allow Content Indexing save to disk

[Shooter3k]

Is it possible to add the option to turn on Content Indexing for files and allow the index to be saved (and reused) to a local database file(s) similar to how the Everything db works?

[void]

No.

Searching the Windows Index for content.
Content Indexing.
minimize_on_close or hide_on_close (help keep the content information in cache -cache is cleared when the window is closed)

[Shooter3k]

Pitty.

There's really no good software for indexing multi gigabyte text files or lots of files like Everything does so well.

[phil2search]

Hello,

In case that helps, there is https://www.recoll.org/ which is open source and only asks for a small donation for the windows exe.
(I am NOT related to the project, just a user)

It deals better than what I've tried before (windows search, docfetcher, X1, etc.) with hundreds of thousands of files, some of which large text (several hundreds MB for me), pdf to ocr, custom parsing filters in python etc. It's highly configurable and the developer is also very nice answering questions. So a lot of similarities with everything .

I prefer not to update the index continuously as it may slow down the machine for no good reason: I have a lot of large text files with a very short life (generate, check, delete) and I know what I saved 5 minutes ago. I can access recent files quickly with everything, along with what I did in the past few weeks and more if using reasonable names (which is desirable). If I don't find it or if I don't know that I have it (large collection of pdf), I launch recoll which is fast.

Somehow linking the two could be nice (send search to or plugin as recoll has a server option) but it's no big deal.

Good luck

[Herkules97]

Is it possible to add the option to turn on Content Indexing for files and allow the index to be saved (and reused) to a local database file(s) similar to how the Everything db works?

I initially never sent my question a month ago, but I am too curious now..What do you mean the db itself can be reused?
There is a way to read the db outside of Everything(EBV)?
I don't know anything about database stuff, I have DB Browser for SQLite which works fine for the types it can read of which EBV is not one from when I've tried.

[Shooter3k]

Right now the content is indexed into ram. I was asking if it could be stored to disk instead so that when you reboot your machine, it could continue where it left off, just like the normal database

[NotNull]

I was asking if it could be stored to disk

It can.
When Everything exits (manually through menu> File > Exit or on a reboot), it will write the indexed content to disk too.
Do you see different behaviour?

[rgbigel]

I was researching AI about this subject, and I got an interesting hint:
Why NotNull’s answer is technically correct
Everything does write something to disk on exit:

The main database (Everything.db)

The content indexing progress state (so it knows which files need rescanning)

The configuration (INI or registry)

This allows Everything to:

➤ Avoid re‑scanning unchanged files
So after reboot, it may appear that the content index “survived”, because Everything:

sees that files have not changed

skips re‑extracting content

instantly rebuilds the in‑RAM index from scratch by scanning only changed files

this partially explains to me why even after there significant content indexing was done, the situation after rebooting the system is NOT that the indixing starts all over. In my experience, something carries over. But i may be wrong.

Anyway, it seems such a waste to not allow everything to keep the content index as well.

[Herkules97]

I was asking if it could be stored to disk

It can.
When Everything exits (manually through menu> File > Exit or on a reboot), it will write the indexed content to disk too.
Do you see different behaviour?

Re-reading this thread because of the latest reply, I think I brainfarted before.
I think he wants a separate database file for specifically indexed file content.
I guess you could then load the content index file with any instance like how you can with file list exports.

But that seems like it'd would require some amount of re-writes for how EBV works.
How much of a benefit would that provide?

Best-case for having an instance that only indexes file content is to either have a massive exclusion filter with NTFS-monitoring or use folder-monitoring. I have 2 folder-only instances that indexes a bunch of properties and one that indexes file content for only smaller txt files.
My issue with NTFS-monitoring is that it would be an up-to-date index. So if I delete files, they are removed from being indexed. Beneficial if you want to always see files that exist but I just want everything indexed.
Speaking of, I wonder if a mechanic to (verb)offline files could be made..So a folder index can have files counted as offline or online depending on which you've toggled to be one or the other..That way you could have something similar to NTFS where files that no longer exist can be set to offline and you can exclude them with online:..But you get to keep all the indexed data.