Build wo xml

[zaibacu]

Ok, so in general these changes can be a bit controversial, but hear me out.

In general OpenCCG is designed in such way, that it relies on provided grammar XML to build all of the structures. In our case - XML generation was huge bottleneck and additional hassle.

What these changes does - they allow to fully construct Grammar, Lexicon, Rules and Types (if you know what you're doing, of course) at runtime without any need of parsing XML files.

[dmhowcroft]

This sounds like an interesting idea to me; do you have a set of tests or documentation that demonstrate (1) that it doesn't break existing functionality and (2) how it helps in other use cases?

[zaibacu]

As far as not breaking standard functionallity, current tests should be sufficient to verify, will do that my self just to be sure

As for use cases, this is valuable when OpenCCG is being used as a library and rules are generated dynamically at runtime

[mwhite14850]

Sounds good! Please confirm when you’ve run current tests. Hope to have time to review this next week.

…

On Wed, Sep 11, 2019 at 10:20 AM Šarūnas Navickas ***@***.***> wrote: As far as not breaking standard functionallity, current tests should be sufficient to verify, will do that my self just to be sure As for use cases, this is valuable when OpenCCG is being used as a library and rules are generated dynamically at runtime — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#29?email_source=notifications&email_token=AA6G7TOAHJ5GMKFU3AUDKTDQJD5DZA5CNFSM4IVSMVFKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD6OU32Q#issuecomment-530402794>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA6G7TJ73D2EX5NBCPG6LN3QJD5DZANCNFSM4IVSMVFA> .

[zaibacu]

Testing results:

[INFO] Results:
[INFO] 
[INFO] Tests run: 141, Failures: 0, Errors: 0, Skipped: 0