Hello everyone,
The jmmmp library of abstractions has been updated to version 0.60.
Added [jp.coll], an abstraction to store/edit message collections VERY SIMILAR to cyclone/coll, but in vanilla.
You can download it soon through deken, or in http://bit.ly/jmmmp_060 https://bit.ly/jmmmp_060?fbclid=IwAR1AmnRc3wBfBjjdD0RjUtxIMFOJmMLrhfkpoAjXzJE_CyBg8zjywoI2y1s
Best,
João Pais
Pd-announce mailing list Pd-announce@lists.iem.at https://lists.puredata.info/listinfo/pd-announce
Em sáb., 29 de ago. de 2020 às 20:10, João Pais jmmmpais@gmail.com escreveu:
Added [jp.coll], an abstraction to store/edit message collections VERY SIMILAR to cyclone/coll, but in vanilla.
sorry for nitpicking, but I checked the abstraction and it is not really "vanilla" as it takes some externals (from your library I suppose), not sure if you really mean to make it 100% vanilla, but I thought I'd bring it up anyway.
cheers
actually, just one external (f+), but used multiple times
Em qui., 3 de set. de 2020 às 19:09, Alexandre Torres Porres < porres@gmail.com> escreveu:
Em sáb., 29 de ago. de 2020 às 20:10, João Pais jmmmpais@gmail.com escreveu:
Added [jp.coll], an abstraction to store/edit message collections VERY SIMILAR to cyclone/coll, but in vanilla.
sorry for nitpicking, but I checked the abstraction and it is not really "vanilla" as it takes some externals (from your library I suppose), not sure if you really mean to make it 100% vanilla, but I thought I'd bring it up anyway.
cheers
João, you say it can't "embed" and save contents with the patch, but [text] allows you to do so with the "-k" flag, why don't you use it?
Em qui., 3 de set. de 2020 às 19:18, Alexandre Torres Porres < porres@gmail.com> escreveu:
Em qui., 3 de set. de 2020 às 19:11, Alexandre Torres Porres < porres@gmail.com> escreveu:
actually, just one external (f+), but used multiple times
Oh, I see that's also a vanilla abstraction :) sorry
You also have not implemented assoc and deassoc, I'm thinking about that.
coll saves the text separating with a comma the index and the actual message. You can also do that with [text] and if you do so you could potentially use it to define and detect an address, and have something else associated with it - I hope you know what I mean and that I'm not too confused.
Em qui., 3 de set. de 2020 às 19:31, Alexandre Torres Porres < porres@gmail.com> escreveu:
João, you say it can't "embed" and save contents with the patch, but [text] allows you to do so with the "-k" flag, why don't you use it?
Em qui., 3 de set. de 2020 às 19:18, Alexandre Torres Porres < porres@gmail.com> escreveu:
Em qui., 3 de set. de 2020 às 19:11, Alexandre Torres Porres < porres@gmail.com> escreveu:
actually, just one external (f+), but used multiple times
Oh, I see that's also a vanilla abstraction :) sorry
João, you say it can't "embed" and save contents with the patch, but [text] allows you to do so with the "-k" flag, why don't you use it?
[text define] is the base for [jp.coll], and it has to be created dynamically at startup. also, apparently the contents are saved with the pd file in a line such as "#A set this is a message ; this is another 1 ... ;;" From what I tested this is saved with the patch where [text define] resides, which in this case it's the abstraction and not in the main patch. This would need to be changed in Pd's workings so that -k contents are saved on the main patch file.
You also have not implemented assoc and deassoc, I'm thinking about that.
coll saves the text separating with a comma the index and the actual message. You can also do that with [text] and if you do so you could potentially use it to define and detect an address, and have something else associated with it - I hope you know what I mean and that I'm not too confused.
this is the main reason why I didn't name it [jmmmp/coll] but [jmmmp/jp.coll]. Adding commas to a [text] messes up the number of lines - can be seen already in the help of [text get]. Now I remembered I could put the commas back in, and do an internal conversion for the line number. If this works, it would allow for assoc/deassoc, and make both coll versions practically compatible (except for missing features).
Btw all (?) jmmmp objects are dependent on other jmmmp objects - mostly [f+]. Some of them have to rely on other libraries as well, but usually it's better to have the whole jmmmp together.
Joao
Em qui., 3 de set. de 2020 às 19:31, Alexandre Torres Porres < porres@gmail.com> escreveu:
João, you say it can't "embed" and save contents with the patch, but [text] allows you to do so with the "-k" flag, why don't you use it?
Em qui., 3 de set. de 2020 às 19:18, Alexandre Torres Porres < porres@gmail.com> escreveu:
Em qui., 3 de set. de 2020 às 19:11, Alexandre Torres Porres < porres@gmail.com> escreveu:
actually, just one external (f+), but used multiple times
Oh, I see that's also a vanilla abstraction :) sorry
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list
Em sáb., 5 de set. de 2020 às 09:04, João Pais jmmmpais@gmail.com escreveu:
João, you say it can't "embed" and save contents with the patch, but [text] allows you to do so with the "-k" flag, why don't you use it?
[text define] is the base for [jp.coll], and it has to be created dynamically at startup. also, apparently the contents are saved with the pd file in a line such as "#A set this is a message ; this is another 1 ... ;;" From what I tested this is saved with the patch where [text define] resides, which in this case it's the abstraction and not in the main patch. This would need to be changed in Pd's workings so that -k contents are saved on the main patch file.
hmmm, I see. Then, what I'm thinking of is the [savestate] object, which is used for that. It should work very easily, right?
You also have not implemented assoc and deassoc, I'm thinking about that.
coll saves the text separating with a comma the index and the actual message. You can also do that with [text] and if you do so you could potentially use it to define and detect an address, and have something else associated with it - I hope you know what I mean and that I'm not too confused.
this is the main reason why I didn't name it [jmmmp/coll] but [jmmmp/jp.coll]. Adding commas to a [text] messes up the number of lines - can be seen already in the help of [text get]. Now I remembered I could put the commas back in, and do an internal conversion for the line number.
[text] will also tell you if the line ends with ";" or with "," - so you could use this to differentiate if it's an address or the data.
Btw all (?) jmmmp objects are dependent on other jmmmp objects - mostly
[f+]. Some of them have to rely on other libraries as well, but usually it's better to have the whole jmmmp together.
The thing about offering a vanilla patch is that it can be used in stuff like libpd, camomile, etc... so someone may not want or can't have all of the library. Anyway, [f+] is very simple, and it can be easily replaced with [f] & [+] to become 100% full vanilla and dependency free (an external vanilla abstraction is still an external dependency).
cheers