These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

EVE Technology Lab

 
  • Topic is locked indefinitely.
123Next page
 

Dev blog: Industry Changes for Phoebe

First post First post
Author
CCP Phantom
C C P
C C P Alliance
#1 - 2014-10-13 16:31:30 UTC  |  Edited by: CCP Phantom
With the industry changes coming with Phoebe as outlined here, the Static Data Export (SDE) and the API will be updated as well. To give you wonderful third party developers enough time to update your tools and sites, CCP Nullarbor announced the pre-release version of the Phoebe SDE.

The pre-release version of the SDE can be found here: http://cdn1.eveonline.com/data/Phoebe_0.9_106316_db.zip

Read all about the SDE and the industry related XML API changes in CCP Nullarbor's blog Industry Changes for Phoebe.

CCP Phantom - Senior Community Developer

CCP Nullarbor
C C P
C C P Alliance
#2 - 2014-10-13 16:35:09 UTC
Will be watching this thread for questions, so fire away Bear

CCP Nullarbor // Senior Engineer // Team Game of Drones

Querns
GoonWaffe
Goonswarm Federation
#3 - 2014-10-13 16:49:27 UTC
I have a question, but it's kind of out of scope for third party development. If you can't answer it, it's cool.

With Reverse Engineering being merged into Invention, does this mean that you'll be able to do Reverse Engineering jobs in all Invention lines? Currently, no Reverse Engineering facilities exist in empire:

Quote:

sqlite> select count(*) from ramAssemblyLineStations a left join ramAssemblyLineTypes b on (a.assemblyLineTypeID = b.assemblyLineTypeID) left join mapSolarSystems c on (a.solarSystemID = c.solarSystemID) where b.activityID = 8 AND c.security > 0;
434
sqlite> select count(*) from ramAssemblyLineStations a left join ramAssemblyLineTypes b on (a.assemblyLineTypeID = b.assemblyLineTypeID) left join mapSolarSystems c on (a.solarSystemID = c.solarSystemID) where b.activityID = 7 AND c.security > 0;
0

(7 = reverse engineering, 8 = invention)

This post was crafted by the wormhole expert of the Goonswarm Economic Warfare Cabal, the foremost authority on Eve: Online economics and gameplay.

CCP Nullarbor
C C P
C C P Alliance
#4 - 2014-10-13 16:54:40 UTC
Querns wrote:
I have a question, but it's kind of out of scope for third party development. If you can't answer it, it's cool.

With Reverse Engineering being merged into Invention, does this mean that you'll be able to do Reverse Engineering jobs in all Invention lines? Currently, no Reverse Engineering facilities exist in empire:

Quote:

sqlite> select count(*) from ramAssemblyLineStations a left join ramAssemblyLineTypes b on (a.assemblyLineTypeID = b.assemblyLineTypeID) left join mapSolarSystems c on (a.solarSystemID = c.solarSystemID) where b.activityID = 8 AND c.security > 0;
434
sqlite> select count(*) from ramAssemblyLineStations a left join ramAssemblyLineTypes b on (a.assemblyLineTypeID = b.assemblyLineTypeID) left join mapSolarSystems c on (a.solarSystemID = c.solarSystemID) where b.activityID = 7 AND c.security > 0;
0

(7 = reverse engineering, 8 = invention)


Existing reverse engineering facilities are being converted to invention facilities but restricted to inventing just ancient relics.

So no, the labs will stay the same.

CCP Nullarbor // Senior Engineer // Team Game of Drones

Querns
GoonWaffe
Goonswarm Federation
#5 - 2014-10-13 17:13:49 UTC
CCP Nullarbor wrote:
Existing reverse engineering facilities are being converted to invention facilities but restricted to inventing just ancient relics.

So no, the labs will stay the same.

Cool -- thanks for the followup :)

This post was crafted by the wormhole expert of the Goonswarm Economic Warfare Cabal, the foremost authority on Eve: Online economics and gameplay.

Arronicus
Caldari Provisions
Caldari State
#6 - 2014-10-13 21:18:30 UTC
Sad to see interfaces go, even if they didn't add much, but the changes look good mostly. I can't help but feel like the new improved invention success probabilities might become the new status quo on big ticket items like jump freighters though? Gotta run the math again on invention cost vs savings... Nice to see multiple invention run jobs coming.
Retar Aveymone
DJ's Retirement Fund
Goonswarm Federation
#7 - 2014-10-13 21:18:55 UTC
Does that mean outpost upgrades that affect invention will now affect RE as well? (sorry this is also not an SDE question)
Altessa Post
Midnight special super sexy
#8 - 2014-10-13 22:03:34 UTC
Changing from ids to strings is probably ok. Although I do not see the benefit.

Removing the consume attribute is surely a good thing. Yet, there are other questions left open by the yaml file: invention can lead to more than one outcome. Every outcome has its own probability. Is this intentional? It artificially complicates things...

I am really no fan of yaml. Yet, parsing the existing file turned out to be unexpectedly simple. I think this is mostly due to the fact that you stick to only two constructs of the yaml standard and you use these repeatedly. Excellent for recursive parsing. I was really positively surprised! There are not so many trustworthy parsers out there. Especially in the .NET area.

By bringing in the "hyphen" notation for lists, you make things unnecessary complicated. To be honest, I have not given it too much thought yet but I already cringe at the idea...


Trying not to sound like a drama queen, but the industry has already gone through this learning curve with SoA. You do not change a public interface when it is not absolutely necessary. With every change you risk to lose 3rd party developers. So, please, consider carefully if it is worth the risk. Stability of the contract is more important than convenience...

On the internet, you can be whatever you want to be. It is amazing that so many people chose to be stupid.

Querns
GoonWaffe
Goonswarm Federation
#9 - 2014-10-13 22:50:08 UTC
Altessa Post wrote:
I am really no fan of yaml. Yet, parsing the existing file turned out to be unexpectedly simple. I think this is mostly due to the fact that you stick to only two constructs of the yaml standard and you use these repeatedly. Excellent for recursive parsing. I was really positively surprised! There are not so many trustworthy parsers out there. Especially in the .NET area.

By bringing in the "hyphen" notation for lists, you make things unnecessary complicated. To be honest, I have not given it too much thought yet but I already cringe at the idea...

Have you considered just using a pre-built library for parsing YAML? http://yaml.codeplex.com/ comes up on google for .net specifically. There's little reason to write your own parser while open source software is a thing. Any YAML library worth its salt should be able to handle the multi-line array format.

This post was crafted by the wormhole expert of the Goonswarm Economic Warfare Cabal, the foremost authority on Eve: Online economics and gameplay.

Steve Ronuken
Fuzzwork Enterprises
Vote Steve Ronuken for CSM
#10 - 2014-10-14 00:06:43 UTC
When I get an appropriate free moment, I'll be converting it again. (and shifting the yaml into database tables)

Woo! CSM XI!

Fuzzwork Enterprises

Twitter: @fuzzysteve on Twitter

Brianum
Nemesis Logistics
Goonswarm Federation
#11 - 2014-10-14 08:21:43 UTC  |  Edited by: Brianum
Steve Ronuken
Fuzzwork Enterprises
Vote Steve Ronuken for CSM
#12 - 2014-10-14 11:23:51 UTC
Conversions are now up.

I'm not updating the -latest symlinks, as this isn't a release SDE

https://www.fuzzwork.co.uk/dump/mysql56-phoebe-0.9-106316.tbz - mysql
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/ - everything else
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/eve.db.bz2 - Sqlite

Postgres - public schema
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/postgres-phoebe-0.9-106316.sql.bz2

postgres eve schema
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/postgres-phoebe-0.9-106316-withschema.sql.bz2


I've undone the change to meaningful names for the activities (ramActivites if you want them) taking it back to numbers.
I've defaulted the consume flag to 1 for everything, so sql doesn't immediately break.

Woo! CSM XI!

Fuzzwork Enterprises

Twitter: @fuzzysteve on Twitter

EMT Holding
EMT Holding Corporation
#13 - 2014-10-14 12:36:25 UTC
As long as this is the last schema change for a while, fine. I don't like having to rewrite the DB side of my application and test it all again every 2 months.
Altessa Post
Midnight special super sexy
#14 - 2014-10-14 19:16:00 UTC
Querns wrote:
Altessa Post wrote:
I am really no fan of yaml. Yet, parsing the existing file turned out to be unexpectedly simple.

Have you considered just using a pre-built library for parsing YAML? http://yaml.codeplex.com/ comes up on google for .net specifically. There's little reason to write your own parser while open source software is a thing. Any YAML library worth its salt should be able to handle the multi-line array format.

Being a lazy person startpaging (cool kids do not use google) for a solution was my first approach. Yet, the offered parsers looked badly documented or like abandoned projects.
And I did not write a general purpose yaml parser, I just parsed the file at hand. As I said, this worked surprisingly easy. Once I have prettified my code to be not too embarrassed, I can publish it here. It is not rocket science.

On the internet, you can be whatever you want to be. It is amazing that so many people chose to be stupid.

probag Bear
Xiong Offices
#15 - 2014-10-17 11:17:35 UTC  |  Edited by: probag Bear
Echoing the question(s) about Phoebe's "Adding multiple outcomes to Invention"

  • I assume the multiple outcome probabilities will be reflected in the SDE, or will happen according to some general formula somewhere, and you just haven't gotten around to actually talking about this?
  • Sorry, but I don't really see the point of the "successfulRuns" field, on its own, with multiple outcomes being added. If the quality of the successful runs is not also given, the quantity serves very little purpose.
Althalus Stenory
Flying Blacksmiths
#16 - 2014-10-17 22:53:13 UTC  |  Edited by: Althalus Stenory
Steve Ronuken wrote:
Conversions are now up.

I'm not updating the -latest symlinks, as this isn't a release SDE

https://www.fuzzwork.co.uk/dump/mysql56-phoebe-0.9-106316.tbz - mysql
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/ - everything else
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/eve.db.bz2 - Sqlite

Postgres - public schema
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/postgres-phoebe-0.9-106316.sql.bz2

postgres eve schema
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/postgres-phoebe-0.9-106316-withschema.sql.bz2


I've undone the change to meaningful names for the activities (ramActivites if you want them) taking it back to numbers.
I've defaulted the consume flag to 1 for everything, so sql doesn't immediately break.

It looks like there is an issue for IndustryActivityMaterials : all materialTypeID are between 0 and 18 (and they are correct in the yaml blueprint file).
edit: it seems to be the same for
- industryActivityProbabilities with productTypeID
- industryActivityProducts with productTypeID
- industryActivitySkills with skillID

(tested for sqlite and mysql conversion)

Another question, what is the "IndustryRaces" table ? :o

EsiPy - Python 2.7 / 3.3+ Swagger Client based on pyswagger for ESI

Desmont McCallock
#17 - 2014-10-18 07:47:54 UTC
Althalus Stenory wrote:
Another question, what is the "IndustryRaces" table ? :o
For output of RE on T3 items.
Steve Ronuken
Fuzzwork Enterprises
Vote Steve Ronuken for CSM
#18 - 2014-10-18 09:31:47 UTC  |  Edited by: Steve Ronuken
Althalus Stenory wrote:
Steve Ronuken wrote:
Conversions are now up.

I'm not updating the -latest symlinks, as this isn't a release SDE

https://www.fuzzwork.co.uk/dump/mysql56-phoebe-0.9-106316.tbz - mysql
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/ - everything else
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/eve.db.bz2 - Sqlite

Postgres - public schema
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/postgres-phoebe-0.9-106316.sql.bz2

postgres eve schema
https://www.fuzzwork.co.uk/dump/phoebe-0.9-106316/postgres-phoebe-0.9-106316-withschema.sql.bz2


I've undone the change to meaningful names for the activities (ramActivites if you want them) taking it back to numbers.
I've defaulted the consume flag to 1 for everything, so sql doesn't immediately break.

It looks like there is an issue for IndustryActivityMaterials : all materialTypeID are between 0 and 18 (and they are correct in the yaml blueprint file).
edit: it seems to be the same for
- industryActivityProbabilities with productTypeID
- industryActivityProducts with productTypeID
- industryActivitySkills with skillID

(tested for sqlite and mysql conversion)

Another question, what is the "IndustryRaces" table ? :o


Sad

I'll see about kicking the loader code around. I'd modified it a bit to make it parse, but it looks like the change was a little more significant than I thought.

(yes, industryRaces was about the outputs for RE)


This may need to wait till I'm back home from Vegas, as coding while tipsy is less than the best plan in the world, and I'm going to be a trifle busy. But I'll poke at it at least.

Woo! CSM XI!

Fuzzwork Enterprises

Twitter: @fuzzysteve on Twitter

Steve Ronuken
Fuzzwork Enterprises
Vote Steve Ronuken for CSM
#19 - 2014-10-18 09:45:27 UTC
Or maybe it was a little easier than I had thought.

It used to be that the material typeids were the keys for the list of materials. That's no longer the case. Needed a tweak to make it work.

I've not updated the main mysql dump file, but there is now an industryTables.sql.bz2 file to run as a patch for it.

The sqlite file is also updated, as is the excel dump.

As for postgres, if anyone needs it, poke me. (I've been drinking. I need my bed)

Woo! CSM XI!

Fuzzwork Enterprises

Twitter: @fuzzysteve on Twitter

Althalus Stenory
Flying Blacksmiths
#20 - 2014-10-18 13:47:35 UTC  |  Edited by: Althalus Stenory
Steve Ronuken wrote:
Or maybe it was a little easier than I had thought.

It used to be that the material typeids were the keys for the list of materials. That's no longer the case. Needed a tweak to make it work.

I've not updated the main mysql dump file, but there is now an industryTables.sql.bz2 file to run as a patch for it.

The sqlite file is also updated, as is the excel dump.

As for postgres, if anyone needs it, poke me. (I've been drinking. I need my bed)

Great!

Thanks for updating it so fast :D

edit for CCP :

It seems that some lines are duplicated in Blueprint.yaml (and then in industry skill table) :
Quote:

typeID typeName activityID skillID
10630 Rocket Launcher I Blueprint 8 11451
10632 Rocket Launcher II Blueprint 1 11451
10632 Rocket Launcher II Blueprint 3 11451
10632 Rocket Launcher II Blueprint 4 11451
10632 Rocket Launcher II Blueprint 5 11451
13321 Cruise Missile Launcher I Blueprint 8 11451
17939 Core Probe Launcher I Blueprint 8 11451
18640 Expanded Probe Launcher I Blueprint 8 11451
19740 Cruise Missile Launcher II Blueprint 1 11451
19740 Cruise Missile Launcher II Blueprint 3 11451
19740 Cruise Missile Launcher II Blueprint 4 11451
19740 Cruise Missile Launcher II Blueprint 5 11451
20344 50mm Reinforced Steel Plates II Blueprint 5 11442
20352 800mm Reinforced Steel Plates II Blueprint 5 11442
20354 1600mm Reinforced Steel Plates II Blueprint 5 11442
27022 Perpetual Motion Unit I Blueprint 8 3402


Not a big issue, but just saying in case :)

EsiPy - Python 2.7 / 3.3+ Swagger Client based on pyswagger for ESI

123Next page