rfm - RepRapFirmware FileManager [duetbackup successor]

gtj0

Hey Manuel, rfm has been working great for me but is there any chance you could automatically exclude the "System Volume Information" directory or maybe save excludes in the rfm.toml file?

wilriker

@gtj0 I'll think about something.

Meanwhile you can also use rfm to just delete this stupid folder. It doesn't serve any purpose in the first place. I don't think that removable media are supposed to even have this folder.

wilriker

@gtj0 Here you go

Release v1.1.0-RC1

Get it at the GitHub Releases page.

This release implements the persisting of -exclude parameter values in rfm.toml. These parameters will be persisted separately for backup and upload command so they do not confuse each other.

A new keyword reset has been added as a recognized value for the -exclude parameter to delete all persisted excludes for the current command, i.e.

rfm backup -exclude reset ...

will remove all persisted excludes for the backup command. Further -exclude parameter after that (even in the same command) will add new excludes.

Note that order is relevant, i.e.

rfm backup -exclude 0:/gcodes -exclude reset backup-dir 0:/

will backup everything in 0:/ since the given exclude for 0:/gcodes was reset by the following -exclude reset parameter.

gtj0

Is -removeLocal working?

./rfm backup -domain duet0 -removeLocal /tmp/duet0 /
...
Removing no longer existing files in /tmp/duet0/gcodes

But it doesn't actually remove any files from the local filesystem.

wilriker

@gtj0 Last time I checked it worked. It will only remove local files that do not exist anymore on remote (like -delete for rsync). I will check tomorrow if something is broken with it.

wilriker

@gtj0 You were right, -removeLocal was broken for removed files (it worked for removed directories). So here is

Release v1.1.0-RC2

This fixes broken -removeLocal parameter of the backup sub-command that did not delete files locally.

Get it at the GitHub Releases page.

3mm

Hi wilriker,

I hate to sound stupid, but how do I use it? Where do I put the files? And how do I access it? My system is Windaze 7 PC and a DUET-EtherNet controller.

Thanks in advance...3mm

gtj0

@wilriker

Yep, -removeLocal now works.
You can't specify the remote directory as 0:/ any more though. Not sure if this is intentional or not.

$ ./rfm backup -domain duet0 /tmp/duet0 0:/
2019/07/30 05:49:53 Fetching filelist for 0:/0:
2019/07/30 05:49:53 Directory not found

Doesn't matter to me, not sure if it's matter to others.

wilriker

@3mm You download the release for your operating system. For Windows that would be rfm-windows_amd64.zip. Then you unpack that to a location of your choice. It is a command-line tool so you have to open cmd and navigate to the directory where you unpacked rfm to (alternatively: in Explorer right click while holding shift in the directory where rfm.exe is located - this should provide a context-menu entry to open a command line in this directory).

There you can simply run

rfm help

to get an overview and the available sub-commands. With

rfm help <subcommand>

you can see individual parameters of each sub-command.

wilriker

@gtj0 said in rfm - RepRapFirmware FileManager [duetbackup successor]:

@wilriker

Yep, -removeLocal now works.
You can't specify the remote directory as 0:/ any more though. Not sure if this is intentional or not.
$ ./rfm backup -domain duet0 /tmp/duet0 0:/
2019/07/30 05:49:53 Fetching filelist for 0:/0:
2019/07/30 05:49:53 Directory not found
Doesn't matter to me, not sure if it's matter to others.

No, that's a bug. I'll look into that.

wilriker

@gtj0 Found it and fixed it.

Release v1.1.0-RC3

This can be found as usual on GitHub Releases page.

Another bug fix. There was a regression introduced in v1.1.0-RC1 where it was no longer possible to specify the root directory of the SD card as 0:/. This is fixed in this release.

Also this release candidate is intended to be the last one before final release of this version.

3mm

@wilriker

Thank you very much!

3mm

wilriker

Arch User Repository Package

For Arch Linux users: I have submitted a package to install rfm via AUR: https://aur.archlinux.org/packages/rfm/

This is available for x86_64, arm, armv6h, armv7h and aarch64.

wilriker

Release v1.1.0

v1.1.0 has existed Release Candidate stage and is now available at GitHub Releases page or via Arch User Repository.

Reiterating the changes:
This release implements the persisting of -exclude parameter values in rfm.toml. These parameters will be persisted separately for backup and upload command so they do not confuse each other.

A new keyword reset has been added as a recognized value for the -exclude parameter to delete all persisted excludes for the current command, i.e.

rfm backup -exclude reset ...

will remove all persisted excludes for the backup command. Further -exclude parameter after that (even in the same command) will add new excludes.

Note that order is relevant, i.e.

rfm backup -exclude 0:/gcodes -exclude reset backup-dir 0:/

will backup everything in 0:/ since the given exclude for 0:/gcodes was reset by the following -exclude reset parameter.

Bug Fixes

-removeLocal was broken for files (working for directories though)
Sub-command upload would fail most of the times (this was fixed between RC3 and this version)

fma

I'm trying to use your script, but I always get 'invalid character '<' looking for beginning of value' error. What am I doing wrong?

I'm using the amd64 version...

Thanks.

wilriker

@fma This happens if the Duet responded with an HTML error page instead of a JSON error. I've seen this happen more often recently but so far have not found the cause.

Are you using the latest v1.1.0 final release above?

Also which sub-command does this happen for? Also you can use the parameter -debug to see the actual response coming from the Duet. This might give a hint what's going wrong.

fma

Yes, I'm using the final release.

The error happens with all commands, and -debug does not give much info:

./rfm ls 0:/ -debug

0::
2019/08/18 12:52:40 invalid character '<' looking for beginning of value

BTW, does it require a specific version of RRF? Im' running 2.02RC2...

wilriker

@fma It should work with any version of RRF. Though the -debug parameter has to come before the location to list (in this case) or more general: before any parameter that does not have its own -something prefix.

fma

My bad, I made a mistake in the device name! All seems to work fine

Thanks for your help.

wilriker

@fma I'm glad it was that simple.