TalkChess.com

Question to HGM: Is the current source code of your ICS available somewhere, so that people could take a look at it and see whether they could find some weak spots?

He means whether source is a public Github project. Or not?

All these communists trying to steal your source code.
Or is it possible to give limited access.

Don't know much about github

Yes, or maybe a download link at HGM's website.

You can find the source on HGM's git:

http://hgm.nubati.net/cgi-bin/gitweb.cg ... ;a=summary

It is a modified version of:

https://chess.samba.org/

You can also find it here:

https://github.com/ddugovic/capablanca

Indeed, the most recent source is in the repository at http://hgm.nubati.net/cgi-bin/gitweb.cgi .

As far as I understood the lowest-level input is in network.c ( readline2(), called from select_loop() ). The input line is then passed to timeseal_parse() (in timeseal.c) and proces_input() (in command.c).

hgm wrote: ↑Fri Aug 30, 2019 3:15 pm Indeed, the most recent source is in the repository at http://hgm.nubati.net/cgi-bin/gitweb.cgi .

As far as I understood the lowest-level input is in network.c ( readline2(), called from select_loop() ). The input line is then passed to timeseal_parse() (in timeseal.c) and proces_input() (in command.c).

Do you send your changes upstream? Or can I clone the repo somehow?

I guess this repo is as much upstream as you can get. No one else has been developing the public ICS code for years.

The github repo has 11 newer commits compared to the hgm.nubati.net repo. So I assume the binary running on the HGM server is build from the nubati version but nevertheless the github repo has the most recent source, am I right?

Sven wrote: ↑Sat Aug 31, 2019 1:22 am The github repo has 11 newer commits compared to the hgm.nubati.net repo. So I assume the binary running on the HGM server is build from the nubati version but nevertheless the github repo has the most recent source, am I right?

As far as I can tell is the github repo a fork of the hgm.nubati.net repo from before the time-seal fix.

I wasn't even aware there was a fork of the ICS code on Github. The commits there seem to do little more than deleting some unused files (remnants of the pre-git CVS version control), replace some hard-coded constants by macros of the same value, and suppress some compiler warnings in mamer.