Hi Morgan ,

This html or PDF text handled by DTSeacrh so we haven't any control on that. the dtsearch have to provided the formatted text.

Ok. Thanks Harsh .

Radomir , is there anything you could do over the next 2 days to resolve this issue with the excerpts?

Thanks,

Morgan

Harsh this is on the web application side. The search service returns you the highlighted paragraph as HTML. The paragraph character is properly encoded as an HTML entity - see https://dev.w3.org/html5/html-author/charref
I guess you're adding the paragraph to the page as text instead of raw html.

Hi Radomir ,

As discussed in today's call, The JSON request from dtsearch provide us unicode text. so, please look into this and provide feedback.

Harsh please send me a sample HTML for which this happens. I don't see the problem with content from PDF so it might be something with the HTML or paragraph extraction.

Hi Radomir ,

Here, I have attached sample html where we get Unicode latter while click on paragraph number.

IC-0012-10.html 788 KB • Download

Harsh the indexer I sent you today also addresses the HTML entity issue in paragraph highlights. Please recreate FTS index and test.

Hi Radomir ,

After updating the FTS Indexes, the html excerpt Unicode issue is resolved.

But, when I pass the following Search request to get text then it is doesn't fetch the text from PDF (Page 4)

{"searchRequest":"tribunal","SearchType":"3","Stemming":"false","WordNetSynonyms":"false","Fuzzy":"false","Fuzziness":"1","paraId":"EE073142D5FC08C5509D1A05E23D7D27#NA=="}

But in Same Document if I pass this search request to get text then it display text. (Page 8)

{"searchRequest":"tribunal","SearchType":"3","Stemming":"false","WordNetSynonyms":"false","Fuzzy":"false","Fuzziness":"1","paraId":"EE073142D5FC08C5509D1A05E23D7D27#OA=="}

Here, I have attached PDF File. The text is fetching from this PDF file. In old indexing it was works for above both search request.

UN-0040-44 - Yukos v. Russia - Respondent Appeal.pdf 31.5 MB • Download

Harsh I sent you a fix for the missing paragraph in folder 2021-03-27/search

Hi Morgan ,

This issue is resolved on staging.islg. Please check and confirm.

Looks great Harsh . The only outstanding item in the FTS is changing the default setting for the search so that Stemming and Fuzzy Typo are enabled for 1 letter:

image.png 194 KB • Download

Thanks,

Morgan

Hi Morgan ,

The above change has been done on staging.islg. Please check and confirm.

Looks good Harsh on both staging.islg and app.islg.

Note the issue I experienced with the searches in the Subject Navigator, FTS and Dispute Documents noted here: Re: 2021-03-25 08.19.13 Weekly Tologix-DevIT-Industrial Meeting 81175607260.mp4 - TOLOGIX - ISLG App Rebuild. However, I will assume this is just a indexing issue for the time being, and will mark this to-do complete.

Thanks,

Morgan