some help for writing your http://puredata.org/dev/PdDefinitions ... those are surely not the best wordings, but it adds detail.
comma (atom type): a comma character that is not backslashed, and thus is effective as a message separator in a messagebox.
semicolon (atom type): a semicolon character that is not backslashed, and thus is effective as a message separator in any context. if there was any implicit context (as in a messagebox), semicolon forgets it, and thus you get to write an explicit receiver after every semicolon.
dollar (in general): a nonbackslashed dollar sign followed by one or more digits. in an evaluation, this can be worth a float, symbol or pointer, whatever the corresponding argument in the argument list is.
dollar (atom type): a dollar that is parsed individually; is replaced by the corresponding atom upon evaluation.
dollsym (atom type) (missing): in pd 0.39 and earlier: a dollar (subtitutor) that is at the beginning of an atom, followed by symbol characters. Upon substitution it becomes a symbol, as the corresponding atom gets cast to characters and the symbols are merged to make one resulting symbol.
dollsym (atom type) (missing): in pd 0.40 and earlier: any alternation of dollar (subtitutor) and normal symbol characters. All those dollar substitutors get replaced by corresponding atoms cast to characters and all the parts are merged to make one resulting symbol.
(The backslashing stuff may be important to understand if you expect to modify or generate patches using [textfile] like some people do; in general it's good for understanding the fileformat)
Personally, I called "float" "symbol" "pointer" the stable atoms, whereas I call "comma" "semi" "dollar" "dollsym" the radioactive atoms, because they decay upon being used by messageboxes. the radioactive atoms also only really stay alive inside of messageboxes, so in general in pd you will only see functions/methods for handling stable atoms, not radioactive ones.
I could've put this in the wiki directly, but wanted to make sure it gets noticed (and also I'd like some feedback and to know what you'd change in this). Are there any automatic notifications of changes of wiki pages on puredata.info (as a configurable option, per user) ? I mean by email, just like for sf's bug tracker. It would be good to have this so that all parties involved in modifying pages really know about what's going on. It gives the advantages of mailing-lists to wikis.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801, Montréal QC Canada
Has the backlash been integrated as an escape character in the pd's editing environment? Or is it still only useful in the file description format?
Tom
On 9/29/07, Mathieu Bouchard matju@artengine.ca wrote:
some help for writing your http://puredata.org/dev/PdDefinitions ... those are surely not the best wordings, but it adds detail.
comma (atom type): a comma character that is not backslashed, and thus is effective as a message separator in a messagebox.
semicolon (atom type): a semicolon character that is not backslashed, and thus is effective as a message separator in any context. if there was any implicit context (as in a messagebox), semicolon forgets it, and thus you get to write an explicit receiver after every semicolon.
dollar (in general): a nonbackslashed dollar sign followed by one or more digits. in an evaluation, this can be worth a float, symbol or pointer, whatever the corresponding argument in the argument list is.
dollar (atom type): a dollar that is parsed individually; is replaced by the corresponding atom upon evaluation.
dollsym (atom type) (missing): in pd 0.39 and earlier: a dollar (subtitutor) that is at the beginning of an atom, followed by symbol characters. Upon substitution it becomes a symbol, as the corresponding atom gets cast to characters and the symbols are merged to make one resulting symbol.
dollsym (atom type) (missing): in pd 0.40 and earlier: any alternation of dollar (subtitutor) and normal symbol characters. All those dollar substitutors get replaced by corresponding atoms cast to characters and all the parts are merged to make one resulting symbol.
(The backslashing stuff may be important to understand if you expect to modify or generate patches using [textfile] like some people do; in general it's good for understanding the fileformat)
Personally, I called "float" "symbol" "pointer" the stable atoms, whereas I call "comma" "semi" "dollar" "dollsym" the radioactive atoms, because they decay upon being used by messageboxes. the radioactive atoms also only really stay alive inside of messageboxes, so in general in pd you will only see functions/methods for handling stable atoms, not radioactive ones.
I could've put this in the wiki directly, but wanted to make sure it gets noticed (and also I'd like some feedback and to know what you'd change in this). Are there any automatic notifications of changes of wiki pages on puredata.info (as a configurable option, per user) ? I mean by email, just like for sf's bug tracker. It would be good to have this so that all parties involved in modifying pages really know about what's going on. It gives the advantages of mailing-lists to wikis.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801, Montréal QC Canada _______________________________________________ PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Hallo, Thomas O Fredericks hat gesagt: // Thomas O Fredericks wrote:
Has the backlash been integrated as an escape character in the pd's editing environment?
Not that I know of.
Or is it still only useful in the file description format?
It's also used for example in symbols generated by [openpanel] and can be made with [makefilename]
Frank Barknecht _ ______footils.org_ __goto10.org__
On Sat, 29 Sep 2007, Frank Barknecht wrote:
It's also used for example in symbols generated by [openpanel] and can be made with [makefilename]
Every parsing interprets dollars and removes a layer of backslashes. Therefore when [openpanel] sends backslashes they are immediately removed and they are not actually part of the symbol. If they were, they would've to be sent as double-backslashes.
Btw, the .pd file format could've been made so that it does not evaluate $ at every possible occasion, and if it had been so, dollars wouldn't need to be backslashed by the fileformat. As it is, messageboxes and objectboxes are parsed once but evaluated twice, even though $-args are meaningless in the context of loading a patch: putting a nonbackslashed dollar in a .pd file will just make it disappear or be replaced by a zero. So, if you want to have an actual backslash in a messagebox or objectbox, you'd need to put a triple backslash in the file.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801, Montréal QC Canada
On Sat, 29 Sep 2007, Thomas O Fredericks wrote:
Has the backlash
The "backlash" is only implemented on pd-list ;)
The "backslash" is something else...
been integrated as an escape character in the pd's editing environment? Or is it still only useful in the file description format?
you may find a use for it from using pd, but you will have to either generate it with 92 -> [makesymbol %c] or edit .pd files manually, but in the latter case, pd doesn't resave it properly, so you are better off with [makesymbol]. You will also have to generate dollar (36), comma (44), semicolon (59).
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801, Montréal QC Canada
Go ahead a make a page in the PdDefinitions section and add this
stuff. That's what it's there for.
.hc
On Sep 29, 2007, at 2:54 AM, Mathieu Bouchard wrote:
some help for writing your http://puredata.org/dev/ PdDefinitions ... those are surely not the best wordings, but it
adds detail.comma (atom type): a comma character that is not backslashed, and
thus is effective as a message separator in a messagebox.semicolon (atom type): a semicolon character that is not
backslashed, and thus is effective as a message separator in any
context. if there was any implicit context (as in a messagebox),
semicolon forgets it, and thus you get to write an explicit
receiver after every semicolon.dollar (in general): a nonbackslashed dollar sign followed by one
or more digits. in an evaluation, this can be worth a float, symbol
or pointer, whatever the corresponding argument in the argument
list is.dollar (atom type): a dollar that is parsed individually; is
replaced by the corresponding atom upon evaluation.dollsym (atom type) (missing): in pd 0.39 and earlier: a dollar
(subtitutor) that is at the beginning of an atom, followed by
symbol characters. Upon substitution it becomes a symbol, as the
corresponding atom gets cast to characters and the symbols are
merged to make one resulting symbol.dollsym (atom type) (missing): in pd 0.40 and earlier: any
alternation of dollar (subtitutor) and normal symbol characters.
All those dollar substitutors get replaced by corresponding atoms
cast to characters and all the parts are merged to make one
resulting symbol.(The backslashing stuff may be important to understand if you
expect to modify or generate patches using [textfile] like some
people do; in general it's good for understanding the fileformat)Personally, I called "float" "symbol" "pointer" the stable atoms,
whereas I call "comma" "semi" "dollar" "dollsym" the radioactive
atoms, because they decay upon being used by messageboxes. the
radioactive atoms also only really stay alive inside of
messageboxes, so in general in pd you will only see functions/ methods for handling stable atoms, not radioactive ones.I could've put this in the wiki directly, but wanted to make sure
it gets noticed (and also I'd like some feedback and to know what
you'd change in this). Are there any automatic notifications of
changes of wiki pages on puredata.info (as a configurable option,
per user) ? I mean by email, just like for sf's bug tracker. It
would be good to have this so that all parties involved in
modifying pages really know about what's going on. It gives the
advantages of mailing-lists to wikis._ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801, Montréal QC Canada
All information should be free. - the hacker ethic
On Sun, 30 Sep 2007, Hans-Christoph Steiner wrote:
Go ahead a make a page in the PdDefinitions section and add this stuff. That's what it's there for.
I was asking you about modifying the existing page, not tucking away my contributions in a page subordinate to yours.
If you don't want to address my concerns about wiki collaboration, then I'll leave things as they are now; it'll be less trouble.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801, Montréal QC Canada
On Sep 30, 2007, at 5:59 PM, Mathieu Bouchard wrote:
On Sun, 30 Sep 2007, Hans-Christoph Steiner wrote:
Go ahead a make a page in the PdDefinitions section and add this
stuff. That's what it's there for.I was asking you about modifying the existing page, not tucking
away my contributions in a page subordinate to yours.If you don't want to address my concerns about wiki collaboration,
then I'll leave things as they are now; it'll be less trouble.
There are no proposed definitions on the main page, that's a survey
of existing definitions. Frank and my attempts at definitions are
each on separate pages. I am suggesting you do the same.
.hc
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801, Montréal QC Canada
All information should be free. - the hacker ethic