Wrong naming of the folders under revisions folder

Get help for specific problems
Posts: 7
Joined: 17 May 2023

cyoram

I am using for versioning the time stamp folder convention.
But instead of getting \revisions\YYYY-MM-DD hhmmss\folder\file.doc
I have \revisions\YYYY-MM-DD hhmmss\file.doc

thanks four your help
User avatar
Posts: 3608
Joined: 11 Jun 2019

xCSxXenon

For files in the root of the location, that is what should happen
Posts: 7
Joined: 17 May 2023

cyoram

For files in the root of the location, that is what should happen xCSxXenon, 17 May 2023, 14:41
But the files I speak about are not in the root location
Posts: 944
Joined: 8 May 2006

therube

Are all the files ending up in \YYYY-MM-DD hhmmss\?

As in, might you have "file.doc" in both \YYYY-MM-DD hhmmss\ & also in \YYYY-MM-DD hhmmss\folder\ ?
User avatar
Posts: 3608
Joined: 11 Jun 2019

xCSxXenon

https://freefilesync.org/manual.php?topic=versioning

It's not super clear in the examples provided. I am interpreting the documentation in a way that suggests each root location gets a versioning folder in the location you specify, and then everything deleted/overwritten in that location gets put into the root of the versioning folder, regardless of folder depth/location.

I tested this on my own and the subfolder was retained in the versioning folder, so my interpretation is wrong. Considering it works as expected for me, I am wagering that you may have some incorrect settings. can you post a screenshot of your FFS window with the versioning settings and locations visible?
Posts: 944
Joined: 8 May 2006

therube

What I think is going on, is that for any directory pair, any topmost items are backed up to the root of your specified versioning directory.

For subdirectories to a particular pair, backups go to a same named subdirectory.

PAIR (left side): C:\123\MP -> somewhere\someotherdir
PAIR (left side): C:\LIB -> somewhereelse\somenewlib

C:\123\MP\file1.txt
C:\123\MP\DIR2\file2.txt

C:\LIB\ffs.exe
C:\LIB\FFS\ffs_for_xp.exe

VERSIONS = C:\OUT

C:\OUT\20230517.123456.file1.txt
C:\OUT\20230517.246809.ffs.exe

C:\OUT\MP\20230517.654321.file2.txt
C:\OUT\LIB\20230517.369280.ffs_for_xp.exe

So file1.txt gets stuck into the root of VERSIONS
where file2.txt ends up in VERSIONS/MP/file2.txt.

It is odd, in that respect.

Note that files1.txt & ffs.exe, though sourced from different trees, are at the route of each tree (directory pair) & as such, get stuck "together" in the root of the VERSIONS directory - which is very odd, & really unexpected, given that there is no relation between anything in MP & anything in LIB.

You would expect (& it would be much more logical) if the tree structure was recreated as is, but instead it is (i just lost the word, & not) "compressed" to the topmost (& that's the wrong word too) [& which ends up in VERSIONS] & subdirectories thereof.

(Oh, that was a confusing explanation, no ;-).)
Posts: 944
Joined: 8 May 2006

therube

(
fastcopy [a different program] mentions something similar, dealing with destination directory paths & how the trees are recreated therein:

1. The last character of DestDir is '\' Copy Source directory, including itself, to DestDir.
(DestDir\SourceDir\Contents_of_SourceDir)

2. The last character of DestDir is NOT '\' Copy Source directory's contents to DestDir.
(DestDir\Contents_of_SourceDir)
However, if multiple directories are specified in Source, the behavior will be changed to 1.

(mini tips) If you want to copy Source directory's contents to the DestDir that is specified drive root, please append "*" to Source directory.
For example, if Source is "C:\Folder1\*" and DestDir "D:\", the result is D:\(Contents_of_Folder1)


It always takes me a bit of thinking to get my head around the meaning, such that I get the end results I want.)
Posts: 944
Joined: 8 May 2006

therube

Where I say "root" above, I guess that should really be "base".
(And I'm sure I've used other terminology incorrectly, too.)

So BASE files from any directory pair, get backed up (together) to the ROOT of the VERSIONS directory.

(heh, lets confuse some more...)

Descendants of the BASE (so subdirectories of BASE of a directory pair) get put into a named subdirectory of the VERSIONS directory.
Posts: 7
Joined: 17 May 2023

cyoram

https://freefilesync.org/manual.php?topic=versioning

It's not super clear in the examples provided. I am interpreting the documentation in a way that suggests each root location gets a versioning folder in the location you specify, and then everything deleted/overwritten in that location gets put into the root of the versioning folder, regardless of folder depth/location.

I tested this on my own and the subfolder was retained in the versioning folder, so my interpretation is wrong. Considering it works as expected for me, I am wagering that you may have some incorrect settings. can you post a screenshot of your FFS window with the versioning settings and locations visible? xCSxXenon, 17 May 2023, 16:40
These are my FFS Windows with the versioning settings and locations.
Image
Attachments
freefilesync_sync_setting.jpg
freefilesync_sync_setting.jpg (455.97 KiB) Viewed 865 times
User avatar
Posts: 3608
Joined: 11 Jun 2019

xCSxXenon

Looks fine... Can you provide an example of the issue? For my test, I simply made a text file in a subfolder, synced it, changed the text file, synced again. It put the deleted copy in the correct subfolder in the version folder I entered.
Posts: 944
Joined: 8 May 2006

therube

- if you have something like this:
d:/inter/                       interview.txt
d:/inter/pass/                  pass-inter.txt
d:/inter/fail/                  fail-inter.txt
d:/users/family/pictures/       picture.jpg
d:/users/family/pictures/negev/ oldest.jpg
d:/users/family/videos/         video.mp4
d:/users/family/videos/disney/  wewontgothere.mp4

- i suspect you're seeing something (at least similar to)
  this (where interview.txt & picture.jpg & video.mp4 all end up in /revisions/
d:/ffs/revisions/                interview.txt
d:/ffs/revisions/                picture.jpg
d:/ffs/revisions/                video.mp4

d:/ffs/revisions/inter/pass/     pass-inter.txt
d:/ffs/revisions/inter/fail/     fail-inter.txt

d:/ffs/revisions/pictures/negev/ oldest.txt
d:/ffs/revisions/videos/disney/  wewontgothere.mp4

- where you want, all your interview texts to be in the /inter/ directory,
  all your pictures to be in the /pictures/ directory, &
  all your videos with the /videos/ directory
d:/ffs/revisions/inter/                 interview.txt
d:/ffs/revisions/inter/pass/            pass-inter.txt
d:/ffs/revisions/inter/fail/            fail-inter.txt

d:/ffs/revisions/family/pictures/       picture.jpg
d:/ffs/revisions/family/pictures/negev/ oldest.jpg

d:/ffs/revisions/videos/                video.mp4
d:/ffs/revisions/videos/disney/         wewontgothere.mp4
(Albeit, in your case, this would actually happen in dated subdirectories.)
Posts: 7
Joined: 17 May 2023

cyoram

https://freefilesync.org/manual.php?topic=versioning

It's not super clear in the examples provided. I am interpreting the documentation in a way that suggests each root location gets a versioning folder in the location you specify, and then everything deleted/overwritten in that location gets put into the root of the versioning folder, regardless of folder depth/location.

I tested this on my own and the subfolder was retained in the versioning folder, so my interpretation is wrong. Considering it works as expected for me, I am wagering that you may have some incorrect settings. can you post a screenshot of your FFS window with the versioning settings and locations visible? xCSxXenon, 17 May 2023, 16:40
These are my FFS Windows with the versioning settings and locations.
Image cyoram, 18 May 2023, 08:14
Here is an example of what I have
d:/inter/family/example.doc <== I deleted this file
d:/freefilesync/revisions/example.doc <== This is what I get
d:/freefilesync/revisions/2023-05-18 224100 <== This is what I expect to get according to the product documentation when the delete was done at 23-05-18 22:14:00
Posts: 7
Joined: 17 May 2023

cyoram

Looks fine... Can you provide an example of the issue? For my test, I simply made a text file in a subfolder, synced it, changed the text file, synced again. It put the deleted copy in the correct subfolder in the version folder I entered. xCSxXenon, 18 May 2023, 15:20
Here is an example of what I have
d:/inter/family/example.doc <== I deleted this file
d:/freefilesync/revisions/example.doc <== This is what I get
d:/freefilesync/revisions/2023-05-18 224100 <== This is what I expect to get according to the product documentation when the delete was done at 23-05-18 22:14:00
User avatar
Posts: 3608
Joined: 11 Jun 2019

xCSxXenon

What you are getting and what you expecting are both wrong...
In your example, deleting 'example.doc' should put a copy here:
"d:/freefilesync/revisions/2023-05-18 224100/family/example.doc"

Is "d:/freefilesync/revisions/example.doc" maybe a remnant from something else? Have you deleted that file from there and then confirmed FFS puts it back if deleted again?
Last edited by xCSxXenon on 21 May 2023, 17:15, edited 1 time in total.
Posts: 7
Joined: 17 May 2023

cyoram

I had a typing mistake, here is the right message: Here is an example of what I have
d:/inter/family/example.doc <== I deleted this file
d:/freefilesync/revisions/2023-05-18 224100/example.doc <== This is what I get
<== This is what I expect to get according to the product documentation when the delete was done at 23-05-18 22:41:00
d:/freefilesync/revisions/2023-05-18 224100/inter/family/example.doc
User avatar
Posts: 2287
Joined: 22 Aug 2012

Plerry

Your expectation is not correct.
When not using local sync settings (the right of the two gear-icons in between each left-right pair), all versions of files in any of the base-directories are written to the root of the versioning location, or when using date-coded directories for versioning, to the root of a date-coded directory in the versioning location.

When you define to use local sync settings for any left-right pair, you can define a versioning location per left-right pair and thereby
if your left base location is
D:\Users\Family\Documents
create e.g.
D:\freefilesync\revisions\Family\Documents\2023-05-18 224100\example.doc
note that you will not get your
D:\freefilesync\revisions\2023-05-18 224100\Family\Documents\example.doc

But for your setup it would mean quite a lot of work, with your many left-right pairs.
Given your setup, a simpler approach would be to define a first left-right pair of
left D:\ and right [MyPassport]\FamilyPC Backup\ for all your D: base folder locations, use local Filter settings for that left-right pair (the funnel-icon in between the left and right base location, and use local Include Filter to define all D-locations that should be included in the sync or, if more convenient, use the Exclude Filter to exclude the few(?) D-locations that do not need to be synced.
And define local sync settings, in which you specify the versioning location (e.g. D_revisions) for D-based files.
And create a second left-right pair for your C-location(s) with local sync settings, in which you specify the versioning location (e.g. C_revisions) for C-based files.
This should give you the desired structure for C-based files, albeit in separate versioning locations.

The only disadvantage of above approach is that this will then sync e.g. the content of
D:\Users\Family\Documents into
[MyPassport]\FamilyPC Backup\Users\Family\Documents
rather than into
[MyPassport]\FamilyPC Backup\Documents
You can overcome this by not including D:\Users\Family (or anything in it) in above first left-right pair, but rather create a separate, third left-right pair for D:\Users\Family with its own local filter settings to define which folders to include (or, conversely, exclude), and local sync settings with yet again its own (e.g. Family_revisions) revision location.
Posts: 944
Joined: 8 May 2006

therube

Your expectation is not correct.
But why :-) ?

Rather then saying ones expectation is not correct (IMO, it is),
I'd say the operation of the program is not correct (in a logical sense).

The program may be operating in a manner that is not "wrong", but it is also not working in a manner that is "logical".
User avatar
Posts: 2287
Joined: 22 Aug 2012

Plerry

You wrote earlier
This is what I expect to get according to the product documentation
And the program does exactly what is described in section 2B.
So, if you expect something else based on the product documentation, your expectation is not correct.
Whether you find the behavior logical is another matter.
But that was not up for discussion.
Posts: 7
Joined: 17 May 2023

cyoram

I base my expectation on help documentation section 2B.
https://freefilesync.org/manual.php?topic=versioning

Example: Last versions of the file Folder\File.txt inside folder D:\Revisions
D:\Revisions\2020-12-11 111111\Folder\File.txt
D:\Revisions\2020-12-12 122222\Folder\File.txt
D:\Revisions\2020-12-13 133333\Folder\File.txt

What is the basis for your determination ?

"When not using local sync settings (the right of the two gear-icons in between each left-right pair), all versions of files in any of the base-directories are written to the root of the versioning location, or when using date-coded directories for versioning, to the root of a date-coded directory in the versioning location."
User avatar
Posts: 2287
Joined: 22 Aug 2012

Plerry

> What is the basis for your determination

Not sure which determination you mean.

If you mean my determination that your expectation based on the FFS manual page on versioning, section 2B is not correct:
That is, as stated, based on the fact that despite that the FFS program does exactly what is described in the FFS manual page on versioning, section 2B, you apparently expect different behavior.

If you mean the quote from my earlier reply in this thread:
It is not based on the FFS manual page on versioning, section 2B, as that does page does not describe what happens when using versioning when having multiple left-right base-folder pairs.
Instead, it is based on my own experience, your own experience/observation (as described in this thread), and is also fully in line with the reply from therube above.