Chip 2000 May

home *** CD-ROM | disk | FTP | other *** search

/ Chip 2000 May / Chip_2000-05_cd1.bin / zkuste / Perl / ActivePerl-5.6.0.613.msi / 䆊䌷䈹䈙䏵-䞅䞆䞀㡆䞃䄦䠥 / _e9da619a6e0b83f0e15a069ddba89cec < prev next >

Wrap

Text File | 2000-03-23 | 26.8 KB | 553 lines

<HTML> <HEAD> <TITLE>Tk2portableTk - how to make your B<Tk> source portable to other interpreted languages.</TITLE> <LINK REL="stylesheet" HREF="../../../Active.css" TYPE="text/css"> <LINK REV="made" HREF="mailto:"> </HEAD> <BODY> <TABLE BORDER=0 CELLPADDING=0 CELLSPACING=0 WIDTH=100%> <TR><TD CLASS=block VALIGN=MIDDLE WIDTH=100% BGCOLOR="#cccccc"> Tk2portableTk - how to make your B<Tk> source portable to other interpreted languages. </TD></TR> </TABLE> <A NAME="__index__"></A>  <UL> <LI><A HREF="#name">NAME</A></LI><LI><A HREF="#supportedplatforms">SUPPORTED PLATFORMS</A></LI> <LI><A HREF="#author">Author</A></LI> <LI><A HREF="#description">DESCRIPTION</A></LI> <LI><A HREF="#structure of ptk, porting your code">Structure of pTk, porting your code</A></LI> <LI><A HREF="#portabletk api">PortableTk API</A></LI> <UL> <LI><A HREF="#checking what you are running under">Checking what you are running under</A></LI> <LI><A HREF="#new types of configuration options">New types of configuration options</A></LI> <LI><A HREF="#language data">Language data</A></LI> <LI><A HREF="#conversion">Conversion</A></LI> <LI><A HREF="#callbacks">Callbacks</A></LI> <LI><A HREF="#setting variables">Setting variables</A></LI> <LI><A HREF="#language functions">Language functions</A></LI> <LI><A HREF="#translation of some tcl functions">Translation of some TCL functions</A></LI> </UL> <LI><A HREF="#translation back to tcl">Translation back to TCL</A></LI> <UL> <LI><A HREF="#new types of events ">New types of events ????</A></LI> </UL> <LI><A HREF="#checking for trouble">Checking for trouble</A></LI> <LI><A HREF="#additional api">Additional API</A></LI> <UL> <LI><A HREF="#listfactory"><CODE>ListFactory</CODE></A></LI> <LI><A HREF="#dstrings">DStrings</A></LI> <LI><A HREF="#accessing args">Accessing <CODE>Arg</CODE>s</A></LI> <LI><A HREF="#assigning numbers to args">Assigning numbers to <CODE>Arg</CODE>s</A></LI> <LI><A HREF="#creating new args">Creating new <CODE>Arg</CODE>s</A></LI> <LI><A HREF="#evaluating a list">Evaluating a list</A></LI> <LI><A HREF="#getting result as arg">Getting result as <CODE>Arg</CODE></A></LI> </UL> </UL>  <HR> <H1><A NAME="name">NAME</A></H1> Tk2portableTk - how to make your Tk source portable to other interpreted languages. <HR> <H1><A NAME="supportedplatforms">SUPPORTED PLATFORMS</A></H1> <UL> <LI>Linux</LI> <LI>Solaris</LI> <LI>Windows</LI> </UL> <HR> <H1><A NAME="author">Author</A></H1> Ilya Zakharevich <<A HREF="mailto:ilya@math.ohio-state.edu">ilya@math.ohio-state.edu</A>> has contributed most of this document. Many thanks. <HR> <H1><A NAME="description">DESCRIPTION</A></H1> PortableTk is an attempt to make Tk useful from other languages. Currently tk4.0 runs under Perl using this approach. Below, Lang is the notation for an external language to which PortableTk glues Tk code. The main problem with using the code developed for TCL with different languages is the absence of data types: almost anything is <CODE>char*</CODE>. It makes automatic translation hopeless. However, if you <CODE>typedef</CODE> several new symbols to be <CODE>char*</CODE>, you can still use your code in TCL, and it will make the automatic translation possible. Another problem with the approach that ``everything is a string'' is impossibility to have a result that says ``NotApplicable'' without setting an error. Thus different Tk command return different string values that mean ``error happened'', like <CODE>""</CODE>, <CODE>" "</CODE> or <CODE>"??"</CODE>. Other languages can be more flexible, so in portableTk you should inform the compiler that what you want to return means ``error'' (see <A HREF="#setting variables">Setting variables</A>). Currently PortableTk uses several different approachs to simplify translation: several TCL functions that are especially dangerous to use are undefined, so you can easily find places that need to be updated to use Language-independent functions based on compiler warnings. Eventually a way to use these Language-independent functions under proper TCL will be also provided. The end of this document provides a starting point for such a project. <HR> <H1><A NAME="structure of ptk, porting your code">Structure of pTk, porting your code</A></H1> pTk, that is a port of Tk, is very special with respect to porting of other code to portableTk. The problem is that currently there is very little hope to merge the modifications back into Tk, so a special strategy is needed to maintain this port. Do not use this strategy to port your own code. pTk is produced from Tk via a two-step process: first, some manual editing (the result is in the subdirectory <CODE>mTk</CODE>), and second, automatic conversion by the <CODE>munge</CODE> script (written in Perl). Thus the subdirectory <CODE>pTk/mTk</CODE> contains code with minimal possible difference from the virgin Tk code, so it is easier to <CODE>merge(1)</CODE> the differences between Tk versions into modified code. It looks like the strategy for a portable code should be exactly opposite: starting from TCL-based code, apply <CODE>munge</CODE>, and then hand-edit the resulting code. Probably it is also possible to target your code to portableTk from scratch, since this will make it possible to run it under a lot of Languages. The only reason anyone would like to look into contents of <CODE>pTk/mTk</CODE> directory is to find out which constructs are not supported by <CODE>munge</CODE>. On the other hand, <CODE>pTk</CODE> directory contains code that is conformant to portableTk, so you can look there to find example code. <CODE>munge</CODE> is the script that converts most common Tk constructs to their <CODE>portableTk</CODE> equivalent. For your code to qualify, you should follow Tk conventions on indentation and names of variables, in particular, the array of arguments for the <CODE>...CmdProc</CODE> should be called <CODE>argv</CODE>. For details on what <CODE>munge</CODE> can do, see <A HREF="#translation of some tcl functions">Translation of some TCL functions</A>. <HR> <H1><A NAME="portabletk api">PortableTk API</A></H1> <H2><A NAME="checking what you are running under">Checking what you are running under</A></H2> PortableTk provides a symbol <CODE>????</CODE>. If this symbol is defined, your source is compiled with it. <H2><A NAME="new types of configuration options">New types of configuration options</A></H2> PortableTk defines several new types of configuration options: <PRE> TK_CONFIG_CALLBACK TK_CONFIG_LANGARG TK_CONFIG_SCALARVAR TK_CONFIG_HASHVAR TK_CONFIG_ARRAYVAR TK_CONFIG_IMAGE</PRE> You should use them instead of TK_CONFIG_STRING whenever appropriate. This allows your application to receive a direct representation of the corresponding resource instead of the string representation, if this is possible under given language. ???? It looks like <CODE>TK_CONFIG_IMAGE</CODE> and <CODE>TK_CONFIG_SCALARVAR</CODE> set variables of type <CODE>char*</CODE>. <H2><A NAME="language data">Language data</A></H2> The following data types are defined: <DL> <DT><A NAME="item_Arg"><CODE>Arg</CODE></A> <DD> is the main datatype of the language. This is a type that your C function gets pointers to for arguments when the corresponding Lang function is called. The corresponding config type is <CODE>TK_CONFIG_LANGARG</CODE>. This is also a type that keeps information about contents of Lang variable. <DT><A NAME="item_Var"><CODE>Var</CODE></A> <DD> Is a substitute for a <CODE>char *</CODE> that contains name of variable. In Lang it is an object that contains reference to another Lang variable. <DT><A NAME="item_LangResultSave"><CODE>LangResultSave</CODE></A> <DD> ???? <DT><A NAME="item_LangCallback"><CODE>LangCallback</CODE></A> <DD> <CODE>LangCallback*</CODE> a substitute for a <CODE>char *</CODE> that contains command to call. The corresponding config type is <CODE>TK_CONFIG_CALLBACK</CODE>. <DT><A NAME="item_LangFreeProc"><CODE>LangFreeProc</CODE></A> <DD> It is the type that the <CODE>Lang_SplitList</CODE> sets. Before you call it, declare <PRE> Args *args; LangFreeProc *freeProc = NULL; ... code = Lang_SplitList(interp, value, &argc, &args, &freeProc);</PRE> After you use the split values, call <PRE> if (args != NULL && freeProc) (*freeProc)(argc,args);</PRE> It is not guaranteed that the <CODE>args</CODE> can survive deletion of <CODE>value</CODE>. </DL> <H2><A NAME="conversion">Conversion</A></H2> The following macros and functions are used for conversion between strings and the additional types: <PRE> LangCallback * LangMakeCallback(Arg) Arg LangCallbackArg(LangCallback *) char * LangString(Arg)</PRE> After you use the result of LangCallbackArg(), you should free it with <CODE>freeProc</CODE> <CODE>LANG_DYNAMIC</CODE> (it is not guaranteed that any change of <A HREF="#item_Arg"><CODE>Arg</CODE></A> will not be reflected in <LangCallback>, so you cannot do LangSet...() in between, and you should reset it to <CODE>NULL</CODE> if you want to do any further assignments to this <A HREF="#item_Arg"><CODE>Arg</CODE></A>). The following function returns the <A HREF="#item_Arg"><CODE>Arg</CODE></A> that is a reference to <A HREF="#item_Var"><CODE>Var</CODE></A>: <PRE> Arg LangVarArg(Var)</PRE> ???? It is very anti-intuitive, I hope the name is changed. <PRE> int LangCmpCallback(LangCallback *a,Arg b)</PRE> (currently only a stub), and, at last, <PRE> LangCallback * LangCopyCallback(LangCallback *)</PRE> <H2><A NAME="callbacks">Callbacks</A></H2> Above we have seen the new datatype <A HREF="#item_LangCallback"><CODE>LangCallback</CODE></A> and the corresponding Config option <CODE>TK_CONFIG_CALLBACK</CODE>. The following functions are provided for manipulation of <A HREF="#item_LangCallback"><CODE>LangCallback</CODE></A>s: <PRE> void LangFreeCallback(LangCallback *) int LangDoCallback(Tcl_Interp *,LangCallback *, int result,int argc, char *format,...)</PRE> The argument <A HREF="../../../lib/Pod/perlfunc.html#item_format"><CODE>format</CODE></A> of <CODE>LangDoCallback</CODE> should contain a string that is suitable for <A HREF="#item_sprintf"><CODE>sprintf</CODE></A> with optional arguments of <CODE>LangDoCallback</CODE>. <A HREF="#item_result"><CODE>result</CODE></A> should be false if result of callback is not needed. <PRE> int LangMethodCall(Tcl_Interp *,Arg,char *method, int result,int argc,...)</PRE> ???? Conceptually, <CODE>LangCallback*</CODE> is a substitute for ubiquitous <CODE>char *</CODE> in TCL. So you should use <CODE>LangFreeCallback</CODE> instead of <CODE>ckfree</CODE> or <CODE>free</CODE> if appropriate. <H2><A NAME="setting variables">Setting variables</A></H2> <PRE> void LangFreeArg (Arg, Tcl_FreeProc *freeProc) Arg LangCopyArg (Arg); void Tcl_AppendArg (Tcl_Interp *interp, Arg) void LangSetString(Arg *, char *s) void LangSetDefault(Arg *, char *s)</PRE> These two are equivalent unless s is an empty string. In this case <CODE>LangSetDefault</CODE> behaves like <CODE>LangSetString</CODE> with <CODE>s==NULL</CODE>, i.e., it sets the current value of the Lang variable to be false. <PRE> void LangSetInt(Arg *,int) void LangSetDouble(Arg *,double)</PRE> The Lang functions separate uninitialized and initialized data comparing data with <CODE>NULL</CODE>. So the declaration for an <A HREF="#item_Arg"><CODE>Arg</CODE></A> should look like <PRE> Arg arg = NULL;</PRE> if you want to use this <CODE>arg</CODE> with the above functions. After you are done, you should use <CODE>LangFreeArg</CODE> with <CODE>TCL_DYNAMIC</CODE> as <CODE>freeProc</CODE>. <H2><A NAME="language functions">Language functions</A></H2> Use <DL> <DT><A NAME="item_LangNull"><CODE>int LangNull(Arg)</CODE></A> <DD> to check that an object is false; <DT><A NAME="item_LangStringMatch"><CODE>int LangStringMatch(char *string, Arg match)</CODE></A> <DD> ???? <DT><A NAME="item_LangExit"><CODE>void LangExit(int)</CODE></A> <DD> to make a proper shutdown; <DT><A NAME="item_LangEval"><CODE>int LangEval(Tcl_Interp *interp, char *cmd, int global)</CODE></A> <DD> to call Lang <A HREF="../../../lib/Pod/perlfunc.html#item_eval"><CODE>eval</CODE></A>; <DT><A NAME="item_Lang_SetErrorCode"><CODE>void Lang_SetErrorCode(Tcl_Interp *interp,char *code)</CODE></A> <DD> <DT><A NAME="item_Lang_GetErrorCode"><CODE>char *Lang_GetErrorCode(Tcl_Interp *interp)</CODE></A> <DD> <DT><A NAME="item_Lang_GetErrorInfo"><CODE>char *Lang_GetErrorInfo(Tcl_Interp *interp)</CODE></A> <DD> <DT><A NAME="item_LangCloseHandler"><CODE>void LangCloseHandler(Tcl_Interp *interp,Arg arg,FILE *f,Lang_FileCloseProc *proc)</CODE></A> <DD> currently stubs only; <DT><A NAME="item_LangSaveVar"><CODE>int LangSaveVar(Tcl_Interp *,Arg arg,Var *varPtr,int type)</CODE></A> <DD> to save the structure <CODE>arg</CODE> into Lang variable <CODE>*varPtr</CODE>; <DT><A NAME="item_LangFreeVar"><CODE>void LangFreeVar(Var var)</CODE></A> <DD> to free the result; <DT><A NAME="item_LangEventCallback"><CODE>int LangEventCallback(Tcl_Interp *,LangCallback *,XEvent *,KeySym)</CODE></A> <DD> ???? <DT><A NAME="item_LangEventHook"><CODE>int LangEventHook(int flags)</CODE></A> <DD> <DT><A NAME="item_LangBadFile"><CODE>void LangBadFile(int fd)</CODE></A> <DD> <DT><A NAME="item_LangCmpConfig"><CODE>int LangCmpConfig(char *spec, char *arg, size_t length)</CODE></A> <DD> unsupported????; <DT><A NAME="item_Tcl_AppendArg"><CODE>void Tcl_AppendArg (Tcl_Interp *interp, Arg)</CODE></A> <DD> </DL> Another useful construction is <PRE> Arg variable = LangFindVar(interp, Tk_Window tkwin, char *name);</PRE> After using the above function, you should call <PRE> LangFreeVar(Var variable);</PRE> ???? Note discrepancy in types! If you want to find the value of a variable (of type <A HREF="#item_Arg"><CODE>Arg</CODE></A>) given the variable name, use <CODE>Tcl_GetVar(interp, varName, flags)</CODE>. If you are interested in the string value of this variable, use <CODE>LangString(Tcl_GetVar(...))</CODE>. To get a C array of <A HREF="#item_Arg"><CODE>Arg</CODE></A> of length <CODE>n</CODE>, use <PRE> Arg *args = LangAllocVec(n); ... LangFreeVec(n,args);</PRE> You can set the values of the <A HREF="#item_Arg"><CODE>Arg</CODE></A>s using <CODE>LangSet...</CODE> functions, and get string value using <CODE>LangString</CODE>. If you want to merge an array of <A HREF="#item_Arg"><CODE>Arg</CODE></A>s into one <A HREF="#item_Arg"><CODE>Arg</CODE></A> (that will be an array variable), use <PRE> result = Tcl_Merge(listLength, list);</PRE> <H2><A NAME="translation of some tcl functions">Translation of some TCL functions</A></H2> We mark items that can be dealt with by <CODE>munge</CODE> by Autoconverted. <DL> <DT><A NAME="item_Tcl_AppendResult"><CODE>Tcl_AppendResult</CODE></A> <DD> does not take <CODE>(char*)NULL</CODE>, but <CODE>NULL</CODE> as delimiter. Autoconverted. <DT><A NAME="item_Tcl_CreateCommand%2C_Tcl_DeleteCommand"><CODE>Tcl_CreateCommand</CODE>, <CODE>Tcl_DeleteCommand</CODE></A> <DD> <CODE>Tk_CreateWidget</CODE>, <CODE>Tk_DeleteWidget</CODE>, the second argument is the window itself, not the pathname. Autoconverted. <DT><A NAME="item_sprintf"><CODE>sprintf(interp->result, "%d %d %d %d",...)</CODE></A> <DD> <CODE>Tcl_IntResults(interp,4,0,...)</CODE>. Autoconverted. <DT><A NAME="item_result"><CODE>interp->result = "1";</CODE></A> <DD> <CODE>Tcl_SetResult(interp,"1", TCL_STATIC)</CODE>. Autoconverted. <DT>Reading <CODE>interp->result</CODE> <DD> <CODE>Tcl_GetResult(interp)</CODE>. Autoconverted. <DT><A NAME="item_Tk_PathName"><CODE>interp->result = Tk_PathName(textPtr->tkwin);</CODE></A> <DD> <CODE>Tk_WidgetResult(interp,textPtr->tkwin)</CODE>. Autoconverted. <DT><A NAME="item_Sequence_Tcl_PrintDouble%2C_Tcl_PrintDouble%2C_%2E">Sequence <CODE>Tcl_PrintDouble, Tcl_PrintDouble, ..., Tcl_AppendResult</CODE></A> <DD> Use a single command <PRE> void Tcl_DoubleResults(Tcl_Interp *interp, int append, int argc,...);</PRE> <CODE>append</CODE> governs whether it is required to clear the result first. A similar command for <A HREF="../../../lib/Pod/perlfunc.html#item_int"><CODE>int</CODE></A> arguments is <CODE>Tcl_IntResults</CODE>. <DT><A NAME="item_Tcl_SplitList"><CODE>Tcl_SplitList</CODE></A> <DD> Use <CODE>Lang_SplitList</CODE> (see the description above). </DL> <HR> <H1><A NAME="translation back to tcl">Translation back to TCL</A></H1> To use your portableTk program with TCL, put <PRE> #include "ptcl.h"</PRE> before inclusion of <CODE>tk.h</CODE>, and link the resulting code with <CODE>ptclGlue.c</CODE>. These files currently implement the following: <DL> <DT><A NAME="item_Additional_config_types%3A">Additional config types:</A> <DD> <PRE> TK_CONFIG_CALLBACK TK_CONFIG_LANGARG TK_CONFIG_SCALARVAR TK_CONFIG_HASHVAR TK_CONFIG_ARRAYVAR TK_CONFIG_IMAGE</PRE> <DT><A NAME="item_Types%3A">Types:</A> <DD> <PRE> Var, Arg, LangCallback, LangFreeProc.</PRE> <DT><A NAME="item_Functions_and_macros%3A">Functions and macros:</A> <DD> <PRE> Lang_SplitList, LangString, LangSetString, LangSetDefault, LangSetInt, LangSetDouble Tcl_ArgResult, LangCallbackArg, LangSaveVar, LangFreeVar, LangFreeSplitProc, LangFreeArg, Tcl_DoubleResults, Tcl_IntResults, LangDoCallback, Tk_WidgetResult, Tcl_CreateCommand, Tcl_DeleteCommand, Tcl_GetResult.</PRE> </DL> Current implementation contains enough to make it possible to compile <CODE>mTk/tkText*.[ch]</CODE> with the virgin Tk. <H2><A NAME="new types of events ">New types of events ????</A></H2> PortableTk defines following new types of events: <PRE> TK_EVENTTYPE_NONE TK_EVENTTYPE_STRING TK_EVENTTYPE_NUMBER TK_EVENTTYPE_WINDOW TK_EVENTTYPE_ATOM TK_EVENTTYPE_DISPLAY TK_EVENTTYPE_DATA</PRE> and a function <PRE> char * Tk_EventInfo(int letter, Tk_Window tkwin, XEvent *eventPtr, KeySym keySym, int *numPtr, int *isNum, int *type, int num_size, char *numStorage)</PRE> <HR> <H1><A NAME="checking for trouble">Checking for trouble</A></H1> If you start with working TCL code, you can start convertion using the above hints. Good indication that you are doing is OK is absence of <A HREF="#item_sprintf"><CODE>sprintf</CODE></A> and <CODE>sscanf</CODE> in your code (at least in the part that is working with interpreter). <HR> <H1><A NAME="additional api">Additional API</A></H1> What is described here is not included into base portableTk distribution. Currently it is coded in TCL and as Perl macros (core is coded as functions, so theoretically you can use the same object files with different interpreted languages). <H2><A NAME="listfactory"><CODE>ListFactory</CODE></A></H2> Dynamic arrays in TCL are used for two different purposes: to construct strings, and to construct lists. These two usages will have separate interfaces in other languages (since list is a different type from a string), so you should use a different interface in your code. The type for construction of dynamic lists is <CODE>ListFactory</CODE>. The API below is a counterpart of the API for construction of dynamic lists in TCL: <PRE> void ListFactoryInit(ListFactory *) void ListFactoryFinish(ListFactory *) void ListFactoryFree(ListFactory *) Arg * ListFactoryArg(ListFactory *) void ListFactoryAppend(ListFactory *, Arg *arg) void ListFactoryAppendCopy(ListFactory *, Arg *arg) ListFactory * ListFactoryNewLevel(ListFactory *) ListFactory * ListFactoryEndLevel(ListFactory *) void ListFactoryResult(Tcl_Interp *, ListFactory *)</PRE> The difference is that a call to <CODE>ListFactoryFinish</CODE> should precede the actual usage of the value of <CODE>ListFactory</CODE>, and there are two different ways to append an <A HREF="#item_Arg"><CODE>Arg</CODE></A> to a <CODE>ListFactory</CODE>: <CODE>ListFactoryAppendCopy()</CODE> guarantees that the value of <CODE>arg</CODE> is copied to the list, but <CODE>ListFactoryAppend()</CODE> may append to the list a reference to the current value of <CODE>arg</CODE>. If you are not going to change the value of <CODE>arg</CODE> after appending, the call to ListFactoryAppend may be quicker. As in TCL, the call to <CODE>ListFactoryFree()</CODE> does not free the <CODE>ListFactory</CODE>, only the objects it references. The functions <CODE>ListFactoryNewLevel()</CODE> and <CODE>ListFactoryEndLevel()</CODE> return a pointer to a <CODE>ListFactory</CODE> to fill. The argument of <CODE>ListFactoryEndLevel()</CODE> cannot be used after a call to this function. <H2><A NAME="dstrings">DStrings</A></H2> Production of strings are still supported in portableTk. <H2><A NAME="accessing args">Accessing <A HREF="#item_Arg"><CODE>Arg</CODE></A>s</A></H2> The following functions for getting a value of an <A HREF="#item_Arg"><CODE>Arg</CODE></A> may be provided: <PRE> double LangDouble(Arg) int LangInt(Arg) long LangLong(Arg) int LangIsList(Arg arg)</PRE> The function <CODE>LangIsList()</CODE> is supported only partially under TCL, since there is no data types. It checks whether there is a space inside the string <CODE>arg</CODE>. <H2><A NAME="assigning numbers to args">Assigning numbers to <A HREF="#item_Arg"><CODE>Arg</CODE></A>s</A></H2> While <CODE>LangSetDouble()</CODE> and <CODE>LangSetInt()</CODE> are supported ways to assign numbers to assign an integer value to a variable, for the sake of efficiency under TCL it is supposed that the destination of these commands was massaged before the call so it contains a long enough string to <A HREF="#item_sprintf"><CODE>sprintf()</CODE></A> the numbers inside it. If you are going to immediately use the resulting <A HREF="#item_Arg"><CODE>Arg</CODE></A>, the best way to do this is to declare a buffer in the beginning of a block by <PRE> dArgBuffer;</PRE> and assign this buffer to the <A HREF="#item_Arg"><CODE>Arg</CODE></A> by <PRE> void LangSetDefaultBuffer(Arg *)</PRE> You can also create the <CODE>buffer(s)</CODE> manually and assign them using <PRE> void LangSetBuffer(Arg *, char *)</PRE> This is the only choice if you need to assign numeric values to several <A HREF="#item_Arg"><CODE>Arg</CODE></A>s simultaneously. The advantage of the first approach is that the above declarations can be made <CODE>nop</CODE>s in different languages. Note that if you apply <CODE>LangSetDefaultBuffer</CODE> to an <A HREF="#item_Arg"><CODE>Arg</CODE></A> that contains some value, you can create a leak if you do not free that <A HREF="#item_Arg"><CODE>Arg</CODE></A> first. This is a non-problem in real languages, but can be a trouble in <CODE>TCL</CODE>, unless you use only the above API. <H2><A NAME="creating new args">Creating new <A HREF="#item_Arg"><CODE>Arg</CODE></A>s</A></H2> The API for creating a new <A HREF="#item_Arg"><CODE>Arg</CODE></A> is <PRE> void LangNewArg(Arg *, LangFreeProc *)</PRE> The API for creating a new <A HREF="#item_Arg"><CODE>Arg</CODE></A> is absent. Just initialize <A HREF="#item_Arg"><CODE>Arg</CODE></A> to be <CODE>NULL</CODE>, and apply one of <CODE>LangSet...</CODE> methods. After you use this <A HREF="#item_Arg"><CODE>Arg</CODE></A>, it should be freed thusly: <CODE>LangFreeArg(arg, freeProc)</CODE>. <H2><A NAME="evaluating a list">Evaluating a list</A></H2> Use <PRE> int LangArgEval(Tcl_Interp *, Arg arg)</PRE> Here <CODE>arg</CODE> should be a list to evaluate, in particular, the first element should be a <A HREF="#item_LangCallback"><CODE>LangCallback</CODE></A> massaged to be an <A HREF="#item_Arg"><CODE>Arg</CODE></A>. The arguments can be send to the subroutine by reference or by value in different languages. <H2><A NAME="getting result as arg">Getting result as <A HREF="#item_Arg"><CODE>Arg</CODE></A></A></H2> Use <CODE>Tcl_ArgResult</CODE>. It is not guaranteed that result survives this operation, so the <A HREF="#item_Arg"><CODE>Arg</CODE></A> you get should be the only mean to access the data from this moment on. After you use this <A HREF="#item_Arg"><CODE>Arg</CODE></A>, you should free it with <CODE>freeProc</CODE> <CODE>LANG_DYNAMIC</CODE> (you can do LangSet...() in between). <TABLE BORDER=0 CELLPADDING=0 CELLSPACING=0 WIDTH=100%> <TR><TD CLASS=block VALIGN=MIDDLE WIDTH=100% BGCOLOR="#cccccc"> Tk2portableTk - how to make your B<Tk> source portable to other interpreted languages. </TD></TR> </TABLE> </BODY> </HTML>