Bug - Fixed r8795-Item errors

[Fluxxdog]

Retrieving character data...
Unknown item found: a rumpled paper strip (s1
Bad item for monster "Evil Trumpet-playing Mariachi": a rumpled paper strip (s1
Unknown item found: a torn paper strip (s1
Bad item for monster "Evil Trumpet-playing Mariachi": a torn paper strip (s1
Unknown item found: a creased paper strip (s1
Bad item for monster "Evil Vihuela-playing Mariachi": a creased paper strip (s1
Unknown item found: a crinkled paper strip (s1
Bad item for monster "Evil Vihuela-playing Mariachi": a crinkled paper strip (s1
Unknown item found: a crumpled paper strip (s1
Bad item for monster "Evil Vihuela-playing Mariachi": a crumpled paper strip (s1
Unknown item found: a folded paper strip (s1
Bad item for monster "Evil Vihuela-playing Mariachi": a folded paper strip (s1
Unknown item found: a ragged paper strip (s1
Bad item for monster "Evil Vihuela-playing Mariachi": a ragged paper strip (s1
Unknown item found: tin cup of mulligan stew (s1
Bad item for monster "Hodgman, The Hoboverlord": tin cup of mulligan stew (s1
Unknown item found: Hodgman's journal #1: The Lean Times (s1
Bad item for monster "Hodgman, The Hoboverlord": Hodgman's journal #1: The Lean Times (s1
Unknown item found: Hodgman's journal #2: Entrepreneurythmics (s1
Bad item for monster "Hodgman, The Hoboverlord": Hodgman's journal #2: Entrepreneurythmics (s1

This was very odd, especially since I'm playing a Disco Bandit. Got this as soon as I logged on. I have the paper strips, but none of the Hodgman stuff in my inventory. Mafia did not create an error log.

 

[Winterbay]

That seems to answer my question regarding s1 or 0 being the new unkown droprate definition at least... Only it appears to be odd.

 

[Theraze]

Appears that the system didn't properly do the unspaded part of the monster update... the s1 is supposed to be unspaded, 1% drop rate. Works on mine, Veracity's update post seemed to have it. I'll have to update to that and see what got missed.

 

[Theraze]

Lovely... appears that those 8-10 items are missing the right parenthesis. No clue what happened there, since it didn't give me any item errors yesterday. *mutters* I'll post a fix to this.

 

[Fluxxdog]

Hey, at least we it got quickly. When the RSS bot posted 8795, it still wasn't up on the builds list.

 

[Theraze]

Anyways, here's the right parenthesis fix. Sorry about that. Makes sense that the one patch that gets implemented is the one missing 10 parenthesis. Sorry Veracity.

And no, I don't just try to paste all the wiki-crap in. I actually started by going through the big list of spaded and looked for patterns in how it diverged... It started that the underwater ones were generally 10% lower defence (useful defence, not actual). Then I started wandering on to where mafia had no data at all, and then I started looking through the spots where the meat spade flag said that the drop should have been impossible. I'll need to run some of those areas later (when I'm done with BM completely) with a flat 0% meat modifier for a week or two to make them certain.

Edit: Tested, this patch should provide no item errors.

 

[Theraze]

So yeah, patch is in the post above. Question... some of the data files are formatted in *nix style, some are formatted in DOS style. Should they get standardized as one or the other eventually? I'd probably suggest the DOS style because, while it does involve an extra character per line, it's more compatible... regardless of OS, any user should be able to browse it properly. But if we standardize on *nix, that should work as well...

 

[Bale]

Well, this explains why my script was getting an unknown monster error for Hodgman, The Hoboverlord. Thanks for the quick patch. My script works again.

 

[Veracity]

Revision 8796

 

[Theraze]

Thanks, and sorry to screw things up for you, Veracity.

 

[xKiv]

So yeah, patch is in the post above. Question... some of the data files are formatted in *nix style, some are formatted in DOS style. Should they get standardized as one or the other eventually? I'd probably suggest the DOS style because, while it does involve an extra character per line, it's more compatible... regardless of OS, any user should be able to browse it properly. But if we standardize on *nix, that should work as well...

I don't know about SVN, but CVS had some flag to make line endings check out (and in) according to what is the checking users OS (or maybe another preference), so that everybody could have it as they need, while it being consistent on the server and not showing as a difference.

Quick google:
Set the property 'svn:eol-style' to 'native' - this will automatically convert newlines to whatever is used on your platform (use 'CRLF', 'CR' or 'LF' to get those regardless of what the OS wants).

You can use auto-properties so that all future files you create will have this property set (auto props are handled client-side, so you'd have to set this up for each user).

---
That will probably not change line endings in already committed files though ...

 

[Theraze]

Probably screws up when I make patches though, since the diffs won't match. It'll end up making whatever lines I have with my style of line endings, regardless of what the base (SVN) version was.

It's all well and good for me to have weird stuff on my client. It's bad if I make patches and spread my weirdness in ways that breaks other people in non-replicable ways.

 

[xKiv]

Yeah, I have seen too many ^M-before-end-of-line-on-some-but-not-all-lines-of-the-file, I should really know better than that.