Tx 21: refine sub-actions and the effect of "cancel-everything"

[dexX7]

This commit refines the effects of the meta-dex sub-actions and adds further information about validity checks.

In particular the behavior of "cancel-everything" was not fully defined, because cross-ecosystem effects are usually not allowed.

[dacoinminster]

I believe this reflects changes @dexX7 already made to the code, so I plan to merge this ASAP. I've reviewed the changes. Any objections?

[dexX7]

I believe this reflects changes @dexX7 already made to the code, ....

The initial PR was split for easier auditing, whereby all necessary prerequisites are now merged or ACKed. The core logic, namely the behavior of "cancel-everything" was not yet merged. To finalize the split, I'll push the rest in just a few minutes, after the regtests finished.

IIRC later this day there is a meeting, which might be a good opertunity to get consensus on this topic.

[zathras-crypto]

@dexx that's correct mate, meeting today in Sococco in 20 mins from this
message, if you can attend - fantastic :)

On Wed, Dec 10, 2014 at 2:26 AM, dexX7 [email protected] wrote:

I believe this reflects changes @dexX7 https://github.com/dexX7 already
made to the code, ....

The initial PR was split for easier auditing, whereby all necessary
prerequisites are now merged or ACKed. The core logic, namely the behavior
of "cancel-everything" was not yet merged. To finalize the split, I'll push
the rest in just a few minutes, after the regtests finished.

IIRC later this day there is a meeting, which might be a good opertunity
to get consensus on this topic.

—
Reply to this email directly or view it on GitHub
#292 (comment).

[dacoinminster]

Er, since the work is already done, I support merging everything. Do we
actually have an objection against doing this, or do people want to talk
about it more first? If so, maybe we should do that in the Skype chat . . .

On Tue, Dec 9, 2014 at 12:42 PM, zathras-crypto [email protected]
wrote:

@dexx that's correct mate, meeting today in Sococco in 20 mins from this
message, if you can attend - fantastic :)

On Wed, Dec 10, 2014 at 2:26 AM, dexX7 [email protected] wrote:

I believe this reflects changes @dexX7 https://github.com/dexX7
already
made to the code, ....

The initial PR was split for easier auditing, whereby all necessary
prerequisites are now merged or ACKed. The core logic, namely the
behavior
of "cancel-everything" was not yet merged. To finalize the split, I'll
push
the rest in just a few minutes, after the regtests finished.

IIRC later this day there is a meeting, which might be a good opertunity
to get consensus on this topic.

—
Reply to this email directly or view it on GitHub
#292 (comment).

—
Reply to this email directly or view it on GitHub
#292 (comment).

[dexX7]

No, I don't object. If you think it's ready and there is consensus this is the right approach, then please go ahead.

[dacoinminster]

@dexX7 are we certain we're merging the code? If not, I'll wait on this.

[dexX7]

This is something I'm actually not sure about. I splitted the rest and the core logic is isolated in mastercoin-MSC/mastercore#223. So ping: @zathras-crypto @m21

The process should probably be: proposal -> consensus -> specification -> integration. Only exception: something is already live on mainnet where a spec change is made to catch up with "reality". I provided the code proactively to speed things up, but having the code ready should not have much weight here. The consensus part is crucial.

To me it sounded as if this is the desired behavior, based on mastercoin-MSC/mastercore#219.

[dacoinminster]

Merged, based on your comments above :)