| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
Special-case technic machines
|
|
This makes them work correctly with filters.
|
|
|
|
Add digilines support to autocrafter
|
|
"on" and "off" messages turn it on or off, "single" crafts one item, and sending nested tables in the shape of the crafting grid sets the craft. Example message:
{
{"default:wood","default:wood","default:wood"},
{"default:wood","","default:wood"},
{"default:wood","default:wood","default:wood"}
}
|
|
|
|
for priority tube, instead of [colorize
(except for inv image). Saves a tiny bit of RAM.
|
|
You can now make reqyests like `{group="stick"}`, `"default:pick_wood 1
30000"`, and `"mod:block <count> <exact wear> <meta>"` to match
items precisely.
If you don't specify a field, that field won't be checked. If you
specify a field in an invalid way, that rule will match nothing.
You can also specify wear as a table `wear={min, max}` to specify
a range `[min, max)` of acceptable wear values. For example,
`{name="default:pick_wood", wear={0, 32768}}` matches only wooden
pickaxes that have at least half of their life left.
You can even do things like `{count=2, metadata="whatever")}`, which
will match any item at all, as long as its metadata matches, and will
retrieve at most two of those items.
|
|
* Made digiline filter-injectors not pull a whole stack if the count is exactly 1
* Made digiline filter-injectors pull a whole stack if no count specified
* `default:dirt` still has a count of 1, but `{name="name"}` has no count
|
|
Add list rings and enable sorting tube reordering
|
|
|
|
|
|
|
|
* Add digiline detector tube
The digiline detector tube outputs an itemstring of every stack
that passes through it on the channel specified in its formspec.
* Don't store digiline detector tube's formspec in a temporary local
|
|
The table lookup will fail if node.param2 is outside [0-3] which
is easily possible since there are several ways to modify param2
values of nodes. Force truncating param2 to always be 0-3 before
using it in a table lookup.
|
|
This shouldn't be this complex. For me, both syntaxes work,
but I bet it breaks others.
|
|
Before, init.lua called io.open on
pipeworks.worldpath..'/pipeworks_settings.txt'
to see if it existed, but did not close the resulting file handle if
it was found to exist. It instead erroneously called io.close() with
no argument, which does nothing if the default output file is set to
stdout, which it is.
Now, the result of io.open is saved to a local variable. If that value
is not nil (i.e. if the world settings file exists), the file handle is
passed to io.close before calling dofile.
Also, this saves pipeworks.worldpath..'/pipeworks_settings.txt' to a
local variable to reduce redundancy.
|
|
This adds a new type of Filter-Injector that waits for a digiline
message on its channel and then pulls the items described by the
message out of the inventory. It is basically a Stackwise Injector
that, on receiving a digiline message, sets its filter to the contents
of the digiline message and then activates itself.
Sending the message {name="default:brick", count=2} should do the
same thing as setting the filter of a Stackwise Filter-Injector to
two Brick Blocks and then punching it.
If no count is specified, it defaults to 1. Since this is based off
of the Stackwise Injector, it might make more sense if the default
were an entire stack. I can change this trivially.
You can also send requests like {{name="default:brick", count=1},
{name="default:dirt", count=1}}, which acts the same as setting the
filter to one Brick Block and one Dirt Block and then punching it.
If you send a string "default:dirt" instead of a table
{name="default:dirt"}, the string is passed to ItemStack and the
name and count are extracted from the resulting ItemStack. You can
also send a list of strings instead of tables: {"default:dirt",
"default:brick"}, and the first item found will be pulled.
Punching this or activating it with Mesecons currently does
nothing. I'm not really sure what would be the right thing to do in
either of those two cases, so I made it do nothing. I guess I could
make it use the previously-used filter, but I can't really see any
usefulness in that.
The recipe is probably too cheap. The darker of the two blue texture
colors could probably be better.
|
|
Fixed mixup between enables for conductor and detector tubes
|
|
pipeworks.enable_detector_tube would define detector tubes but the
recipe for conductor tubes, and vice versa
|
|
|
|
|
|
|
|
item_drop makes Minetest 0.4.13 crash, but add_item works in all
versions and the behavior is identical when looking at the blocks.
|
|
|
|
|
|
Work with NodeTimer based furnaces.
|
|
If we insert items through tubes, we must start the furnace timer
otherwise it will never start operating. This shouldn't break
older versions, as not having a timer function should just cancel
out.
|
|
|
|
|
|
reduce texture sizes
|
|
|
|
allow_metadata_* functions
|
|
|
|
Changed hud_channge to hud_change
|
|
Implement two functions for fake player used by the hunger mod.
Also, add a list of functions to be implemented for an overview.
|
|
|
|
|
|
|
|
|
|
This reverts commit d0952a97fcff08328328d05f630d2b0036a48163.
|
|
luaentity.add_entity() are nil
|
|
|
|
just use minetest's standard protection functions instead
(leave it up to those protection mods to do their job properly)
|
|
The default mese block has been changed to default:mese, so a change of the recipe of the teleport_tube_1 had to be made
|
|
side effect left over from new_flow_logic tests
|
|
take any available ownership into account before deciding on area protection
|
|
|
