Age | Commit message (Collapse) | Author |
|
Adjust PKG_MIRROR_HASH to the filename change due to .zst adoption.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
|
|
Adjust PKG_MIRROR_HASH to the filename change due to .zst adoption.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
|
|
Adjust PKG_MIRROR_HASH to the filename logic change due to
APK preparations.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
|
|
Adjust PKG_MIRROR_HASH to the filename logic change due to
APK preparations.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
|
|
Older gcc versions raise a `label at end of compound statement` error due
to the empty default case. Fix the problem by adding an explicit `break`
statement.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
9b5b683 multipart-parser: properly handle parts with trailing CR
Fixes: #6282
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
The ucode-mod-html library provides assorted utility functions for dealing
with HTML markup data.
Example usage:
#!/usr/bin/ucode
'use strict';
import { tokenize, striptags, entitydecode, entityencode,
OPEN, ATTR, TEXT, CLOSE, RAW, COMMENT, CDATA, PROCINST, EOF } from 'html';
tokenize('<div class="example">Hello world!</div>...',
function(type, text, value) {
switch (type) {
case OPEN: print(`Opening tag: ${text}\n`); break;
case ATTR: print(`Attribute: ${text}${value ? `=${value}`}\n`; break;
case TEXT: print(`Text data: ${text}\n`); break;
case CLOSE: print(`Closing tag: ${text}\n`); break;
case RAW: print(`Script/CSS: ${text}\n`); break;
case COMMENT: print(`Comment: ${text}\n`); break;
case CDATA: print(`CDATA text: ${text}\n`); break;
case PROCINST: print(`<!...> tag: ${text}\n`); break;
case EOF: print(`End of input\n`); break;
}
}
);
print(striptags('<p>This is some <b>text</b> with <br> markup</p>\n'));
print(entitydecode('<   & ä'));
print(entityencode('1 < 2 && "foo"'));
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
Avoid redundancies in generated exception messages and include Lua
tracebacks when catching exceptions in protected calls.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
Properly handle accesses to properties of the userdatum itself in the
lua_uv_index() __index metamethod and treat integer keys as array indexes
in case of wrapped ucode array values. Also fix an incorrect refcount
decrement in the function.
Also fix uc_lua_vm_get() and uc_lua_lv_getraw() to gracefully handle
accesses to not defined or non-table values and ensure that those functions
properly reset the Lua stack after they complete.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
Expose ucode arrays and objects with prototypes as userdata proxy objects
to Lua and extend the userdata metadatable with an __index metamethod to
lookup not found properties in the ucode values prototype chain.
Also extend the __call metamethod implementation to infer method call
status from the activation record in order to invoke ucode functions with
the correct `this` context when called as method from Lua.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
Reopen self with dlopen(RTLD_GLOBAL) in order to export liblua symbols for
runtime loading of dynamic Lua extensions.
Reported-by: Stijn Tintel <stijn@linux-ipv6.be>
Tested-by: Stijn Tintel <stijn@linux-ipv6.be>
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
6e68a10 utils: also compare attribute name length in lh_header_attribute()
7b721af testcases: adjust urldecode tests
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
The ucode-mod-lua library provides an ucode-to-Lua bridge and a set of
functions to instantiate Lua VMs, invoke Lua functions as well as
exchanging data structures between ucode and Lua.
Example usage:
#!/usr/bin/ucode
'use strict';
const lua = require("lua");
let vm = lua.create();
vm.set({
hello: function(...args) {
print(`A ucode "Hello world" function called from Lua! Got arguments: ${args}\n`);
},
data_from_ucode: {
bool: true,
float: 1.3,
int: 0x11223344,
string: "Hello from ucode!",
array: [ 1, 2, 3, null, 5 ],
object: {
apple: "green",
banana: "yellow",
[5]: "foo",
[-1]: null,
nested: {
a: [ 5, 6 ],
b: { c: NaN }
}
},
regexp: /foo/
}
});
vm.invoke("hello", true, 123, "Foo");
vm.eval('print("Print from Lua!", data_from_ucode.int * data_from_ucode.float);');
try {
vm.invoke("error", "Throwing a Lua exception...");
}
catch (e) {
print(`Caught exception: ${e}\n`);
}
print(`Lua VM version is: ${vm.get('_G', '_VERSION').value()}\n`);
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
cc85183 ucode: add ucode library binding
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
707feae parse_css: do not omit white space in url() property values
Ref: https://github.com/openwrt/luci/commit/69608199973651baf585d24a095edf8c0c42e21f#commitcomment-52088358
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
Fixes compilation under 32bit systems by using the proper printf format
specifier for size_t values.
Also fixes compilation with Ninja by appending instead of
overwriting CMAKE_OPTIONS.
Fixes: #5116
Suggested-by: Rosen Penev <rosenp@gmail.com>
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
a34a17d src: allow overriding buffer size from cli in multipart tester
730a46f lib: fix potentially lost bytes in boundary parsing across buffer limits
8734af2 lib: add buffer tracing to multipart parser
913051b src: add file dump option to multipart test utility
c419539 src: allow specifying custom buffer sizes in multipart testcases
f6e0564 lib: fix handling of empty multipart fields
Fixes: #2816
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
f6e0564 lib: fix handling of empty multipart fields
91c01c3 lib: fix multipart state transition in boundary parsing
Fixes: #2737
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
Even Freifunk was one of the major factory to create the LuCI-system, it's
now only a very small part of LuCI. LuCI has become a much bigger thing
and it seems that it's time to move the packages only relating to Freifunk
into it's own feed.
On the mailinglist it was discussed [1] and a repo below the general
Freifunk team on github was created.
This commit removes all packages that will be hosted in the new repo [2]
1 - http://lists.freifunk.net/pipermail/wlannews-freifunk.net/2019-February/004818.html
2 - https://github.com/freifunk/openwrt-packages
Signed-off-by: Sven Roederer <freifunk@it-solutions.geroedel.de>
|
|
1afbdcc build: add soversion to library
Also adjust ABI_VERSION and install recipe accordingly.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
1d56201 prepare: do not consider "overflow" to be a shorthand for -x, -y
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
33594b4 ("csstidy: do not dequote selector strings")
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
Package a fork of the CSSTidy C++ implementation for CSS minification
use in a later commit.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
New profile for Fürstenwalde. At the moment we use parts of the Berlin-
infrastructure. Thus I have not changed the ip-address-related things.
Signed-off-by: Martin Hübner <martin.hubner@web.de>
[reword and rewrap commit message]
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
Signed-off-by: Philipp Borgers <borgers@mi.fu-berlin.de>
|
|
Signed-off-by: David Yang <mmyangfl@gmail.com>
|
|
Signed-off-by: David Yang <mmyangfl@gmail.com>
|
|
Signed-off-by: pmelange <isprotejesvalkata@gmail.com>
|
|
The mesh_network option is used to check the user input. The wizard
checks if the input ip address is part of the mesh_network. We use
multiple /16 networks. There is no support for multiple ranges so we the
10.0.0.0/8.
For reference the ip list:
https://wiki.freifunk.net/IP-Netze
Signed-off-by: Philipp Borgers <borgers@mi.fu-berlin.de>
|
|
ff_olsr_watchdog: delete stale tunnels
|
|
When OLSRd crashes, the old tunnels still exist. This can lead to unexpected behaviour.
The tunnels to be removed start with "tnl_"
See freifunk-berlin/firmware#522
Signed-off-by: pmelange <isprotejesvalkata@gmail.com>
|
|
6ddea4c utils: fix crash with zero length input string in lh_header_attribute()
cb119de lib: add support for setting the maximum allowed data size
Fixes #1784.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
See issue openwrt/luci#1757
Signed-off-by: Perry Melange <isprotejesvalkata@gmail.com>
|
|
ccc685e lua: expose LH_URLDECODE_PLUS flag in Lua library
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
c7c9c66 src: extend multipart parser test program
5071efb testcases: add multipart parsing edge cases
689e3d0 lib: multipart-parser: fix various edge cases
Fixes #1754.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
8617997 lib: cast size_t values in printf() to prevent compielr warnings
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
b7470d1 lua: back out early when instantiating parser with bad boundary
e1b1b1f testcases: remove stray .swp file
b46a6ca utils: introduce new LH_URLDECODE_PLUS flag
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
Package liblucihttp, a utility library providing HTTP parsing and data
decoding helpers.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
|
|
After some discussion in our meetings and on the maling list we decided to
change the SSID to a fixed one.
Signed-off-by: Hannes Fuchs <hannes.fuchs@gmx.org>
|
|
Remove freifunk-p2pblock and luci-app-p2pblock as they have been BROKEN
since 2016. Packages depend on the removed l7-protocols as well as the
deprecated l7 support in the iptables kernel modules.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
|
|
|
|
inside OLSR some plugins recently got newer version numbers (and of course improvements)
Signed-off-by: Ufo <ufo@rund.freifunk.net>
|
|
- fixed ssid
- disable IPv6
- added more interface settings
- added settings for 5GHz wifi
- added ssidscheme
- added dhcp leasetime setting
- added LinkQualityAlgorithm for olsrd
Signed-off-by: Hannes Fuchs <hannes.fuchs@gmx.org>
|
|
Remove the code related to the deprecated madwifi driver.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
|
|
Remove the code related to the deprecated madwifi driver.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
|
|
Remove the ancient 'remote-update' package that has not been
updated for years. It e.g. references two non-existing targets.
Reference to discussion in #995
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
|
|
the updates for olsr-0.9.5 compatibility should have changed the release-number
Signed-off-by: Sven Roederer <freifunk@it-solutions.geroedel.de>
|
|
since Kathleen-0.2.0 Release there is a new owm-api-url
Signed-off-by: Sven Roederer devel-sven@geroedel.de
|
|
Older OLSRd-version outputs JSON humanreadable over multiple lines.
The introduced change in 524439cd1647e5d6d7b48f9c75ec5800a5444109 broke
the output for those users. Fix that be a finer filter regex.
Thanks to Thomas Huehn <thomas@net.t-labs.tu-berlin.de> for spotting that.
|