DP2 Related Queries

[SirishaGorasa]

1. We didn't find anything related to Word to DTBook conversion in the below link.

https://daisy.github.io/pipeline/Get-Help/User-Guide/Scripts/
Is it the case like you would add information related to this conversion in the next release?

2. We have tried converting an Adobe exported XML (Using SaveAsXML) to braille and have got the below error
   
   How can we resolve this?
3. We have tried converting a PDF to HTML using a java library called jPdfWeb, and tried to convert to braille and have got the below error
   
   How can we resolve this?
4. What could be the maximum number of pages which can be translated to EPUB3/Braille Format?
5. What is the maximum number of documents which can be processed in Parallel?
6. What is the SLA to convert one pdf document of 10 pages to EPUB3 format and Braille Format?
7. Is it possible to add PDF as one of the input formats in dp2?

[bertfrees]

Hi @SirishaGorasa. I will answer your questions one comment at a time. If you wish to respond, please "Write a reply" to the comment.

We didn't find anything related to Word to DTBook conversion in the below link.

That is now fixed, see: https://daisy.github.io/pipeline/Get-Help/User-Guide/Scripts/word-to-dtbook

[SirishaGorasa]

Sure, We would try the same.
Thanks.

[bertfrees]

We have tried converting an Adobe exported XML (Using SaveAsXML) to braille and have got the below error

Could you please attach the source XML, or send it to [email protected], so I can check if there are any issues with the file?

[SirishaGorasa]

We will share the source XML to mail.

[bertfrees]

I have gotten the source XML by email. Thanks.

None of the 3 XML documents are DTBook XML. Sampletagged1.xml and Sampleuntagged1.xml (created by SaveAsXML) are in a special "tagged PDF" format. It is not a valid input for DTBook to braille, and I can't find much documentation about this format by the way. Sampletagged2.xml is also in this format, but without the metadata.

[bertfrees]

We have tried converting a PDF to HTML using a java library called jPdfWeb, and tried to convert to braille and have got the below error

Could you please attach the source HTML, or send it to [email protected], so I can check if there are any issues with the file?

[SirishaGorasa]

We will share the source HTML to mail.

[bertfrees]

Thanks for sending me the sample HTML documents. I tried converting them to PEF, and got a result for the Sampleuntagged1.html one. Note that I'm using the latest version of Pipeline for testing. The result is probably not super useful because it lacks formatting, but it is braille and it can be read.

Sampleuntagged2.html, the one created by jPDFWeb, is not accessible, so can never produce any really accessible output. The HTML just contains a big embedded SVG that has all the content in it.

Pipeline didn't produce a good result at all, because a lot of the styling information was transcribed to braille. In a more conventional HTML, this text would be automatically ignored, but this is a very special type of HTML, basically it's an SVG.

I was not able to reproduce the error that you got (about "Error translating inputs").

[bertfrees]

What could be the maximum number of pages which can be translated to EPUB3/Braille Format?

The larger the source file, the longer the conversion will take, but there isn't a maximum number of pages.

[SirishaGorasa]

Ok !

[bertfrees]

What is the maximum number of documents which can be processed in Parallel?

This is determined by the org.daisy.pipeline.procs configuration. If you want help with setting properties, please let me know the platform you are on, and how you are running the Pipeline server.

[SirishaGorasa]

Sure, Thanks !
I will check this configuration link and in case of any help needed, will respond here.

[bertfrees]

What is the SLA to convert one pdf document of 10 pages to EPUB3 format and Braille Format?

I'm sorry, what does SLA mean?

[SirishaGorasa]

https://fastercapital.com/content/Performance-Metrics--Service-Level-Agreements--The-Metrics-of-Commitment.html#:~:text=In%20the%20realm%20of%20performance%20metrics%2C%20the%20commitment,instrument%20known%20as%20a%20Service%20Level%20agreement%20%28SLA%29.
This can help

[bertfrees]

So what is it you are asking exactly?

[SirishaGorasa]

Are there any benchmarks for the solution of converting from Word/HTML to Braille and EPUB3??

[bertfrees]

Oh I see. No, we don't have any benchmark tests.

[bertfrees]

Is it possible to add PDF as one of the input formats in dp2?

It is not in our plans at the moment. Depending on the interest of our DAISY members, this could change, but at the moment it's not a priority. Adding such a feature is possible, but it will surely come with some challenges. Just like was the case with the DOCX input support, reusing off-the-shelve solutions in Pipeline is often not possible because accessibility is usually not taken into account enough in these solutions.

[SirishaGorasa]

Ok !

[NandyJ]

Dear Bert,
For Daisy3 to MP3 conversion Iam providing the book.opf file has the input but Iam facing the issue

Below is the folder structure we received for daisy3 from dtbook to daisy3 conversion.

Please confirm me that .opf should the input for daisy3 to mp3? if not then what extension file i should provide as an input.

[bertfrees]

The .opf file is indeed the correct input. However I can see in your print screen that the input is a DAISY 3 of type "textNCX", which means that the book has no audio. It is text and navigation only. Currently, the DAISY 3 to MP3 script requires that the input already has audio. It will not automatically perform text to speech.

[NandyJ]

OK bert will do the changes as you suggested.

[NandyJ]

For dtbook to daisy3 when i try to click on the switch to CLI view link i find that the synopsis is empty. Could please check and let us know what type command i should use and the option available.

[bertfrees]

The CLI view seems to be broken. Sorry about that. To get the available options, you can also use. the dp2 help command.

[NandyJ]

Ok Bert Thanks!