I don’t see python on your image :)

I finally manage to encrypt the body trough ptyhon-gnupg ( warning their documentation is still in alpha stage. )

now, remain to encrypt the subject (ThunderBird compatible) if you have any clues I’m all ears

When time permit I will publish my code in a pastbin.

Wubba Lubba dub-dub**

indeed, but a lot of Linux distribution come with it :)
otherwise it’s installable.

instead of using a library I can directly use subprocess with gnupg but in both case it seem gnupg require to import the public key to the keyring !? I don’t want that.

Thanks all for your input !

So yes /sdcard was /storage/emulated/0 and it’s a real sd-card.

That was indeed formatted as FAT, I reformatted into EXT4 (from a desktop Linux) put the SD-card back into the phone and started the phone.

Ive tried (in the phone)

su
chmod 777 /storage/emulated/0 #yes it's still mounted as emulated !
stat /storage/emulated/0
#...
#Access: (0771)
#....

So I’m still blocked… any ideas ?

Thanks.

I’ve found the most simple way (for my case) adb shell 'a command'

example

adb shell ls /

Thank you all for your input… but it seem my question is still not fully answered…

let me rephrase, I’m not looking to have a GUI to transfer files, but I would like to execute terminal command remotely (from my computer) to my android phone. Like SSH .

So I’ve read that I can install a SSH server on my android phone… (If you know some’s (FLOSS), I’m all ears) Or if you know a better way than SSH I’m all ears too.

Thanks.

setfacl -m m:r aFile
#re set the mask

solve the problem, but the question is: why the F**** this is happening !?

Update, this is only happening when I copy files from a ~SMB share… :'(
Otherwise it’s correct…

Proxmox seem powerfull

It’s a Type1, not Type2
https://en.wikipedia.org/wiki/Hypervisor#Classification

Thank you all !

Indeed setting execute perm on example, sub1, sub2, static

The program/user have now access to the directory.

In order words all the parents directory need at least execute in order to have access in the targeted directory…

Now I gave 751 for static. Meaning than others (here nginx) cannot list the files within. But never the less it works
the static files are appearing when requested (HTTP) but forbidding nginx to list the directory is changing something ? (performance/security)

Thanks

ok I’ve found something that ~works !

	server {
		listen 443 ssl;
		server_name _;
		ssl_certificate /etc/nginx/ssl/catchall.crt;
		ssl_certificate_key /etc/nginx/ssl/catchall.key;

		error_page 404 /404.html; #at /var/www/html/

		location /404.html {internal;}

		return 404;
	}

so i get the default 404 html from nginx. but not the one that I specified error_page 404 /404.html; any ideas ?

Something worth reading regarding Systemd https://www.devuan.org/os/announce/ Cheers.

Thank you all for your quick reactions !!

To summarize if I want to use the PDF built-in signing I will need to convert my OpenPGP into a X.509 cert otherwise I can simply use the OpenPGP file signing

I want to stick to the UNIX Philosophy especially:

Write programs that do one thing and do it well.

So I will use the OpenPGP signing tool :)

Thanks !

Thank @[email protected]

it’s look like that that code only verify the form when submitting… I need to check the validity when the element has been loaded.

btw checkValidity() do not solve this issue as it return True when the length is longer than maxlength ! O_o I guess this is were is lying the “bug”

Thank you @[email protected] that exactly my question… how can I trigger the validation process again ?

It will not be digest to send all what I use for you to see it (it will lead to TL;DR)

but here an elements (loaded with XHR that give the problem

<input id="something" name="something" type="text" placeholder="aPlaceholder"  minlength="3" maxlength="32" value="this value is more than the limiation of 32 char" autofocus required>

So normally this element should be invalid and a CSS selector of :invalid should match. but it’s not the case because the browser seem to not run the validation check on loaded elements… !?

if we edit manually the input , for example removing one character then the validation process kick-in and the CSS selector work etc…