maxice8

On Thu, Mar 08, 2018 at 10:41:12PM -0800, Ben Boyter wrote: > OK so for this #39 #40 #41 its all the same issue. Going to close the others as...

found another one. ykpers has BSD-2-Clause ``` $ lc masterdir/builddir/ykpers-1.19.0 -------------------------------------------------------------------------------- Directory File License Confidence Size -------------------------------------------------------------------------------- masterdir/builddir/ykpers-1.19.0/ COPYING BSD-3-Clause-Clear 89.26% 1.6K -------------------------------------------------------------------------------- ``` ``` Copyright (c) 2008-2014 Yubico AB...

looks amazing :o

ping

Here is [feed json](https://raw.githubusercontent.com/maxice8/vespa-bug-repro/main/data.json) of the document, the field in question is `content`. I modified the schema from the [repro repository](https://github.com/maxice8/vespa-bug-repro) as follows: ``` schema diary_page { stemming: none document...

Thanks, I have modified the schema and tried again. Unfortunately, the `phrase("JOAO", "PAULO", "MARTINS", "DE", "SOUZA", "BARBOZA")` produces no summary as it has no matches, so I tried the one...

Hello, We currently have the language set to Brazillian Portuguese in the language across the schema. ``` schema base_definition { document base_definition { field language type string { indexing: "pt-BR"...

I modified the content of `data.json` to only include the content and it indexed the `de` correctly. ```json { "id": "toplevel", "relevance": 1.0, "fields": { "totalCount": 1 }, "coverage": {...

Found another instance of this bug in the following document (see put for id:diary_page:diary_page::2) of the following Vespa Feed file https://github.com/maxice8/vespa-bug-repro/blob/main/data.json. The query using `phrase("MATEUS", "DO", "VALE", "MENDES")` matches and...

With 8.283.46 and the new `match { max-occurrences: }` setting bumped it now stores all the tokens that were skipped before.