Nettongia PDF Editor


Nettongia PDF Editor — Free Download. Offline PDF editing

Nettongia PDF Editor is a free offline desktop application for Windows that edits real PDF content without uploading documents to a server. It changes existing text, adds movable and resizable text boxes, manages pages through draggable thumbnails, rotates pages left or right, and deletes pages with the Delete key or a context menu. Users can work with images that are inserted or already embedded in the PDF, moving, resizing and rotating them while preserving original quality where possible. The program includes offline optical character recognition for Czech, Slovak, Polish, German and English, adds an invisible searchable text layer to image-only pages, and skips pages that already contain usable text. A Save a Copy command guarantees that the source file is not overwritten, and the editor supports undo and redo across page, text and object changes, automatic crash recovery, a recent files list, document compatibility inspection, and background saving and compression in an isolated process.

★★★★★
5.0(1 ratings)
File size: 98.1 MB
The latest version of Nettongia PDF Editor is: 0.22.0
Operating system: Windows
Languages: English, German, French, Spanish, Portuguese
Price: $0.00 USD (Open Source (AGPL-3.0))
  • Real text editing. Nettongia PDF Editor modifies text that already exists inside a PDF rather than covering it with annotations. A blue outline identifies an editable text object when the pointer moves over it; a double-click opens direct typing on the page, and Ctrl+Enter or a click outside confirms the change while Esc cancels it. The text toolbar controls installed font, size, bold, italic, underline and color. Entering an empty replacement removes the selected source text, and the modified content is written back as genuine searchable PDF text. Editing works on one visual text run at a time, usually a single line or one styled part of a line, and automatic reflow of surrounding paragraphs is not performed.
  • Page management. The Pages panel lists document thumbnails that can be dragged to a new position like slides in a presentation. Page rotation is reversible: the Page menu, the thumbnail context menu, or the Ctrl+Shift+Left and Ctrl+Shift+Right shortcuts rotate the selected page left or right, and page contents, pending text, images and visual signatures rotate together. Undo and Redo restore the complete earlier state. The Delete key removes the selected thumbnail, and the right-click menu offers move-earlier, move-later and delete commands. Bookmarks, internal links and pending editor objects follow their logical page during reordering.
  • Image transformation. Inserted images become interactive after placement. A single click selects an image, dragging moves it, the lower handle scales it proportionally, and the upper handle rotates it freely. The same direct-click workflow applies to images already embedded in the PDF. The first click only selects and outlines a source image without rewriting, rasterising or lowering its quality. Moving, resizing or rotating and releasing the mouse creates one undoable replacement. JPEG streams and orthogonal PDF rotations remain in their original form whenever possible, and only an arbitrary-angle transform requires a new lossless raster surface.
  • Offline OCR. Optical character recognition runs for the current page or the complete document and uses five bundled models: Czech, Slovak, Polish, German and English. Recognition executes in a separate process and can be cancelled without changing the open document. Only pages without usable text are processed; each receives an invisible searchable text layer while original page pixels, links, bookmarks and document structure remain intact. The result is validated and applied as one undoable document change. No Tesseract installation and no internet connection are required because the language models and their SHA-256 manifest ship with the package.
  • Text search. The Edit menu and Ctrl+F open a search strip with up and down buttons that move through every result in the document. The current match is highlighted on the page and the counter shows its position, F3 advances to the next result, Shift+F3 returns to the previous one, and Esc closes the strip. Search uses the current working version of the PDF, including unsaved edited and newly inserted text. Queries run in a worker thread so a large document does not freeze the editor, and starting a new query cancels the previous result.
  • Undo and redo. Every page, text and object change enters the same undoable history, whether the action came from a toolbar button, a menu command or a context menu. Opening a menu or selecting its target does not create a history entry; only choosing a command does. Rotation, image transforms, text replacement, page deletion and OCR application can all be reversed. Undo and redo also survive crash recovery, so a restored session keeps its previous states available.
  • Crash recovery. A confirmed unsaved state is captured after a short debounce in a background worker. The capture includes page changes, original and inserted text, images, image deletions, signatures, the current page and zoom. If the process or computer stops unexpectedly, the editor offers to restore that state on the next start. Recovery files use a non-executable ZIP and JSON format with strict field, size, member-name and SHA-256 integrity checks, and both recovery and normal saves use atomic temporary-file replacement.
  • Save a Copy. The File menu provides Save a Copy through Ctrl+Alt+S for an independent safety copy. It uses the same isolated, validated and atomic write path as a normal save but never changes the active document path, its normal Save target, the saved history point or the unsaved-work marker. A PDF opened from disk is not overwritten automatically: the first Save opens Save As with an _edited.pdf suggestion. After a target has been chosen, later Save commands write atomically to that target.
  • Background saving and compression. Save, Save As and Compress PDF run in a separate helper process, so long writes do not monopolize the Python interpreter or freeze the editor. A non-modal progress window remains visible while the document can still be read or edited. If editing continues during Save, the output contains the consistent snapshot taken when Save was pressed and the newer state remains marked with an asterisk. Cancel terminates the isolated worker, and an incomplete write cannot truncate an existing destination because the target is replaced only after a complete temporary file has been written.
  • Document compatibility inspection. After a document opens, a separate helper scans its structure and performs low-memory preflight renders of the first, middle and last pages while the editor remains responsive. The summary in File > Document compatibility uses three status levels: green for a normal document, yellow for features that can change during rewriting, and red for high-risk structures such as signed digital signatures, XFA, JavaScript or PDF portfolios. Existing signed signatures, forms, attachments, layers, portfolios, tagged structure, encryption, annotations and unusually large pages are identified, and saving a document with a signed signature requires explicit confirmation.
  • Password-protected PDFs. A protected PDF opens normally with a masked password dialog and allows another attempt after an incorrect password. The same workflow applies when importing pages from another protected PDF. The password unlocks only the source and is not retained in the editor state or recovery data. The working document is decrypted in memory, and edited or recovered copies are saved without password protection, which the status bar reports after a protected document is opened.
  • Recent files. File > Recent files reopens one of the ten most recently opened or saved PDFs. Opening the same file moves it to the top instead of creating a duplicate. The list is stored only in the local application settings and can be cleared from the submenu. If a file has been moved or deleted, selecting it removes the stale entry and reports the missing path.
  • PDF document tree. When a PDF contains bookmarks or an EPLAN navigation structure, the left sidebar opens on the Tree tab with expandable branches. A single click navigates to the linked page, and EPLAN device links also center and enlarge the destination area. The tree loads lazily, so only visible portions are read into the view; a 798-page EPLAN reference with 28,971 entries in nine levels can be browsed without creating all entries at once. PDFs without a document tree keep the Pages tab active.
  • Tiled rendering performance. Large or highly zoomed pages first appear as a memory-bounded full-page preview, then the visible area is refined in 768 x 768 pixel tiles by an isolated worker process. Tiles nearest the viewport centre render first and scrolling prefetches one tile beyond the viewport. A bounded 96 MB LRU cache keeps revision-keyed tiles reusable across page and zoom changes. Generation and revision checks ensure cached or late worker results can never replace newer page content, and a 20-second timeout discards obsolete work while the existing preview stays usable.
  • Portable Windows package. The release is distributed as a portable ZIP for Windows 10 and 11 x64. Extracting it to a folder and running NettongiaPDFEditor.exe requires no installation. The packaged build already contains its Python dependencies and offline OCR models, and the verification value for the current preview archive is published on the download page alongside the platform information. A PDF can also be dragged onto the window or passed as a path to run_editor.py in a source checkout.

Nettongia PDF Editor was created by the Nettongia project, a small independent development group that publishes the application together with a public community page showing aggregate usage totals, Windows test results and OCR language counts. The project distributes its releases through the official site at nettongia.com, maintains a GitHub Issues tracker, and accepts feedback at support@nettongia.com. Development follows a rapid public preview cycle: version 0.11.0 introduced automatic crash recovery and a persistent Recent files menu, version 0.12.0 added password-protected PDF opening, version 0.13.0 moved normal saves and compression into a separate process, version 0.14.1 hardened progressive tiled rendering for large engineering documents, and version 0.15.0 added process-isolated compatibility inspection. Version 0.16.0 relocated high-detail tile rasterisation into a dedicated helper process, version 0.17.0 added process-isolated OCR together with interactive inserted images, version 0.18.0 introduced reversible page reordering and direct transformation of images already embedded in a PDF, and version 0.19.0 added reversible page rotation. The application is written in Python and runs on Python 3.11 or newer, using the PySide binding for its graphical interface and the PyMuPDF and MuPDF library for PDF parsing, rendering and writing.

Alternatives to Nettongia PDF Editor:

TruePage — Free Download. PDF workspace

TruePage

TruePage is a Windows 11 desktop application for assembling PDFs and images in a visual page workspace.
Price: $20   Size: 269.2 MB   Version: 1.0   OS: Windows
SwifDoo PDF — Free Download. PDF editor

SwifDoo PDF

Looking for methods to edit, merge, split, convert, rotate, annotate, and perform other manipulations on PDF files? SwifDoo PDF matches these needs and handles most PDF-related jobs.
Price: $16.99   Size: 32.3 MB   Version: 1.0.0   OS: Windows, Mac OS, Linux
JOPDF — Free Download. PDF editing tool

JOPDF

JOPDF is a desktop application for managing PDF documents.
Price: Free   Size: 1.3 MB   Version: 2.4.0   OS: Windows, Mac OS, Linux