EVE Technology Lab

Ok so...stepping through asset list. Each asset has a flag that shows its exact location type. That flag relates to an invFlags or config.Flags table. Is reverance blue.config not capable of using that table? I don't see it in table list anywhere and it bitches if I try to use it. Any help would be greatly appreciated. Thanks.

Cheers,
-PL

Hey thanks for quick response. Sorry for long response here.

I was not aware of those in the const.py. Those could be useful to me later.

You're right. I should be much more specific. Here is a snippet of code to illustrate what I'm after:


import os
import time
import cPickle as pickle
import eveapi
from reverence import blue

USER_ID = 0
API_KEY = ""
CHAR_NAME = ""
EVE_PATH = ""

class APICacher(object):
def __init__(self):
self.file_ext = ".api_result"
self.file_path = "./API_Cache/"
if not os.path.exists(self.file_path):
os.makedirs(self.file_path)

def retrieve(self, host, path, params):
hash_string = str(hash(str(host) + str(path) + str(params)))
file_name = self.file_path + hash_string + self.file_ext
cache_file= None
try:
cache_file= open(file_name, 'rb')
except IOError:
return None
else:
cache_object = pickle.load(cache_file)
cache_file.close()
if time.time() < cache_object.cachedUntil:
return cache_object
else:
return None

def store(self, host, path, params, doc, obj):
hash_string = str(hash(str(host) + str(path) + str(params)))
file_name = self.file_path + hash_string + self.file_ext
cache_file = open(file_name, 'wb')
pickle.dump(obj, cache_file)
cache_file.close()


eve = blue.EVE(EVE_PATH)
invtypes = eve.cfg.invtypes
api = eveapi.EVEAPIConnection(cacheHandler=APICacher())
auth = api.auth(userID=USER_ID, apiKey=API_KEY)
chars = auth.account.Characters().characters
char_id = None
for char in chars:
if char.name == CHAR_NAME:
char_id = char.characterID
corp_auth = auth.corporation(char_id)

for i in corp_auth.AssetList().assets:
print invtypes.get(i.typeID).typeName, "| flag= ", i.flag
if hasattr(i,"contents"):
for j in i.contents:
print " "*4, invtypes.get(j.typeID).typeName, "| flag= ", j.flag
if hasattr(j,"contents"):
for k in j.contents:
print " "*8, invtypes.get(k.typeID).typeName



The flag field gives me a number here that relates to this table: link here

How do I look up the flag number in this table and get either flagName or flagText?


I'm going to apologize in advance if this is stupid or obvious but humor me anyway, if you would, please.

Thanks for any help!


Cheers,
-PL


*edit*
unfortunately, the forum destroys the indentations in the code. Oddly, if I go to Edit it...all the indentations are shown.

Ah, not entirely sure that invFlags table exists in the client. A quick glance over the code suggests it's not there.

I would suggest hardcoding that table as a dict.

The script you have there is one to read the price history, not the current prices. I assume thats what you want? There is also :

key[1]=="GetNewPriceHistory"

GetNewPriceHistory(regionID, typeID)

Returns: CRowset (historyDate, lowPrice, highPrice, avgPrice, volume, orders)

This returns only one list entry, not multiples as you would expect for market histories.

historyDate apparently is the number of 100-nanosecond intervals since January 1st, 1600, which a common Microsoft way of storing time values. To represent it in the more familar "seconds since 1-1-1970" 32-bit timestamp:

secondsSinceUnixEpoch = (historyDate - 116444736000000000) / 10000000

(info from http://wiki.eve-id.net/Cache_Resources#GetNewPriceHistory.28regionID.2C_typeID.29)

for current prices you need the GetOrders key.

Something in my current script to read current prices from cache is broken since the changes to the bulkdata so thats all I can suggest till I find out what that issue is :(

Addendum : and I have it - there are / and " characters in the apparel names that are killing my script. Nerfed by space barbie o\

Updated version 1.4.0 for Crucible now available (see OP).

Looping over a large amount of files without calling close() after each tends to cause this issue.
The garbage collector simply kicks in too late for it to stop you running out of filehandles that way. Just explicitly close a file before moving on to the next one and you will have no issue.

Also, avoid using LoadCacheFolder() for this sort of thing (if you're using it), I kinda regret putting it in the published version.

Documentation is possibly not my strong point ;) Though, the deeper stuff (like cacheMgr) is there mostly for use in reverence itself anyway.
I'm fixing the filehandle issue though. Will post an update sometime Soon™

Today I started to get following error from blue.marshal.Load(open(filename,"rb").read()) call (full code).

Error is: "Attempted store of NULL at shared position #3 - type:0x2 ctype:0x14 len:23 share:64 pos:212 size:13151"

EDIT: I'm using reverence-1.4.1.win32-py2.6. Error was coming from one cache file and when I deleted it I was able to continue using my script again.

Yep... I was so eager to get it working that I just deleted the file. If I get same error again I'll save the file.

Since simple XML is basicly same as CSV except is uses a lot of crap instead of simple commas, this might work: http://pastie.org/private/7ssdk0mde7gdsfqeqkwjw

(and no, I don't know how to split that big line into smaller parts, might need to rewrite section following it to do that.)