In
http://ask.wireshark.org/questions/5726/newbie-trying-to-get-lua-scripts-to-execute
it says
Wireshark tries to load Lua files from all plugins directories
(specified in the Wireshark manual). The file extensions must be ".lua" (case
sensitive). Directories under plugins are searched recursively for Lua scripts.
In the following tree, for example, files a through e are loaded at startup:
* $HOME/.wireshark/plugins/a.lua
* $HOME/.wireshark/plugins/foo/b.lua
* $HOME/.wireshark/plugins/foo/bar/c.lua
* $HOME/.wireshark/plugins/foo/bar/d.lua
* $HOME/.wireshark/plugins/foo/bar/foo2/e.lua
* $HOME/.wireshark/plugins/foo/bar/foo2/f.Lua (ignored due to
ext mismatch)
* $HOME/.wireshark/plugins/foo/bar/foo2/g.LUA (ignored due to
ext mismatch)
If we're on a case-insensitive file system, *should* those be ignored?
Similarly, we look for ".dll", but not ".DLL" or..., for plugins on Windows.
Should we have a ws_dir_get_suffix_matching_name() routine that's like
ws_dir_get_name() but that takes a suffix as an argument, skips files whose
names don't have that suffix and, on case-insensitive file systems, does a
case-insensitive match? On Windows, we might just assume case-insensitivity;
on UN*X, we might assume case-sensitivity except on Mac OS X, in which case, at
least in Snow Leopard and later (I forget when it was introduced), we could use
pathconf() and _PC_CASE_SENSITIVE (I forget which file systems supported it in
which releases; for AFP/SMB/NFS, it might not be possible to query the server,
so they might just say "case-insensitive" for AFP and SMB and "case-sensitive"
for NFS).
___________________________________________________________________________
Sent via: Wireshark-dev mailing list <[email protected]>
Archives: http://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
mailto:[email protected]?subject=unsubscribe
