Subj : Newb; script library
To   : netscape.public.mozilla.jseng
From : Lyndon
Date : Mon Oct 18 2004 12:31 pm

Setup: I have an application which is XSLT-like in nature; it transforms
large XML datasets into other formats.  I've added SpiderMonkey to it, and
can successfully call a single function out of my library.  It is when I try
to have more than one function that I'm encountering trouble.

At application initialization, I set up the run time, context, and global
object (copy and paste from SpiderMonkey docs).
Then, I truck through my script library, loading things up, using
load_js_files() below.
an example of a file is:
function ucase( ucase_string )
{
    return( ucase_string.toUpperCase() );
}
//-------------------------------
void load_js_files()
{
        extern char *sz_jsdir;
        jsval rval;
        char tfnam[MAX_STR_LEN];
        DIR *d;
        FILE *f;
        char *b;
        struct dirent *de;
        struct stat sb;
        JSScript *jsp;

        if ( NULL != ( d = opendir( sz_jsdir ) ) )
        {
                while( NULL != ( de = readdir( d ) ) )
                {
                        if ( 0 == strcmp( &de->d_name[de->d_namlen-3],
".js" ) )
                        {
                                sprintf( (char *)&tfnam, "%s/%s%c",
sz_jsdir, de->d_name, 0 );
                                stat( tfnam, &sb );
                                if ( NULL != ( f = fopen( tfnam, "r" ) ) )
                                {
                                        if ( NULL != ( b = (char *)malloc(
sb.st_size + 1 ) ) )
                                        {
                                                memset( b, 0, sb.st_size +
1 );
                                                fread( b, sb.st_size, 1,
f );
                                                if ( NULL == ( jsp =
JS_CompileScript( jscx, jsglob, b, sb.st_size, de->d_name, 0 ) ) )
                                                {
                                                        /* bad thing */
                                                        fprintf( stderr,
"Error compiling script %s\n", de->d_name );
                                                }
                                                else if ( JS_FALSE ==
JS_ExecuteScript( jscx, jsglob, jsp, &rval ) )
                                                {
                                                        /* bad thing */
                                                        fprintf( stderr,
"Error executing script %s\n", de->d_name );
                                                }
                                                free( b );
                                                b = NULL;
                                        }
                                        fclose( f );
                                }
                        }
                }
                closedir( d );
        }
}

//-------------------------------


For each datum I encounter, I add it as a variable in the global.  This
allows me to reference data using the tagnames in the original XML...  e.g.
<data>
<myname>Lyndon</myname>
</data>
would, in the below snippit, produce
"var myname = \"Lyndon\";"
//-------------------------------
        char jstmp[MAX_STR_LEN];
        jsval rval;
        sprintf( jstmp, "%s = \"%s\";", tag, tbuf );
        if ( JS_FALSE == JS_EvaluateScript( jscx, jsglob, jstmp,
strlen(jstmp), jstmp, 0, &rval ) )
        {
                fprintf( stderr, "Error creating variable %s with value
%s\n", tag, tbuf );
        }

//-------------------------------



Of course I have a Processing Instructions file, something like a
stylesheet.  For the above example, it may
do something like this:
<PI>
<callFunction>ucase( myname );</callFunction>
</PI>
When I encounter this tag, I call into the Monkey:
//-------------------------------
                        case SEGMENT_TYPE_JSFUNC:
                                /* javascript function */
                                if ( JS_TRUE == JS_EvaluateScript( jscx,
jsglob, segments[seg_i].data, segments[seg_i].size, segments[seg_i].data, 0,
&rval ) )
                                {
                                        JS_ConvertValue( jscx, rval,
JSTYPE_STRING, &rval );
                                        rstr = JS_ValueToString(jscx, rval);
                                        strncat( mbuf, JS_GetStringBytes(
rstr ), JS_GetStringLength( rstr ) );
                                }
                                break;

//-------------------------------


So, that all runs through fine, and I was actually pretty pleased with how
easy it was to add this limited support that I have.  To the point, though,
if I have another JS file in my library (or even another function in the
same file), that function fails to execute properly.  To wit:
<datafile>
<record><firstname>Lyndon</firstname><hobby>billiards</hobby></record>
</datafile>

<PI>
Name:&#x20;<callFunction>ucase(
hobby )</callFunction>&#x20;<callFunction>lcase( firstname )</callFunction>
</PI>

Output should be:
Name: LYNDON billiards

but I get:
Name: billiards
when lcase() is loaded first, or
Name: LYNDON
when ucase() is loaded first.

Please help; I'm confused about things like JS_CompileFunction, which partly
appears to be for JavaScript, and partly appears to be for C callbacks.  I'd
like to just have all my functions in a global way, without worrying too
much about keeping track of tons of JSScript* or other pointers.  Any
suggestions greatly appreciated.

Thanks,
<:)  Lyndon

.