If the PDFIUM function FPDFText_GetLooseCharBox returns an empty rectangle, then the function FPDFText_GetCharBox is used as a fallback. If this happens, then the text selection was not visualized correctly in older versions. With the fallback, however, this visualization is restored, not as pretty, but better than nothing.

When saving a snapshot to a file, a message was always displayed saying that the file could not be saved, even though it was saved correctly. This issue has been fixed.

At a distance of more than 100 cm, the measuring tool displays an incorrect size. The calculation was carried out incorrectly. This problem has now been resolved.

The measuring tool now also has new setting options for controlling which unit is displayed in the measuring tool. These can be SI units and/or inches.

The text selection did not work correctly in certain PDFs. This was the case, for example, with PDFs that were created using the pages per sheet function. Forms are used internally here and the char boxes were calculated incorrectly. The problem has been fixed.

Tabs can now be moved between instances. This currently only works if the top-level windows are in the same process. If the windows are in different processes, moving the tabs is not possible.

In the sidebar there is a view of the attachments of a PDF file. If attachments are present, you can now open them by double-clicking on an attachment.

The print status display could display the wrong page number for certain print settings. This problem has been fixed.

The entire UI of the PDF24 Reader has been redesigned to include the dark mode that many users have requested. The default mode is derived from the system mode. The mode can be selected in the settings.

Tabs can be opened in a separate frame instance by dragging and dropping them from the current instance. To do this, you have to grab the tab and drag it out of the window and then release the mouse. This will open the tab in its own window.

We were able to further optimize the way a PDF is displayed and the way it starts. The result is an even faster start of the user interface. The PDF24 Reader should be fast and we don't like a slow start of the UI. This improvement makes using the PDF24 Reader even more fun.

Locally installed language files could no longer be used via the command line. This has been fixed.

Double-clicking a file will open the final PDF file with recognized text if the file has already been processed. If the file has not yet been processed, the original file will open.

The „Add Files“ button now displays a pop-up menu that allows the user to select where to add files from. Available options are „From File“ and „From Scanner“.

Image optimization now supports more image formats, including binary images. Images are now also analyzed to select an optimal output format. This often produces smaller PDF output files than before. The reason for this improvement is that sometimes PDFs were generated that were significantly larger than the input files.

Some components or executables can also be used on the command line. Switching to console mode has been further improved to make command line usage close to what you would expect.

If you run pdf24.exe --version on the command line, the version number is displayed.

Some users had problems with the PDF24 standard folder for temporary files. This folder is sometimes not writable and a corresponding message appears. This has been further improved and alternative folders are now used.

When saving PDF files as PNG, JPG, and other image files, the destination file was not overwritten even though the user accepted the overwrite confirmation. This issue has been fixed.

• Ghostscript updated to version 10.04.0
• Webview2 updated to version 130.0.2849.80
• OpenJDK updated to version 17.0.13.11

In addition to the user password or the password for opening a PDF, there is now also a new password field for the owner password. All other input fields in this tool are now also always visible. For some users, the owner password is also important.

The original PDF file size is now displayed instead of the file size of the unlocked PDF file. Loading a PDF involves the unlocking process. The PDF created in this way can sometimes differ significantly from the original. To make the before/after comparison easier, it is important that the file size of the original PDF is displayed.

There is now special handling for additional image formats. These include 1bpp, 4bpp and 8bpp index images with RGB or grayscale color space. The effect is that PDFs created from these image formats are now much smaller than before.

New option assistant.openDirAfterSave that allows the assistant to open the folder and mark the file after a file is saved. This is disabled by default. This new option can also be configured via the settings interface.

Added profile "NONE" to the "Convert to PDF" shell handler to skip profile application, which is sometimes useful.

The Ghostscript option ColorConversionStrategy is automatically set to RGB if it is LeaveColorUnchanged when creating PDF/A files. Otherwise, an invalid PDF/A file may be created.

The Ghostscript option ColorConversionStrategy is automatically set to CMYK if it is LeaveColorUnchanged when creating PDF/X files. Otherwise, an invalid PDF/X file may be created.

There are now additional invoice types in the BT-3 selection field. A correction has also been made here.

Fields for BT-130, BT-134, BT-135 and others were added.

There have been some changes to the user interface. There are now many drop-down menus that hide various functions for the corresponding areas.

Storage has been moved from localStorage to IndexDB, so that much more data can now be stored than before. This allows the user to manage and use many data sets.

There are now more tooltips for various elements to help the user with input. More will follow. The position of these tooltips has also been improved for some elements so that certain content is not covered.

When selecting a record, there is now a search field, which is very helpful when managing many records.

There is now a data manager that allows you to manage data stored in the data stores. You can view each data store, list records, update names, delete entries, and import and export data. You can now also perform data backups this way.

The OK button of the confirmation message has focus, so you can use the spacebar to press the button. This is often a bit faster than having to click the button with the mouse.

Some notes and descriptions may contain line breaks, which are now added to the generated PDF. Previously, new lines were removed.

ZUGFeRD invoices can be created in different languages, regardless of the user interface language. The output language can be selected by the user.

The order of the parameters in the XRechnung for attachments was wrong, so a testing tool marked this as a problem. The order is now correct.

CrossIndustryInvoice xml files and hybrid ZUGFeRD PDF invoices can now be imported. All fields that we currently support are read and displayed. Importing the UBL type of e-invoices is not yet supported. However, this will also be integrated.

You can now save an attachment to the invoice using the dropdown menu of an attachment. This is particularly useful if you have imported an e-invoice.

Output options have been added to control the number and date format.

If select field entries are missing, the user can enter custom select field options.

The invoice generator now supports various data sources. Currently, it supports local data storage and external files that can be loaded from the Internet or the file system. External files are very useful in a multi-user environment.

A settings area has been created in which the user can make certain presets. Data sources that the e-invoice generator now supports can also be created here.

There is now also a field for the intended purpose and there is now also the SEPA direct debit payment method.

You can now enter allowances and charges for line-items. Allowances and charges at invoice level will follow.