Copy old .db and keep index journal entries?

[Herkules97]

I should point out I sometimes can't load into a page on this website and I'm unsure if it's my browser or this website specifically.
I then wait until it goes to an unvisited state and click enter in address bar and it works.

Anyway to my point, I've been looking around on Google and on the first page of this forum and in neither did I find an answer.(But I'm lazy so maybe an answer could've been found)
I have old databases, I wanted to copy them over and I finally did. Made a folder for temporarily copied ones and would use that folder when needed.
But it resets the new index journal feature, new to me, when loading in any database.
Is this because it only loads correctly in the exact location it was or because of USN journal being on?
E1: Oh right, I forgot to also mention I tried opening them in DB Browser for SQLite, but I guess they're not that.
I believe I've exported all .csv before the index journal replaces the bottom that have not been exported, but it would be easier to backup the entire .db so I could then go from the index journal to the normal entries and search a frozen state for whatever I may be looking for if any and also see more tracked stats like all the date [verb?] in both places.

Found related posts:
viewtopic.php?p=35549#p35549
Re: Index Journal Suggestions - post by void » Mon Mar 15, 2021 11:58 am
"Keep item IDs, noted.
Currently, if you change your index, the Journal will be reset and the next item ID will start at 0.
I will look into keeping the journal through index setting changes.
I can't do this due to the Everything Server, clients use the Index Journal to synchronize their local indexes.
An index setting change may just have to force log rolling."

viewtopic.php?p=35149#p35149
Re: Whats the difference between indexing and USN Journal? - post by NotNull » Mon Mar 01, 2021 9:20 pm
"Include in Database means that Everything will do an initial scan and include all found files/folders from this drive it in it's database.

The USN Journal is needed to read file changes of your drive, so the database can be updated to reflect those changes.
The USN Journal is part of Windows; Everything uses it.
If you disable the USN Journal, Windows will re-create it at boot time, so it will be enabled again.


Long story short: don't disable the USN Journal. Everything needs it to get filesystem updates ."

LONG ,MAYBE POINTLESS, SO I PUT IT ON THE BOTTOM BUT IT BELONGS AT THE TOP.
I realised too late that I should back up my .db files so I can load them in later and look into the past. Not late because of the normal entries, but because of the new index journal.
I had previously been exporting as .csv but it lacks some of the useful sorting possible in normal entries like date created and date accessed.
This was an issue in 1.4 too, like date accessed missing in an export. I think normal entry exports also lack date accessed in 1.5, but the index journal exports lack even more.
I probably made a post about that, but let's ignore that because this is about the .db not the exports.

[void]

I should point out I sometimes can't load into a page on this website and I'm unsure if it's my browser or this website specifically.
I then wait until it goes to an unvisited state and click enter in address bar and it works.

Do you see a timeout error page? or any error page?

Anyway to my point, I've been looking around on Google and on the first page of this forum and in neither did I find an answer.(But I'm lazy so maybe an answer could've been found)
I have old databases, I wanted to copy them over and I finally did. Made a folder for temporarily copied ones and would use that folder when needed.
But it resets the new index journal feature, new to me, when loading in any database.
Is this because it only loads correctly in the exact location it was or because of USN journal being on?

Please load the db with -read-only
For example:

Everything.exe -db "My Everything.db" -read-only

Without -read-only, any changes to your indexes will cause Everything to rebuild.
A rebuild will reset your Index journal.
Tools -> Debug -> Statistics -> Build -> "Last Rebuild Reason" will show the last rebuild reason.



Date accessed changes are not logged in the USN Journal.
Everything cannot monitor changes to the date accessed timestamps.
Everything will rescan date accessed timestamps when the file is "changed". (modified, renamed, attributes-change, etc..)

The Index Journal only exports visible columns.
Right click the column header to show all columns.

[Herkules97]

I should point out I sometimes can't load into a page on this website and I'm unsure if it's my browser or this website specifically.
I then wait until it goes to an unvisited state and click enter in address bar and it works.

Do you see a timeout error page? or any error page?

Fortunately recorded it on video, no it's Firefox probably giving up loading a page. It loads for maybe a minute then the page stops and it's blank like a not found message but nothing.
If you haven't noticed anything from others, it's probably just my Firefox on that specific session messing up.


Ok so read-only..

It's unfortunate, I made a template folder so I could have everything ready and then I only added all the columns for one of the databases after it was made. Fortunately only the system database gets enough changes to go above the 3.5GB limit I set so if I don't have that terrible of a memory every other database can still be scrolled down to the first number in the journal and have an improved export.

Also there is a redundant quote thingie at the end .
Nevermind, it went away now

[Herkules97]

Ok so read-only..

Should I report successes or is it assumed that success was had?
Because it worked.

Still curious, if I want to add a new folder or drive there is no way to keep the index journal intact and have it start adding from the new addition?
At least as the program is right now with 1367a.

Also I was staring at my old answer and edited it and realised..Is there a point to me exporting into a text file versus storing the whole database if we ignore space as an issue?
And is there a way to read the databases outside of the program or no point? Like with Firefox places.sqlite I don't have to start Firefox to read it using the DB Browser or Mozilla History View. There is less of a point with this program given it launches the way it does and there is no extra like when opening a browser, but I am still curious.

[void]

if I want to add a new folder or drive there is no way to keep the index journal intact and have it start adding from the new addition?

The index journal will be lost when adding the new volume.

Also I was staring at my old answer and edited it and realised..Is there a point to me exporting into a text file versus storing the whole database if we ignore space as an issue?

Exporting as TXT with all columns shown should export all index journal data.
You can save the whole database if you have room.
Consider zipping/7zipping the db files.

I have plans to add index journal support to ES (the command line interface) at some stage.

And is there a way to read the databases outside of the program or no point?

ES can read your loaded Everything.db
db2efu will eventually be updated for Everything 1.5.

[Herkules97]

Also I was staring at my old answer and edited it and realised..Is there a point to me exporting into a text file versus storing the whole database if we ignore space as an issue?

Exporting as TXT with all columns shown should export all index journal data.
You can save the whole database if you have room.
Consider zipping/7zipping the db files.

It turns out exporting as csv this entire time included date accessed.

Cleaned up, here's a much simpler post:

This should apply to both index journal and dynamic database, any way to export .csv with the UNIX or Microsoft filetime format that the .efu has?
And for only the dynamic database, any way to load in a file list from .csv with all the stuff working? Unlike .efu, it doesn't load the dates for me.
I presume you're meant to use other software, but would be very convenient if it all worked within Everything.

I don't mention loading in .json because the time change to the .csv exporting can apply to both index journal and dynamic database.

This would be a toggle in advanced settings or maybe even in the format selection when exporting so those that use it for excel or whatever .csv normally is used for can still be used in case it's imported to something that can not automatically convert the time.

[harryray2]

David, you mentioned that exporting as a txt file keeps the index journal..Gave it a try, but I couldn't see where the index journal entries are.

Also, is it possible, when saving as an EFU, to also save the index journal?

[void]

David, you mentioned that exporting as a txt file keeps the index journal..Gave it a try, but I couldn't see where the index journal entries are.

You will need to export from the Index Journal: (not from the main File menu -> Export)

Index -> Index Journal -> Export button.

Only the visible columns are exported.

Also, is it possible, when saving as an EFU, to also save the index journal?

No.
EFU files try to avoid any personal information.

Exporting the index journal from ES might be possible in the future.