Re: Passing parameters by reference
On Mar 12, 2012, at 11:18 PM, Mark Wieder wrote: You'd have to somehow prevent the dereferencing of myArray[myKey] in order to pass it by reference. Just as the dereferencing of x would have be be prevented to pass it as reference. Perhaps any mutable place is the same. Dar ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
That is what passing an element to an array amounts to was my point. Bob On Mar 12, 2012, at 5:39 PM, Pete wrote: I'm not sure whose post you're responding to Bob. Where do you see something that amounts to a statement being passed as a referenced parameter? Pete On Mon, Mar 12, 2012 at 3:04 PM, Bob Sneidar b...@twft.com wrote: Just weighing in here, that would be a bit confusing. Passing by reference means that the command or function has access to the variable passed to it. By passing what amounts to a statement, there is nothing for LC to manipulate on the other end. Statements have to have some place to put the results. In this case, there is no place for LC to put the statement when passed by reference. Even a reference to an element of an array is a statement of sorts. That the command is in essence the characters for key delimiters [] doesn't change that. The array that is an element in a multidimensional array is not itself a container. The array is the container. To work with it you have to put it into it's own container then pass the new array by reference. I hope that makes sense. At least it does to me. :-) Bob On Mar 10, 2012, at 11:53 AM, Dar Scott wrote: Thanks for the tip, Dick, on using the list of keys. One can think of arrays as nested or multidimensional. On Mar 10, 2012, at 1:06 AM, Dick Kriesel wrote: I agree it'd be good if LC could accept any array reference for invoking a handler that specifies pass-by-reference. Though is is probably more work, one might also consider chunks in pass-by-reference. Maybe any thing the subtract command can take. However, this might be a problem: doSomethingToTheseTwo char 1 to 2 of it, char 2 to 3 of it command doSomethingToTheseTwo @a, @b put butter into a put cheese into b put empty into a end doSomethingToTheseTwo That might also have a problem with this call: doSomethingToTheseTwo x, x[t] I immagine LiveCode folks can come up with a semantics that makes sense for weird cases. The subtract command does not have the the problem because it modifies only one thing. Dar ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode -- Pete Molly's Revenge http://www.mollysrevenge.com ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
Hi, Bob! On Mar 13, 2012, at 9:37 AM, Bob Sneidar wrote: That is what passing an element to an array amounts to was my point. I think what you are calling a statement many will call an expression. This concern is understandable. However, I think we can distinguish between an expression that means a value and an expression that means a mutable place. We have this distinction in LiveCode. The expression '4 + a' can be used only as a value. The expression a[3] can be used as a value or a mutable place. That value is what is contained in that mutable place. That mutable place is called a container in HyperCard. We can't say 'put 7 into 4+a' but we can say 'put 7 into a[3]'. The concept has strong HyperCard and LiveCode precedence. We can say 'put 3 into char 2 of item 6 of a'. A chunk is a container. Allowing @ parameters to refer to containers and not just variables can create some implementation challenges, perhaps, even some documentation challenges, but it does open the door to commands that can be used to, say, capitalize the value of containers. I would hope the conceptual challenges in use would only apply to advanced users doing weird things. In the simplest case, the extension would replace variable with container in the description of @. I come from a background in functional programming in which just about everything is a value and not a container, but I have embraced the language used in LiveCode and the ability to modify things in ways besides replacing variable values and replacing properties. Dar ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
Hi Dick, Thanks again for this tip. I ended up using it to solve my problem by passing the whole array variable along with an optional second parameter which is a list of comma-separated key values. The function turns the key values list into an array then uses it to access the data array. If the first parameter is a simple variable, then the second parameter isn't present and the function can access it directly.. I'm wondering how you found about about this key values array feature? I can't find any mention of it in the dictionary or the reference manual and it sure seems like something that should be known! Thanks, Pete On Sat, Mar 10, 2012 at 12:06 AM, Dick Kriesel dick.krie...@mail.comwrote: LC has a way to get and set a node at any level of an array, without code that concatenates [ ] to specify the node. In an array reference a single [ ] can enclose an array of keys, -- Dick script on mouseUp local tArray, tString put 4 into tArray[ 1 ][ 2 ][ 3 ] put getNodeViaListOfKeys( tArray, ( 1, 2, 3 )) after tString -- that is, get tArray[ 1 ][ 2 ][ 3 ] setNodeViaListOfKeys tArray, ( 5, 6, 7, 8 ), 9 -- that is, put 9 into tArray[ 5 ][ 6 ][ 7 ][ 8 ] put comma getNodeViaListOfKeys( tArray, ( 5, 6, 7, 8 )) after tString -- 4,9 breakpoint end mouseUp function getNodeViaListOfKeys @pArray, pList local tKeys put pList into tKeys split tKeys by comma and null return getNodeViaArrayOfKeys( pArray, tKeys ) end getNodeViaListOfKeys function getNodeViaArrayOfKeys @pArray, @pKeysArray return pArray[ pKeysArray ] end getNodeViaArrayOfKeys command setNodeViaListOfKeys @rArray, pList, pValue local tKeys put pList into tKeys split tKeys by comma and null setNodeViaArrayOfKeys rArray, tKeys, pValue end setNodeViaListOfKeys command setNodeViaArrayOfKeys @rArray, @pKeysArray, @pValue put pValue into rArray[ pKeysArray ] end setNodeViaArrayOfKeys /script ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode -- Pete Molly's Revenge http://www.mollysrevenge.com ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
Pete- Monday, March 12, 2012, 10:19:14 AM, you wrote: I'm wondering how you found about about this key values array feature? I can't find any mention of it in the dictionary or the reference manual and it sure seems like something that should be known! Dick is well known as the master of arrays. I've learned to just accept his advice without questioning. -- -Mark Wieder mwie...@ahsoftware.net ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
On Mar 12, 2012, at 10:19 AM, Pete wrote: I'm wondering how you found about about this key values array feature? I can't find any mention of it in the dictionary or the reference manual and it sure seems like something that should be known! I found it while looking through bug reports for arrays, in http://quality.runrev.com/show_bug.cgi?id=7166 dated 2008-11-28. So, thanks be to Mark Waddingham and David Bovill for that. For convenience, here's Mark's reply to David: Thanks for the suggestion. This is on the list to include for the next version. In that version with: put tValue into tArray[tKey] (indeed anywhere you can currently do tArray[tKey]) The meaning of 'tKey' will be extended: 1) If it is a string then it is as it is now 2) If it is a numerically keyed array starting at 1, then it will mean: put tValue into tArray[tKey[1]][tKey[2]]...[tKey[n]] where n is the number of elements in tKey 3) Otherwise it is an error. This will allow you to construct an array containing the path to the key you want to modify/fetch/delete dynamically. I agree, the feature's worth adding into the dictionary and manual. -- Dick ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
On Mar 12, 2012, at 11:12 AM, Mark Wieder wrote: Dick is well known as the master of arrays. I've learned to just accept his advice without questioning. I didn't know that! But, I make so many misteaks that I have to recomend that you resume questoining. -- Dick ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
Just weighing in here, that would be a bit confusing. Passing by reference means that the command or function has access to the variable passed to it. By passing what amounts to a statement, there is nothing for LC to manipulate on the other end. Statements have to have some place to put the results. In this case, there is no place for LC to put the statement when passed by reference. Even a reference to an element of an array is a statement of sorts. That the command is in essence the characters for key delimiters [] doesn't change that. The array that is an element in a multidimensional array is not itself a container. The array is the container. To work with it you have to put it into it's own container then pass the new array by reference. I hope that makes sense. At least it does to me. :-) Bob On Mar 10, 2012, at 11:53 AM, Dar Scott wrote: Thanks for the tip, Dick, on using the list of keys. One can think of arrays as nested or multidimensional. On Mar 10, 2012, at 1:06 AM, Dick Kriesel wrote: I agree it'd be good if LC could accept any array reference for invoking a handler that specifies pass-by-reference. Though is is probably more work, one might also consider chunks in pass-by-reference. Maybe any thing the subtract command can take. However, this might be a problem: doSomethingToTheseTwo char 1 to 2 of it, char 2 to 3 of it command doSomethingToTheseTwo @a, @b put butter into a put cheese into b put empty into a end doSomethingToTheseTwo That might also have a problem with this call: doSomethingToTheseTwo x, x[t] I immagine LiveCode folks can come up with a semantics that makes sense for weird cases. The subtract command does not have the the problem because it modifies only one thing. Dar ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
I'm not sure whose post you're responding to Bob. Where do you see something that amounts to a statement being passed as a referenced parameter? Pete On Mon, Mar 12, 2012 at 3:04 PM, Bob Sneidar b...@twft.com wrote: Just weighing in here, that would be a bit confusing. Passing by reference means that the command or function has access to the variable passed to it. By passing what amounts to a statement, there is nothing for LC to manipulate on the other end. Statements have to have some place to put the results. In this case, there is no place for LC to put the statement when passed by reference. Even a reference to an element of an array is a statement of sorts. That the command is in essence the characters for key delimiters [] doesn't change that. The array that is an element in a multidimensional array is not itself a container. The array is the container. To work with it you have to put it into it's own container then pass the new array by reference. I hope that makes sense. At least it does to me. :-) Bob On Mar 10, 2012, at 11:53 AM, Dar Scott wrote: Thanks for the tip, Dick, on using the list of keys. One can think of arrays as nested or multidimensional. On Mar 10, 2012, at 1:06 AM, Dick Kriesel wrote: I agree it'd be good if LC could accept any array reference for invoking a handler that specifies pass-by-reference. Though is is probably more work, one might also consider chunks in pass-by-reference. Maybe any thing the subtract command can take. However, this might be a problem: doSomethingToTheseTwo char 1 to 2 of it, char 2 to 3 of it command doSomethingToTheseTwo @a, @b put butter into a put cheese into b put empty into a end doSomethingToTheseTwo That might also have a problem with this call: doSomethingToTheseTwo x, x[t] I immagine LiveCode folks can come up with a semantics that makes sense for weird cases. The subtract command does not have the the problem because it modifies only one thing. Dar ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode -- Pete Molly's Revenge http://www.mollysrevenge.com ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
Pete- Monday, March 12, 2012, 5:39:16 PM, you wrote: I'm not sure whose post you're responding to Bob. Where do you see something that amounts to a statement being passed as a referenced parameter? Here's the problem: a parameter of the form array[key] is not a pointer to the key element of the array but rather the value stored in the array. So saying myArray[myKey] = hello doSomethingWith myArray[myKey] gives a parameter of hello to on doSomethingWith @arg You'd have to somehow prevent the dereferencing of myArray[myKey] in order to pass it by reference. -- -Mark Wieder mwie...@ahsoftware.net ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
On Mar 9, 2012, at 5:05 PM, Dar Scott wrote: Maybe the array could be global (or passed as a parameter) and the subscript passed. ... There might be problems I don't see right off, such as access to the same array at two different levels at the same time. Hi, Dar. LC has a way to get and set a node at any level of an array, without code that concatenates [ ] to specify the node. In an array reference a single [ ] can enclose an array of keys, as you can see in the code below. Put the following into a button and click it. The code invokes the debugger so you can see the array and the result string on the variables tab. I agree it'd be good if LC could accept any array reference for invoking a handler that specifies pass-by-reference. Pete, I'd be glad to vote for your enhancement request. -- Dick script on mouseUp local tArray, tString put 4 into tArray[ 1 ][ 2 ][ 3 ] put getNodeViaListOfKeys( tArray, ( 1, 2, 3 )) after tString -- that is, get tArray[ 1 ][ 2 ][ 3 ] setNodeViaListOfKeys tArray, ( 5, 6, 7, 8 ), 9 -- that is, put 9 into tArray[ 5 ][ 6 ][ 7 ][ 8 ] put comma getNodeViaListOfKeys( tArray, ( 5, 6, 7, 8 )) after tString -- 4,9 breakpoint end mouseUp function getNodeViaListOfKeys @pArray, pList local tKeys put pList into tKeys split tKeys by comma and null return getNodeViaArrayOfKeys( pArray, tKeys ) end getNodeViaListOfKeys function getNodeViaArrayOfKeys @pArray, @pKeysArray return pArray[ pKeysArray ] end getNodeViaArrayOfKeys command setNodeViaListOfKeys @rArray, pList, pValue local tKeys put pList into tKeys split tKeys by comma and null setNodeViaArrayOfKeys rArray, tKeys, pValue end setNodeViaListOfKeys command setNodeViaArrayOfKeys @rArray, @pKeysArray, @pValue put pValue into rArray[ pKeysArray ] end setNodeViaArrayOfKeys /script ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
Dick: Thanks for the script, another little nugget of knowledge added to my LC store! I have entered an enhancement request - it's # 10070 Dar: The function in question is a general purpose one that gets called from many places. Sometimes I need to pass an array element and other times just a simple variable so I'd rather not complicate things by passing in an extra parameter that holds the array key if it is an array that's being passed. But that is a possible solution so thanks for the suggestion. Bob: Creating a temporary array works, of course, and that's what I'm doing for now. But it defeats the whole purpose of passing the parameter by reference since now I have to put the temporary array back into the real array after the call. I may as well just change the function to return the result of its code and pass the parameter by value instead of reference. Pete On Sat, Mar 10, 2012 at 12:06 AM, Dick Kriesel dick.krie...@mail.comwrote: Hi, Dar. LC has a way to get and set a node at any level of an array, without code that concatenates [ ] to specify the node. In an array reference a single [ ] can enclose an array of keys, as you can see in the code below. Put the following into a button and click it. The code invokes the debugger so you can see the array and the result string on the variables tab. I agree it'd be good if LC could accept any array reference for invoking a handler that specifies pass-by-reference. Pete, I'd be glad to vote for your enhancement request. -- Dick -- Pete Molly's Revenge http://www.mollysrevenge.com ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
On Mar 10, 2012, at 11:21 AM, Pete wrote: I may as well just change the function to return the result of its code and pass the parameter by value instead of reference. Hmm. Like this? put modified(a[s]) into a[s] This might be the best for now. Dar --- Dar Scott dba Dar Scott Consulting 8637 Horacio Place NE Albuquerque, NM 87111 Lab, home, office phone: +1 505 299 9497 For Skype and fax, please contact. d...@swcp.com Computer Programming and tinkering, often making LiveCode libraries and externals, sometimes writing associated microcontroller firmware. --- ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Passing parameters by reference
I think I already know the answer to this so just confirming in case I'm missing something Let's say I have a handler with a parameter that is passed by reference, eg command xyz @parm1. If I try to pass an element of an array, eg xyz theArray[2], I get a runtime error. If, however, I pass the whole array, eg xyz theArray, all seems to work fine. That seems somewhat strange to me but apparently that's the way it is? -- Pete Molly's Revenge http://www.mollysrevenge.com ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
Oh, that would be cool! Maybe the array could be global (or passed as a parameter) and the subscript passed. This works: add 1 to a[x] ... it seems like a reasonable thing to do. There might be problems I don't see right off, such as access to the same array at two different levels at the same time. Dar On Mar 9, 2012, at 4:30 PM, Pete wrote: I think I already know the answer to this so just confirming in case I'm missing something Let's say I have a handler with a parameter that is passed by reference, eg command xyz @parm1. If I try to pass an element of an array, eg xyz theArray[2], I get a runtime error. If, however, I pass the whole array, eg xyz theArray, all seems to work fine. That seems somewhat strange to me but apparently that's the way it is? -- Pete Molly's Revenge http://www.mollysrevenge.com ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing parameters by reference
I have an update on this. If the parameter passed is simply the name of the array with no key qualifications, all is fine. LC only kicks up a fuss when you pass a qualified array like myArray[myKey]. Weird. Pete On Fri, Mar 9, 2012 at 5:05 PM, Dar Scott d...@swcp.com wrote: Oh, that would be cool! Maybe the array could be global (or passed as a parameter) and the subscript passed. This works: add 1 to a[x] ... it seems like a reasonable thing to do. There might be problems I don't see right off, such as access to the same array at two different levels at the same time. Dar On Mar 9, 2012, at 4:30 PM, Pete wrote: I think I already know the answer to this so just confirming in case I'm missing something Let's say I have a handler with a parameter that is passed by reference, eg command xyz @parm1. If I try to pass an element of an array, eg xyz theArray[2], I get a runtime error. If, however, I pass the whole array, eg xyz theArray, all seems to work fine. That seems somewhat strange to me but apparently that's the way it is? -- Pete Molly's Revenge http://www.mollysrevenge.com ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode -- Pete Molly's Revenge http://www.mollysrevenge.com ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing Parameters
Pete, Most programmes were not written for socket communication. Instead, they write to stdout, which can be read from using the read from process command. Writing to stdout is much easier than writing to a socket. However, if you are writing the app you want to communicate with yourself, then you might as well use sockets. IIRC writing to stdout doesn't work correctly in LiveCode, which leaves sockets as a good alternative. -- Best regards, Mark Schonewille Economy-x-Talk Consulting and Software Engineering Homepage: http://economy-x-talk.com Twitter: http://twitter.com/xtalkprogrammer KvK: 50277553 Become our partner in sales http://qery.us/16r Start selling Color Converter today. 20% commission! On 9 nov 2011, at 08:38, Pete wrote: Very useful lesson. I guess I'm still interested in the differences/pros/cons of using sockets versus open process/read from process/write to process LC commands (assuming both processes are LC). Do those commands use sockets under the covers? Pete Molly's Revenge http://www.mollysrevenge.com ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing Parameters
If you have an existing program that reads from stdin and writes to stdout - and you want to control / monitor it from another app - then it'd be natural to just use 'read from process'. Or if you have a program that you don't own / develop you might have to use 'read from process'. In general, using sockets is much more flexible and (once you've got over the initial hurdle of building socket communications once or twice) very simple - start with the lesson Phil pointed to, look at Chatrev, etc. 'read from process' has many limitations - in particular you need to poll for input, using the 'read from process... in x msecs form. If you use sockets then you can have the programs operate asynchronously, and have a handler that is called when data is ready for you. Or you can communicate with multiple other apps. Or move one of the apps to another machine (with some care over security). Or almost anything. -- Alex. On 09/11/2011 07:38, Pete wrote: Very useful lesson. I guess I'm still interested in the differences/pros/cons of using sockets versus open process/read from process/write to process LC commands (assuming both processes are LC). Do those commands use sockets under the covers? Pete Molly's Revengehttp://www.mollysrevenge.com On Tue, Nov 8, 2011 at 9:27 PM, Phil Davis There is a nice intro to LC sockets here: http://lessons.runrev.com/s/**lessons/m/4071/l/12924-How-to-** communicate-with-other-**applications-using-socketshttp://lessons.runrev.com/s/lessons/m/4071/l/12924-How-to-communicate-with-other-applications-using-sockets For inter-app communication, both apps are client AND server. Phil ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing Parameters
Note that these days most OSes will put up a security warning when you start to accept sockets, which you will need to do for inter-process communication. On corporate windows environments, accepting sockets might be disabled altogether for user processes. On 9 Nov 2011, at 12:17, Alex Tweedly wrote: If you have an existing program that reads from stdin and writes to stdout - and you want to control / monitor it from another app - then it'd be natural to just use 'read from process'. Or if you have a program that you don't own / develop you might have to use 'read from process'. In general, using sockets is much more flexible and (once you've got over the initial hurdle of building socket communications once or twice) very simple - start with the lesson Phil pointed to, look at Chatrev, etc. 'read from process' has many limitations - in particular you need to poll for input, using the 'read from process... in x msecs form. If you use sockets then you can have the programs operate asynchronously, and have a handler that is called when data is ready for you. Or you can communicate with multiple other apps. Or move one of the apps to another machine (with some care over security). Or almost anything. -- Alex. On 09/11/2011 07:38, Pete wrote: Very useful lesson. I guess I'm still interested in the differences/pros/cons of using sockets versus open process/read from process/write to process LC commands (assuming both processes are LC). Do those commands use sockets under the covers? Pete Molly's Revengehttp://www.mollysrevenge.com On Tue, Nov 8, 2011 at 9:27 PM, Phil Davis There is a nice intro to LC sockets here: http://lessons.runrev.com/s/**lessons/m/4071/l/12924-How-to-** communicate-with-other-**applications-using-socketshttp://lessons.runrev.com/s/lessons/m/4071/l/12924-How-to-communicate-with-other-applications-using-sockets For inter-app communication, both apps are client AND server. Phil ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode -- Watch live presentations every Saturday: http://livecode.tv Use an alternative Dictionary viewer: http://bjoernke.com/bvgdocu/ Chat with other RunRev developers: http://bjoernke.com/chatrev/ ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing Parameters
Thanks for the explanations - stopped me from going down the wrong path. Pete Molly's Revenge http://www.mollysrevenge.com On Wed, Nov 9, 2011 at 3:17 AM, Alex Tweedly a...@tweedly.net wrote: If you have an existing program that reads from stdin and writes to stdout - and you want to control / monitor it from another app - then it'd be natural to just use 'read from process'. Or if you have a program that you don't own / develop you might have to use 'read from process'. In general, using sockets is much more flexible and (once you've got over the initial hurdle of building socket communications once or twice) very simple - start with the lesson Phil pointed to, look at Chatrev, etc. 'read from process' has many limitations - in particular you need to poll for input, using the 'read from process... in x msecs form. If you use sockets then you can have the programs operate asynchronously, and have a handler that is called when data is ready for you. Or you can communicate with multiple other apps. Or move one of the apps to another machine (with some care over security). Or almost anything. -- Alex. On 09/11/2011 07:38, Pete wrote: Very useful lesson. I guess I'm still interested in the differences/pros/cons of using sockets versus open process/read from process/write to process LC commands (assuming both processes are LC). Do those commands use sockets under the covers? Pete Molly's Revengehttp://www.**mollysrevenge.comhttp://www.mollysrevenge.com On Tue, Nov 8, 2011 at 9:27 PM, Phil Davis There is a nice intro to LC sockets here: http://lessons.runrev.com/s/lessons/m/4071/l/12924-How-to-http://lessons.runrev.com/s/**lessons/m/4071/l/12924-How-to-** communicate-with-other-applications-using-socketshtt** p://lessons.runrev.com/s/**lessons/m/4071/l/12924-How-to-** communicate-with-other-**applications-using-socketshttp://lessons.runrev.com/s/lessons/m/4071/l/12924-How-to-communicate-with-other-applications-using-sockets For inter-app communication, both apps are client AND server. Phil __**_ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/**mailman/listinfo/use-livecodehttp://lists.runrev.com/mailman/listinfo/use-livecode __**_ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/**mailman/listinfo/use-livecodehttp://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing Parameters
On 11/7/11 12:47 AM, Phil Davis wrote: If you need to send a message from app1 to app2, and app2 is already running, you could use socket communication between them. Don't have time to pencil it out, but maybe someone else will. If not, I'll do it tomorrow. Phil There is a nice intro to LC sockets here: http://lessons.runrev.com/s/lessons/m/4071/l/12924-How-to-communicate-with-other-applications-using-sockets For inter-app communication, both apps are client AND server. Phil On 11/7/11 12:28 AM, Phil Davis wrote: On Mac OS X, you can do it from the command line. 1) create a standalone app (named Untitled1 in this example) that has this stack script: on startup dispatch $1 to me quit end startup on incoming beep end incoming 2) create a different stack with this BUTTON script: on mouseUp answer file Pick the app: if it = empty then exit to top get shell (it /Contents/MacOS/Untitled1 incoming) put the result after msg end mouseUp 3) click the button. It should start the Untitled1 app and you should hear the beep. At least it works here. I haven't tried this on Windows, but I bet something similar would work there. Phil Davis On 11/6/11 7:35 PM, Cal Horner wrote: Is there any command that will allow me to pass a simple parameter to a compiled LC program? ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode -- Phil Davis PDS Labs Professional Software Development http://pdslabs.net ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing Parameters
Hi Cal, You can start up a standalone from the commandline with C:/path/to/standalone.exe param1 param2,... ~/applications/standalone.app param1 param2... In your standalone you can retrieve the values in $1, $2,... anywhere in your scripts: put $1 into fld 1 get item 2 of $2 If your standalone is already running, you can use sockets to communicate. Here's http://qery.us/1a2 an example of using sockets for telnet communication. Once you understand how sockets work, you can use them for inter-application communication. On Mac OS X, you can also use AppleScript: set myScript to answer \hi\ tell application Revolution to do script myScript Of course, the script limits apply, but if you only want to set the value of a variable, this should be sufficient. -- Best regards, Mark Schonewille Economy-x-Talk Consulting and Software Engineering Homepage: http://economy-x-talk.com Twitter: http://twitter.com/xtalkprogrammer KvK: 50277553 Become our partner in sales http://qery.us/16r Start selling Color Converter today. 20% commission! On 7 nov 2011, at 04:35, Cal Horner wrote: Is there any command that will allow me to pass a simple parameter to a compiled LC program? ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing Parameters
Phil Davis wrote: On Mac OS X, you can do it from the command line. 1) create a standalone app (named Untitled1 in this example) that has this stack script: on startup dispatch $1 to me quit end startup As an example of how to use environment variables that's an excellent one, but for production work it may pay to be a little paranoid, taking the view that an API like a command line is a potential point of security exposure. Rather than execute anything that comes in on the command line, we could instead parse it to ensure it meets a list of acceptable commands: on startup switch word 1 of $1 case ExampleCommand1 case ExampleCommand2 case ExampleCommand3 dispatch $1 to me default put Invalid command: word 1 of $1 end switch quit end startup For the relatively low cost of double-checking the command in the string $1 we can limit the range of things that the app can be used for. It's a little more work, but prevents using the full range of LiveCode to manipulate the app in unintended ways. -- Richard Gaskin Fourth World LiveCode training and consulting: http://www.fourthworld.com Webzine for LiveCode developers: http://www.LiveCodeJournal.com LiveCode Journal blog: http://LiveCodejournal.com/blog.irv ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing Parameters
I've never tried the open.read from/write process commands - would those be another way to do this? Back in the days when I worked on HP3000 computers, they had something called message files. They were basically files that any number of processes could write to and one process could read from and as a record was read, it was deleted from the file. It was a great and very easy way to implement client/server communications. Pete Molly's Revenge http://www.mollysrevenge.com On Mon, Nov 7, 2011 at 3:41 AM, Mark Schonewille m.schonewi...@economy-x-talk.com wrote: Hi Cal, You can start up a standalone from the commandline with C:/path/to/standalone.exe param1 param2,... ~/applications/standalone.app param1 param2... In your standalone you can retrieve the values in $1, $2,... anywhere in your scripts: put $1 into fld 1 get item 2 of $2 If your standalone is already running, you can use sockets to communicate. Here's http://qery.us/1a2 an example of using sockets for telnet communication. Once you understand how sockets work, you can use them for inter-application communication. On Mac OS X, you can also use AppleScript: set myScript to answer \hi\ tell application Revolution to do script myScript Of course, the script limits apply, but if you only want to set the value of a variable, this should be sufficient. -- Best regards, Mark Schonewille Economy-x-Talk Consulting and Software Engineering Homepage: http://economy-x-talk.com Twitter: http://twitter.com/xtalkprogrammer KvK: 50277553 Become our partner in sales http://qery.us/16r Start selling Color Converter today. 20% commission! On 7 nov 2011, at 04:35, Cal Horner wrote: Is there any command that will allow me to pass a simple parameter to a compiled LC program? ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing Parameters
Yes! My example was mostly built on a 'lab experiment' model. Phil On 11/7/11 7:52 AM, Richard Gaskin wrote: Phil Davis wrote: On Mac OS X, you can do it from the command line. 1) create a standalone app (named Untitled1 in this example) that has this stack script: on startup dispatch $1 to me quit end startup As an example of how to use environment variables that's an excellent one, but for production work it may pay to be a little paranoid, taking the view that an API like a command line is a potential point of security exposure. Rather than execute anything that comes in on the command line, we could instead parse it to ensure it meets a list of acceptable commands: on startup switch word 1 of $1 case ExampleCommand1 case ExampleCommand2 case ExampleCommand3 dispatch $1 to me default put Invalid command: word 1 of $1 end switch quit end startup For the relatively low cost of double-checking the command in the string $1 we can limit the range of things that the app can be used for. It's a little more work, but prevents using the full range of LiveCode to manipulate the app in unintended ways. -- Richard Gaskin Fourth World LiveCode training and consulting: http://www.fourthworld.com Webzine for LiveCode developers: http://www.LiveCodeJournal.com LiveCode Journal blog: http://LiveCodejournal.com/blog.irv ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode -- Phil Davis PDS Labs Professional Software Development http://pdslabs.net ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing Parameters
On 11/7/11 8:42 AM, Pete wrote: I've never tried the open.read from/write process commands - would those be another way to do this? Yes. See 'open process for neither' in the docs - it accomplishes the same thing as 'launch'. Back in the days when I worked on HP3000 computers, they had something called message files. They were basically files that any number of processes could write to and one process could read from and as a record was read, it was deleted from the file. It was a great and very easy way to implement client/server communications. Pete Molly's Revengehttp://www.mollysrevenge.com On Mon, Nov 7, 2011 at 3:41 AM, Mark Schonewille m.schonewi...@economy-x-talk.com wrote: Hi Cal, You can start up a standalone from the commandline with C:/path/to/standalone.exe param1 param2,... ~/applications/standalone.app param1 param2... In your standalone you can retrieve the values in $1, $2,... anywhere in your scripts: put $1 into fld 1 get item 2 of $2 If your standalone is already running, you can use sockets to communicate. Here's http://qery.us/1a2 an example of using sockets for telnet communication. Once you understand how sockets work, you can use them for inter-application communication. On Mac OS X, you can also use AppleScript: set myScript to answer \hi\ tell application Revolution to do script myScript Of course, the script limits apply, but if you only want to set the value of a variable, this should be sufficient. -- Best regards, Mark Schonewille Economy-x-Talk Consulting and Software Engineering Homepage: http://economy-x-talk.com Twitter: http://twitter.com/xtalkprogrammer KvK: 50277553 Become our partner in sales http://qery.us/16r Start selling Color Converter today. 20% commission! On 7 nov 2011, at 04:35, Cal Horner wrote: Is there any command that will allow me to pass a simple parameter to a compiled LC program? ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode -- Phil Davis PDS Labs Professional Software Development http://pdslabs.net ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Passing Parameters
Is there any command that will allow me to pass a simple parameter to a compiled LC program? ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: Passing Parameters
Hi Parameters are passed along with command or function calls. Check the dictionary. Read the entries on command and function. Look up the params and paramCount. Note that there are two methods of passing parameters, by value and by reference. Check the user gude as well. You must make a simple stack and experiment. The natural LC syntax is easy to learn and fun to use. And incredibly powerful. Write back if you need more... Craig Newman -Original Message- From: Cal Horner calhor...@xtra.co.nz To: use-livecode use-livecode@lists.runrev.com Sent: Sun, Nov 6, 2011 5:37 pm Subject: Passing Parameters Is there any command that will allow me to pass a simple parameter to a compiled LC program? ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
