Page 15 of 16

Re: apgsearch v1.0

Posted: May 13th, 2016, 5:38 am
by calcyman
muzik wrote:Looks like my apgsearch glitched out and submitted a mutant Siamese twin haul:

Image


I mean, not that that's a major problem or anything, just kinda interesting
It does that whenever the Internet connection is temporarily down.

Re: apgsearch v1.0

Posted: May 13th, 2016, 6:34 am
by muzik
Makes sense. My internet connection is unpredictable overnight

Re: apgsearch v1.0

Posted: May 14th, 2016, 5:14 am
by muzik

Re: apgsearch v1.0

Posted: May 14th, 2016, 7:11 am
by Apple Bottom
muzik wrote:Looks like apgcodes for Seeds patterns actually work:

https://catagolue.appspot.com/object/xq1_69/b2

https://catagolue.appspot.com/object/xp2_12/b2
Of course -- why wouldn't they? The site only verifies that the code actually describes the object named.

Fun fact 1: the site also accepts what you might call "secondary" codes: ones that result from different readings of the same object, but which are not the canonical code on account of being too long, or (if not that) lexicographically sorting after it. For instance, xs6_2552 appears as a perfectly ordinary still life. This is a good reason to always use a script to figure out an object's apgcode. ;)

Fun fact 2: xp2_12, contrary to what the comment on its page says, has been seen -- it's appeared in a number of semitotalistic rules, e.g. b2-a4s23. The moon, on the hand, has not, and in fact to my knowledge there's no photons known to Catagolue at all. Are there any rules which have photons but do not explode?

Re: apgsearch v1.0

Posted: May 14th, 2016, 8:08 am
by muzik
Apple Bottom wrote:Are there any rules which have photons but do not explode?
Indeed: viewtopic.php?f=11&t=803&start=85

It has some pretty crazy rakes and breeders, though. (maybe it could answer my question on the MMMM breeder in the questions thread?)

Re: apgsearch v1.0

Posted: May 14th, 2016, 8:23 am
by Apple Bottom
muzik wrote:
Apple Bottom wrote:Are there any rules which have photons but do not explode?
Indeed: viewtopic.php?f=11&t=803&start=85
I should've been more specific, I reckon. ;) Are there any semi-totalistic or outer-totalistic rules which have photons but do not explode, i.e. rules that are supported by Catagolue and that are (in principle at the very least) soup-searchable with Calcyman's apg* tools, and/or A for Awesome's hacked version of apgsearch?

Put more succinctly, is there a chance we'll see a photon on Catagolue anytime soon?

Re: apgsearch v1.0

Posted: May 14th, 2016, 10:21 am
by A for awesome
Apple Bottom wrote:Put more succinctly, is there a chance we'll see a photon on Catagolue anytime soon?
Probably not too soon. Currently, my hacked version (and indeed any of the python versions) will break upon encountering any p1 spaceship, due to a small detail in the implementation of the bijoscar() function. Unless someone finds a B2 outer-totalistic rule that doesn't explode (which doesn't seem likely; the best I've been able to find are rules that sometimes only grow linearly (in the form of wickstretchers), rather than quadratically) and is searchable by apgmera, it'll probably be a while.

Re: apgsearch v1.0

Posted: May 14th, 2016, 12:31 pm
by muzik
A for awesome wrote: Currently, my hacked version (and indeed any of the python versions) will break upon encountering any p1 spaceship, due to a small detail in the implementation of the bijoscar() function.
Can you not fix that?

Re: apgsearch v1.0

Posted: May 14th, 2016, 4:01 pm
by A for awesome
muzik wrote:
A for awesome wrote: Currently, my hacked version (and indeed any of the python versions) will break upon encountering any p1 spaceship, due to a small detail in the implementation of the bijoscar() function.
Can you not fix that?
I probably can, but I have a lot of other bugs and things I have to fix before I release a new version, so it's going to take a while. I also don't entirely know whether the apgsearch algorithm would even work with rules like cb2, given all of the natural puffers, rakes, guns, and breeders present.

Re: apgsearch v1.0

Posted: May 14th, 2016, 5:16 pm
by muzik
b3s12/C1 hasn't updated in hours and there's a bunch of hauls lined up


EDIT: seems many other rules are doing this

EDIT2: including normal life....the fuuuuu...?

Re: apgsearch v1.0

Posted: July 8th, 2016, 12:46 am
by drc
It's having a hard time searching B2e3/S23-j, an extremely stable rule. What's going on (and I said heyyeyaaeyaaaeyaeyaa)

Re: apgsearch v1.0

Posted: July 8th, 2016, 1:35 am
by wildmyron
drc wrote:It's having a hard time searching B2e3/S23-j, an extremely stable rule. What's going on (and I said heyyeyaaeyaaaeyaeyaa)
I think you meant to post this on the Hacking apgsearch thread, but in any case something is going wrong with the rule parsing. When I tried a search with apgsearch-2016-2-06-v0.54+0.21i-update.py it was actually trying to search B23/S23 - doomed to failure. Searching works fine when I used my version of non-totalistic apgsearch (apgsearch-isotropic-v0.2.py) The search runs as expected. (No upload capability though.) I have no idea what's going wrong there.

Btw, that's a nice rule, lots of variety in the natural spaceships.

Re: apgsearch v1.0

Posted: July 8th, 2016, 12:59 pm
by A for awesome
wildmyron wrote:
drc wrote:It's having a hard time searching B2e3/S23-j, an extremely stable rule. What's going on (and I said heyyeyaaeyaaaeyaeyaa)
I think you meant to post this on the Hacking apgsearch thread, but in any case something is going wrong with the rule parsing. When I tried a search with apgsearch-2016-2-06-v0.54+0.21i-update.py it was actually trying to search B23/S23 - doomed to failure. Searching works fine when I used my version of non-totalistic apgsearch (apgsearch-isotropic-v0.2.py) The search runs as expected. (No upload capability though.) I have no idea what's going wrong there.

Btw, that's a nice rule, lots of variety in the natural spaceships.
Did either of you generate the rule table before searching the rule? This is a drawback of 0.21i: it does not autogenerate the rule table for you; you have to use isotropic-rulegen.py to make the rule and enter it using an underscore, not a slash. I'll try to improve this when I get back from vacation in three weeks, or sooner if I can.

Re: apgsearch v1.0

Posted: July 8th, 2016, 1:08 pm
by drc
A for awesome wrote:Did either of you generate the rule table before searching the rule? This is a drawback of 0.21i: it does not autogenerate the rule table for you; you have to use isotropic-rulegen.py to make the rule and enter it using an underscore, not a slash. I'll try to improve this when I get back from vacation in three weeks, or sooner if I can.
I did, but it still searches like 3 soups a second

Re: apgsearch v1.0

Posted: August 29th, 2016, 7:49 pm
by Moth-Wingthane
muzik wrote:
Kazyan wrote:Add apgsearch.py to Golly's Scripts folder, then double-click the file from within Golly, instead of your file explorer.
When I do this, and click on it, it just gives a dialog box "Could not load the Python library" and asks me to type in the file adress for python27.dll and even when I do, the same box just comes up again.

Tried both with python27.dll at the end and not.


Where exactly should the file be located? When I searched for it, I found it in system32, but I'm not sure if that's the correct file. Is there a way to generate one?
This issue is exactly what I have with trying to get v1 running on Raspbian on a RPi. Everything is 32bit, so no issue there but I just can't see what needs to be entered here to make it happy.

Re: apgsearch v1.0

Posted: August 30th, 2016, 12:56 pm
by _zM
Moth-Wingthane wrote:
muzik wrote:
Kazyan wrote:Add apgsearch.py to Golly's Scripts folder, then double-click the file from within Golly, instead of your file explorer.
When I do this, and click on it, it just gives a dialog box "Could not load the Python library" and asks me to type in the file adress for python27.dll and even when I do, the same box just comes up again.

Tried both with python27.dll at the end and not.


Where exactly should the file be located? When I searched for it, I found it in system32, but I'm not sure if that's the correct file. Is there a way to generate one?
This issue is exactly what I have with trying to get v1 running on Raspbian on a RPi. Everything is 32bit, so no issue there but I just can't see what needs to be entered here to make it happy.
On my (Ubuntu) laptop I had the same problem - turns out that sometimeswhen Python updated the file name changes. The old version was(as expected) libpython2.7.so (which didn't work), while the new file was named libpython2.7.so.1. Maybe you have a similar problem

Re: apgsearch v1.0

Posted: August 31st, 2016, 6:48 am
by Rhombic
I'm extremely sorry if this question is trivial, but I have found no information after searching (site:conwaylife.com).

How do you (or is it possible to) set apgsearch v1.0 to non-totalistic rules?

Re: apgsearch v1.0

Posted: August 31st, 2016, 6:49 am
by _zM
Use one of the hacked versions.

Re: apgsearch v1.0

Posted: August 31st, 2016, 7:26 am
by Rhombic
_zM wrote:Use one of the hacked versio0ns.
Sorry... what do you mean?

Re: apgsearch v1.0

Posted: August 31st, 2016, 7:29 am
by _zM

Re: apgsearch v1.0

Posted: August 31st, 2016, 10:51 am
by Rhombic
When I search for B37/S2-i34q, it just goes on with B37/S234 :(

Re: apgsearch v1.0

Posted: August 31st, 2016, 11:40 am
by Apple Bottom
Rhombic wrote:When I search for B37/S2-i34q, it just goes on with B37/S234 :(
Non-isotropic rules aren't supported directly; you have to create a rules file first, e.g. using the isotropic-rule.py script (which I think is also floating around on the forums here somewhere). Running that script and entering "B37/S2-i34q" will create a file called B37_S2-i34q.rule; when then asked by apgsearch (Aidan's version) what rule to search, enter "B37_S2-i34q".

Re: apgsearch v1.0

Posted: August 31st, 2016, 11:46 am
by dvgrn
Apple Bottom wrote:Non-isotropic rules aren't supported directly; you have to create a rules file first, e.g. using the isotropic-rule.py script (which I think is also floating around on the forums here somewhere).
The isotropic-rule-gen.py script is hiding on the Rule Request thread.

Re: apgsearch v1.0

Posted: August 31st, 2016, 11:50 am
by Rhombic
Thank you very much everyone, your help is greatly appreciated!
A bit disappointed at the slower searching speeds compared to totalistic rules (due to it being RuleTable) but better than having nothing at all.

- Rhombic

Re: apgsearch v1.0

Posted: September 1st, 2016, 6:00 am
by Rhombic
I am apgsearching B2-a3i4aijk/S2a3-i4 and the census counts xp2_5 and xp2_7 as different objects, when they are actually the two phases of the same oscillator.