), before only some tools did this.

v2.12.8 — 2026-05-25

Fixed

  • single-format converter pages (e.g. JPG to PNG) were silently outputting WebP instead of the locked format.

Added

  • Add: lightweight client-side browser compatibility check. If the visitor’s browser is too old, a static banner is shown — no server cost, no redirects, no asset blocking.

v2.12.7 — 2026-05-20

Fixed

  • real root cause of image resizer adding the picked image to the queue twice.

v2.12.5 — 2026-05-20

Fixed

  • image cropper — the initial crop box wasn’t centred on the image when the loaded photo was narrower than the stage. The canvas was being centred by the wrapper’s flexbox while the crop box was absolutely positioned relative to that wrapper, so the box appeared offset to the left of the image and couldn’t be lined up properly. Added an inline-block inner container so the crop box now positions relative to the canvas itself and stays correctly aligned regardless of image width.
  • image resizer — clicking browse files added the same image to the queue twice.

v2.12.4 — 2026-05-20

Improved

  • Polish: image background remover no longer pre-paints the dropped photo into the Result panel before the user clicks Convert image. The result panel now shows a small placeholder (“Click ‘Convert image’ to remove the background”) on the checkerboard until processing finishes, so it’s obvious the tool hasn’t run yet. Previously the source appeared identically in both panes, making the “after” preview look like nothing had happened.

v2.12.3 — 2026-05-20

Fixed

  • image background remover was returning the original image unchanged (the foreground “cutout” was the whole picture).
  • image background remover was still returning the original image after the 2.12.2 version bump.

v2.12.0 — 2026-05-20

Added

  • New: image backround remover — remove the background from any photo entirely in the visitor’s browser. Live split-pane before/after preview, one-click background swap (transparent / white / black / accent / custom colour).
  • New: image cropper — visual crop tool with 8 drag handles, rule-of-thirds grid, live pixel-dimension readout, and aspect-ratio presets (Free, 1:1, 4:3, 3:2, 16:9, 9:16, 2:3). Exports as PNG, JPG, or WebP.
  • New: image resizer — batch resize up to 25 images in one go with three modes: exact width/height (with optional ratio lock), percentage scale (10–200%), or social-media presets (Instagram square / portrait / story, Facebook OG, X, YouTube, Full HD, Web). Preset mode adds Fit / Cover / Stretch fitting. Quality slider for JPG/WebP. Per-file download + bulk “Download all as ZIP”.
  • Note: the first background removal takes 30–60 s while the ~17 MB model downloads from the CDN; subsequent runs are fast (cached). The model requires a modern desktop browser with WebAssembly + sufficient RAM (≥ 2 GB free).

v2.11.2 — 2026-05-19

Added

  • Added live preview to the Watermark PDF tool.
  • Renamed all PDF tool downloads to branded resizelab.app-XXXXXX-XXXXXX.* filenames.

v2.11.1 — 2026-05-17

Removed

  • Removed the Language dropdown from the OCR UI for now due to a bug.
  • OCR is now hard-coded to English.
  • We need to revisit multi-language support in a future release.

Fixed

  • Removed the cramped “Higher = more accurate, slower” badge from inside the label.
  • Replaced it with a clear help line under the 1.5× / 2× / 3× buttons: “Higher scales produce more accurate OCR but take longer per page.”

v2.11.0 — 2026-05-17

Improved

  • pdf ocr now also accepts JPG, PNG, and WebP images (single file or batch), not just PDFs — useful for phone photos, scanned receipts, whiteboard shots, etc.
  • Image inputs go straight to the OCR engine (no PDF.js rasterisation step), so they’re faster than the equivalent PDF.
  • All three output modes (Searchable PDF / Plain .txt / Both) work for image input. The searchable PDF wraps each image as a single page with the OCR text layer; batches of images produce a single multi-page searchable PDF in queue order.
  • Long OCR jobs now show a rolling wait-time estimate like Processing page X of Y · ~Ns remaining so users can see progress at a glance.

v2.10.0 — 2026-05-17

Added

  • New: pdf ocr — extract text from scanned / image-only PDFs entirely in the visitor’s browser, powered by Tesseract.js (vendored). Per-page progress, 12 built-in languages (English, French, German, Spanish, Italian, Portuguese, Dutch, Russian, Simplified Chinese, Japanese, Korean, Arabic), and three output modes:
    • Searchable PDF — keeps the original page image and adds an invisible OCR text layer so the result is copy- and search-able in any PDF reader.
    • Plain text (.txt) — just the recognised text, one blank line between pages.
    • Both — searchable PDF and .txt packaged as a single ZIP.
  • Render scale (1.5× / 2× / 3×) trades speed for OCR accuracy.
  • Tesseract’s language training data (~10–15 MB per language) is fetched on first use from the jsDelivr / tessdata CDN and cached by the browser; the visitor’s PDF content itself still never leaves their device.
  • Existing tools, favicon maker, word to jpg, and the image converter are unchanged.

v2.9.0 — 2026-05-17

Added

  • New: nine PDF tools, all running entirely in the visitor’s browser — nothing is uploaded:
    • pdf merge — combine multiple PDFs (drag rows to reorder)
    • pdf split — split into individual pages, or by custom page ranges (e.g. 1-3, 5, 7-9); always packaged as a ZIP
    • pdf to jpg — render every page to a JPG or PNG image (multiple PDFs supported), packaged as a ZIP
    • jpg to pdf — combine JPG, PNG, or WebP images into a single PDF (Fit / A4 / Letter, Auto / Portrait / Landscape)
    • pdf compress — re-encode pages as JPEG with Low / Medium / High presets to shrink image-heavy or scanned PDFs
    • pdf rotate — rotate every page 90/180/270°, or pick individual pages from a thumbnail grid
    • pdf reorder — drag page thumbnails to rearrange or remove pages
    • pdf watermark — text or image watermark with adjustable colour, position, angle, opacity, and font size
    • pdf page numbers — add page numbers in six positions and three formats, with custom start number and skip-first-N
  • Each PDF tool supports the same accent_color, max_mb, and heading attributes as the image tools. Powered by the bundled pdf-lib and pdf.js.
  • Note: the nine PDF tools, favicon maker, and word to jpg are 100% client-side. image converter is client-first but still falls back to PHP/GD for GIF, BMP, and AVIF on browsers without AVIF encode support.
  • Known limitations / gotchas:
    • Very large or page-heavy PDFs can exhaust browser memory — try a smaller scale (for pdf to jpg) or a lower compress preset, or split the file first.
    • pdf compress re-encodes pages as JPEG, so it’s most effective on image-heavy / scanned PDFs and will introduce some visible quality loss at the High preset.
    • Password-protected (encrypted) PDFs are not supported — users see a friendly “remove the password and try again” message.
    • pdf to jpg always returns a ZIP, even for a single-page input, so the download UX is consistent across files.

v2.8.2 — 2026-05-15

Fixed

The Convert / Quality / Resize section had been falling back to bland native form controls. It now matches the rest of the tool’s design:

  • Format chooser renders as proper pills — white with a subtle border by default, gradient purple with a soft shadow when active, gentle hover lift.
  • Quality slider has a custom white-on-accent thumb with a soft glow, and the percentage now sits in a small accent-tinted badge next to the label.
  • Resize fields are unified card-style inputs with a tucked-in “px” suffix, focus halo in your accent colour, and a small pill-style “Ratio kept” indicator instead of plain body text.

No functional changes.

v2.8.1 — 2026-05-15

Improved

Now that processing happens in the visitor’s browser, default upload limits have been raised:

  • image converter — 200 MB per file (was 50 MB)
  • favicon maker — 250 MB per file (was 50 MB)
  • word to jpg — 250 MB per file (was 100 MB)

v2.8.0 — 2026-05-15

Overhaul

A major architectural shift: the heavy work moved from our server into visitor’s browsers.

  • favicon maker is now fully client-side. The source image, all six PNG sizes (16, 32, 48, 180, 192, 512), the multi-resolution favicon.ico, and the final ZIP are all generated locally. Nothing is uploaded.
  • image converter is now client-first with automatic server fallback.
    • JPEG, PNG, and WebP → always processed in the browser (instant).
    • AVIF → processed in the browser on Chrome / Firefox / Edge; falls back to PHP on Safari.
    • GIF and BMP → continue to use the server (browsers can’t encode them natively).
  • “Download all as ZIP” is now built in the browser for the image converter. In mixed batches, server-converted files are fetched and bundled locally so the ZIP always contains every file.
  • Why this matters: dramatically faster conversions for the common case, no upload bandwidth used, and effectively zero PHP/GD load for most visitors. Privacy improves too — typical files never leave the device.

v2.7.2 — 2026-05-15

Improved

  • Raised default upload limits: image converter and favicon maker bumped from 20 MB to 50 MB; Word-to-JPG to 100 MB.

v2.7.1 — 2026-05-15

Added

  • New: when converting two or more Word documents at once, a “Download all as ZIP” button appears so you can grab them in a single archive instead of one by one. The ZIP is built in the browser — your files still never leave your device.

v2.7.0 — 2026-05-15

Added

  • New: Tool that converts Word documents (.docx) to a single JPG image — runs entirely in a browser.
  • Friendly error messages when a non-Word file or a legacy .doc file is dropped in (legacy .doc isn’t supported — visitors are asked to save as .docx).
  • Hardened HTML sanitization on the converted document (strips scripts, event handlers, and dangerous URLs) before rendering, so a malicious .docx can’t run code on your page.

v2.6.4 — 2026-05-15

Fixed

  • favicon set now uses your uploaded image instead of a placeholder blue circle. Each size is resampled from the source while preserving aspect ratio and transparency.
  • Improvement: clearer error message when an uploaded file can’t be decoded as an image.

v2.6.3 — 2026-05-15

Fixed

  • the favicon “Download ZIP” button pointed to the upload endpoint and saved a JSON error message as the ZIP. It now points to the actual generated archive.

v2.6.2 — 2026-05-15

Fixed

  • favicon ZIP downloads stayed corrupted on hosts with gzip/output compression enabled; downloads now force identity encoding and disable compression at the PHP and Apache layers.

v2.6.1 — 2026-05-15

Fixed

  • favicon ZIP downloads were corrupted on some hosts due to nested output buffers; download handlers now drain all buffers before streaming.

v2.6.0 — 2026-05-15

Added

  • Added a favicon maker tool.
  • Added favicon maker AJAX generation and ZIP download (favicon.ico, 16×16, 32×32, 48×48, apple-touch-icon 180×180, 192×192, 512×512).
  • Added favicon-specific template with a chip grid showing every size that will be created.
  • Matched favicon maker styling to the main converter tool.
  • Updated plugin assets and documentation.

v2.5.0 — 2026-05-13

Added

  • Redesigned the converter with a more polished interface
  • Added drag-and-drop upload support
  • Added paste-from-clipboard support
  • Added file size savings stats
  • Added success confetti after conversion

Improved

  • Updated button styling and spacing
  • Improved visual feedback during upload and conversion
  • Refined mobile layout and overall presentation

v2.4.0 — 2026-05-06

Added

  • Expanded image format support
  • Improved conversion flow for common image types

Fixed

  • Minor layout issues on smaller screens
  • Upload edge cases for certain file types

v2.3.0 — 2026-04-28

Added

  • Initial support for more image conversion formats

Fixed

  • Early conversion and UI bugs