diff options
Diffstat (limited to 'detoxk/detox.c')
| -rw-r--r-- | detoxk/detox.c | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/detoxk/detox.c b/detoxk/detox.c new file mode 100644 index 0000000..9889e45 --- /dev/null +++ b/detoxk/detox.c @@ -0,0 +1 @@ +/*__________________________________________________________________________
detox á extract value/content from tag-structured symbol
Copyright (C) 2003 - 2006 jan schacher
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
pd-port 20060517
revised tree output 20040712
initial build 20030502
____________________________________________________________________________*/
#include "m_pd.h"
#include <string.h>
#include <stdlib.h>
#include <ctype.h>
#include <locale.h>
#include <stdio.h>
#define VERSION "1.3"
#define CLIP(x,a,b) (x)=(x)<(a)?(a):(x)>(b)?(b):(x)
#define MAXLENGTH 4096
typedef struct _detox
{
t_object ob;
t_atom t_tree[257], t_attrpair[2];
t_atom t_comp, t_blip;
t_atom t_checkat[2];
char t_charbuf[MAXLENGTH];
long t_treecount;
short t_debug;
short t_mode;
void *s_outlet, *s_outlet2, *s_outlet3, *s_outlet4;
} t_detox;
t_symbol *ps_nothing;
void *detox_class;
void *detox_new(t_symbol *s, short argc, t_atom *argv);
void detox_bang(t_detox *x);
void detox_float(t_detox *x, t_float f);
void detox_list(t_detox *x, t_symbol *s, short argc, t_atom *argv);
void detox_anything(t_detox *x, t_symbol *s, short argc, t_atom *argv);
void detox_free(t_detox *x);
void detox_reset(t_detox *x);
void detox_debug(t_detox *x, float f);
void detox_mode(t_detox *x, float f);
void detox_anything(t_detox *x, t_symbol *s, short argc, t_atom *argv);
void detox_action(t_detox *x);
void detox_fixsymbol(t_detox *x);
void version(t_detox *x);
void detox_setup(void)
{
detox_class = class_new(gensym("detox"), (t_newmethod)detox_new,
(t_method)detox_free, sizeof(t_detox), 0, A_GIMME, 0);
class_addsymbol(detox_class, (t_method)detox_anything);
class_addmethod(detox_class, (t_method)detox_list, gensym("list"), A_GIMME,0);
class_addmethod(detox_class, (t_method)detox_debug, gensym("debug"), A_FLOAT, 0);
class_addmethod(detox_class, (t_method)detox_mode, gensym("mode"), A_FLOAT, 0);
class_addmethod(detox_class, (t_method)detox_reset, gensym("reset"), 0);
class_addmethod(detox_class, (t_method)version, gensym("version"), 0);
post("- detox - jasch - "__DATE__" ");
ps_nothing = gensym("");
}
void *detox_new(t_symbol *s, short argc, t_atom *argv)
{
t_detox *x = (t_detox *)pd_new(detox_class);
x->s_outlet = outlet_new(&x->ob, NULL);
x->s_outlet2 = outlet_new(&x->ob, NULL);
x->s_outlet3 = outlet_new(&x->ob, NULL);
x->s_outlet4 = outlet_new(&x->ob, gensym("float"));
x->t_treecount = 0;
x->t_debug = 0;
if((argc >= 1)&&(argv[0].a_type == A_FLOAT)){
if(argv[0].a_w.w_float == 0){
x->t_mode = 0;
}else{
x->t_mode = 1;
}
}else{
x->t_mode = 1;
}
x->t_debug = 0;
if(x->t_debug) post("mode is %ld", x->t_mode);
x->t_checkat[0].a_type = A_SYMBOL;
x->t_checkat[0].a_w.w_symbol = gensym("none");
x->t_checkat[1].a_type = A_SYMBOL;
x->t_checkat[1].a_w.w_symbol = gensym("none");
return (x);
}
void detox_free(t_detox *x)
{
// notify_free((t_object *)x);
}
void detox_bang(t_detox *x)
{
detox_action(x);
}
void detox_float(t_detox *x, t_float f)
{
sprintf(x->t_charbuf, "%.6f", f);
detox_action(x);
}
void detox_list(t_detox *x, t_symbol *s, short argc, t_atom *argv)
{
long i;
char temp[MAXLENGTH];
x->t_charbuf[0] = 0;
for(i = 0; i < argc; i++){
if(i) strcat(x->t_charbuf, " ");
temp[0] = 0;
if(argv[i].a_type == A_FLOAT){
sprintf(temp, "%.6f", argv[i].a_w.w_float);
}else if(argv[i].a_type == A_SYMBOL){
// check for whitespace in symbol and add back doublequotes
if(strchr(argv[i].a_w.w_symbol->s_name, 32) == NULL) { // no whitespace : normal proc
strcpy(temp, argv[i].a_w.w_symbol->s_name);
} else { // with whitespace : fix it
x->t_checkat[0] = argv[i];
detox_fixsymbol(x);
strcpy(temp, x->t_checkat[1].a_w.w_symbol->s_name);
}
}
if((strlen(x->t_charbuf) + strlen(temp)) < MAXLENGTH){
strcat(x->t_charbuf, temp);
}else{
return;
}
}
detox_action(x);
}
void detox_anything(t_detox *x, t_symbol *s, short argc, t_atom *argv)
{
long i;
char temp[MAXLENGTH];
x->t_charbuf[0] = 0;
strcpy(x->t_charbuf, s->s_name);
for(i = 0; i < argc; i++){
strcat(x->t_charbuf, " ");
temp[0] = 0;
if(argv[i].a_type == A_FLOAT){
sprintf(temp, "%.6f", argv[i].a_w.w_float);
}else if(argv[i].a_type == A_SYMBOL){
// check for whitespace in symbol and add back doublequotes
if(strchr(argv[i].a_w.w_symbol->s_name, 32) == NULL) { // no whitespace : normal proc
strcpy(temp, argv[i].a_w.w_symbol->s_name);
} else { // with whitespace : fix it
x->t_checkat[0] = argv[i];
detox_fixsymbol(x);
strcpy(temp, x->t_checkat[1].a_w.w_symbol->s_name);
}
}
// post("temp is %s", temp);
if((strlen(x->t_charbuf) + strlen(temp)) < MAXLENGTH){
strcat(x->t_charbuf, temp);
}else{
return;
}
}
detox_action(x);
return;
}
void detox_reset(t_detox *x)
{
short i;
x->t_treecount = 0;
for(i=0;i<256; i++) x->t_tree[i].a_w.w_symbol = ps_nothing;
}
void detox_debug(t_detox *x, float f)
{
if(f == 0.0){
x->t_debug = 0;
}else if(f != 0){
x->t_debug = 1;
}
}
void detox_mode(t_detox *x, float f)
{
if(f == 0.0){
x->t_mode = 0;
}else if(f != 0){
x->t_mode = 1;
}
}
void detox_action(t_detox *x)
{
t_atom *outlist, *attrpair; // *comp,
char local[MAXLENGTH], local2[MAXLENGTH];
char temp[MAXLENGTH];
char tempstring[2]; // for strtok
char *ptr1 = NULL, *ptr2 = NULL, *ptr3 = NULL;
long i, j, k, last, len, tempcount = 0;
short tagtype = 0; // 0 = closed, 1 = open, 2 = closing
short outsize = 0;
char quotetype[1];
short digitflag = 0;
short punctflag = 0;
short quoteflag = 0;
// short attrpresent = 0;
quotetype[0] = '\"'; // init to standard double quotes
i = 0;
strcpy(tempstring, " ");
outlist = x->t_tree;
attrpair = x->t_attrpair;
x->t_attrpair[0].a_type = A_SYMBOL;
x->t_attrpair[1].a_type = A_SYMBOL; // check type and recast on output
strcpy(local2, x->t_charbuf);
last = strlen(local2); // test for tag
// if(x->t_debug) post("input is %s", x->t_charbuf);
k = 0;
while(isspace(local2[k])){ // remove whitespace/tab/cr/linefeed
k++;
}
// if(x->t_debug) post("k is %ld, last is %ld", k, last);
for(i=k, j=0; i<last; i++, j++){ // remove whitespace/tab/cr/linefeed
local[j] = local2[i];
}
local[j] = 0; // terminate string
if((local[0] != '<') || (local[j-1] != '>')){
tagtype = 0; // not a well formed tag
if(x->t_debug) post("tagtype 0");
goto content;
}
if((local[1] == '?') || (local[1] == '!')){
tagtype = 4; // this might be a metatag
if(x->t_debug) post("tagtype 4");
goto content;
}
for(i = 0, ptr1 = NULL, quoteflag = 0; i < (long)(strlen(local) -1); i++) {
if((local[i] == '\"') || (local[i] == '\'')){
quoteflag = !quoteflag;
}
if((local[i] == '/') && (quoteflag == 0)) {
ptr1 = &local[i];
}
}
//ptr1 = strchr(local, '/'); // is it a closing tag ? // bug: if slash is within quotes: should ignore
if(x->t_debug) post("ptr1 contains %s", x->t_charbuf);
if(ptr1 == NULL){ // we have new open tag
tagtype = 2;
if(x->t_debug) post("tagtype 2");
}else if((local[1] == '/')){ // is it at beginning
tagtype = 3;
if(x->t_debug) post("tagtype 3");
}else if(local[1] != '/'){ // is it elsewhere
if(x->t_debug) post("tagtype 1");
tagtype = 1;
}
switch(tagtype){
case 1: // closed tag :: add temporarily to tree
ptr1 = strchr(local, '>');
ptr2 = strchr(local, ' ');
if(ptr2 != NULL) *ptr2 = 0;
if(ptr1 != NULL){
*ptr1 = 0;
}else{
return;
}
len = strlen(local);
ptr3 = (char *)memmove((local + 0),(local + 1), len-1);
local[len-1] = 0;
ptr3 = local;
x->t_blip.a_type = A_SYMBOL;
x->t_blip.a_w.w_symbol = gensym(ptr3);
tempcount = 1;
break;
case 2: // opening tag :: add to tree, increment count
ptr2 = strchr(local, ' ');
if(ptr2 != NULL){
*ptr2 = 0;
if(x->t_debug) post("inside opening tag w/ whitespace, %s", local);
}else{
local[strlen(local) - 1] = 0;
// local[strlen(ptr2) -1] = 0;
if(x->t_debug) post("inside opening tag no whitespace, %s", local);
// return;
}
len = strlen(local);
ptr3 = (char *)memmove((local + 0),(local + 1), len-1);
if(x->t_debug) post("ptr3 is %s", ptr3);
local[len-1] = 0;
ptr3 = local;
SETSYMBOL(x->t_tree + x->t_treecount, gensym(ptr3));
x->t_treecount++;
x->t_treecount = CLIP(x->t_treecount, 0, 255);
break;
case 3: // closing tag :: remove from tree, decrement count
ptr2 = strchr(local, '>');
if(ptr2 != NULL){
*ptr2 = 0;
}else{
return;
}
len = strlen(local);
ptr3 = (char *)memmove((local + 0),(local + 2), len-2);
local[len-2] = 0;
ptr3 = local;
x->t_comp.a_type = A_SYMBOL;
x->t_comp.a_w.w_symbol = gensym(ptr3);
if(x->t_comp.a_w.w_symbol->s_name == x->t_tree[x->t_treecount-1].a_w.w_symbol->s_name){
x->t_treecount -= 1;
x->t_treecount = CLIP(x->t_treecount, 0, 255);
}else{
outlet_float(x->s_outlet4, 3);
outlet_float(x->s_outlet3, -1);
return;
}
break;
}
content:
outlet_float(x->s_outlet4, tagtype);
if(tagtype == 0) return;
// the tree
if(x->t_treecount > 0){
if(tempcount){
outsize = x->t_treecount + 1;
x->t_tree[outsize-1] = x->t_blip;
}else{
outsize = x->t_treecount;
}
outlet_list(x->s_outlet3, 0L, outsize, outlist); // ouput tree
}else if(x->t_treecount == 0){
if(tempcount){
outlet_anything(x->s_outlet3, x->t_blip.a_w.w_symbol, 0, 0L); // ouput tree
}else{
outlet_float(x->s_outlet3, 0); // ouput tree
}
}
strcpy(local, x->t_charbuf);
ptr1 = strchr(local, '>'); // parse content between enclosing tags <tag>blah</tag>
if(ptr1 == NULL) goto attributes;
*ptr1 = 0;
++ptr1;
ptr2 = strchr(ptr1, '<');
if(ptr2 == NULL) goto attributes;
*ptr2 = 0;
outlet_anything(x->s_outlet2, gensym(ptr1), 0, 0L);
attributes:
strcpy(local, x->t_charbuf); // retrieve full buffer again
switch(x->t_mode) { // the old way of attribute parsing (mode 0)
case 0:
if(x->t_debug) post("content: before strtok %s", local);
ptr1 = local;
ptr2 = strchr(local, '>'); // find tag end
ptr2++; // terminate local buffer after tag end
ptr2[0] = 0;
if(strcspn(ptr1, "=\"\'") == strlen(ptr1)) { // check for attributes inside tag
break; // no attr-chars found, bail ouit
}
ptr1 = strchr(local, ' '); // find first whitespace
if(ptr1 == NULL) {
break; // no whitespace found bail
}
while((ptr1[0] == ' ') && (ptr1 != NULL)) { // find next non-whitespace char -> potential start of attr
ptr1++;
}
ptr2 = ptr1; // setup second pointer from end of attr name
while( !((ptr2[0] == ' ') || (ptr2[0] == '=')) && (ptr2 != NULL)) { // find next non-white or not equal-sign -> what if not well formed ??
ptr2++;
}
if(ptr2 == NULL) {
error("detox: mode 0, attrloop ptr2 reached NULL looking for \" \" and \"=\"");
return;
}
/*len = 0;
len = strlen(ptr1) - strlen(ptr2); // determine length of attr name: name minus whitespace/equal sign
for(i = 0; i < len; i++) {
temp[i] = ptr1[i];
}
temp[len] = 0; // terminate temp_buffer
*/
ptr1 = ptr2; // copy start pointer to position of end pointer -> remainder
while(ptr1 != NULL) { // move start pointer until single or double quote
if ((ptr1[0] == '\'') || (ptr1[0] == '\"')) {
quotetype[0] = ptr1[0];
break;
}
ptr1++;
}
if(ptr1 == NULL) { // bail if no quotes found and end reached
error("detox: mode 0, attrloop ptr1 reached NULL");
return;
}
// store quote char for later comparison
if (x->t_debug) post("first quotetype found is %c", quotetype[0]);
ptr2 = ++ptr1; // copy end pointer to pos of startpointer + 1, right after the quote
while(ptr2 != NULL) { // looking for next quote of stored type
if(ptr2[0] == quotetype[0]) {
break;
}
ptr2++;
}
// post("quotetype is %c and ptr2 is %c", quotetype[0], ptr2[0]);
if(ptr2 == NULL) {
error("detox: mode 0, attrloop ptr2 reached NULL");
return;
}
len = 0;
digitflag = 1; // reset check for long
punctflag = 0; // reset check for float
len = strlen(ptr1) - strlen(ptr2); // get the length of the attr content
// post("ptr2 ends scan at %c", ptr2[0]);
for(i = 0; i < len; i++) { // copy attr-content into temp-buffer
temp[i] = ptr1[i];
punctflag += (temp[i] == '.'); //
digitflag = digitflag && (isdigit(temp[i]) || ispunct(temp[i])); // are all digits?
}
temp[len] = 0;
if (x->t_debug) post("temp is %s, digitflag is %ld, punctflag is %ld", temp, digitflag, punctflag);
if(digitflag == 0){
attrpair[0].a_type = A_SYMBOL; // to do: check atom type and cast accordingly js 20061202
attrpair[0].a_w.w_symbol = gensym(temp);
outlet_anything(x->s_outlet, x->t_attrpair[0].a_w.w_symbol, 0, 0L);
} else if ((digitflag == 1)&&(punctflag == 0)){
attrpair[1].a_type = A_FLOAT; // to do: check atom type and cast accordingly js 20061202
attrpair[1].a_w.w_float = atoi(temp);
outlet_float(x->s_outlet, x->t_attrpair[2].a_w.w_float);
} else if ((digitflag == 1)&&(punctflag == 1)){
attrpair[1].a_type = A_FLOAT; // to do: check atom type and cast accordingly js 20061202
attrpair[1].a_w.w_float = atof(temp);
outlet_float(x->s_outlet, x->t_attrpair[1].a_w.w_float);
}
break;
case 1: // the new way of attribute parsing (mode 1) ¥¥¥¥ fixed 20070322
ptr1 = local;
ptr2 = strchr(local, '>'); // find tag end
ptr2++; // terminate local buffer after tag end
ptr2[0] = 0;
if(strcspn(ptr1, "=\"\'") == strlen(ptr1)) { // check for attributes inside tag
break; // no attr-chars found, bail ouit
}
ptr1 = strchr(local, ' '); // find first whitespace
if(ptr1 == NULL) {
break; // no whitespace found bail
}
for(;;) { // ¥¥ attribute loop; example: <person name = 'Alan Turing' died = '07/06/1954' born = "1912-06-23" />
while((ptr1[0] == ' ') && (ptr1 != NULL)) { // find next non-whitespace char -> potential start of attr
ptr1++;
}
ptr2 = ptr1; // setup second pointer from end of attr name
while( !((ptr2[0] == ' ') || (ptr2[0] == '=')) && (ptr2 != NULL)) { // find next non-white or not equal-sign -> what if not well formed ??
ptr2++;
}
if(ptr2 == NULL) {
error("detox: mode 1, attrloop ptr2 reached NULL looking for \" \" and \"=\"");
return;
}
len = 0;
len = strlen(ptr1) - strlen(ptr2); // determine length of attr name: name minus whitespace/equal sign
for(i = 0; i < len; i++) {
temp[i] = ptr1[i];
}
temp[len] = 0; // terminate temp_buffer
attrpair[0].a_w.w_symbol = gensym(temp); // put tempbuffer into output symbol
// ¥ done with the attr-name
ptr1 = ptr2; // copy start pointer to position of end pointer -> remainder
while(ptr1 != NULL) { // move start pointer until single or double quote
if ((ptr1[0] == '\'') || (ptr1[0] == '\"')) {
quotetype[0] = ptr1[0]; // store quote char for later comparison
break;
}
ptr1++;
}
if(ptr1 == NULL) { // bail if no quotes found and end reached
error("detox: mode 1, attrloop ptr1 reached NULL");
return;
}
if (x->t_debug) post("first quotetype found is %c", quotetype[0]);
ptr2 = ++ptr1; // copy end pointer to pos of startpointer + 1, right after the quote
while(ptr2 != NULL) { // looking for next quote of stored type
if(ptr2[0] == quotetype[0]) {
break;
}
ptr2++;
}
// post("quotetype is %c and ptr2 is %c", quotetype[0], ptr2[0]);
if(ptr2 == NULL) {
error("detox: mode 1, attrloop ptr2 reached NULL");
return;
}
len = 0;
digitflag = 1; // reset check for long
punctflag = 0; // reset check for float
len = strlen(ptr1) - strlen(ptr2); // get the length of the attr content
// post("ptr2 ends scan at %c", ptr2[0]);
for(i = 0; i < len; i++) { // copy attr-content into temp-buffer
temp[i] = ptr1[i];
punctflag += (temp[i] == '.'); //
digitflag = digitflag && (isdigit(temp[i]) || ispunct(temp[i])); // are all digits?
}
temp[len] = 0;
if (x->t_debug) post("temp is %s, digitflag is %ld, punctflag is %ld", temp, digitflag, punctflag);
if(digitflag == 0){
attrpair[1].a_type = A_SYMBOL; // to do: check atom type and cast accordingly js 20061202
attrpair[1].a_w.w_symbol = gensym(temp);
} else if ((digitflag == 1)&&(punctflag == 0)){
attrpair[1].a_type = A_FLOAT; // to do: check atom type and cast accordingly js 20061202
attrpair[1].a_w.w_float = atoi(temp);
} else if ((digitflag == 1)&&(punctflag == 1)){
attrpair[1].a_type = A_FLOAT; // to do: check atom type and cast accordingly js 20061202
attrpair[1].a_w.w_float = atof(temp);
}
outlet_anything(x->s_outlet, x->t_attrpair[0].a_w.w_symbol, 1, &x->t_attrpair[1]);
ptr1 = ++ptr2; // move start pointer to end of atrr-content
while(ptr1 != NULL) { // find next non-whitespace char
if (ptr1[0] != ' ') {
break;
}
ptr1++;
}
if((ptr1[0] == '/') || (ptr1[0] == '>') || (ptr1[0] == '?') || (ptr1 == NULL)) { // if char is "/" or ">" or "?" or end of string finish up
if (x->t_debug) post("ptr1 reached ending meta characters");
return;
}
if (x->t_debug) post("ptr1 reached loop point");
}
break;
}
}
void detox_fixsymbol(t_detox *x) // this is an ugly hack for max stripping double quotes from symbols with whitespace in them (language problem)
{
long i = 0;
long len = 0;
x->t_checkat[1].a_w.w_symbol->s_name[0] = '\"'; // first char in output is "
len = strlen(x->t_checkat[0].a_w.w_symbol->s_name);
len = CLIP(len, 0, (MAXLENGTH-3));
for(i = 0; i < len; i++) {
if ((x->t_checkat[0].a_w.w_symbol->s_name[i] == '/') || (x->t_checkat[0].a_w.w_symbol->s_name[i] == '>')) { // check for syntax differences
break;
}
x->t_checkat[1].a_w.w_symbol->s_name[i+1] = x->t_checkat[0].a_w.w_symbol->s_name[i];
}
x->t_checkat[1].a_w.w_symbol->s_name[i+1] = '\"'; // last char in output is "
x->t_checkat[1].a_w.w_symbol->s_name[i+2] = 0; // terminate string
}
void version(t_detox *x)
{
post("á detox á jasch á v. "VERSION" á compiled "__DATE__" - "__TIME__ );
}
\ No newline at end of file |