The road to Rivet 2.1.0

[Massimo Manghi <mx...@apache.org>]

I'd like to lay down a list of todos that would lead us to eventually
release Rivet 2.1.0. I have in mind a few issues still to be addressed,
everyone is welcome to add theirs

- Code

* command ::rivet::debug relies on the RivetServerConf and RivetUserConf
arrays, but they are now created only on demand. The problem has to be
addressed
* GlobalInitScript (conf->rivet_global_init_script) as a child
initialization script could be merged with ServerInitScript. Rationale:
GlobalInitScript role is more efficiently played by ServerInitScript and
GlobalInitScript code is run only when SeparateVirtualInterps is off.
The configuration directive could still be supported (maybe deprecated)
and its code lines collected separately and eventually appended to
conf->rivet_server_init_script, thus preserving the current execution
sequence for rivet_server_init_script and rivet_global_init_script
* In general code needs some trimming and better comments. mod_rivet.c
still bears code excluded from compilation whose purpose has never been
evident to me.

- Test suite

* new tests needed for ServerInitScript initialization code
* new command ::rivet::http_accept and package entities need a test
* no doubt ::rivet::inspect command needs one or more tests to be
implemented

- Documentation expansion and revision:

* There are a few pages in the docs that need further development; many
manual pages relating to the command set would benefit from simple
examples.
* Package 'entities' still needs a manual page
* If anyone with an good insight into Docbook and with a gift for
styling pages wants to elaborate a new CSS for the manual. I found a
basic CSS for docbook that looks to me already an improvement even
though I haven't changed a single definition of it.

-- Massimo


---------------------------------------------------------------------
To unsubscribe, e-mail: rivet-dev-unsubscribe@tcl.apache.org
For additional commands, e-mail: rivet-dev-help@tcl.apache.org