Script Registry -- for script authors; track your script usage
- Thread starter zarqon
- Start date
bumcheekcity
Active member
That worked perfectly. Perhaps you should say that on the error, but not to sound ungrateful. Thanks, the script's great!
bumcheekcity
Active member
Zarqon, what do you do about people who re-name your scripts before they use them? It's pretty obvious they're users of the script, but they named it hccheck-cc.ash or whatever, so it won't work with the script.
Bale
Minion
I can tell you what I do when that happens. I add a line to register_user() that looks like this:
if(scriptname == "chamoix.ash") scriptname = "chamois.ash";
You should be able to figure out exactly where in the function to insert it.
Then I just change the specifics based on the kmail.
if(scriptname == "chamoix.ash") scriptname = "chamois.ash";
You should be able to figure out exactly where in the function to insert it.
Then I just change the specifics based on the kmail.
zarqon
Well-known member
I originally planned a kind of "aliasing" idea, where you could register aliases of scripts by typing [register secondaryscriptname alias of primaryscriptname]. This is supported in the ASH script, but is still not supported server-side.
For now, Bale's solution ought to do you nicely. Personally, I usually just delete those notifications. But it is becoming common knowledge that I am not as OCD as he is.
For now, Bale's solution ought to do you nicely. Personally, I usually just delete those notifications. But it is becoming common knowledge that I am not as OCD as he is.
Bale
Minion
LoL! Yeah, I like certain things to be done precisely. If someone sends me a notification for one of my scripts I want it sorted properly even if they change the name.
Anyway, I'd say that anyone who has a valid need for this script should be able to figure out such a minor detail as how to use that information I provided in the previous post so it isn't important for you to update the script for such eventualities. This is a script for script writers after all.
Anyway, I'd say that anyone who has a valid need for this script should be able to figure out such a minor detail as how to use that information I provided in the previous post so it isn't important for you to update the script for such eventualities. This is a script for script writers after all.
mredge73
Member
hello Zarqon
I have a low priority request:
Can this be updated so an author can remove scripts no longer supported?
Personally, I have changed the name of 3 of my scripts that I have registered and would like to be able to remove them since I no longer support them by that name.
Again, this is low-priority. They are not hurting anything by being up there, just making a mess.
I have a low priority request:
- [unregister scriptname] - Use this command to delete a script you have registered from the registry. At present, scripts cannot be unregistered if they have more than one registered user. Example: [unregister myscript.ash]
Can this be updated so an author can remove scripts no longer supported?
Personally, I have changed the name of 3 of my scripts that I have registered and would like to be able to remove them since I no longer support them by that name.
Again, this is low-priority. They are not hurting anything by being up there, just making a mess.
Bale
Minion
What might be better and more useful is: [changename oldscriptname newscriptname]
which would merge the user lists of the two scripts into newscriptname while removing the entry for oldscriptname. I think that is really what you want. It would tidy up your list of scripts without losing any relevant information.
which would merge the user lists of the two scripts into newscriptname while removing the entry for oldscriptname. I think that is really what you want. It would tidy up your list of scripts without losing any relevant information.
zarqon
Well-known member
If you want to change the name of a script, just kmail/PM me, and I'll edit the data file directly. Then all your previously registered users will persist. In your case, if you've already added the renamed scripts to the registry, that presents more of a difficulty. Let me know what you'd like to do.
PS -- yay for the recent "script" directive, eh? I should probably include this in the Wiki page I made for notify.
PS -- yay for the recent "script" directive, eh? I should probably include this in the Wiki page I made for notify.
zarqon
Well-known member
Certainly!
But don't thank me too much -- I made it first for myself.
If I'm away from my computer for a few days, or if mafia makes some change that appears to clear a lot of people's "used scripts" preference, I'm very glad for this script. I've had upwards of 50 notify messages get processed in one batch.
But don't thank me too much -- I made it first for myself.
If I'm away from my computer for a few days, or if mafia makes some change that appears to clear a lot of people's "used scripts" preference, I'm very glad for this script. I've had upwards of 50 notify messages get processed in one batch.
bumcheekcity
Active member
Btw zarqon, I'm getting lots of "Unexpected error, debug log printed." messages when there are messages to be processed. I think it's due to trying to store things that aren't integers in the integer type:
That does look what it's doing. Unfortunately, the stack trace won't help him find out which line in his script is buggy. I'll look into trapping such errors in to_int() and giving a more helpful message, including file and line of code.
Edit: we used to do exactly that, but the new code in StringUtilities.parseInt pre-empted it and forced a stack trace to be printed. to_int() now calls a version which intentionally throws an exception if it can't parse, traps the errors, and gives a usual ASH runtime error.
Last edited:
Yes. This has been heavily discussed all over these forums. You can no longer pass in non-integers and get KoLmafia to, miraculously, give you an integer.
Used to be, as in this script, if you gave it "sel1234", it would return 1234. I suppose that might make a little bit of sense. But, you could also give it "1 for me and 2 for you" and get back 12.
Bottom line: if you give a non-integer to to_int(), it is now considered to be an error.
zarqon
Well-known member
Fixed this and the other problem: print(integer + "string") now evaluates the whole thing as an integer for some reason. Changing it to print(to_string(integer) + "string") fixes that, which I had in quite a few of my scripts.
I'm still a bit sad about the apparent regression in to_int()'s functionality, but I've had some time over the holidays so I was able to go through most of my scripts that stopped working and "fix" them. Sigh, I miss the old functionality though.
I'm still a bit sad about the apparent regression in to_int()'s functionality, but I've had some time over the holidays so I was able to go through most of my scripts that stopped working and "fix" them. Sigh, I miss the old functionality though.
It was a bug. Fixed in revision 7941.