There seems to be invalid patterns on the site

Has something gone haywire? Let us know about it!
Post Reply
dbell
Posts: 302
Joined: June 27th, 2013, 12:47 am
Contact:

There seems to be invalid patterns on the site

Post by dbell » March 27th, 2023, 11:02 pm

I have recently noticed some Life Viewer errors on some of the old patterns.

For example, this old pattern from the 'Can we substantiate this claim' thread generates an error:

Code: Select all

x = 89, y = 53
#C A fly-by deletion of a pond using an unusual spaceship.
#C From a collection by Jason Summers in 2001.
23boo16bo5bo18bo5bo5bo$22bobbo14b3o3b3o7b3o6b3o3b3o3b3o$22bobbo14boboo
boobbo5bobbo6bobooboobbobobboo5bo$23boo6bo9bobo4b3o3b3obbo6bobobobboob
obboo4b3o$30b3o8b3obobobo8b6obbo3bobobobobbo5boboo$bo15bo11bobboo3b8ob
5obobbo4bobo3b3o5b3o3bo5b3o$3o6bo6b3o10bobboo3boboobbo5bobobbo9bo4boob
o7bo5b3o$oboo4b3o5boboo11bobbobbobo4bobbo15bo7bo8b4obboo$b3o3bobboo5b
3o7bo3bobbo9boo30bo3bo$b3obbobobbo5b3o11b4o$boo7bob3obboo12booboo$4boo
6boo3boo14bobbo$bbobbo6bo3boo12bobbobbob3o$bo11bo17bo6boboo$bo35bobbob
oo$40bobo$19bo12bo6booboboo$18b3o10b3o8boboo$18boboo8boobo8bobbo$19b3o
8b3o9booboo$19b3o8b3o6booboobbo$19b3o9boo8boobbobo$19boo20boo$$42bobo$
17bo23bob3o$16b3o21boobobbo$15boobo20bobo$15b3o21bo5bo$15b3o20bo4bobo$
16boo19boo4bobo$37boo$38bobbo$$38b3o$37bobbo$37b5o$37boo$38boobbo$39bo
5bo$38bo5b3o$37boo5boboo$38bo6b3o$38bo6boo$39boboo$40b3o$$43bo$44bo$
44boo$44b3o$44b3o$44boo!
The problem seems to be the inclusion of comment lines after the initial 'x =' line.
This doesn't cause problems for golly, but does for other Life programs (such as my own).

I don't know if the RLE standard allows embedded comment lines, but I thought it didn't.

Someone must have run some script over the patterns on the site to change them in this way. Please change them back so that the patterns run again. Thanks!

BCNU,
-dbell

User avatar
muzik
Posts: 5897
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: There seems to be invalid patterns on the site

Post by muzik » March 28th, 2023, 5:46 am

On a related note, the pattern in the linked post also doesn't seem to be runnable anymore:
viewtopic.php?f=3&t=1622&p=157568#p157568

It was fixed a bit over a month ago, but seems to have broken again since (almost certaintly due to changes to handling Niemiec cells).
Parity Replicator Collection v1.6 is now live - please send all relevant discoveries here.

User avatar
rowett
Moderator
Posts: 4019
Joined: January 31st, 2013, 2:34 am
Location: UK
Contact:

Re: There seems to be invalid patterns on the site

Post by rowett » March 28th, 2023, 5:57 am

dbell wrote:
March 27th, 2023, 11:02 pm
I have recently noticed some Life Viewer errors on some of the old patterns.

The problem seems to be the inclusion of comment lines after the initial 'x =' line.
This has been fixed. Thanks for reporting!

No need to change the patterns it was a different issue. LifeViewer is happy with comment lines in the RLE body.

User avatar
rowett
Moderator
Posts: 4019
Joined: January 31st, 2013, 2:34 am
Location: UK
Contact:

Re: There seems to be invalid patterns on the site

Post by rowett » March 28th, 2023, 5:57 am

muzik wrote:
March 28th, 2023, 5:46 am
On a related note, the pattern in the linked post also doesn't seem to be runnable anymore:
Fixed, thanks.

dbell
Posts: 302
Joined: June 27th, 2013, 12:47 am
Contact:

Re: There seems to be invalid patterns on the site

Post by dbell » March 28th, 2023, 7:15 am

Thanks for the quick fix! It also fixes the other examples I found that didn't work before.

So the real problem wasn't the comment lines after all; that was just a coincidence.

(However, now I have to update my own Life program to accept embedded comments too.)

BCNU,
-dbell

dbell
Posts: 302
Joined: June 27th, 2013, 12:47 am
Contact:

Re: There seems to be invalid patterns on the site

Post by dbell » March 30th, 2023, 2:43 am

Following up on the previous post:

Some of the patterns on the web site have embedded comment lines, which the viewer is happy to accept.

However, the 'Run Length Encoded` wiki article does not indicate that this is allowed - It only mentions comment lines before the header line.

If embedded comment lines are really allowed, then the wiki article needs updating to say so.

BCNU,
-dbell

User avatar
rowett
Moderator
Posts: 4019
Joined: January 31st, 2013, 2:34 am
Location: UK
Contact:

Re: There seems to be invalid patterns on the site

Post by rowett » March 30th, 2023, 5:20 am

dbell wrote:
March 30th, 2023, 2:43 am
Following up on the previous post:

Some of the patterns on the web site have embedded comment lines, which the viewer is happy to accept.

However, the 'Run Length Encoded` wiki article does not indicate that this is allowed - It only mentions comment lines before the header line.

If embedded comment lines are really allowed, then the wiki article needs updating to say so.

BCNU,
-dbell
My post was misleading: LifeViewer isn't happy with comments in the RLE body.

LifeViewer is happy with comments before or after the header line but before the RLE body. Additionally anything after the ! is treated as a comment (without the need for the #C prefix).

Golly is happy with comments anywhere: Extended RLE Format.

I may well align LifeViewer's support with Golly's to ensure better interoperability of patterns between them. Having said that I'm not sure I've ever seen a pattern with comments embedded in the RLE.

So I think the wiki article is correct because it's referencing the original Run Length Encoding format and not extensions to it.

User avatar
dvgrn
Moderator
Posts: 11047
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: There seems to be invalid patterns on the site

Post by dvgrn » March 30th, 2023, 7:19 am

The pattern posted in dbell's message is pretty definitely the only example I've ever noticed of comment lines coming after the header line but before the RLE body. I've never ever seen comments in the middle of the RLE body, that I can recall -- not sure what use that would be, since Golly doesn't display them, and RLE is not, as a rule (heh), naturally chunked up at the line level into commentable pieces. If people want to comment their RLE, they're going to use in-pattern labels and/or LifeViewer labels.

I also am not aware of any scripts being run or edits being done in any systematic way, to change the locations of comments like that. The number of people who could conceivably do this is very low -- only Nathaniel has the necessary access, I think, and he would have zero motivation to do such a thing.

Are there any other examples of this? If not, I think Occam's Razor would suggest that this was a one-time editing error on dbell's part, in putting the message together.

dbell
Posts: 302
Joined: June 27th, 2013, 12:47 am
Contact:

Re: There seems to be invalid patterns on the site

Post by dbell » March 30th, 2023, 8:11 am

> If not, I think Occam's Razor would suggest that this was a one-time editing error on dbell's part, in putting the message together.

This is probably true, but I can't find anything in my files about it. I probably pasted the pattern in by hand and then added the comments. Since the viewer showed it fine, I didn't notice that it was wrong.

Sorry about the hassle.

Anyway, I am happy that we agree that the RLE format shouldn't have embedded comments.

BCNU,
-dbell

hkoenig
Posts: 271
Joined: June 20th, 2009, 11:40 am

Re: There seems to be invalid patterns on the site

Post by hkoenig » April 1st, 2023, 11:38 pm

Posted today were several patterns that appear to have been attempts at jokes. Those need to be removed, as they will clutter up future searches.

hotdogPi
Posts: 1774
Joined: August 12th, 2020, 8:22 pm

Re: There seems to be invalid patterns on the site

Post by hotdogPi » April 2nd, 2023, 7:13 am

hkoenig wrote:
April 1st, 2023, 11:38 pm
Posted today were several patterns that appear to have been attempts at jokes. Those need to be removed, as they will clutter up future searches.
For my "9 oscillator discoveries" post, I made an edit at the beginning of April 2 saying which two of the nine were real. However, the other seven are still valid partials.
User:HotdogPi/My discoveries

Periods discovered: 5-16,⑱,⑳G,㉑G,㉒㉔㉕,㉗-㉛,㉜SG,㉞㉟㊱㊳㊵㊷㊹㊺㊽㊿,54G,55G,56,57G,60,62-66,68,70,73,74S,75,76S,80,84,88,90,96
100,02S,06,08,10,12,14G,16,17G,20,26G,28,38,44,47,48,54,56,72,74,80,92,96S
217,300,486,576

S: SKOP
G: gun

Post Reply