I have rewritten large parts of the UdgerParser from the dev_v3 branch to make it threadsafe, faster and cleanup the code. It passes all the tests, and from my own testing is much faster than the previous version.

Compared to the previous implementation it caches most of the database in memory, except IP locations and IPv6 data center mainly because the IP location lookup table is huge. It adds a flag to do Data Center parsing (DataCenterParserEnabled) so that you enable data center parsing at the same time as IP location can be disabled (so you can run it fully in memory without disk reads and still parse IPv4 data centers).

We have been running it in production for a while, and it seems to be working well.

Currently its a fork available here: https://github.com/nordicfactory/udger-dotnet
I would be happy to make a PR if you think its appropriate.

Hi,
I've tried your app. I don't understand how it finds details about user agent. Examples in the Console test don't show any details and if I run the code step-by-step the app jump always a part when you check browser details.

I think generated regex is wrong.

When a User Agent contains an apostrophe ('), the Udger client fails with the exception:

System.Data.SQLite.SQLiteException (0x80004005): SQL logic error or missing database
unrecognized token: ":"
DataReader.cs(62,0): at Udger.Parser.DataReader.selectQuery(String query)
UdgerParser.cs(106,0): at Udger.Parser.UdgerParser.parseUA(String _userAgent)
UdgerParser.cs(73,0): at Udger.Parser.UdgerParser.parse()
Program.cs(46,0): at ConsoleTest.Program.Main(String[] args)

This is because query parameters are being added to the sql query by concatenation.
SqliteParameters should be used instead. Here are some of the reasons why: http://stackoverflow.com/a/20017707/390819

I have made a rewrite of udger-dotnet into .Net Core and with some other cleanup and improvements. Doing this I used code from udger-dotnet, udger-dotnet v3 branch and also udger-java.

Can you clarify what licenses these are under? Seems like Java is using BSD: https://udger.com/legal/java_parser but the .NET code is very unclear to me (LGPL, also BSD?).

I hope to contribute my rewrite, but first I need to know what kind of license it becomes.

I am starting a new project in dotnet core, I would need a compatible version.
If nothing is planned/started on your side, I will probably start a fork (or total new version) to make it work.

Waiting any feedback before developing stuff for nothing ;).

There are three queries in UdgerParser.cs that have no filters. This means that the same data (and all data) is selected from the DB every single time a detection is being made:

//client
DataTable clientRegex = dt.selectQuery(@"SELECT class_id,client_id,regstring,name,name_code,homepage,icon,icon_big,engine,vendor,vendor_code,vendor_homepage,uptodate_current_version,client_classification,client_classification_code
                      FROM udger_client_regex
                      JOIN udger_client_list ON udger_client_list.id = udger_client_regex.client_id
                      JOIN udger_client_class ON udger_client_class.id = udger_client_list.class_id
                      ORDER BY sequence ASC")

DataTable osRegex = dt.selectQuery(@"SELECT os_id,regstring,family,family_code,name,name_code,homepage,icon,icon_big,vendor,vendor_code,vendor_homepage
                      FROM udger_os_regex
                      JOIN udger_os_list ON udger_os_list.id = udger_os_regex.os_id
                      ORDER BY sequence ASC")

DataTable device = dt.selectQuery(@"SELECT deviceclass_id,regstring,name,name_code,icon,icon_big
                      FROM udger_deviceclass_regex
                      JOIN udger_deviceclass_list ON udger_deviceclass_list.id=udger_deviceclass_regex.deviceclass_id
                      ORDER BY sequence ASC")

Using the test database, these queries bring 141, 165 and 42 records respectively.

Given the fact that the DB is essentially read-only from the perspective of the Udger Client and that the same data is retrieved every time, I think that these records could be cached to avoid 3 DB roudtrips on every detection.

With a user-agent such as

Mozilla/5.0 (Linux; Android 4.4.4; eZee' Tab10010-S Build/KTU84Q) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/33.0.0.0 Safari/537.36

We get an exception :

Unhandled Exception: System.Data.SQLite.SQLiteException: SQL logic error or missing database
near "Tab10010": syntax error
   at Udger.Parser.DataReader.selectQuery(String query)
   at Udger.Parser.UdgerParser.parseUA(String _userAgent)
   at Udger.Parser.UdgerParser.parse()

Looks like the user agent string is not sanitized when creating the query for sqlite and the single quote after eZee' causes the problem.