Traceback (most recent call last):
File "C:\Users\Justin Tang\Desktop\apgsearch\apgsearch-2014-09-08.py", line 27, in <module>
import golly as g
ImportError: No module named 'golly'
Don't run it from IDLE or a Python command line -- use File > Run Script in Golly.Hdjensofjfnen wrote:When I run it, it turns up an error:
Traceback (most recent call last):
File "C:\Users\Justin Tang\Desktop\apgsearch\apgsearch-2014-09-08.py", line 27, in <module>
import golly as g
ImportError: No module named 'golly'
The Python library does not have this symbol: Py_InitModule4
Well, we'd better start from the beginning then: what version of Golly? 32-bit or 64-bit? What version of Python -- is it 2.7.x? 32-bit or 64-bit? Make sure Golly and Python match. If you download the default version of Golly from SourceForge, I believe it's still giving you the 32-bit version.Hdjensofjfnen wrote:Now it turns up this error:The Python library does not have this symbol: Py_InitModule4
Won't work. Python 2.x is what's compatible with Golly. The Golly Gang hasn't made the jump to Python 3.x yet (and we're still in good company -- might take another decade to get everyone converted over, at this rate...)Hdjensofjfnen wrote:* Python 3.4 32 bit
* Golly 1.7 32 bit
I typed in "python34" for the system file query
I've mentioned this in other threads before, and you need to first run isotropic-rulegen.py to allow the RuleLoader algorithm to run the rule, and then enter the rule name with an underscore instead of a slash when you run the actual search. As I have said, I hope to fix this soon.Rhombic wrote:On a similar note, I'm having trouble with non-totalistic rules, even with the hacked version of apgsearch. It deletes every neighbourhood specification so tLife becomes B3/S234 and so on. Quite unnerving.
Which other threads? I've tried to search and the forum's search tells me "isotropic-rulegen" is too common a term to search on. Okay then.A for awesome wrote: I've mentioned this in other threads before, and you need to first run isotropic-rulegen.py to allow the RuleLoader algorithm to run the rule, and then enter the rule name with an underscore instead of a slash when you run the actual search. As I have said, I hope to fix this soon.
This entire thread is about the Python versions (vv. 0.x and 1.x) for running directly inside Golly.Rich Holmes wrote:Which other threads? I've tried to search and the forum's search tells me "isotropic-rulegen" is too common a term to search on. Okay then.A for awesome wrote: I've mentioned this in other threads before, and you need to first run isotropic-rulegen.py to allow the RuleLoader algorithm to run the rule, and then enter the rule name with an underscore instead of a slash when you run the actual search. As I have said, I hope to fix this soon.
Basically if there's an easy to follow recipe (with a little more clear detail than the above) for doing non totalistic rules in apgmera, I'd like to know where to find it.
Yes, the forum's search algorithm specializes in being annoying -- won't pay attention to quotes around phrases, ignores words that it shouldn't ignore, mysteriously doesn't find things sometimes, you name it. I usually give up after a few tries and just use Google, but sometimes (as in this case) that doesn't work either.Rich Holmes wrote:I've tried to search and the forum's search tells me "isotropic-rulegen" is too common a term to search on. Okay then.
There are no such easy recipes, at least for that specific problem as you've stated it.Rich Holmes wrote:Basically if there's an easy to follow recipe (with a little more clear detail than the above) for doing non totalistic rules in apgmera, I'd like to know where to find it.
(Offtopic warning) I just found google custom search and it looks nice.dvgrn wrote:Yes, the forum's search algorithm specializes in being annoying -- won't pay attention to quotes around phrases, ignores words that it shouldn't ignore, mysteriously doesn't find things sometimes, you name it. I usually give up after a few tries and just use Google, but sometimes (as in this case) that doesn't work either.Rich Holmes wrote:I've tried to search and the forum's search tells me "isotropic-rulegen" is too common a term to search on. Okay then.
Code: Select all
$ git clone https://gitlab.com/apgoucher/apgmera.git
-bash: git: command not found
If your Cygwin is telling you "command not found", you may just not have installed the git plugin. Run the Cygwin installer again and double-check that you asked for the right packages:ygh wrote:Mine won't even download using Cygwin.
Code: Select all
$ git clone https://gitlab.com/apgoucher/apgmera.git -bash: git: command not found
In the apgmera install instructions, Adam wrote:Install Cygwin64 (from http://cygwin.com), ensuring that 'gcc-g++', 'git' and 'make' are checked in the list of plugins to install.
Code: Select all
Me@DESKTOP-HD6VFF2 ~
$ git clone https://gitlab.com/apgoucher/apgmera.git
Cloning into 'apgmera'...
remote: Counting objects: 271, done.
remote: Compressing objects: 100% (135/135), done.
remote: Total 271 (delta 167), reused 227 (delta 136)
Receiving objects: 100% (271/271), 388.55 KiB | 0 bytes/s, done.
Resolving deltas: 100% (167/167), done.
Checking connectivity... done.
Me@DESKTOP-HD6VFF2 ~
$ bash recompile.sh
bash: recompile.sh: No such file or directory
Did you go into the apgmera directory before entering the bash command? recompile.sh should be in the top level of the apgmera directory. If it's not, git failed you badly but it looks like it did what it should.ygh wrote:Hmm. Yeah, guess that would help.
Edit: Wait, still not working.I downloaded literally every possible package, by the way. Took like 10 minutes. So I ruled that possible point of failure out.Code: Select all
Me@DESKTOP-HD6VFF2 ~ $ git clone https://gitlab.com/apgoucher/apgmera.git Cloning into 'apgmera'... remote: Counting objects: 271, done. remote: Compressing objects: 100% (135/135), done. remote: Total 271 (delta 167), reused 227 (delta 136) Receiving objects: 100% (271/271), 388.55 KiB | 0 bytes/s, done. Resolving deltas: 100% (167/167), done. Checking connectivity... done. Me@DESKTOP-HD6VFF2 ~ $ bash recompile.sh bash: recompile.sh: No such file or directory
Commonly called a "folder". In this case, it's the place where recompile.sh and your other apgmera files are stored.ygh wrote:What's a directory?
Code: Select all
cd "FILE_PATH"Code: Select all
cd "C:\Users\Matthias\Documents\Life"Code: Select all
Me@DESKTOP-HD6VFF2 /cygdrive/c/Users/Me/Documents/Life/apgmera
$ bash recompile.sh
Skipping updates; use --update to update apgmera automatically.
Rule unspecified; assuming b3s23.
Symmetry unspecified; assuming C1.
Configuring rule b3s23; symmetry C1
Valid rulestring: b3s23
Valid symmetry: C1
Rule integer: 6152
Rule circuit: [-131-124-450-014-672]
Rule integer: 6152
Rule circuit: [-131-124-450-014-672]
Rule integer: 6152
Rule circuit: [-131-124-450-014-672]
Success!
recompile.sh: line 66: make: command not found
Me@DESKTOP-HD6VFF2 /cygdrive/c/Users/Me/Documents/Life/apgmera
$ ./apgmera -n 10000000
-bash: ./apgmera: No such file or directory
Code: Select all
Cygwin Package Information
Package Version Status
make 4.2.1-1 OK
That rule is submitted as http://catagolue.appspot.com/census/b37-cs23/C1Rich Holmes wrote:Is there some trick to getting Catagolue to post non totalistic results?
I've been running B37e_S23 and it seems to work fine. But nothing on Catalogue.
http://catagolue.appspot.com/census/b23es23/C1
Ah, thanks. Never thought to look for that rule name.wildmyron wrote:That rule is submitted as http://catagolue.appspot.com/census/b37-cs23/C1Rich Holmes wrote:Is there some trick to getting Catagolue to post non totalistic results?
I've been running B37e_S23 and it seems to work fine. But nothing on Catalogue.
http://catagolue.appspot.com/census/b23es23/C1
That's a quirk of the way rule names are canonicalised with the isotropic version of apgsearch.