Accessibility Tags in Hebrew Text are Incorrect After DOCX to PDF Conversion using .NET

When creating a PDF there is a problem in the way the tags structure is created, this create an issue when using a read aloud program, as the order of the tags is generated in a LTR ways instead the expected RTL for hebrew. Maybe I am missing a property or do to much. What I need is an accesible document so the tags are readable. i will like to save the paragraph for example, i also found in the properties something strange (not the main issue probably)
So I will say, Do I have control in the order of the tags from LTR? Can it be creatd as Paragraph? Can i change the Properies to hebrew and RTL in Properies? Thanks

The code:

var options = new PdfSaveOptions
            {
                ExportDocumentStructure = true,
                OptimizeOutput = true,
                FontEmbeddingMode = PdfFontEmbeddingMode.EmbedAll,
                JpegQuality = 100,
                ColorMode = ColorMode.Grayscale,
                UseCoreFonts = false,
                Compliance = PdfCompliance.PdfA1a,

                OutlineOptions =
                {
                    CreateMissingOutlineLevels =true,//outlineOptions.CreateMissingOutlineLevels,
                    CreateOutlinesForHeadingsInTables=true,//outlineOptions.CreateMissingOutlineLevels,
                    DefaultBookmarksOutlineLevel=  0,//outlineOptions.DefaultBookmarksOutlineLevel,
                    ExpandedOutlineLevels = 6,//outlineOptions.ExpandedOutlineLevels,
                    HeadingsOutlineLevels = 6,//outlineOptions.HeadingsOutlineLevels,
                }
            };

            doc.Save(instructions.Target, options);

871085.pdf (166.5 KB)

2020-11-16_12h50_44.png (142.6 KB)

2020-11-16_12h55_36.png (45.1 KB)

@kushnir_l-1

Could you please attach your input Word document here for testing? We will investigate the issue on our side and provide you more information.

871085.zip (340.8 KB)

@kushnir_l-1

We have logged this problem in our issue tracking system as WORDSNET-21421. You will be notified via this forum thread once this issue is resolved.

We apologize for your inconvenience.

Do you know when this problem will be corrected? more or less? Is a matter of weeks, months?
For us is a critical Bug in the system
Many thanks

@kushnir_l-1,

WORDSNET-21421 is currently pending for analysis and is in the queue. We will inform you via this forum thread as soon as this issue will get resolved in future or any time frame (ETA) may be available. We apologize for any inconvenience.

A very unpleasant situation!

Any News we are in huge troubles due the facts that the documents have to be Accessible to the population and as that your solution was perfect for us and right now due the fact that the bug is not corrected and there is not even a compromise we have to check alternatives. It’s a petty that we are in this situation and I don’t really understand how after 45 days we are absolutely in the beginning of the list. I really expected some days to have an evaluation and compromise of the solution, the situation is not acceptable and of course I ask to elevate this request to the appropriate level of management to attend our complain ASAP. In case this is not possible, please let me know so I will decide what to do.

@kushnir_l-1

We try our best to deal with every customer request in a timely fashion, we unfortunately cannot guarantee a delivery date to every customer issue. Our developers work on issues on a first come, first served basis. We feel this is the fairest and most appropriate way to satisfy the needs of the majority of our customers.

Currently, your issue is under analysis phase. Once our product team completes the analysis of your issue, we will then be able to provide you an estimate.

Please note that you reported this issue in free support forum and it will be treated with normal priority.

February and still no information about the issue! Do you have any information about if/ when and how the bug will be solved?

@kushnir_l-1

Your issue is now under analysis phase. We will share the ETA of this issue after completing the analysis. Thanks for your patience and understanding.

Hi, Any news after the analysis?
Leo

@kushnir_l-1

Unfortunately, there is no ETA available for this issue at the moment. We will be sure to inform you via this forum thread once there is any update available on it.

@kushnir_l-1

Could you please check the attached output PDF and let us know if you are satisfied with output? out.new.aw.pdf (137.5 KB)

Thanks for your cooperation.

There are no tags in the document
From the elements order seems not fixed
New Bitmap Image.jpg (125.4 KB)
The cells order in the table should be RTL
But Only after there is tag structure I can tell you for sure

Thanks
Leo

@kushnir_l-1

The layout of tables in document looks fine. MS Word and Aspose.Words generates the same output. Could you please share some more detail of current issue? Please also share how are you checking accessibility issues.

based in word saving a document to pdf it creates tags (also aspose words supports the save with tags)
Table-hebrew.pdf (62.6 KB)
Table-hebrew.docx (12.0 KB)
What you can see:
1- TAGS
2- The order of the TAGS, this is used by the screen reader in order to read loud the text in the tables so it should be in the same form
If you provide me the pdf with tags I will be able to provide you with more details
Right now the pdf you gave is flat, must be PDF/A compliant or t least tag structure included, aspose words support this

options.jpg (77.8 KB)

@kushnir_l-1

We have logged your concerns in our issue tracking system. We will inform you once there is an update available on it.

@kushnir_l-1

Can you please check the attached PDF in the read aloud program and let us know if it is your desired output?
871085-logical-order.aw.pdf (143.0 KB)

The reading order of the tables is incorrect - the left cell in the row is read first. In Hebrew, the right cell in the row is required to be read first. This is a definition in Word - of the direction of the table.
image.png (10.0 KB)
This can be seen both in the structure of the table tags(see image) and in the demo of the reading of the screen reader
image.png (64.6 KB)
See video related
https://drive.google.com/file/d/1INxM4y16nCHl287zIecHTEe_H8B8G6oe/view?usp=sharing
This was checked by an Accessibility Professional
I hope now is clearer
The order of the TAGS in a Table is opposite to the LTR

1- Word is able to do it perfectly
image.png (28.5 KB)

image.png (36.7 KB)

2- Aspose do it LTR
image.png (18.0 KB)

So in that case to test it you just need to create the pdf from the original docx in the zip and then open it with Acrobat and test the TAGS in the tables, if the order is from Up to Down: Right To Left you are in the right direction

Thanks In advance!
Leo