Add missing "cancel" checks, "cancel-everything" ecosystem limitation

[dexX7]

• trade_MP can be used with zero amounts
• trade_MP can be used with non existing properties
• fixes behavior where raw encoded offer with zero units for sale fails, but zero units desired passes etc.
• guards and fillins of todos for "meta dex logic math"

... and cancel-everything that behaves as follows:

• both properties in main ecosystem -> cancel main properties
• both properties in test ecosystem -> cancel test properties
• cross ecosystem values (...) cancel offers of both ecosystems

Consider the last part as debatable of course.

[dexX7]

If desired, I can split this one up, but I was too tired yesterday. Sub parts:

• flexible input validation of trade_MP to remove constraints related to the fields that are simply junk for some actions
• don't always reject Bitcoin as property in any case, because the property field could just have been abused as filler and come with "zero" values
• ensure sufficient values are provided for "cancel-at-price" and "cancel-currency-pair"
• cross ecosystem behavior and support for "cancel-everything" limited to one ecosystem

Imho, related to the last part, there is consensus for the general idea. I'd want a limitation of "cancel-everything" to one ecosystem, Marv seems to support it too, as the comment linked in the other thread implies, Faiz supports it for consistency, zathras supports it as well. So if you don't hate it, let's do it. ;)

JR's objection was the work needed to adjust the code - which I tackled with this PR. I can push a PR to refine the spec on this as well and imho it's not a spec violation, but the spec is rather ambigious and not clearly defined in this context.

Opinions?

Edit: not to mention all the regtests that don't fail anymore. ;)

[dacoinminster]

heh. Nice. If the other devs don't object and you are willing to do the
spec work too, this is okay by me.

On Wed, Dec 3, 2014 at 12:16 PM, dexX7 notifications@github.com wrote:

If desired, I can split this one up, but I was too tired yesterday. Sub
parts:

• flexible input validation of trade_MP to remove constraints related
  to the fields that are simply junk for some actions
• don't always reject Bitcoin as property in any case, because the
  property field could just have been abused as filler and come with "zero"
  values
• ensure sufficient values are provided for "cancel-at-price" and
  "cancel-currency-pair"
• cross ecosystem behavior and support for "cancel-everything" limited
  to one ecosystem

Imho, related to the last part, there is consensus for the general idea.
I'd want a limitation of "cancel-everything" to one ecosystem, Marv seems
to support it too, as the comment linked in the other thread implies, Faiz
supports it for consistency, zathras supports it as well. So if you don't
hate it, let's do it. ;)

JR's objection was the work needed to adjust the code - which I tackled
with this PR. I can push a PR to refine the spec on this as well and imho
it's not a spec violation, but the spec is rather ambigious and not clearly
defined in this context.

Opinions?

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

[dexX7]

@dacoinminster: PR pending: OmniLayer/spec#292

[dexX7]

@m21:
Consider this as finalized - unless you or anyone else has anything to add of course.. ;)

With #223 test support is back and all tests pass.

[m21]

TBD after the tag