Trying to write simple example following your description
...
typedef struct {
int screen;
int views;
int active;
int storage;
} Watch_t;
...
Pvoid_t dict = (Pvoid_t) NULL;
int set_to_storage(char *key, Watch_t *val) {
PWord_t *PV;
JSLI(PV, dict, key); //first warning goes here
if (PV) {
*PV = (PWord_t) val;
return 0;
} else {
return 1;
}
}
int get_from_storage(char *key, Watch_t *val) {
PWord_t *PV;
JSLG(PV, dict, key); //first warning goes here
if (PV) {
*val = (Watch_t) *PV; //error goes here.
// *val = (Watch_t) PV; //error goes here.
return 0;
} else {
return 1;
}
}
It doesn't compile with following issues:
1). warning: pointer targets in passing argument 2 of ‘JudySLIns’
differ in signedness. - I suppose, i should cast somewhere to
unsigned, but not sure where exactly.
2). error: conversion to non-scalar type requested.
What I'm doing wrong?
2009/5/12 Alan Silverstein <[email protected]>:
> Rauan et al,
>
>> I want to create kind of associative array or hashmap and due to lack
>> of simple examples (quickstarts)...
>
> Sorry about that.
>
>> ...don't know what type of judy array to use, and how to implement it.
>
> I'll try to help.
>
>> My task is: I have unique string hash, e.g. "asdczxtdfdgk" (if it
>> does make sense, length of hash is persistent). By this hash i want
>> to get object/struct or array of values.
>>
>> Could you provide simple example?
>
> Yes, but let's talk about your hash string first. Why use a hash at all
> with libJudy, when it takes CPU time to compute the hash key, and
> there's some risk of collision no matter how good is your algorithm?
>
> Let's say instead that you have an actual C string as your key/index,
> and you want to map it to a malloc()'d object of whatever kind you like.
> The "liberating" concept here is that storing a very long string can
> still be very fast and cheap. The code (using the macros and ignoring
> error handling) looks like this:
>
> mydata_pt mydatap; // your pre-malloc()'d data record.
> char * index; // a key to save.
> char index2[MAXSIZE]; // for traversing JudySL array.
> Pvoid_t jsl = NULL; // the JudySL array.
> PWord_t valuep; // current value.
> PWord_t rcw; // for JSLFA().
>
> // Save a value:
>
> ...
> JSLI(valuep, jsl, index);
> *valuep = (PWord_t) mydatap;
> ...
>
> // Traverse all keys (strings) in simple-sort order:
>
> index2[0] = '\0';
> JSLF(valuep, jsl, index2);
>
> while (valuep != NULL)
> {
> DoSomethingWith(index2, (mydata_pt) *valuep);
> JSLN(valuep, jsl, index2);
> }
>
> // Erase array; must visit and free() each value record first:
> // (Not necessary if the data fits directly in the value word.)
>
> index2[0] = '\0';
> JSLF(valuep, jsl, index2);
>
> while (valuep != NULL)
> {
> free((void *) (*valuep));
> JSLN(valuep, jsl, index2);
> }
> JSLFA(rcw, jsl);
>
> ----------------------
>
> Now, does that help?
>
> Alan Silverstein
>
------------------------------------------------------------------------------
The NEW KODAK i700 Series Scanners deliver under ANY circumstances! Your
production scanning environment may not be a perfect world - but thanks to
Kodak, there's a perfect scanner to get the job done! With the NEW KODAK i700
Series Scanner you'll get full speed at 300 dpi even with all image
processing features enabled. http://p.sf.net/sfu/kodak-com
_______________________________________________
Judy-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/judy-devel
