Add support for resource id to the archiver

[gmgigi96]

Before the archiver only supported resources provided by a path.
Now also the resources ID are supported in order to specify the content of the archive to download. The parameters accepted by the archiver are two:

• an optional list of path (containing the paths of the resources)
• an optional list of id (containing the resources IDs of the resources)

Closes #2097

[update-docs]

Thanks for opening this pull request! The maintainers of this repository would appreciate it if you would create a changelog item based on your changes.

[glpatcern]

With the latest fixes it looks good to go for me.

[kulmann]

On a mac I get the following error when trying to extract the downloaded tarball with 7z x download.tar:

7-Zip [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21
p7zip Version 16.02 (locale=utf8,Utf16=on,HugeFiles=on,64 bits,12 CPUs x64)

Scanning the drive for archives:
1 file, 453241 bytes (443 KiB)

Extracting archive: download.tar

ERRORS:
Headers Error


WARNINGS:
There are data after the end of archive

--
Path = download.tar
Type = tar
ERRORS:
Headers Error
WARNINGS:
There are data after the end of archive
Physical Size = 250880
Tail Size = 202361
Headers Size = 1536
Code Page = UTF-8



Archives with Errors: 1

Warnings: 1

Open Errors: 1

It extracts it, the folder structure is correct, but the files are corrupted. Does this work for you?

[pascalwengerter]

Can confirm that on Linux/Ubuntu archives that contain PNG and JPG files don't work as expected - top level files can't be opened, while a folder containing images didn't even make it into the *.tar

[gmgigi96]

@kulmann @pascalwengerter I tried on both Mac and Arch and I didn't manage to reproduce the errors you had. For me it works on both systems, could you please provide me the steps to reproduce it? Thanks

[kulmann]

Maybe it's an issue with the current state of owncloud/web#5832
I'm running an ocis with
a) a local replace to reva which is checked out to your branch AND rebased to reva master
b) with WEB_ASSET_PATH=path/to/my/web/dist/folder

[gmgigi96]

Can you please try doing an HTTP GET request directly to the archiver service?

[gmgigi96]

@kulmann @pascalwengerter I added some unit tests into the archiver and all are passing, are you having the same issues?

[kulmann]

The id based version now works like a charm 🥳

[kulmann]

@kulmann @pascalwengerter I added some unit tests into the archiver and all are passing, are you having the same issues?

Resolved in the web PR. Sorry for the noise 😅

[gmgigi96]

@kulmann @pascalwengerter I added some unit tests into the archiver and all are passing, are you having the same issues?

Resolved in the web PR. Sorry for the noise 😅

Glad it works! @labkode @glpatcern can we merge it now?

[glpatcern]

Great, thanks all! Merging right away

[individual-it]

@gmgigi96 when using both id and path in the same request e.g. curl -k -u admin:admin 'https://localhost:9200/archiver?path=/home/image.jpg&id=MTI4NGQyMzgtYWE5Mi00MmNlLWJkYzQtMGIwMDAwMDA5MTU3OmZlZTVhNjZlLWI4MzYtNDY2OS05ZDJhLTliNGEwMDQwYmY0OQ==' --output /tmp/6.tar

I get whatever is requested by path in the folder called home and whatever is requested by id in the folder users/<uuid>/
is that intended?

$ tar -tvf /tmp/6.tar 
drwxr-xr-x 0/0               0 2021-11-11 16:10 users/ddc2004c-0977-11eb-9d3f-a793888cd0f8/my_data
-rw-r--r-- 0/0          472800 2011-03-07 21:41 users/ddc2004c-0977-11eb-9d3f-a793888cd0f8/my_data/11-02.JPG
-rw-r--r-- 0/0         1278511 2020-08-17 15:39 home/image.jpg