I would like to report an issue regarding how PDF-XChange Editor generates page labels (logical page numbers) when edited directly via the Thumbnails pane or page properties.
Issue Summary
- When changing a page label for any page other than the first page using the Thumbnails pane or page properties, PDF-XChange Editor fails to include the mandatory "page index 0" in the /Nums array of the Number Tree. This causes Adobe Acrobat to ignore the page labels entirely, as it violates the PDF specification.
- Open a multi-page PDF document.
- Go to the Thumbnails pane.
- Select the 2nd page (physical page 2, index 1) and change its page number/label directly (e.g., set it to "two"), or modify it via Page Properties.
- Save the file.
- Open the saved PDF in Adobe Acrobat.
- The custom page label is displayed correctly in Adobe Acrobat (as it does when using the "Number Pages" dialog).
- Adobe Acrobat ignores the page labels completely and only shows the default physical numbers (1, 2, 3...).
(Note: Foxit PDF Editor and PDF Annotator display it correctly because they have more permissive parsers, but Acrobat strictly enforces the spec).
- When inspecting the internal structure of the generated PDF, the /Nums array looks like this:
- /Nums [ 1 << /P (two) >> 2 << /S /D /St 3 >> ]
According to ISO 32000-1:2008, Section 12.4.2 (Page Labels), it explicitly states:
- "The tree shall include a value for page index 0."
Please look into this implementation inconsistency between the two UI methods.
- For your reference, I have recorded verification videos and provided the output PDF files for three different methods of setting page labels.
The ultimate goal for all three methods is to have the page labels displayed as "1 two 3 4 5 6 7 8 9 10" in the Thumbnails pane.
Method 1: Direct edit of the 2nd page label via Thumbnails pane (Buggy)
I changed the page label under the 2nd page thumbnail directly to "two". As described above, this method fails to display the custom labels in Adobe Acrobat.
Method 2: Direct edit with a workaround for Index 0 (Succeeds)
In addition to Method 1, I temporarily cleared the page label of the 1st page (making it blank) and then changed it back to "1". Doing this forces the editor to explicitly write the definition for "index 0". As a result, the issue from Method 1 is bypassed, and Acrobat can display the labels correctly.
Method 3: Using the "Number Pages" dialog (Succeeds)
I used the "Number Pages" dialog to set a section starting from page 2 with the prefix "two", and another section starting from page 3 with an initial value of 3. With this method, a section for page 1 (index 0) is automatically generated without any manual intervention. Acrobat displays the labels perfectly.
Below, you will find the verification videos, the resulting PDF files, and the extracted /Nums array of the Number Tree for each method.
- Extracted Code: /Nums [ 1 << /P (two) >> 2 << /S /D /St 3 >> ]
- Extracted Code: /Nums [ 0 << /S /D >> 1 << /P (two) >> 2 << /S /D /St 3 >> ]
- Extracted Code: /Nums [ 0 << /S /D >> 1 << /P (two) >> 2 << /S /D /St 3 >> ]
Thank you for your hard work on this great software.
Best regards,
rakunavi
- PDF-XChange Editor PRO Version: 11.0.0
- Adobe Acrobat Reader Continuous Release Version 2026.001.21563
- Foxit PDF Editor Pro 13.1.2.22442
- OS Version: Windows 11 Pro / Home 25H2 Build 26200.8457
- PC Model: GMKtec Nucbox M7 Pro with HUION Kamvas Pro 19 / Lenovo IdeaPad C340-15IWL