[Hamlib-developer] ts2k branch

[Stephane F. <f4...@fr...>]

Dale,

I've created the branch, labelled 'branch_ts2k'.
You can go ahead and commit to this branch, and stop checking in the trunk.

To select the branch, change directory to the parent dir, and do
a "cvs checkout -r branch_ts2k hamlib". Be careful as the checkout may
overwrite some of your local files. Either make a backup first or rename
your hamlib directory to perform a fresh checkout.

The 'branch_ts2k' tag is a sticky tag that will automatically be present
in any commands issued under this checkout. So you can commit as usual,
the tag will be there. Read cvs man page for more information.
To resync with trunk, "cvs update -r HEAD <file>", and commit
eventually (to your branch).

*I* will take care of the ts2k merge to the trunk when times will look better.


Meanwhile, I'm analyzing the past commits, and I'm going to cleanup rig.h 
after you answered couple of questions.


* What is  "RIG_RPT_SHIFT_1750      // ts2000 E-type '='" ??
  To me, a shift is either + or -. 1750 sounds like the tone we're using in EU.
  Could it be an automatic shift? Anyway, it's misnamed.


* in  struct channel
  char lock;              /* added, I use this --kd7eni */
  => inefficient. This will be stored in the flags field.

  tone_t tone;            /* added --Dale kd7eni */
  tone_t tone_sql;        /* added --Dale kd7eni */
  => duplicates. These are actually ctcss_tone and ctcss_sql.

* About new RIG_MTYPE's. Tell me why are they needed. Do they store 
  any _frequency_ ?
       RIG_MTYPE_MENU,         /* menus A or B */
       RIG_MTYPE_SETUP,        /* bank 0-5; seems to save only menus */
       RIG_MTYPE_PCT,          /* Packet cluster buffered memories */

* ts2k_menu:
  This is to be implemented by ts2k_set_conf and ts2k_get_conf.
  Combo fields are a must...

* A quick glance at ts2000.c showed me this:

     # define TS2000_PARM_OP (RIG_PARM_EXCLUDE(RIG_PARM_BAT | RIG_PARM_TIME))
                              ^^^^^^^^^^^^^^^^

      This is a wrong approach. Can you imagine the mmelter adding a
      PARM after the ts2k is developped? This doesn't work. Be explicit
      in capabilities definition.



Thanks for your attention,
Stephane