Jump to content
  • 0

How does Evernote Search feature work?


Vidalia

Idea

7 replies to this idea

Recommended Posts

  • 0
  • Level 5

In general yes: Search is based on a search index. There are two versions of it: One is locally installed, the master is on the server. Now it depends on the client you use (especially if legacy or one of the v10-clients) how it searches in detail. Common words that will be found in many places are excluded from the index. Search is not Boolean, but follows an own syntax.

Some more can be found here:

Link to comment
  • 0
  • Level 5*
1 hour ago, Vidalia said:

When I search for few words, does Evernote scan my notes in real time or it checks for those words in an index type database?

If it is database driven, does it mean EN indexes every single word in my every note?

If you're online, searches are executed on the server

Yes, there is a search index of "every single word"
- it excludes "stop words" like the/and/or    
- it excludes special characters

The search index also covers attachments, including OCR of images

 

Link to comment
  • 0

So that means every time I modify a note, in the background EN updates its database with keywords?

For example, if I have just 2 notes with following content.

[NOTE 1] Jill had a little lamb.

[NOTE 2] Jack and Jill went up the little hill.

EN search database will have values like this?
 

Quote

 

Jill = 1, 2

Jack = 2

little = 1,2

went = 2

hill = 2

lamb = 1

 

 

 

Link to comment
  • 0
  • Level 5*
10 minutes ago, Vidalia said:

So that means every time I modify a note, in the background EN updates its database with keywords?

For example, if I have just 2 notes with following content.

[NOTE 1] Jill had a little lamb.

[NOTE 2] Jack and Jill went up the little hill.

EN search database will have values like this?

Simplistic, but the index is something like that   
I should have mentioned it's not case sensitive; so jill, jack, little

Link to comment
  • 0
  • Level 5

Yes sure, that is how it works. It holds an indexed list of key words, and behind it a table to where these words are found.

Every note has an unique denominator (a code that is unique to that note), so it searches very fast from key word to all notes where it occurs.

Because of the OCR (that can be fuzzy) the list is probably longer than all real key words, because there are options of words in the list to make search work when the OCR is not exactly on the spot. This is why sometimes there are search results in the list where actually the key word is not found. The index is hidden from the user.

Link to comment
  • 0
  • Level 5

Whoever knows ...

EN is hosted on Google servers (but not inside of Google, they run their own environment), so there is enough computing power behind it. Today there are so many advanced data base designs that I doubt I would understand it. And probably EN holds it pretty close anyhow, because it is a main advantage point that search is universal, reliable and very fast.

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...