Search missing many results
Completed
I'm searching for an exact string that occurs in many of my files, and it's not returning the results. It's a substring of a tag.
-
NvUltra indexes words in documents for faster searching and to avoid having to read every file on disk with every keystroke. So searching for a suffix of a word won’t work.
It does incrementally index the first x characters of a word to allow for filtering as you type (Eg b, bo, boo, book). I’d have to to look back at the code to remember exactly what x is, but it’s in the neighborhood of 10.
If you are searching for the prefix of a word that isn’t matching, let me know the entire word and the prefix you are searching so I can test it.
0 -
Thanks, Fletcher Penney. I have some Bible study files tagged with both book names and chapter names. So for example, some files are tagged with #2corinthians while some are tagged with #2corinthians01 and #2corinthians02 etc.
I would think that when I search for #2corinth it should match all the examples above.
0 -
No. You're searching for the **tag** `2corinth` which doesn't exist. You would have to search for the tag that you want as an exact match (the search bar will help you autocomplete tags that exist in your folder). For more complex searches, nvUltra supports Boolean logic, so you could also do `#2corinthians01 OR #2corinthians03` or `#2corinthians03 NOT #2corinthians01`, for example.
Searching for plain text (not tags) works on partial matching (e.g. `2corinth` matches the **text** `2corinthians01` or `2corinthians02`), so that would work. (With the caveat that currently the tag `#2corinthians01` is indexed as a tag, not as text. I updated code so that it is indexed as both, and this will be included in a future update.)
To summarize:
* Tags are indexed "as is" and represent a way to search for an exact subset of files that you define.
* Text is indexed as is and for incremental matches, which allows matching words that contain the prefix you typed
* Future updates will index inline tags (`#foo`) as text in addition to indexing as a tag. Metadata tags are already indexed that way.0 -
Fletcher Penney Awesome! Having it indexed as both will be great. Thanks
0
Please sign in to leave a comment.
Comments
4 comments