Mass-adding items?

[BMR]

Hiya, was wondering if there was a way to mass-add items in Custom? Like around 100+ items that I have charted out on a spreadsheet. I've been digging around trying to figure it out, is there a cunning to use a combination of unlumping, editing, and relumping to do what I want? Or am I over-complicating things?

[Spoonweaver]

There MAY be a way to do what you want but the likely hood of your chart being in the exact right format is very very low. Even if your format was pretty good, and the process of importing it wasn't shaky at best, you'd still have to go through all the items to make sure they came out right which chances are they wouldn't.

You're best bet is to just start entering them in to the editor.

[Bob the Hamster]

No, unfortunately there is no way to mass-add items.

[BMR]

Hmm, righto then. No way at all to edit the unlumped files in a hex-editor or something then?

[Bob the Hamster]

Oh, well, I meant there is no easy built-in way. If you are talking about programming your own tool for editing the .ITM lump, yes that is possible.

The documentation for the ITM lump is here: http://rpg.hamsterrepublic.com/ohrrpgce/ITM

Also, the nohrio python library already has at least partial support for the ITM lump, so that might be the easiest starting point. https://gitorious.org/nohrio

[TMC]

I had a play around, and found that nohrio's description of the ITM lump was wrong (which was the fault of the wiki page, which didn't mention an unused field. A fix is available from my nohrio clone

I wrote a script to convert an .itm lump to a .csv file:

Code: Select all

import numpy as np
from nohrio.ohrstring import *
from nohrio.wrappers import OhrData
from nohrio.dtypes import dt

def get_field_decoder(dtype, name):
    """
    Depack a (possibly compound) field of a dtype, returning a (header, decoder) tuple

    A header is a list of one or more field names.
    A decoder takes a data field and returns a list (same length as header) of ints/floats/strings
    """
    ty = dtype[name]
    if len(ty):
        # compound field
        if ty.names[0] == 'length':
            # it's a string. Assume vstr2, since that's the only one used in ITM
            decoder = lambda x: ['"' + get_str16(x) + '"']
            header = [name]
        else:
            # a tuple
            decoder = lambda x: x
            header = [name + ':' + n for n in ty.names]
    elif len(ty.shape):
        # array
        assert len(ty.shape) == 1
        decoder = list
        header = [name + ' ' + str(i) for i in range(ty.shape[0])]
    else:
        # simple integer/float
        decoder = lambda x: [x]
        header = [name]

    return header, decoder

def lump2csv(lump, fields):
    headers2, decoders = [], []
    for name in fields:
        headers, decoder = get_field_decoder(lump.dtype, name)
        headers2.extend(headers)
        decoders.append(decoder)
    ret = [','.join(headers2)]
    for item in lump:
        stuff = []
        for i, name in enumerate(fields):
            stuff.extend(decoders[i](item[name]))
        ret.append(','.join(str(a) for a in stuff))

    return '\n'.join(ret)

itm = np.memmap("viking.itm", dtype = dt['itm'], mode = 'r+')

# If you're interested in ALL data
#fields = dt['itm'].names

# Only interested in some...
fields = (
 'name',
 'info',
 'value',
 'attack',
 'weaponattack',
 'equippable',
# 'teach',
# 'oobuse',
 'weaponpic',
 'weaponpal',
 'bonuses',
# 'equippableby',
# 'bitsets',
 'consumability',
# 'own_tag',
# 'in_inventory_tag',
# 'equipped_tag',
# 'equippedby_active_tag',
# 'frame2handle',
# 'frame1handle',
# 'elemdmg'
)

print lump2csv(itm, fields)

General enough to be used for most other lumps too.
If you're interested, I can write the csv to itm direction.

[BMR]

Sweet. That would be great if you could do the csv to itm as well, if it's not too much trouble. I can read through that code and get a basic idea of how it works, but the programming level just goes way above my head, hehe... Thanks.