Legend of the Invincibles

Discussion and development of scenarios and campaigns for the game.

Moderator: Forum Moderators

Post Reply
User avatar
Dugi
Posts: 4965
Joined: July 22nd, 2010, 10:29 am
Location: Carpathian Mountains
Contact:

Re: Legend of the Invincibles

Post by Dugi »

Hello.

Fixed the bug in the github version. Its cooldown is mentioned in unit_information_part_1().
Ashes
Posts: 95
Joined: July 8th, 2015, 8:59 am

Re: Legend of the Invincibles

Post by Ashes »

Dugi wrote:Fixed the bug in the github version. Its cooldown is mentioned in unit_information_part_1().
Hello. I did not see it in unit_information_part_1() because it does not work!
The test should be

Code: Select all

    if from_the_ashes_used then
      result = result .. span .. _"Turns until From The Ashes will be usable:</span> "
      .. from_the_ashes_cooldown .. " \n"
    end
User avatar
Dugi
Posts: 4965
Joined: July 22nd, 2010, 10:29 am
Location: Carpathian Mountains
Contact:

Re: Legend of the Invincibles

Post by Dugi »

Committed it, I believe you that you tested it right.
User avatar
Celtic_Minstrel
Developer
Posts: 2375
Joined: August 3rd, 2012, 11:26 pm
Location: Canada
Contact:

Re: Legend of the Invincibles

Post by Celtic_Minstrel »

I think that putting partial markup in your translatable string is probably not a good idea - ie, that </span> should probably be untranslatable.
Author of The Black Cross of Aleron campaign and Default++ era.
Former maintainer of Steelhive.
User avatar
Dugi
Posts: 4965
Joined: July 22nd, 2010, 10:29 am
Location: Carpathian Mountains
Contact:

Re: Legend of the Invincibles

Post by Dugi »

I think that pango parses the markup. The string has the markup complete when it's displayed (before that translatable string is appended right after the string named span that starts the span block). I don't know how could these symbols cause trouble when doing the translation. There are also similar occurences nearby and neither of them causes display issues.
Ashes
Posts: 95
Joined: July 8th, 2015, 8:59 am

Re: Legend of the Invincibles

Post by Ashes »

Hello again. In 1x10 "The Poison", the dragon Arsenius is not vulnerable to misdraconia, because he is a "Fire Dragon" and not "Fire Dragon loti".
I am not sure that it is on purpose.
Having noticed this by trying to use a Dragonslayer bow at that moment, I looked in LotI files and there are other "Fire Dragon" in chapter 3.
User avatar
Dugi
Posts: 4965
Joined: July 22nd, 2010, 10:29 am
Location: Carpathian Mountains
Contact:

Re: Legend of the Invincibles

Post by Dugi »

Fixed in the github version.
User avatar
Celtic_Minstrel
Developer
Posts: 2375
Joined: August 3rd, 2012, 11:26 pm
Location: Canada
Contact:

Re: Legend of the Invincibles

Post by Celtic_Minstrel »

I wasn't considering display when I said that. Rather, I was considering that putting partial markup in a translatable string leads to the potential of it being broken when translated, if some translator changes the partial markup for some reason (maybe they don't understand what it is or why it's there).
Author of The Black Cross of Aleron campaign and Default++ era.
Former maintainer of Steelhive.
danielkwinsor
Posts: 5
Joined: August 20th, 2016, 9:39 pm

Re: Legend of the Invincibles

Post by danielkwinsor »

Hello Dugi,

I love Wesnoth, played LotI back in 1.12.x, but now I'm attempting to play in 1.13.x and no version is working fully.
Looks like there are UML and/or Lua changes in the game engine.
Perhaps you are already aware, but I wanted to bring to your attention and moreover get your input as to whether this is actively being worked on, and whether fixes should come from the game engine or your UMC.

I have a few logs if you need them.
I like to pretend to be a software architect, since that is my job title, so I'm capable of helping in any capacity. In fact, this has made me decide to contribute. Do you have any problems with me submitting to Call for Code Samples / WML Unit Tests based upon what I reverse engineer from LotI, or perhaps some other idea?
User avatar
Dugi
Posts: 4965
Joined: July 22nd, 2010, 10:29 am
Location: Carpathian Mountains
Contact:

Re: Legend of the Invincibles

Post by Dugi »

@Celtic_Minstrel
Aha, that could be a problem. Translators are usually well-versed in literature and computer science skills may not come together with it. I seem to be suffering from Dunning-Kruger effect, not everything I consider trivial is trivial :doh:

@danielkwinsor
I am aware that LotI doesn't work perfectly on 1.13. If I understand it correctly, all occurrences of the string advance should be replaced by advancement, for 1.13 versions but not for 1.12 versions (which are the main concern), but that definitely isn't a trivial find'n'replace task and I haven't figured out how to write the proper regex for that (I could make a C++ program for that, but that looks like an overkill). 1.13 version is a developing version and if you aren't quite into bleeding edge, 1.12 is for you (as it's currently my main focus).

No problem if you go on and use pieces of my code for that thing. I might not be able to give you hints how is that supposed to work because I might not remember what were some pieces of code supposed to do, especially if it's something that was edited ten times in the past. But although I haven't been doing much WML recently (note: it's WML, not UML, Wesnoth Markup Language, not Unified Modelling Language), but I should remember how most fundamental tricks worked.
danielkwinsor
Posts: 5
Joined: August 20th, 2016, 9:39 pm

Re: Legend of the Invincibles

Post by danielkwinsor »

I noticed all odd numbers are dev releases, and yes, I love some of those new features in 1.13.
There's some more problems than just 'advancement', such as AI doesn't move, unit disappears when it picks up an item, etc. So I've included the logs.
Attachments
wesnoth errors.txt
(14.83 KiB) Downloaded 167 times
User avatar
Dugi
Posts: 4965
Joined: July 22nd, 2010, 10:29 am
Location: Carpathian Mountains
Contact:

Re: Legend of the Invincibles

Post by Dugi »

danielkwinsor wrote:~add-ons/Legend_of_the_Invincibles/lua.lua:494: variable 'result' must be assigned before being used
This one is mine. Easy to fix.
danielkwinsor wrote:Mandatory WML child missing yet untested for. Please report.
This is an error in wesnoth. It means that the C++ part of te code expects some property of a WML tag, the tag lacks it and there is nothing to handle it, no default option, no warning, just nothing. It's also important to know when this happens. Report it on GNA.
danielkwinsor wrote:[modifications][advance] is deprecated, use [advancement] instead
This is the one I mentioned previously that it is quite hard to fix. It has nasty side effects.
danielkwinsor wrote:Unknown attribute 'animate' discarded.
Please ignore this one, it's a trick that it does not like, it has no side effect, just this complain.
danielkwinsor wrote:error ai/actions: Return value of AI ACTION was not checked. This may cause bugs!
I will need to consult my AI advisor about this.
danielkwinsor wrote:[next_item_name] not supported at scenario toplevel
If you aren't using old save files, this is really weird. Any idea where it appears?
danielkwinsor wrote:distributor mouse button left [wml_message_left]: SDL left button down. The mouse button is already down, we missed an event.
Looks like a bug in Wesnoth.

I suggest you to ignore the stuff it throws into stderr, most of it means nothing, and consult it only if some problems appear.
danielkwinsor
Posts: 5
Joined: August 20th, 2016, 9:39 pm

Re: Legend of the Invincibles

Post by danielkwinsor »

[next_item_name] not supported at scenario toplevel
If you aren't using old save files, this is really weird. Any idea where it appears?
Appears upon loading a game with items. However, I was mixing save / load versions.


For [modifications][advance] I can write a regex for you, or send pull request. I looked at the change, and maybe I'm missing something, but looks like simple find and replace.
Tad_Carlucci
Inactive Developer
Posts: 503
Joined: April 24th, 2016, 4:18 pm

Re: Legend of the Invincibles

Post by Tad_Carlucci »

danielkwinsor wrote:error ai/actions: Return value of AI ACTION was not checked. This may cause bugs!
I see this one when I interrupt the AI's turn, such as deciding I've seen enough and going off to load another save, or cancelling the game altogether. I just ignore it, because the game is going down anyway. If you see it in normal play you might want to try for a reproducible example and report it.
danielkwinsor wrote:distributor mouse button left [wml_message_left]: SDL left button down. The mouse button is already down, we missed an event.
This one I used to see a LOT and now rarely see. It seems to be all-but-gone in 1.13.5+dev.
I forked real life and now I'm getting merge conflicts.
User avatar
Dugi
Posts: 4965
Joined: July 22nd, 2010, 10:29 am
Location: Carpathian Mountains
Contact:

Re: Legend of the Invincibles

Post by Dugi »

Appears upon loading a game with items. However, I was mixing save / load versions.
Invalid something at scenario toplevel is typical for mixing save file versions.
For [modifications][advance] I can write a regex for you, or send pull request. I looked at the change, and maybe I'm missing something, but looks like simple find and replace.
It's not so easy. The string 'advance' is a substring of 'advancement', and if it's used in a dialogue or in event name, it's correct (evens named pre advance, advance and post advance have defined functionality and are crucial). The 'advance' string has to be either a WML tag or used in a variable name/path. Also, I need it to remain intact for version 1.12, so the whole line must be left unchanged inside a version check preprocessor blocks for 1.12, while the new version would be in its #else part.

I can do a simple find and replace in many files too.
Post Reply