dan m

I am using pattstorage to save the states and content of objects. However, in my patch the objects are created and named with the script new object message to thispatcher. While pattstorage can see these new objects in the client window, it doesnt register any value changes (or anything at all) for these objects in the client window or in the storage window.

I have noticed that if I close the patch, save it and open it again , then pattstorage sees and registers the behaviour of the previously 'invisible' objects...but this is not helpfull in my case really.

I include an attachment as an example (not sure if it's there or not!).I also include as text a tiny abstraction needed in the patch.

I am using pattstorage to save the states and content of objects. However, in my patch the objects are created and named with the script new object message to thispatcher. While pattstorage can see these new objects in the client window, it doesnt register any value changes (or anything at all) for these objects in the client window or in the storage window.
I have noticed that if I close the patch, save it and open it again , then pattstorage sees and registers the behaviour of the previously 'invisible' objects...but this is not helpfull in my case really.
I include an attachment as an example (not sure if it's there or not!).I also include as text a tiny abstraction needed in the patch.
Any help would be great
Thanks

pattrstorage-doesnt-see-scripted-objects

you have to "add" the autopattr object after the creation of new objects!

every time you need to add or remove objects!

actually, there is a difficulty: if I use that method in a patch, it works fine (although I have to refresh the storage window by other means after re-creating the autopattr object). However if I open that patch as an abstraction within another patch, as soon as I create a new object, the client list disappears and stays blank from then on...(?)

actually i am pretty sure, that exactly this was fixed by jb in some 

anyways the patch below works for me; meaning i can script-create a 

and its value will be seen from pattrstorage.

actually i am pretty sure, that exactly this was fixed by jb in some  
update.
anyways the patch below works for me; meaning i can script-create a  
number
and its value will be seen from pattrstorage.
best
klaus

perhaps it's the update. I am using 4.5.5

what's wrong with this one then?
perhaps it's the update. I am using 4.5.5

It's the update. Update to v. 4.5.7 and this patch will work fine.

got the update and now pattrstorage sees all the changes.

Now I am trying to save and recall data from a xml file. I have managed to save the xml, but I am having problems recalling the xml and pattrstorage presets. THe content of the xml's files is correct (shows the data acurately), but I cannot make it appear in the patch. surely I am missing something...

Please, check the patch below. It is a simplified version of the patch I am working on. I tried to simulate the functioning of the original patch, but I havent automated all the tasks, sorry...

Open the main patch 'pattrfolders' first. The other two patches are abstractions.

Hi,
got the update and now pattrstorage sees all the changes.

Any help would be really appreciated.
thanks
d 

sorry, you may need the name of the abstractions. i forgot to include them:

Main: pattrfolders
first abstraction: abst
second abstraction: abstraction


So here's what my max window shows when the write message is sent (I 

print: write Bellyache:/Users/jeremydb/Desktop/abst/ pah.xml soundbank.xml

So here's what my max window shows when the write message is sent (I 
added some prints here)

print: write Bellyache:/Users/jeremydb/Desktop/abst/ pah.xml soundbank.xml


Sorry - message seems to have gotten cut off...

So here's what my max window shows when the write message is sent (I

So you're using [ sprintf %s %s soundbank.xml ] which is 100% _not_ what 

you want here. However, I'm not really clear what it is that you do 

want, since you seem to want to give the xml file 2 different names, or 

save it in some special place or something, but it's not clear from the 

However, I'm guessing that you would probably like [ sprintf %s%s ]

So here's what my max window shows when the write message is sent (I
added some prints here)

So you're using [ sprintf %s %s soundbank.xml ] which is 100% _not_ what 
you want here. However, I'm not really clear what it is that you do 
want, since you seem to want to give the xml file 2 different names, or 
save it in some special place or something, but it's not clear from the 
comments or the patch.

I dont know how you get that print message (?)

If I print after the 'pack write xml' object , this is what I get:

pack: write C:/Documents and Settings/DV Customer/Desktop/amen/ mine soundbank.xml

and also an xml file in the amen folder called 'mine soundbank.xml'

where 'mine' is the name given into the text box. 'soundbank' in sprintf is just to make it easier for me to identify the type of xml file , as there are a few being saved from different parts of the (original) patch. But it is not necessary, so I guess I could get rid of it if it was sttoping pattrstorage from working...

However, I think the .xml in the after sprintf %s%s seems necessary to add the file extension at the end of the name, since the xml in the pack object is just providing the second inlet to the pack object , and never sems to be added to the name...

Jeremy,
I dont know how you get that print message (?)

oh,I guess you entered 'pah.xml' in the name box...

perhaps I should explain what I want to do in my patch in general.

As you can see, you can create several drop boxes that are connected to the abstraction 'abst'. In my original patch, those abstractions include a buffer/js.folderiter/coll combination that plays back the files included in the folder that you dropped in the drop area. I would like to save and recall the xml file to be able to recall the folder path into the text box, and then bang it to reload the files into abst's buffer.

It gets more complicated, as the patch we are working in is itself an abstraction (called 'abstraction'). The reason for this is that in my original patch you can script/create as many as these as you want from the top 'pattrfolder' patch. It is here that you name each of the 'abstraction' patches, and where you define the save path for all the xml files. By the way, I do need to save each xml file with two names: first, the name given to the 'abstraction'; second the type of data saved in the xml file. This is because for every 'abstraction', I am saving more than one xml file...

perhaps I should explain what I want to do in my patch in general.
As you can see, you can create several drop boxes that are connected to the abstraction 'abst'. In my original patch, those abstractions include a buffer/js.folderiter/coll combination that plays back the files included in the folder that you dropped in the drop area. I would like to save and recall the xml file to be able to recall the folder path into the text box, and then bang it to reload the files into abst's buffer.
It gets more complicated, as the patch we are working in is itself an abstraction (called 'abstraction'). The reason for this is that in my original patch you can script/create as many as these as you want from the top 'pattrfolder' patch. It is here that you name each of the 'abstraction' patches, and where you define the save path for all the xml files. By the way, I do need to save each xml file with two names: first, the name given to the 'abstraction'; second the type of data saved in the xml file. This is because for every 'abstraction', I am saving more than one xml file...

The read and write messages to pattrstorage take one optional symbol 

argument, which is the name of the .xml file, partially or fully qualified.

are all valid. Note that symbols containing spaces _must_ be enclosed in 

double-quotes. The "symout" argument to the sprintf object will ensure 

that the output of that object is a single symbol (Max Reference Manual, 

p.550). Nevertheless, whitespace does count in file names and paths, 

won't work either, assuming that the file is located at 

The patch I looked at was inserting spaces into the file name and file 

path due to its use of a [sprintf %s %s] or similar, rather than a 

[sprintf symout %s%s]. This was clear from the excerpt I posted previous 

from the Max window, as well as the one you posted.

So, you need to pay some attention to this detail, at which point, 

pattrstorage will work just as it should.

The read and write messages to pattrstorage take one optional symbol 
argument, which is the name of the .xml file, partially or fully qualified.

write Bellyache:/test/afile.xml
write "Bellyache:/a folder/afile.xml"
write afile.xml

are all valid. Note that symbols containing spaces _must_ be enclosed in 
double-quotes. The "symout" argument to the sprintf object will ensure 
that the output of that object is a single symbol (Max Reference Manual, 
  p.550). Nevertheless, whitespace does count in file names and paths, 
and these:

write Bellyache:/test/ afile.xml
write Bellyache:/ a folder / a file.xml
write "Bellyache:/a folder/ a file.xml"

won't work either, assuming that the file is located at 
Bellyache:/test/afile.xml

The patch I looked at was inserting spaces into the file name and file 
path due to its use of a [sprintf %s %s] or similar, rather than a 
[sprintf symout %s%s]. This was clear from the excerpt I posted previous 
from the Max window, as well as the one you posted.

So, you need to pay some attention to this detail, at which point, 
pattrstorage will work just as it should.

Am 23.05.2006 um 23:36 Uhr schrieb dan:
> hope this gives a bit of context


pattrstorage doesnt see scripted objects