Hello Morgan ,

We have done some changes to resolve above issues:

2.jpg 217 KB • Download

Hello Morgan ,

Waiting for your feedback on above.

Hi Anil ,

Apologies for the delay. I commented on this last week, but it didn't get sent for some reason.

Re formatting the excerpts, its essential that we maintain the formatting/bulletting for the excerpts, because the text loses its meaning without it. Please do some R&D for a solution.

Thanks,

Morgan

OK Thanks Morgan ,

We will do some r & d on this and let you know.

Hello Morgan ,

We did r & d and improved the format but as per your comment in
https://3.basecamp.com/3691577/buckets/3565178/todos/1104177090 point no #4, we need to do further r & d to check its feasibility. As I commented above the bullet numbers are not part of the html document (DOM) they are generated automatically and hence it is difficult to fetch them by coding. 

We will do more r & d and let you know if we found anything.

OK. Thanks Anil . This is an important issue, so please continue with R&D, and let me know if you're unable to come up with a solution by next week.

Thanks,

Morgan 

Hello Morgan ,

We have spent much time for this r & d but still not able to find out any solution. This is something we can consider for post prototype when we develop new PDF to Html conversion tool. We will still continue our r & d on this to find something. Please suggest. 

Hi Anil ,

Understood. However, we'll have to come up with another solution if the current direction is not working, so please continue to work on resolving the problem. Perhaps we need to come up with a different way of producing extracts that can satisfy our requirements.

Note that we can continue to work on the front end while we perform content analysis for the prototype, because it shouldn't affect the work performed by editors on the admin site.

We have to come up with an acceptable solution to this problem, and the current extracts are insufficient for our purposes.

Ryan and Juan , please add your suggestions on possible solutions to this problem.

Thanks,

Morgan 

Hi Morgan and Anil

Juan and I have taken a look and during our test it appears as though the formatting and elements for our test document are appearing as expected (could use some slight style adjustments for the list items).

http://p3lg.tologix.com/Home/ViewTagExcerpts?tagDocumentId=533&tagId=413&oper=tag&keySearch=termination-for-convenience&keyCount=0

See screenshots below.

In tagging mode in admin:

Screen Shot 2018-08-28 at 10.14.15 AM.png 111 KB • Download

Excerpt view on front-end

Screen Shot 2018-08-28 at 10.14.28 AM.png 125 KB • Download

Hi Ryan ,

Thanks for running through those tests. Yes, the styling elements for list items seems to be the last hurdle for this to-do.

Anil , perhaps something could be done on the HTML conversion side of things to deal with this issue? Note that it's very important that the presentation looks good, and we're not there yet.

Thanks,

Morgan

Hi Anil ,

Just following up on my comment above about looking into whether we can resolve the outstanding issues in this to-do through the HTML coding process. 

Thanks,

Morgan 

Hello Morgan ,

Sorry for late reply.

We will upload new converted html document and check whether this can be resolved. 

OK. Sounds good Anil . I'm wondering whether there is changes that we can make to the HTML code within the documents that will compensate for the problem if it persists.

Morgan

Hi Anil ,

Let's move on to this to-do next (assuming not all capacity is needed for the PDF to HTML conversions).

Thanks,

Morgan 

Hi Morgan ,

We will start on this from today. After new HTML conversion we need to look few more things too like TOC, Text highlight etc. as new HTML files' structure are totally different than the old one.

Makes sense to me, Anil .  

Thanks,

Morgan

Hello Morgan ,

We have replaced a new html file on p3lg and did some changes for formatting. Now it looks good but still there is one limitation because of the bullet numbers of the html file. If user will select a paragraph with few bullets between the main list (e.g. from 5 to 8) then the bullet numbers will start from 1 instead of the selected one (so it will become 1 to 4). This issue is because the application cannot identity which bullet is selected between the list. Still, we will do more r & d to find out something.

Please review following URL and feedback:
http://p3lg.tologix.com/Home/ViewTagExcerpts?tagDocumentId=611&tagId=413,411,412,414&oper=doc&keySearch=test-tag-1-or-test-tag-2-or-termination-for-convenience-or-testing-new-tags&keyCount=0

Hi Anil ,

The presentation of the excerpts within the new HTML documents looks great. However, we'll definitely need to resolve the bullets numbering issue, because it will misrepresent what sections within the document are being displayed. Please do more R&D on the issue.

Thanks,

Morgan 

Hello Morgan ,

JFYI:
We are in progress for presentation of the excerpts. We got little success in this and need to check and develop many scenarios. We will keep you updated on this.

Hi Anil ,

Happy to bear progress is being made. Thanks for the update.

Morgan

Hello Morgan ,

We are still in progress with this issue. We are able to handle 70% cases and uploaded on p3lg for your review. We are still working on remaining cases and will update you soon.

Sounds good Anil . Thanks for the update. Looking forward to seeing the final product.

Morgan

Hi Anil ,

Do you have a further update on this to-do?

Thanks,

Morgan

Hello Morgan ,

We are still working on this as there are still many cases that need to be resolved. We hope that this will be completed by mid of next week. I will keep you updated on this.

Sounds good. Thanks for the update Anil .

Do you anticipate this issue being resolved permanently when the immediately problem is resolved? In other words, will this fix apply to any future document uploaded to this system (assuming it is built based the same HTML template)?

Thanks,

Morgan 

Hello Morgan ,

This fix is for all the existing HTML documents only, we might need to change the logic for the new HTML templates which will be decided during the PDF to HTML conversion r & d. This is something which will vary based on the structure of the HTML tags of documents, hope you understand.

Understood. Thanks Anil .

Please ensure that the PDF to HTML R&D explicitly integrates this into the requirements for the project. We need to ensure that any HTML on the system is able to display excerpts with the appropriate formatting.

I've added Devaang to this thread to ensure he understands that this needs to be integrated into the project.

Thanks,

Morgan

Hello Morgan ,

We have uploaded a beta version on p3lg, please check at your end and let us know the feedback.

Hi Anil

Could you please provide a link to a document where I can test this? I'm having issues using search right now.

Hi Ryan ,

I'm still creating the tags necessary for testing. I'll let you know when content is on the system.

Morgan

Great, thanks for the clarification Morgan  

Hi Anil and Ryan ,

The prototype content has been uploaded to the system, and further to the video below, there is a problem in the way the tag excerpts are being displayed for certain documents.

Thanks,

Morgan

Problem with the way tag excerpts are displayed (18-Oct-18).mp4 14.1 MB • Download

Hi Anil ,

Any updates on the issues outlined above?

Thanks,

Morgan 

Hi Morgan ,

This issue has been resolved. Please check and feedback.

Hi Anil ,

Further to the screenshot below, the large gaps have been removed concerning the example above, but the clause numbers are not displaying properly:

image.png 153 KB • Download

Please resolve.

Thanks,

Morgan 

Hello Morgan ,

We have resolved this on p3lg. Please review and feedback.

Hi Anil ,

Looks good to me. Marking to-do complete.

Thanks,

Morgan