http://pubs.opengroup.org/onlinepubs/009695399/functions/setrlimit.html

For cgi and tar

I used wget --continue --level 5 --recursive --accept '*.img' --no-host-directories --no-parent --cut-dirs=1 http://<server>:8080 to get every img file and it works fine. Resuming too. But if the file is already there then wget stops downloading at all and retries again and again. I tested it with a few webservers and most of the time wget skips to the next file.

--2019-03-01 09:00:08--  http://<server>:8080/backup_2019-03-01.img
Connecting to <server>:8080... connected.
HTTP request sent, awaiting response... 206 OK
Retrying.

Attempted to install Qweborf today, totally stocked about it. Upon running pyuic5 qweborf/main.ui > qweborf/main.py , received the oddest error message. /usr/local/bin/pyuic5: 2: exec: /home/phil/usr/bin/python3: not found
I don't know anyone named phil who runs linux. Wondering how I can reset that python path? I have attempted to use the env command, to no avail.

Cheers

Make an option to provide a list of forbidden URLs. Asking of one of those will ban the IP address.

put install_requires in setup.py and remove "python3-pyqt5" from debian deps.

"CFLAGS missing (-fPIE):"
LDFLAGS missing (-fPIE -pie -Wl,-z,now):

Windows mount webdav needs for LOCK/UNLOCK and use it on file move.
Without LOCK and UNLOCK methods in OPTIONS the Windows may decide to mount a folder in read only mode. But I checked and looks like in Windows 7 it should be fine.
Anyway maybe just mock the methods should be improve interopability.

This will simplify usage

Hello,

it would be cool if links open direct as a download dialog. This is possible by using the download attribute (https://stackoverflow.com/a/16316830). Maybe this can be switched on and off by using a parameter.

Thanks.

Create the process ith pidfd_spawn or similar instead of fork and then use pidfd_send_signal to send the signals.

This avoids race conditions.

It is only supported on recent linux, so it will need ifdef and autotools

Hi Great project, please add a weborf.po file to make it possible to translate into other languages

Hi.
May you implement an option to set the group of the weborf process? I've patched weborf to call setgid() with a hardcoded gid before calling setuid(), but a commandline parameter would be nice.

When SOCKETDBG is uncommented in options.h.in, the build process errors out in myio.c in function fd_copy because connection_prop is undeclared.

I used qweborf before with authentication, but now I need to set up weborf on a headless server, with authentication. Could you give some pointers how I can set up credentials? On Ubuntu there's a package weborf-daemon, but I couldn't figure out how to set up the authentication part.

Please change makefile to:

install -D -m644 daemon/weborf.service $(DESTDIR)/usr/lib/systemd/system/weborf.service
install -D -m644 daemon/[email protected] $(DESTDIR)/usr/lib/systemd/system/[email protected]
install -D daemon/weborf-generator $(DESTDIR)/usr/lib/systemd/system-generators/weborf-generator

These files using /usr/lib:

weborf_cgi_wrapper
weborf_py_wrapper

These files using /lib:

weborf.service
[email protected]
weborf-generator

The different paths causing trouble in creating a packet. Thanks.

I really hope this is fixable. When running weborf listing files & directories, it cannot process any names that have a question mark in them. Directories cannot be descended into, files cannot be viewed/downloaded. It says: Error 404, page not found.

System information

OS: Debian 11
GUI: X11 + openbox + LXQT

Problem description

I'm attempting to run weborf from a script using following syntax

weborf -d -m -p 80 -b [here goes the path to my WWW root dir]

and I recieve an error that the typed as the "-b" parameter path need to be a directory (with is and it has propper permissions)

man page
-a, --auth
Must be followed by a unix socket listened by a program who will handle authentication. See the website for more details.
On the website are no details, there is nothing documented anywhere.

It cannot be used without auth. With this Qt-Frontend it works but only if you enable "Webdav" the has nothing to do with auth but its listed under "auth". In the generated command line a "-a /temp/weborf-socke..." appears., that gives you a hint you have to use it anyway, but how without this qt frontend, I dont want this frontend.

Document this software better, it unusable without a description.

Your latest version (0.18.1) requires libc6 >= 2.33. For those of us running Ubuntu 20.04.4 LTS, 0.18.1 won't install, because for Ubuntu 20.04.4, libc6 2.31 is the latest version available.

I am new to Linux, but have the GUI version up and running successfully. Works well.

Is there any way to add a save settings to the GUI, e.g., "Directory to Share". "Try to Share outside LAN", and lastly an automatic "Start Sharing" on startup?

https://pastebin.com/69ieBqHT

The following configuration work out the box for one folder.
Help doc states it is possible to web share more than one folder using /etc/weborf.d/
I wondered for an example and tried some conf files without success.
So I am asking, please, an example :-)

#  /etc/weborf.conf - configuration file for weborf(1)
#  see weborf.conf(5) for details on this file
#Index files. This list is ordered by priority. If not set, the default value will be index.html.
#indexes=index.html,index.php
# Comment this to not start this unit automatically.
# Only works with systemd
start=auto
# Base directory.
basedir=/home/username
# Enables or disables use of CGI support
use-cgi=false
#Enable weborf to send Content-Type header
use-mime=true
user=username
group=usernamegroup
# Directory that will be used for the cached files
cachedir=/var/cache/weborf
# Port used to listen
port=80

I would like to know where are the weborf logs

Hello,

it would be nice to get http status codes and paths on stdout/stderr like a simple type of log entries from http servers. It would create the ability to use weborf in a pipeline.

Thanks.

why?