Mirror/zeek
1
0
Fork 0
mirror of https://github.com/zeek/zeek.git synced 2025-10-02 14:48:21 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

uri parsing function

Browse source
This commit is contained in:
akasza 2014-11-05 20:44:03 -08:00
parent 9045288ad3
commit 3c42350e77
1 changed files with 58 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

58
scripts/base/utils/urls.bro
View file

@ -3,6 +3,16 @@
## A regular expression for matching and extracting URLs. ## A regular expression for matching and extracting URLs.
const url_regex = /^([a-zA-Z\-]{3,5})(:\/\/[^\/?#"'\r\n><]*)([^?#"'\r\n><]*)([^[:blank:]\r\n"'><]*|\??[^"'\r\n><]*)/ &redef; const url_regex = /^([a-zA-Z\-]{3,5})(:\/\/[^\/?#"'\r\n><]*)([^?#"'\r\n><]*)([^[:blank:]\r\n"'><]*|\??[^"'\r\n><]*)/ &redef;
type uri_record: record {
protocol: string &optional;
# this could be a domain name or an IP address
netlocation: string;
portnum: count &optional;
path: string &optional;
file_name: string &optional;
file_ext: string &optional;
};
## Extracts URLs discovered in arbitrary text. ## Extracts URLs discovered in arbitrary text.
function find_all_urls(s: string): string_set function find_all_urls(s: string): string_set
{ {
@ -23,3 +33,51 @@ function find_all_urls_without_scheme(s: string): string_set
return return_urls; return return_urls;
} }
function decompose_uri(s: string): uri_record
{
local parts: string_array;
local u: uri = [$netlocation=""];
if (/:\/\// in s)
{
parts = split1(s, /:\/\//);
u$protocol = parts[1];
s = parts[2];
}
if (/\// in s)
{
parts = split1(s, /\//);
s = parts[1];
u$path = fmt("/%s", parts[2]);
if (|u$path| > 1)
{
local last_token: string = find_last(u$path, /\/.+/);
local full_filename = split1(last_token, /\//)[2];
if (/\./ in full_filename)
{
u$file_name = split1(full_filename, /\./)[1];
u$file_ext = split1(full_filename, /\./)[2];
u$path = subst_string(u$path, fmt("%s.%s", u$file_name, u$file_ext), "");
}
else
{
u$file_name = full_filename;
u$path = subst_string(u$path, u$file_name, "");
}
}
}
if (/:/ in s)
{
parts = split1(s, /:/);
u$netlocation = parts[1];
u$portnum = to_count(parts[2]);
}
else
{
u$netlocation = s;
}
return u;
}