dseq gets to the TX logs.events twice

[arno01]

akash v0.16.4-rc2

Example:

$ akash query tx E0E3548EDDB2CABCA4489BB4B07E1B40173DF7F9B5DBE1AB7F58C73BD24BFBA0 -o json | jq
{
  "height": "6100073",
  "txhash": "E0E3548EDDB2CABCA4489BB4B07E1B40173DF7F9B5DBE1AB7F58C73BD24BFBA0",
...
            {
              "key": "dseq",
              "value": "6100072"
            },
            {
              "key": "module",
              "value": "market"
            },
            {
              "key": "action",
              "value": "order-created"
            },
            {
              "key": "owner",
              "value": "akash130xhrhruhmv65w3awpgzylqkhzj7km5jgfyez9"
            },
            {
              "key": "dseq",
              "value": "6100072"
            },
            {
...

And it was same before the mainnet3 upgrade (before 5629650 height):

$ akash query gov proposal 19 -o json | jq -r '.content.plan.height'
5629650

Note: I'm using pre-mainnet3 akash binary (v0.14.1) to query the pre-mainnet3 TX due to /akash.deployment.v1beta1.MsgCreateDeployment VS /akash.deployment.v1beta2.MsgCreateDeployment TX type. Not sure if this is expected. Have opened a separate issue for this https://github.com/ovrclk/akash/issues/1587

$ akash-v0.14.1 query tx 981D04BB2AC761FB3CABA5C2F7F7261FA75E1969DA489DC706257A7C1FE9AF9C -o json | jq
{
  "height": "5565875",
  "txhash": "981D04BB2AC761FB3CABA5C2F7F7261FA75E1969DA489DC706257A7C1FE9AF9C",
...
            {
              "key": "dseq",
              "value": "5565873"
            },
            {
              "key": "module",
              "value": "market"
            },
            {
              "key": "action",
              "value": "order-created"
            },
            {
              "key": "owner",
              "value": "akash1h24fljt7p0nh82cq0za0uhsct3sfwsfu9w3c9h"
            },
            {
              "key": "dseq",
              "value": "5565873"
            },
...

[boz]

yeah, cosmos-sdk merges all of the events for a transaction, unfortunately.

we basically need to write our own indexer query api.