--- ray/src/common/ezxml.c 2009/06/17 20:41:47 2.1 +++ ray/src/common/ezxml.c 2011/06/01 00:29:40 2.2 @@ -1,5 +1,5 @@ #ifndef lint -static const char RCSid[] = "$Id: ezxml.c,v 2.1 2009/06/17 20:41:47 greg Exp $"; +static const char RCSid[] = "$Id: ezxml.c,v 2.2 2011/06/01 00:29:40 greg Exp $"; #endif /* ezxml.c * @@ -34,32 +34,32 @@ static const char RCSid[] = "$Id: ezxml.c,v 2.1 2009/0 #ifndef EZXML_NOMMAP #include #include -#endif // EZXML_NOMMAP +#endif /* EZXML_NOMMAP */ #include #include "ezxml.h" -#define EZXML_WS "\t\r\n " // whitespace -#define EZXML_ERRL 128 // maximum error string length +#define EZXML_WS "\t\r\n " /* whitespace */ +#define EZXML_ERRL 128 /* maximum error string length */ typedef struct ezxml_root *ezxml_root_t; -struct ezxml_root { // additional data for the root tag - struct ezxml xml; // is a super-struct built on top of ezxml struct - ezxml_t cur; // current xml tree insertion point - char *m; // original xml string - size_t len; // length of allocated memory for mmap, -1 for malloc - char *u; // UTF-8 conversion of string if original was UTF-16 - char *s; // start of work area - char *e; // end of work area - char **ent; // general entities (ampersand sequences) - char ***attr; // default attributes - char ***pi; // processing instructions - short standalone; // non-zero if - char err[EZXML_ERRL]; // error string +struct ezxml_root { /* additional data for the root tag */ + struct ezxml xml; /* is a super-struct built on top of ezxml struct */ + ezxml_t cur; /* current xml tree insertion point */ + char *m; /* original xml string */ + size_t len; /* length of allocated memory for mmap, -1 for malloc */ + char *u; /* UTF-8 conversion of string if original was UTF-16 */ + char *s; /* start of work area */ + char *e; /* end of work area */ + char **ent; /* general entities (ampersand sequences) */ + char ***attr; /* default attributes */ + char ***pi; /* processing instructions */ + short standalone; /* non-zero if */ + char err[EZXML_ERRL]; /* error string */ }; -char *EZXML_NIL[] = { NULL }; // empty, null terminated array of strings +char *EZXML_NIL[] = { NULL }; /* empty, null terminated array of strings */ -// returns the first child tag with the given name or NULL if not found +/* returns the first child tag with the given name or NULL if not found */ ezxml_t ezxml_child(ezxml_t xml, const char *name) { xml = (xml) ? xml->child : NULL; @@ -67,15 +67,15 @@ ezxml_t ezxml_child(ezxml_t xml, const char *name) return xml; } -// returns the Nth tag with the same name in the same subsection or NULL if not -// found +/* returns the Nth tag with the same name in the same subsection or NULL if not */ +/* found */ ezxml_t ezxml_idx(ezxml_t xml, int idx) { for (; xml && idx; idx--) xml = xml->next; return xml; } -// returns the value of the requested tag attribute or NULL if not found +/* returns the value of the requested tag attribute or NULL if not found */ const char *ezxml_attr(ezxml_t xml, const char *attr) { int i = 0, j = 1; @@ -83,16 +83,16 @@ const char *ezxml_attr(ezxml_t xml, const char *attr) if (! xml || ! xml->attr) return NULL; while (xml->attr[i] && strcmp(attr, xml->attr[i])) i += 2; - if (xml->attr[i]) return xml->attr[i + 1]; // found attribute + if (xml->attr[i]) return xml->attr[i + 1]; /* found attribute */ - while (root->xml.parent) root = (ezxml_root_t)root->xml.parent; // root tag + while (root->xml.parent) root = (ezxml_root_t)root->xml.parent; /* root tag */ for (i = 0; root->attr[i] && strcmp(xml->name, root->attr[i][0]); i++); - if (! root->attr[i]) return NULL; // no matching default attributes + if (! root->attr[i]) return NULL; /* no matching default attributes */ while (root->attr[i][j] && strcmp(attr, root->attr[i][j])) j += 3; - return (root->attr[i][j]) ? root->attr[i][j + 1] : NULL; // found default + return (root->attr[i][j]) ? root->attr[i][j + 1] : NULL; /* found default */ } -// same as ezxml_get but takes an already initialized va_list +/* same as ezxml_get but takes an already initialized va_list */ ezxml_t ezxml_vget(ezxml_t xml, va_list ap) { char *name = va_arg(ap, char *); @@ -105,12 +105,12 @@ ezxml_t ezxml_vget(ezxml_t xml, va_list ap) return (idx < 0) ? xml : ezxml_vget(ezxml_idx(xml, idx), ap); } -// Traverses the xml tree to retrieve a specific subtag. Takes a variable -// length list of tag names and indexes. The argument list must be terminated -// by either an index of -1 or an empty string tag name. Example: -// title = ezxml_get(library, "shelf", 0, "book", 2, "title", -1); -// This retrieves the title of the 3rd book on the 1st shelf of library. -// Returns NULL if not found. +/* Traverses the xml tree to retrieve a specific subtag. Takes a variable */ +/* length list of tag names and indexes. The argument list must be terminated */ +/* by either an index of -1 or an empty string tag name. Example: */ +/* title = ezxml_get(library, "shelf", 0, "book", 2, "title", -1); */ +/* This retrieves the title of the 3rd book on the 1st shelf of library. */ +/* Returns NULL if not found. */ ezxml_t ezxml_get(ezxml_t xml, ...) { va_list ap; @@ -122,20 +122,20 @@ ezxml_t ezxml_get(ezxml_t xml, ...) return r; } -// returns a null terminated array of processing instructions for the given -// target +/* returns a null terminated array of processing instructions for the given */ +/* target */ const char **ezxml_pi(ezxml_t xml, const char *target) { ezxml_root_t root = (ezxml_root_t)xml; int i = 0; if (! root) return (const char **)EZXML_NIL; - while (root->xml.parent) root = (ezxml_root_t)root->xml.parent; // root tag - while (root->pi[i] && strcmp(target, root->pi[i][0])) i++; // find target + while (root->xml.parent) root = (ezxml_root_t)root->xml.parent; /* root tag */ + while (root->pi[i] && strcmp(target, root->pi[i][0])) i++; /* find target */ return (const char **)((root->pi[i]) ? root->pi[i] + 1 : EZXML_NIL); } -// set an error string and return root +/* set an error string and return root */ ezxml_t ezxml_err(ezxml_root_t root, char *s, const char *err, ...) { va_list ap; @@ -152,18 +152,18 @@ ezxml_t ezxml_err(ezxml_root_t root, char *s, const ch return &root->xml; } -// Recursively decodes entity and character references and normalizes new lines -// ent is a null terminated array of alternating entity names and values. set t -// to '&' for general entity decoding, '%' for parameter entity decoding, 'c' -// for cdata sections, ' ' for attribute normalization, or '*' for non-cdata -// attribute normalization. Returns s, or if the decoded string is longer than -// s, returns a malloced string that must be freed. +/* Recursively decodes entity and character references and normalizes new lines */ +/* ent is a null terminated array of alternating entity names and values. set t */ +/* to '&' for general entity decoding, '%' for parameter entity decoding, 'c' */ +/* for cdata sections, ' ' for attribute normalization, or '*' for non-cdata */ +/* attribute normalization. Returns s, or if the decoded string is longer than */ +/* s, returns a malloced string that must be freed. */ char *ezxml_decode(char *s, char **ent, char t) { char *e, *r = s, *m = s; long b, c, d, l; - for (; *s; s++) { // normalize line endings + for (; *s; s++) { /* normalize line endings */ while (*s == '\r') { *(s++) = '\n'; if (*s == '\n') memmove(s, (s + 1), strlen(s)); @@ -174,89 +174,89 @@ char *ezxml_decode(char *s, char **ent, char t) while (*s && *s != '&' && (*s != '%' || t != '%') && !isspace(*s)) s++; if (! *s) break; - else if (t != 'c' && ! strncmp(s, "&#", 2)) { // character reference - if (s[2] == 'x') c = strtol(s + 3, &e, 16); // base 16 - else c = strtol(s + 2, &e, 10); // base 10 - if (! c || *e != ';') { s++; continue; } // not a character ref + else if (t != 'c' && ! strncmp(s, "&#", 2)) { /* character reference */ + if (s[2] == 'x') c = strtol(s + 3, &e, 16); /* base 16 */ + else c = strtol(s + 2, &e, 10); /* base 10 */ + if (! c || *e != ';') { s++; continue; } /* not a character ref */ - if (c < 0x80) *(s++) = c; // US-ASCII subset - else { // multi-byte UTF-8 sequence - for (b = 0, d = c; d; d /= 2) b++; // number of bits in c - b = (b - 2) / 5; // number of bytes in payload - *(s++) = (0xFF << (7 - b)) | (c >> (6 * b)); // head - while (b) *(s++) = 0x80 | ((c >> (6 * --b)) & 0x3F); // payload + if (c < 0x80) *(s++) = c; /* US-ASCII subset */ + else { /* multi-byte UTF-8 sequence */ + for (b = 0, d = c; d; d /= 2) b++; /* number of bits in c */ + b = (b - 2) / 5; /* number of bytes in payload */ + *(s++) = (0xFF << (7 - b)) | (c >> (6 * b)); /* head */ + while (b) *(s++) = 0x80 | ((c >> (6 * --b)) & 0x3F); /* payload */ } memmove(s, strchr(s, ';') + 1, strlen(strchr(s, ';'))); } else if ((*s == '&' && (t == '&' || t == ' ' || t == '*')) || - (*s == '%' && t == '%')) { // entity reference + (*s == '%' && t == '%')) { /* entity reference */ for (b = 0; ent[b] && strncmp(s + 1, ent[b], strlen(ent[b])); - b += 2); // find entity in entity list + b += 2); /* find entity in entity list */ - if (ent[b++]) { // found a match + if (ent[b++]) { /* found a match */ if ((c = strlen(ent[b])) - 1 > (e = strchr(s, ';')) - s) { - l = (d = (s - r)) + c + strlen(e); // new length + l = (d = (s - r)) + c + strlen(e); /* new length */ r = (r == m) ? strcpy(malloc(l), r) : realloc(r, l); - e = strchr((s = r + d), ';'); // fix up pointers + e = strchr((s = r + d), ';'); /* fix up pointers */ } - memmove(s + c, e + 1, strlen(e)); // shift rest of string - strncpy(s, ent[b], c); // copy in replacement text + memmove(s + c, e + 1, strlen(e)); /* shift rest of string */ + strncpy(s, ent[b], c); /* copy in replacement text */ } - else s++; // not a known entity + else s++; /* not a known entity */ } else if ((t == ' ' || t == '*') && isspace(*s)) *(s++) = ' '; - else s++; // no decoding needed + else s++; /* no decoding needed */ } - if (t == '*') { // normalize spaces for non-cdata attributes + if (t == '*') { /* normalize spaces for non-cdata attributes */ for (s = r; *s; s++) { if ((l = strspn(s, " "))) memmove(s, s + l, strlen(s + l) + 1); while (*s && *s != ' ') s++; } - if (--s >= r && *s == ' ') *s = '\0'; // trim any trailing space + if (--s >= r && *s == ' ') *s = '\0'; /* trim any trailing space */ } return r; } -// called when parser finds start of new tag +/* called when parser finds start of new tag */ void ezxml_open_tag(ezxml_root_t root, char *name, char **attr) { ezxml_t xml = root->cur; if (xml->name) xml = ezxml_add_child(xml, name, strlen(xml->txt)); - else xml->name = name; // first open tag + else xml->name = name; /* first open tag */ xml->attr = attr; - root->cur = xml; // update tag insertion point + root->cur = xml; /* update tag insertion point */ } -// called when parser finds character content between open and closing tag +/* called when parser finds character content between open and closing tag */ void ezxml_char_content(ezxml_root_t root, char *s, size_t len, char t) { ezxml_t xml = root->cur; char *m = s; size_t l; - if (! xml || ! xml->name || ! len) return; // sanity check + if (! xml || ! xml->name || ! len) return; /* sanity check */ - s[len] = '\0'; // null terminate text (calling functions anticipate this) + s[len] = '\0'; /* null terminate text (calling functions anticipate this) */ len = strlen(s = ezxml_decode(s, root->ent, t)) + 1; - if (! *(xml->txt)) xml->txt = s; // initial character content - else { // allocate our own memory and make a copy - xml->txt = (xml->flags & EZXML_TXTM) // allocate some space + if (! *(xml->txt)) xml->txt = s; /* initial character content */ + else { /* allocate our own memory and make a copy */ + xml->txt = (xml->flags & EZXML_TXTM) /* allocate some space */ ? realloc(xml->txt, (l = strlen(xml->txt)) + len) : strcpy(malloc((l = strlen(xml->txt)) + len), xml->txt); - strcpy(xml->txt + l, s); // add new char content - if (s != m) free(s); // free s if it was malloced by ezxml_decode() + strcpy(xml->txt + l, s); /* add new char content */ + if (s != m) free(s); /* free s if it was malloced by ezxml_decode() */ } if (xml->txt != m) ezxml_set_flag(xml, EZXML_TXTM); } -// called when parser finds closing tag +/* called when parser finds closing tag */ ezxml_t ezxml_close_tag(ezxml_root_t root, char *name, char *s) { if (! root->cur || ! root->cur->name || strcmp(name, root->cur->name)) @@ -266,59 +266,59 @@ ezxml_t ezxml_close_tag(ezxml_root_t root, char *name, return NULL; } -// checks for circular entity references, returns non-zero if no circular -// references are found, zero otherwise +/* checks for circular entity references, returns non-zero if no circular */ +/* references are found, zero otherwise */ int ezxml_ent_ok(char *name, char *s, char **ent) { int i; for (; ; s++) { - while (*s && *s != '&') s++; // find next entity reference + while (*s && *s != '&') s++; /* find next entity reference */ if (! *s) return 1; - if (! strncmp(s + 1, name, strlen(name))) return 0; // circular ref. + if (! strncmp(s + 1, name, strlen(name))) return 0; /* circular ref. */ for (i = 0; ent[i] && strncmp(ent[i], s + 1, strlen(ent[i])); i += 2); if (ent[i] && ! ezxml_ent_ok(name, ent[i + 1], ent)) return 0; } } -// called when the parser finds a processing instruction +/* called when the parser finds a processing instruction */ void ezxml_proc_inst(ezxml_root_t root, char *s, size_t len) { int i = 0, j = 1; char *target = s; - s[len] = '\0'; // null terminate instruction + s[len] = '\0'; /* null terminate instruction */ if (*(s += strcspn(s, EZXML_WS))) { - *s = '\0'; // null terminate target - s += strspn(s + 1, EZXML_WS) + 1; // skip whitespace after target + *s = '\0'; /* null terminate target */ + s += strspn(s + 1, EZXML_WS) + 1; /* skip whitespace after target */ } - if (! strcmp(target, "xml")) { // + if (! strcmp(target, "xml")) { /* */ if ((s = strstr(s, "standalone")) && ! strncmp(s + strspn(s + 10, EZXML_WS "='\"") + 10, "yes", 3)) root->standalone = 1; return; } - if (! root->pi[0]) *(root->pi = malloc(sizeof(char **))) = NULL; //first pi + if (! root->pi[0]) *(root->pi = malloc(sizeof(char **))) = NULL; /*first pi */ - while (root->pi[i] && strcmp(target, root->pi[i][0])) i++; // find target - if (! root->pi[i]) { // new target + while (root->pi[i] && strcmp(target, root->pi[i][0])) i++; /* find target */ + if (! root->pi[i]) { /* new target */ root->pi = realloc(root->pi, sizeof(char **) * (i + 2)); root->pi[i] = malloc(sizeof(char *) * 3); root->pi[i][0] = target; - root->pi[i][1] = (char *)(root->pi[i + 1] = NULL); // terminate pi list - root->pi[i][2] = strdup(""); // empty document position list + root->pi[i][1] = (char *)(root->pi[i + 1] = NULL); /* terminate pi list */ + root->pi[i][2] = strdup(""); /* empty document position list */ } - while (root->pi[i][j]) j++; // find end of instruction list for this target + while (root->pi[i][j]) j++; /* find end of instruction list for this target */ root->pi[i] = realloc(root->pi[i], sizeof(char *) * (j + 3)); root->pi[i][j + 2] = realloc(root->pi[i][j + 1], j + 1); strcpy(root->pi[i][j + 2] + j - 1, (root->xml.name) ? ">" : "<"); - root->pi[i][j + 1] = NULL; // null terminate pi list for this target - root->pi[i][j] = s; // set instruction + root->pi[i][j + 1] = NULL; /* null terminate pi list for this target */ + root->pi[i][j] = s; /* set instruction */ } -// called when the parser finds an internal doctype subset +/* called when the parser finds an internal doctype subset */ short ezxml_internal_dtd(ezxml_root_t root, char *s, size_t len) { char q, *c, *t, *n = NULL, *v, **ent, **pe; @@ -327,92 +327,92 @@ short ezxml_internal_dtd(ezxml_root_t root, char *s, s pe = memcpy(malloc(sizeof(EZXML_NIL)), EZXML_NIL, sizeof(EZXML_NIL)); for (s[len] = '\0'; s; ) { - while (*s && *s != '<' && *s != '%') s++; // find next declaration + while (*s && *s != '<' && *s != '%') s++; /* find next declaration */ if (! *s) break; - else if (! strncmp(s, "'); continue; } for (i = 0, ent = (*c == '%') ? pe : root->ent; ent[i]; i++); - ent = realloc(ent, (i + 3) * sizeof(char *)); // space for next ent + ent = realloc(ent, (i + 3) * sizeof(char *)); /* space for next ent */ if (*c == '%') pe = ent; else root->ent = ent; - *(++s) = '\0'; // null terminate name - if ((s = strchr(v, q))) *(s++) = '\0'; // null terminate value - ent[i + 1] = ezxml_decode(v, pe, '%'); // set value - ent[i + 2] = NULL; // null terminate entity list - if (! ezxml_ent_ok(n, ent[i + 1], ent)) { // circular reference + *(++s) = '\0'; /* null terminate name */ + if ((s = strchr(v, q))) *(s++) = '\0'; /* null terminate value */ + ent[i + 1] = ezxml_decode(v, pe, '%'); /* set value */ + ent[i + 2] = NULL; /* null terminate entity list */ + if (! ezxml_ent_ok(n, ent[i + 1], ent)) { /* circular reference */ if (ent[i + 1] != v) free(ent[i + 1]); ezxml_err(root, v, "circular entity declaration &%s", n); break; } - else ent[i] = n; // set entity name + else ent[i] = n; /* set entity name */ } - else if (! strncmp(s, "")) == '>') continue; - else *s = '\0'; // null terminate tag name + else *s = '\0'; /* null terminate tag name */ for (i = 0; root->attr[i] && strcmp(n, root->attr[i][0]); i++); while (*(n = ++s + strspn(s, EZXML_WS)) && *n != '>') { - if (*(s = n + strcspn(n, EZXML_WS))) *s = '\0'; // attr name + if (*(s = n + strcspn(n, EZXML_WS))) *s = '\0'; /* attr name */ else { ezxml_err(root, t, "malformed ") - 1; - if (*c == ' ') continue; // cdata is default, nothing to do + if (*c == ' ') continue; /* cdata is default, nothing to do */ v = NULL; } - else if ((*s == '"' || *s == '\'') && // default value + else if ((*s == '"' || *s == '\'') && /* default value */ (s = strchr(v = s + 1, *s))) *s = '\0'; else { ezxml_err(root, t, "malformed attr[i]) { // new tag name + if (! root->attr[i]) { /* new tag name */ root->attr = (! i) ? malloc(2 * sizeof(char **)) : realloc(root->attr, (i + 2) * sizeof(char **)); root->attr[i] = malloc(2 * sizeof(char *)); - root->attr[i][0] = t; // set tag name + root->attr[i][0] = t; /* set tag name */ root->attr[i][1] = (char *)(root->attr[i + 1] = NULL); } - for (j = 1; root->attr[i][j]; j += 3); // find end of list + for (j = 1; root->attr[i][j]; j += 3); /* find end of list */ root->attr[i] = realloc(root->attr[i], (j + 4) * sizeof(char *)); - root->attr[i][j + 3] = NULL; // null terminate list - root->attr[i][j + 2] = c; // is it cdata? + root->attr[i][j + 3] = NULL; /* null terminate list */ + root->attr[i][j + 2] = c; /* is it cdata? */ root->attr[i][j + 1] = (v) ? ezxml_decode(v, root->ent, *c) : NULL; - root->attr[i][j] = n; // attribute name + root->attr[i][j] = n; /* attribute name */ } } - else if (! strncmp(s, ""); // comments - else if (! strncmp(s, ""); /* comments */ + else if (! strncmp(s, ""))) ezxml_proc_inst(root, c, s++ - c); } - else if (*s == '<') s = strchr(s, '>'); // skip other declarations + else if (*s == '<') s = strchr(s, '>'); /* skip other declarations */ else if (*(s++) == '%' && ! root->standalone) break; } @@ -420,8 +420,8 @@ short ezxml_internal_dtd(ezxml_root_t root, char *s, s return ! *root->err; } -// Converts a UTF-16 string to UTF-8. Returns a new string that must be freed -// or NULL if no conversion was needed. +/* Converts a UTF-16 string to UTF-8. Returns a new string that must be freed */ +/* or NULL if no conversion was needed. */ char *ezxml_str2utf8(char **s, size_t *len) { char *u; @@ -429,38 +429,38 @@ char *ezxml_str2utf8(char **s, size_t *len) long c, d; int b, be = (**s == '\xFE') ? 1 : (**s == '\xFF') ? 0 : -1; - if (be == -1) return NULL; // not UTF-16 + if (be == -1) return NULL; /* not UTF-16 */ u = malloc(max); for (sl = 2; sl < *len - 1; sl += 2) { - c = (be) ? (((*s)[sl] & 0xFF) << 8) | ((*s)[sl + 1] & 0xFF) //UTF-16BE - : (((*s)[sl + 1] & 0xFF) << 8) | ((*s)[sl] & 0xFF); //UTF-16LE - if (c >= 0xD800 && c <= 0xDFFF && (sl += 2) < *len - 1) { // high-half + c = (be) ? (((*s)[sl] & 0xFF) << 8) | ((*s)[sl + 1] & 0xFF) /*UTF-16BE */ + : (((*s)[sl + 1] & 0xFF) << 8) | ((*s)[sl] & 0xFF); /*UTF-16LE */ + if (c >= 0xD800 && c <= 0xDFFF && (sl += 2) < *len - 1) { /* high-half */ d = (be) ? (((*s)[sl] & 0xFF) << 8) | ((*s)[sl + 1] & 0xFF) : (((*s)[sl + 1] & 0xFF) << 8) | ((*s)[sl] & 0xFF); c = (((c & 0x3FF) << 10) | (d & 0x3FF)) + 0x10000; } while (l + 6 > max) u = realloc(u, max += EZXML_BUFSIZE); - if (c < 0x80) u[l++] = c; // US-ASCII subset - else { // multi-byte UTF-8 sequence - for (b = 0, d = c; d; d /= 2) b++; // bits in c - b = (b - 2) / 5; // bytes in payload - u[l++] = (0xFF << (7 - b)) | (c >> (6 * b)); // head - while (b) u[l++] = 0x80 | ((c >> (6 * --b)) & 0x3F); // payload + if (c < 0x80) u[l++] = c; /* US-ASCII subset */ + else { /* multi-byte UTF-8 sequence */ + for (b = 0, d = c; d; d /= 2) b++; /* bits in c */ + b = (b - 2) / 5; /* bytes in payload */ + u[l++] = (0xFF << (7 - b)) | (c >> (6 * b)); /* head */ + while (b) u[l++] = 0x80 | ((c >> (6 * --b)) & 0x3F); /* payload */ } } return *s = realloc(u, *len = l); } -// frees a tag attribute list +/* frees a tag attribute list */ void ezxml_free_attr(char **attr) { int i = 0; char *m; - if (! attr || attr == EZXML_NIL) return; // nothing to free - while (attr[i]) i += 2; // find end of attribute list - m = attr[i + 1]; // list of which names and values are malloced + if (! attr || attr == EZXML_NIL) return; /* nothing to free */ + while (attr[i]) i += 2; /* find end of attribute list */ + m = attr[i + 1]; /* list of which names and values are malloced */ for (i = 0; m[i]; i++) { if (m[i] & EZXML_NAMEM) free(attr[i * 2]); if (m[i] & EZXML_TXTM) free(attr[(i * 2) + 1]); @@ -469,56 +469,56 @@ void ezxml_free_attr(char **attr) { free(attr); } -// parse the given xml string and return an ezxml structure +/* parse the given xml string and return an ezxml structure */ ezxml_t ezxml_parse_str(char *s, size_t len) { ezxml_root_t root = (ezxml_root_t)ezxml_new(NULL); - char q, e, *d, **attr, **a = NULL; // initialize a to avoid compile warning + char q, e, *d, **attr, **a = NULL; /* initialize a to avoid compile warning */ int l, i, j; root->m = s; if (! len) return ezxml_err(root, NULL, "root tag missing"); - root->u = ezxml_str2utf8(&s, &len); // convert utf-16 to utf-8 - root->e = (root->s = s) + len; // record start and end of work area + root->u = ezxml_str2utf8(&s, &len); /* convert utf-16 to utf-8 */ + root->e = (root->s = s) + len; /* record start and end of work area */ - e = s[len - 1]; // save end char - s[len - 1] = '\0'; // turn end char into null terminator + e = s[len - 1]; /* save end char */ + s[len - 1] = '\0'; /* turn end char into null terminator */ - while (*s && *s != '<') s++; // find first tag + while (*s && *s != '<') s++; /* find first tag */ if (! *s) return ezxml_err(root, s, "root tag missing"); for (; ; ) { attr = (char **)EZXML_NIL; d = ++s; - if (isalpha(*s) || *s == '_' || *s == ':' || *s < '\0') { // new tag + if (isalpha(*s) || *s == '_' || *s == ':' || *s < '\0') { /* new tag */ if (! root->cur) return ezxml_err(root, d, "markup outside of root element"); s += strcspn(s, EZXML_WS "/>"); - while (isspace(*s)) *(s++) = '\0'; // null terminate tag name + while (isspace(*s)) *(s++) = '\0'; /* null terminate tag name */ - if (*s && *s != '/' && *s != '>') // find tag in default attr list + if (*s && *s != '/' && *s != '>') /* find tag in default attr list */ for (i = 0; (a = root->attr[i]) && strcmp(a[0], d); i++); - for (l = 0; *s && *s != '/' && *s != '>'; l += 2) { // new attrib + for (l = 0; *s && *s != '/' && *s != '>'; l += 2) { /* new attrib */ attr = (l) ? realloc(attr, (l + 4) * sizeof(char *)) - : malloc(4 * sizeof(char *)); // allocate space + : malloc(4 * sizeof(char *)); /* allocate space */ attr[l + 3] = (l) ? realloc(attr[l + 1], (l / 2) + 2) - : malloc(2); // mem for list of maloced vals - strcpy(attr[l + 3] + (l / 2), " "); // value is not malloced - attr[l + 2] = NULL; // null terminate list - attr[l + 1] = ""; // temporary attribute value - attr[l] = s; // set attribute name + : malloc(2); /* mem for list of maloced vals */ + strcpy(attr[l + 3] + (l / 2), " "); /* value is not malloced */ + attr[l + 2] = NULL; /* null terminate list */ + attr[l + 1] = ""; /* temporary attribute value */ + attr[l] = s; /* set attribute name */ s += strcspn(s, EZXML_WS "=/>"); if (*s == '=' || isspace(*s)) { - *(s++) = '\0'; // null terminate tag attribute name + *(s++) = '\0'; /* null terminate tag attribute name */ q = *(s += strspn(s, EZXML_WS "=")); - if (q == '"' || q == '\'') { // attribute value + if (q == '"' || q == '\'') { /* attribute value */ attr[l + 1] = ++s; while (*s && *s != q) s++; - if (*s) *(s++) = '\0'; // null terminate attribute val + if (*s) *(s++) = '\0'; /* null terminate attribute val */ else { ezxml_free_attr(attr); return ezxml_err(root, d, "missing %c", q); @@ -528,13 +528,13 @@ ezxml_t ezxml_parse_str(char *s, size_t len) attr[l + 1] = ezxml_decode(attr[l + 1], root->ent, (a && a[j]) ? *a[j + 2] : ' '); if (attr[l + 1] < d || attr[l + 1] > s) - attr[l + 3][l / 2] = EZXML_TXTM; // value malloced + attr[l + 3][l / 2] = EZXML_TXTM; /* value malloced */ } } while (isspace(*s)) s++; } - if (*s == '/') { // self closing tag + if (*s == '/') { /* self closing tag */ *(s++) = '\0'; if ((*s && *s != '>') || (! *s && e != '>')) { if (l) ezxml_free_attr(attr); @@ -543,8 +543,8 @@ ezxml_t ezxml_parse_str(char *s, size_t len) ezxml_open_tag(root, d, attr); ezxml_close_tag(root, d, s); } - else if ((q = *s) == '>' || (! *s && e == '>')) { // open tag - *s = '\0'; // temporarily null terminate tag name + else if ((q = *s) == '>' || (! *s && e == '>')) { /* open tag */ + *s = '\0'; /* temporarily null terminate tag name */ ezxml_open_tag(root, d, attr); *s = q; } @@ -553,23 +553,23 @@ ezxml_t ezxml_parse_str(char *s, size_t len) return ezxml_err(root, d, "missing >"); } } - else if (*s == '/') { // close tag + else if (*s == '/') { /* close tag */ s += strcspn(d = s + 1, EZXML_WS ">") + 1; if (! (q = *s) && e != '>') return ezxml_err(root, d, "missing >"); - *s = '\0'; // temporarily null terminate tag name + *s = '\0'; /* temporarily null terminate tag name */ if (ezxml_close_tag(root, d, s)) return &root->xml; if (isspace(*s = q)) s += strspn(s, EZXML_WS); } - else if (! strncmp(s, "!--", 3)) { // xml comment + else if (! strncmp(s, "!--", 3)) { /* xml comment */ if (! (s = strstr(s + 3, "--")) || (*(s += 2) != '>' && *s) || (! *s && e != '>')) return ezxml_err(root, d, "unclosed