Monster editor?

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • GSN
    Rookie
    • Apr 2007
    • 21

    Monster editor?

    Has someone written a monster _editor_ for Vanilla or any other variant? I have found a monster viewer (BAE) with several useful features, but editing still has be done by hand.
  • d_m
    Angband Devteam member
    • Aug 2008
    • 1517

    #2
    I was thinking about writing one. Initially it would probably run on Linux (which is what I'm developing in).

    Is there any rhyme or reason for how the current file is formatted? Reading the stuff in is easy, but writing out spells, flags and description so they are exact matches of the current monster.txt file is kind of annoying.

    If there really is a hard-and-fast system of how it should be structured I'll definitely support it, but it seems pretty arbitrary right now.
    linux->xterm->screen->pmacs

    Comment

    • Pete Mack
      Prophet
      • Apr 2007
      • 6883

      #3
      In re formatting changes: yes and no. Yes: there's no fundamental
      Reason for the current format, and a canonical (machine generated) format is a good idea.

      No: because any time the format changes, doing revision diffs becomes impossible. Major file changes should be bundled together.

      Comment

      • d_m
        Angband Devteam member
        • Aug 2008
        • 1517

        #4
        Originally posted by Pete Mack
        No: because any time the format changes, doing revision diffs becomes impossible. Major file changes should be bundled together.
        Do you think it would be feasible to agree on a canonical form, and then do one commit which was just reforming the file? Obviously this would only happen if people were convinced that having a canonical format was useful (e.g. if there were tools to manipulate the file and a filter to canonicalize files).
        linux->xterm->screen->pmacs

        Comment

        • Narvius
          Knight
          • Dec 2007
          • 589

          #5
          I know there's a monster *generator* over on thangorodrim.net. Could be combined with that viewer thing for awesomeness.
          If you can convincingly pretend you're crazy, you probably are.

          Comment

          • takkaria
            Veteran
            • Apr 2007
            • 1951

            #6
            Originally posted by d_m
            Do you think it would be feasible to agree on a canonical form, and then do one commit which was just reforming the file? Obviously this would only happen if people were convinced that having a canonical format was useful (e.g. if there were tools to manipulate the file and a filter to canonicalize files).
            There is a canonical form for monster flags:


            Flags should be in order specified there, and each new grouping should start a new line. For each flag, if there are already six flags on the same line, then it should start a new line, too.
            takkaria whispers something about options. -more-

            Comment

            • d_m
              Angband Devteam member
              • Aug 2008
              • 1517

              #7
              Originally posted by takkaria
              There is a canonical form for monster flags:


              Flags should be in order specified there, and each new grouping should start a new line. For each flag, if there are already six flags on the same line, then it should start a new line, too.
              Awesome, thanks! This is exactly what I was looking for.

              EDIT: while you're here, what is the guideline for wrapping descriptions (if any)?
              linux->xterm->screen->pmacs

              Comment

              • takkaria
                Veteran
                • Apr 2007
                • 1951

                #8
                Originally posted by d_m
                Awesome, thanks! This is exactly what I was looking for.

                EDIT: while you're here, what is the guideline for wrapping descriptions (if any)?
                Wrap at 80, and spaces at the end of a line should be moved to the beginning of the next.
                takkaria whispers something about options. -more-

                Comment

                • d_m
                  Angband Devteam member
                  • Aug 2008
                  • 1517

                  #9
                  So here are a few more questions/issues that I have about the file. Free free to defer these questions if they prove too annoying:

                  1. Should spells come before or after flags? Currently it seems to vary by entry. Should this be standardized?

                  2. Is there a canonical order for spells? Should there be? Should spells wrap after the 6th entry like flags? Should they be grouped according to spell type (this might be happening right now)?

                  3. There are some flags missing from the .lua file you referenced (DROP_2, DROP_3, etc). Should these be added?

                  4. What order should attacks be in? Currently it seems alphabetical by attack name (e.g. "bite" before "claw").
                  linux->xterm->screen->pmacs

                  Comment

                  • GSN
                    Rookie
                    • Apr 2007
                    • 21

                    #10
                    1. Should spells come before or after flags? Currently it seems to vary by entry. Should this be standardized?
                    All r_info lines are parsed independently of each other (see parse_r_info in initi1.c). The current convention merely affects human readability.

                    2. Is there a canonical order for spells? Should there be? Should spells wrap after the 6th entry like flags? Should they be grouped according to spell type (this might be happening right now)?
                    No each time.

                    3. There are some flags missing from the .lua file you referenced (DROP_2, DROP_3, etc). Should these be added?
                    Only flags that are defined in defines.h may be added. I don't know much about Lua, but I believe that any flag defined by scripts may be used only in monster definitions contained in scripts, not in r_info.

                    4. What order should attacks be in? Currently it seems alphabetical by attack name (e.g. "bite" before "claw").
                    Attack order is irrelevant except for aesthetic reasons. Currently monsters always do a "full" attack (unless they cast spells), and so does the player. Implementing D&D 3.x style full and base attacks would be more trouble than they are worth. Angband it's not THAT tactical, at least in its current form.

                    Comment

                    • d_m
                      Angband Devteam member
                      • Aug 2008
                      • 1517

                      #11
                      Originally posted by GSN
                      Attack order is irrelevant except for aesthetic reasons. Currently monsters always do a "full" attack (unless they cast spells), and so does the player. Implementing D&D 3.x style full and base attacks would be more trouble than they are worth. Angband it's not THAT tactical, at least in its current form.
                      Thanks for responding. I think maybe I didn't communicate what I was aking. I will try to be more clear:

                      Currently I have a program that can read in the entire monsters.txt file into various Monster objects, and then later write those objects out. The problem is that the way my program chooses to write out the monsters isn't identical to the current format of these monsters (because many of them are arranged in conflicting and arbitrary ways).

                      I would prefer to have the maintainer(s) state a preferred canonical form for the entire file. This way, once the file conforms to the spec, I don't have to worry about having my program save out the "same" monster in a form that looks different (for whatever reason) from the original input.

                      Sorry if I wasn't clear about that. I have read and understood the input format, so I am really asking for an "official" opinion about it, not an explanation for what the fields mean or what order they are usually in.
                      linux->xterm->screen->pmacs

                      Comment

                      • Pete Mack
                        Prophet
                        • Apr 2007
                        • 6883

                        #12
                        correction: monsters almost always do a full set of attacks, unless one attack is an earthquake or theft that moves either the player or the monster. In this case it is important that attack order be preserved.

                        Comment

                        Working...
                        😀
                        😂
                        🥰
                        😘
                        🤢
                        😎
                        😞
                        😡
                        👍
                        👎