Dwarf Fortress Bug Tracker - Dwarf Fortress
View Issue Details
0010581Dwarf FortressDwarf Mode -- Environmentpublic2018-02-18 02:492018-12-23 01:42
0010581: Trees yielding no wood never show up
The two kinds of tree that yield no wood (Abaca and Banana) don't show up either on embark or as saplings later.
I generated a pocket world and selected a tropical location for an embark. Through the use of a DFHack script I removed all trees from that biome except for abaca and banana and embarked. The embark was completely devoid of trees.
I then killed DF and hacked the description of banana to contain a wood section (taken from carambola) instead of "no wood" and replaced "[TREE:NONE:NONE" with "[TREE:LOCAL_PLANT_MAT:WOOD" (again taken from carambola) in the save's copy of plant_new_trees.txt, redid the same thing (embark at the same location after hacking the biome to only retain abaca and banana trees), and the embark is covered in banana trees. Killing DF, hacking abaca to yield wood, and then hacking the biome and embarking again showed both banana and abaca on embark.
For saplings I rely on my observation over the last year or two that I've never gotten any banana saplings on any of the embarks that should have had them (which is what led to this investigation). Using DFHack to add a banana sapling did result in a banana tree after 3 years in a 0.43.X embark (once invaders stopped trampling the saplings hacked in).
I have also hacked my current real save to provide wood for bananas, and there is at least one sapling on the embark after two months (following 13 years of complete absence).

It can be noted that elves occasionally bring bananas as trade goods, but that's probably because those are abstracted rather than actually harvested...
No tags attached.
related to 0010954acknowledged Loci Food/drink preferences are buggy (bugs and references to related reports) 
Issue History
2018-02-18 02:49PatrikLundellNew Issue
2018-05-29 12:42chaosvoltNote Added: 0038398
2018-05-29 12:45chaosvoltNote Edited: 0038398bug_revision_view_page.php?bugnote_id=0038398#r15622
2018-05-29 14:16lethosorNote Added: 0038399
2018-05-29 16:20LociNote Added: 0038404
2018-05-29 16:20LociAssigned To => Loci
2018-05-29 16:20LociStatusnew => confirmed
2018-05-29 16:31lethosorNote Added: 0038405
2018-05-30 03:37PatrikLundellNote Added: 0038407
2018-05-30 09:02lethosorNote Added: 0038409
2018-11-12 12:31LociRelationship addedrelated to 0010954
2018-12-22 01:22FantasticDorfNote Added: 0039037
2018-12-22 01:37PatrikLundellNote Added: 0039038
2018-12-22 12:12FantasticDorfNote Added: 0039039
2018-12-23 01:42PatrikLundellNote Added: 0039041

2018-05-29 12:42   
(edited on: 2018-05-29 12:45)
On the one hand, dfhack REEEE. On the other hand I replicated this issue via raw-only methods. My recent modding shenanigans entailed a non-wood tree being the only tree in a given region and they refused to spawn.

Strangely, there might be something that can work around this, as evidently ZM has had success with non-wood trees. Maybe both DFHack and my flavor of fuckery are failing at this.

Alternatively, it could be that ZM's attempt at it does something in the raws that vanilla woodless trees don't.

2018-05-29 14:16   
We cannot accept DFHack-only bug reports here. If you can get it to reproduce with modded raws (or adjusted vanilla raws), uploading the save after worldgen but without an active fortress (i.e. to allow embarking quickly) would help.
2018-05-29 16:20   
I disagree that this is a DFHack issue. Here's a trivial test world with all surface trees except abaca and banana removed:

http://dffd.bay12games.com/file.php?id=13793 [^]

Modding either abaca or banana trees to produce wood will cause them to appear in tropical biomes.
2018-05-29 16:31   
Yeah, I'm not saying it's a DFHack issue. I'm saying that it's hard to prove it's not a DFHack issue from something like this:

"Through the use of a DFHack script I removed all trees from that biome except for abaca and banana and embarked."

It's possible (in general, not here) that a method used in DFHack fails to take into account things that vanilla DF does take into account, so "I caused an issue using DFHack" on its own doesn't belong here. Given chaosvolt's findings, this specific issue does, though.

Thanks for taking the time to produce a save!
2018-05-30 03:37   
It can be pointed out that I didn't cause the issue using DFHack, but used DFHack to remove the chaff when investigating what I had observed, as it's nearly impossible to see that something doesn't show up when it could be the RNG (consistently) playing tricks on you.

Many thanks to Loci for showing the issue in an intelligent DFHack free way.

It would be useful to know what chaosvolt's reference to ZM's work is.
2018-05-30 09:02   
For all I knew, removing tree types with DFHack could fail to update DF's state in some other way, preventing the remaining trees from showing up at all, while they could show up in vanilla if every other type was removed from the raws. (Again, this is just hypothetical, but it's hard to tell what the side effects of a DFHack method can be.)
2018-12-22 01:22   
Trees will also not reproduce if they don't have a seed growth, bananas have a seed growth within its fruit, and abaca's have a not quite clear but defined seed growth somewhere potentially within its leaves or flowers (unclear to me at the moment)

Glumprongs suffer a problem in which as a singular tree type of evil biomes, they are spawned in but never create saplings meaning that they are sterile and will never re-grow, civs with the relevant [USE_EVIL_WOOD] tags won't be able to access them until renewability is ensured which i've managed to recreate in my modding by making them drop simple nut growths.

So in line with issue report, [HEAVY_BRANCH_DENSITY:0],[BRANCH_DENSITY:0],[MAX_TRUNK_HEIGHT:5],[TRUNK_BRANCHING:0] in which both share no branches and also no wood might constitute that since both the growths are attached to the trunk, without the wood there the problem arises (as a proposed method experiment open to try)
2018-12-22 01:37   
My experience does not match FantasticDorf's description, as I have no problem to get Glumprong (unmodified, but a DFHack script ensures Glumprong exists in the biome) to produce saplings in my evil biome (except, of course, when everything growing there is dead). I've had wood cutters harvest Glumprong regularly over decades.
2018-12-22 12:12   
Using DFhack to spawn saplings doesn't denote anything regarding causation with the OP's tree's to not spread or otherwise generate. My note was to illustrate the effects the individual growth parts had and the reliance on them for trees to be renewable in its biome range.

With nothing to breed with (due to faulty trunk not having wood or other reason) in a busy jungle full of other trees placing saplings, it might just not appear at rough odds 1/100 depending on how wide/small the biome is and where you looked. Typically if you can find some of them, they will be very tall and very old depending on world-age rather than young newly grown ones.

I think i do recall seeing some on 44.12 infrequently.
2018-12-23 01:42   
I do not use DFHack to SPAWN saplings: I use DFHack to ensure Glumprong is among the plants that are available to the biome (if it's not, you won't get any old ones either) in case DF RNG caused it to be excluded. Any saplings appearing appear according to the normal DF rules for plant growth (and I didn't have any problem with Glumprong regrowth [when present] before making sure it was present either).