On Thu, Aug 10, 2017 at 10:55 AM, Jef Driesen <jef@libdivecomputer.org> wrote:
On 2017-07-15 22:39, John Van Ostrand wrote:
-#define COCHRAN_MODEL_COMMANDER_PRE21000 0
-#define COCHRAN_MODEL_COMMANDER_AIR_NITROX 1
-#define COCHRAN_MODEL_EMC_14 2
-#define COCHRAN_MODEL_EMC_16 3
-#define COCHRAN_MODEL_EMC_20 4
+#define COCHRAN_MODEL_COMMANDER_TM 0
+#define COCHRAN_MODEL_COMMANDER_PRE21000 1
+#define COCHRAN_MODEL_COMMANDER_AIR_NITROX 2
+#define COCHRAN_MODEL_EMC_14 3
+#define COCHRAN_MODEL_EMC_16 4
+#define COCHRAN_MODEL_EMC_20 5

These values are now out of sync with the ones in src/descriptor.c:

--- a/src/descriptor.c
+++ b/src/descriptor.c
@@ -297,6 +297,7 @@ static const dc_descriptor_t g_descriptors[] = {
        {"Cochran", "EMC-14",           DC_FAMILY_COCHRAN_COMMANDER, 2},
        {"Cochran", "EMC-16",           DC_FAMILY_COCHRAN_COMMANDER, 3},
        {"Cochran", "EMC-20H",          DC_FAMILY_COCHRAN_COMMANDER, 4},
+       {"Cochran", "Commander TM",     DC_FAMILY_COCHRAN_COMMANDER, 5},
 };

That's not only confusing but also breaks the parser side when using the dc_parser_new2() function. That function obtains the model number from the descriptor, while the dc_parser_new() gets it from the devinfo event.

Okay, I did a re-arrangement, the wrong way again, based on the assumption that no Cochran users are storing dive blobs for later processing. 


--
John Van Ostrand
At large on sabbatical