/**
- * uri.c: set of generic URI related routines
+ * uri.c: set of generic URI related routines
*
- * Reference: RFCs 2396, 2732 and 2373
+ * Reference: RFCs 3986, 2732 and 2373
*
* See Copyright for the status of this software.
*
#include <libxml/globals.h>
#include <libxml/xmlerror.h>
-/************************************************************************
- * *
- * Macros to differentiate various character type *
- * directly extracted from RFC 2396 *
- * *
- ************************************************************************/
+/**
+ * MAX_URI_LENGTH:
+ *
+ * The definition of the URI regexp in the above RFC has no size limit
+ * In practice they are usually relativey short except for the
+ * data URI scheme as defined in RFC 2397. Even for data URI the usual
+ * maximum size before hitting random practical limits is around 64 KB
+ * and 4KB is usually a maximum admitted limit for proper operations.
+ * The value below is more a security limit than anything else and
+ * really should never be hit by 'normal' operations
+ * Set to 1 MByte in 2012, this is only enforced on output
+ */
+#define MAX_URI_LENGTH 1024 * 1024
+
+static void
+xmlURIErrMemory(const char *extra)
+{
+ if (extra)
+ __xmlRaiseError(NULL, NULL, NULL,
+ NULL, NULL, XML_FROM_URI,
+ XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0,
+ extra, NULL, NULL, 0, 0,
+ "Memory allocation failed : %s\n", extra);
+ else
+ __xmlRaiseError(NULL, NULL, NULL,
+ NULL, NULL, XML_FROM_URI,
+ XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0,
+ NULL, NULL, NULL, 0, 0,
+ "Memory allocation failed\n");
+}
+
+static void xmlCleanURI(xmlURIPtr uri);
/*
+ * Old rule from 2396 used in legacy handling code
* alpha = lowalpha | upalpha
*/
#define IS_ALPHA(x) (IS_LOWALPHA(x) || IS_UPALPHA(x))
#define IS_ALPHANUM(x) (IS_ALPHA(x) || IS_DIGIT(x))
/*
- * hex = digit | "A" | "B" | "C" | "D" | "E" | "F" |
- * "a" | "b" | "c" | "d" | "e" | "f"
- */
-
-#define IS_HEX(x) ((IS_DIGIT(x)) || (((x) >= 'a') && ((x) <= 'f')) || \
- (((x) >= 'A') && ((x) <= 'F')))
-
-/*
* mark = "-" | "_" | "." | "!" | "~" | "*" | "'" | "(" | ")"
*/
-#define IS_MARK(x) (((x) == '-') || ((x) == '_') || ((x) == '.') || \
- ((x) == '!') || ((x) == '~') || ((x) == '*') || ((x) == '\'') || \
+#define IS_MARK(x) (((x) == '-') || ((x) == '_') || ((x) == '.') || \
+ ((x) == '!') || ((x) == '~') || ((x) == '*') || ((x) == '\'') || \
((x) == '(') || ((x) == ')'))
+/*
+ * unwise = "{" | "}" | "|" | "\" | "^" | "`"
+ */
+#define IS_UNWISE(p) \
+ (((*(p) == '{')) || ((*(p) == '}')) || ((*(p) == '|')) || \
+ ((*(p) == '\\')) || ((*(p) == '^')) || ((*(p) == '[')) || \
+ ((*(p) == ']')) || ((*(p) == '`')))
/*
* reserved = ";" | "/" | "?" | ":" | "@" | "&" | "=" | "+" | "$" | "," |
- * "[" | "]"
+ * "[" | "]"
*/
-#define IS_RESERVED(x) (((x) == ';') || ((x) == '/') || ((x) == '?') || \
- ((x) == ':') || ((x) == '@') || ((x) == '&') || ((x) == '=') || \
- ((x) == '+') || ((x) == '$') || ((x) == ',') || ((x) == '[') || \
- ((x) == ']'))
+#define IS_RESERVED(x) (((x) == ';') || ((x) == '/') || ((x) == '?') || \
+ ((x) == ':') || ((x) == '@') || ((x) == '&') || ((x) == '=') || \
+ ((x) == '+') || ((x) == '$') || ((x) == ',') || ((x) == '[') || \
+ ((x) == ']'))
/*
* unreserved = alphanum | mark
#define IS_UNRESERVED(x) (IS_ALPHANUM(x) || IS_MARK(x))
/*
- * escaped = "%" hex hex
+ * Skip to next pointer char, handle escaped sequences
*/
-#define IS_ESCAPED(p) ((*(p) == '%') && (IS_HEX((p)[1])) && \
- (IS_HEX((p)[2])))
+#define NEXT(p) ((*p == '%')? p += 3 : p++)
/*
- * uric_no_slash = unreserved | escaped | ";" | "?" | ":" | "@" |
- * "&" | "=" | "+" | "$" | ","
+ * Productions from the spec.
+ *
+ * authority = server | reg_name
+ * reg_name = 1*( unreserved | escaped | "$" | "," |
+ * ";" | ":" | "@" | "&" | "=" | "+" )
+ *
+ * path = [ abs_path | opaque_part ]
*/
-#define IS_URIC_NO_SLASH(p) ((IS_UNRESERVED(*(p))) || (IS_ESCAPED(p)) ||\
- ((*(p) == ';')) || ((*(p) == '?')) || ((*(p) == ':')) ||\
- ((*(p) == '@')) || ((*(p) == '&')) || ((*(p) == '=')) ||\
- ((*(p) == '+')) || ((*(p) == '$')) || ((*(p) == ',')))
+
+#define STRNDUP(s, n) (char *) xmlStrndup((const xmlChar *)(s), (n))
+
+/************************************************************************
+ * *
+ * RFC 3986 parser *
+ * *
+ ************************************************************************/
+
+#define ISA_DIGIT(p) ((*(p) >= '0') && (*(p) <= '9'))
+#define ISA_ALPHA(p) (((*(p) >= 'a') && (*(p) <= 'z')) || \
+ ((*(p) >= 'A') && (*(p) <= 'Z')))
+#define ISA_HEXDIG(p) \
+ (ISA_DIGIT(p) || ((*(p) >= 'a') && (*(p) <= 'f')) || \
+ ((*(p) >= 'A') && (*(p) <= 'F')))
/*
- * pchar = unreserved | escaped | ":" | "@" | "&" | "=" | "+" | "$" | ","
+ * sub-delims = "!" / "$" / "&" / "'" / "(" / ")"
+ * / "*" / "+" / "," / ";" / "="
*/
-#define IS_PCHAR(p) ((IS_UNRESERVED(*(p))) || (IS_ESCAPED(p)) || \
- ((*(p) == ':')) || ((*(p) == '@')) || ((*(p) == '&')) ||\
- ((*(p) == '=')) || ((*(p) == '+')) || ((*(p) == '$')) ||\
- ((*(p) == ',')))
+#define ISA_SUB_DELIM(p) \
+ (((*(p) == '!')) || ((*(p) == '$')) || ((*(p) == '&')) || \
+ ((*(p) == '(')) || ((*(p) == ')')) || ((*(p) == '*')) || \
+ ((*(p) == '+')) || ((*(p) == ',')) || ((*(p) == ';')) || \
+ ((*(p) == '=')) || ((*(p) == '\'')))
/*
- * rel_segment = 1*( unreserved | escaped |
- * ";" | "@" | "&" | "=" | "+" | "$" | "," )
+ * gen-delims = ":" / "/" / "?" / "#" / "[" / "]" / "@"
*/
-
-#define IS_SEGMENT(p) ((IS_UNRESERVED(*(p))) || (IS_ESCAPED(p)) || \
- ((*(p) == ';')) || ((*(p) == '@')) || ((*(p) == '&')) || \
- ((*(p) == '=')) || ((*(p) == '+')) || ((*(p) == '$')) || \
- ((*(p) == ',')))
+#define ISA_GEN_DELIM(p) \
+ (((*(p) == ':')) || ((*(p) == '/')) || ((*(p) == '?')) || \
+ ((*(p) == '#')) || ((*(p) == '[')) || ((*(p) == ']')) || \
+ ((*(p) == '@')))
/*
- * scheme = alpha *( alpha | digit | "+" | "-" | "." )
+ * reserved = gen-delims / sub-delims
*/
-
-#define IS_SCHEME(x) ((IS_ALPHA(x)) || (IS_DIGIT(x)) || \
- ((x) == '+') || ((x) == '-') || ((x) == '.'))
+#define ISA_RESERVED(p) (ISA_GEN_DELIM(p) || (ISA_SUB_DELIM(p)))
/*
- * reg_name = 1*( unreserved | escaped | "$" | "," |
- * ";" | ":" | "@" | "&" | "=" | "+" )
+ * unreserved = ALPHA / DIGIT / "-" / "." / "_" / "~"
*/
-
-#define IS_REG_NAME(p) ((IS_UNRESERVED(*(p))) || (IS_ESCAPED(p)) || \
- ((*(p) == '$')) || ((*(p) == ',')) || ((*(p) == ';')) || \
- ((*(p) == ':')) || ((*(p) == '@')) || ((*(p) == '&')) || \
- ((*(p) == '=')) || ((*(p) == '+')))
+#define ISA_UNRESERVED(p) \
+ ((ISA_ALPHA(p)) || (ISA_DIGIT(p)) || ((*(p) == '-')) || \
+ ((*(p) == '.')) || ((*(p) == '_')) || ((*(p) == '~')))
/*
- * userinfo = *( unreserved | escaped | ";" | ":" | "&" | "=" |
- * "+" | "$" | "," )
+ * pct-encoded = "%" HEXDIG HEXDIG
*/
-#define IS_USERINFO(p) ((IS_UNRESERVED(*(p))) || (IS_ESCAPED(p)) || \
- ((*(p) == ';')) || ((*(p) == ':')) || ((*(p) == '&')) || \
- ((*(p) == '=')) || ((*(p) == '+')) || ((*(p) == '$')) || \
- ((*(p) == ',')))
+#define ISA_PCT_ENCODED(p) \
+ ((*(p) == '%') && (ISA_HEXDIG(p + 1)) && (ISA_HEXDIG(p + 2)))
/*
- * uric = reserved | unreserved | escaped
+ * pchar = unreserved / pct-encoded / sub-delims / ":" / "@"
*/
+#define ISA_PCHAR(p) \
+ (ISA_UNRESERVED(p) || ISA_PCT_ENCODED(p) || ISA_SUB_DELIM(p) || \
+ ((*(p) == ':')) || ((*(p) == '@')))
-#define IS_URIC(p) ((IS_UNRESERVED(*(p))) || (IS_ESCAPED(p)) || \
- (IS_RESERVED(*(p))))
+/**
+ * xmlParse3986Scheme:
+ * @uri: pointer to an URI structure
+ * @str: pointer to the string to analyze
+ *
+ * Parse an URI scheme
+ *
+ * ALPHA *( ALPHA / DIGIT / "+" / "-" / "." )
+ *
+ * Returns 0 or the error code
+ */
+static int
+xmlParse3986Scheme(xmlURIPtr uri, const char **str) {
+ const char *cur;
-/*
-* unwise = "{" | "}" | "|" | "\" | "^" | "`"
-*/
+ if (str == NULL)
+ return(-1);
-#define IS_UNWISE(p) \
- (((*(p) == '{')) || ((*(p) == '}')) || ((*(p) == '|')) || \
- ((*(p) == '\\')) || ((*(p) == '^')) || ((*(p) == '[')) || \
- ((*(p) == ']')) || ((*(p) == '`')))
+ cur = *str;
+ if (!ISA_ALPHA(cur))
+ return(2);
+ cur++;
+ while (ISA_ALPHA(cur) || ISA_DIGIT(cur) ||
+ (*cur == '+') || (*cur == '-') || (*cur == '.')) cur++;
+ if (uri != NULL) {
+ if (uri->scheme != NULL) xmlFree(uri->scheme);
+ uri->scheme = STRNDUP(*str, cur - *str);
+ }
+ *str = cur;
+ return(0);
+}
-/*
- * Skip to next pointer char, handle escaped sequences
+/**
+ * xmlParse3986Fragment:
+ * @uri: pointer to an URI structure
+ * @str: pointer to the string to analyze
+ *
+ * Parse the query part of an URI
+ *
+ * fragment = *( pchar / "/" / "?" )
+ * NOTE: the strict syntax as defined by 3986 does not allow '[' and ']'
+ * in the fragment identifier but this is used very broadly for
+ * xpointer scheme selection, so we are allowing it here to not break
+ * for example all the DocBook processing chains.
+ *
+ * Returns 0 or the error code
*/
+static int
+xmlParse3986Fragment(xmlURIPtr uri, const char **str)
+{
+ const char *cur;
-#define NEXT(p) ((*p == '%')? p += 3 : p++)
+ if (str == NULL)
+ return (-1);
-/*
- * Productions from the spec.
+ cur = *str;
+
+ while ((ISA_PCHAR(cur)) || (*cur == '/') || (*cur == '?') ||
+ (*cur == '[') || (*cur == ']') ||
+ ((uri != NULL) && (uri->cleanup & 1) && (IS_UNWISE(cur))))
+ NEXT(cur);
+ if (uri != NULL) {
+ if (uri->fragment != NULL)
+ xmlFree(uri->fragment);
+ if (uri->cleanup & 2)
+ uri->fragment = STRNDUP(*str, cur - *str);
+ else
+ uri->fragment = xmlURIUnescapeString(*str, cur - *str, NULL);
+ }
+ *str = cur;
+ return (0);
+}
+
+/**
+ * xmlParse3986Query:
+ * @uri: pointer to an URI structure
+ * @str: pointer to the string to analyze
*
- * authority = server | reg_name
- * reg_name = 1*( unreserved | escaped | "$" | "," |
- * ";" | ":" | "@" | "&" | "=" | "+" )
+ * Parse the query part of an URI
*
- * path = [ abs_path | opaque_part ]
+ * query = *uric
+ *
+ * Returns 0 or the error code
*/
+static int
+xmlParse3986Query(xmlURIPtr uri, const char **str)
+{
+ const char *cur;
-#define STRNDUP(s, n) (char *) xmlStrndup((const xmlChar *)(s), (n))
+ if (str == NULL)
+ return (-1);
-/************************************************************************
- * *
- * Generic URI structure functions *
- * *
- ************************************************************************/
+ cur = *str;
+
+ while ((ISA_PCHAR(cur)) || (*cur == '/') || (*cur == '?') ||
+ ((uri != NULL) && (uri->cleanup & 1) && (IS_UNWISE(cur))))
+ NEXT(cur);
+ if (uri != NULL) {
+ if (uri->query != NULL)
+ xmlFree(uri->query);
+ if (uri->cleanup & 2)
+ uri->query = STRNDUP(*str, cur - *str);
+ else
+ uri->query = xmlURIUnescapeString(*str, cur - *str, NULL);
+
+ /* Save the raw bytes of the query as well.
+ * See: http://mail.gnome.org/archives/xml/2007-April/thread.html#00114
+ */
+ if (uri->query_raw != NULL)
+ xmlFree (uri->query_raw);
+ uri->query_raw = STRNDUP (*str, cur - *str);
+ }
+ *str = cur;
+ return (0);
+}
/**
- * xmlCreateURI:
+ * xmlParse3986Port:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
*
- * Simply creates an empty xmlURI
+ * Parse a port part and fills in the appropriate fields
+ * of the @uri structure
*
- * Returns the new structure or NULL in case of error
+ * port = *DIGIT
+ *
+ * Returns 0 or the error code
*/
-xmlURIPtr
-xmlCreateURI(void) {
- xmlURIPtr ret;
+static int
+xmlParse3986Port(xmlURIPtr uri, const char **str)
+{
+ const char *cur = *str;
- ret = (xmlURIPtr) xmlMalloc(sizeof(xmlURI));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlCreateURI: out of memory\n");
- return(NULL);
+ if (ISA_DIGIT(cur)) {
+ if (uri != NULL)
+ uri->port = 0;
+ while (ISA_DIGIT(cur)) {
+ if (uri != NULL)
+ uri->port = uri->port * 10 + (*cur - '0');
+ cur++;
+ }
+ *str = cur;
+ return(0);
}
- memset(ret, 0, sizeof(xmlURI));
- return(ret);
+ return(1);
}
/**
- * xmlSaveUri:
- * @uri: pointer to an xmlURI
+ * xmlParse3986Userinfo:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
*
- * Save the URI as an escaped string
+ * Parse an user informations part and fills in the appropriate fields
+ * of the @uri structure
*
- * Returns a new string (to be deallocated by caller)
+ * userinfo = *( unreserved / pct-encoded / sub-delims / ":" )
+ *
+ * Returns 0 or the error code
*/
-xmlChar *
-xmlSaveUri(xmlURIPtr uri) {
- xmlChar *ret = NULL;
- const char *p;
- int len;
- int max;
+static int
+xmlParse3986Userinfo(xmlURIPtr uri, const char **str)
+{
+ const char *cur;
- if (uri == NULL) return(NULL);
+ cur = *str;
+ while (ISA_UNRESERVED(cur) || ISA_PCT_ENCODED(cur) ||
+ ISA_SUB_DELIM(cur) || (*cur == ':'))
+ NEXT(cur);
+ if (*cur == '@') {
+ if (uri != NULL) {
+ if (uri->user != NULL) xmlFree(uri->user);
+ if (uri->cleanup & 2)
+ uri->user = STRNDUP(*str, cur - *str);
+ else
+ uri->user = xmlURIUnescapeString(*str, cur - *str, NULL);
+ }
+ *str = cur;
+ return(0);
+ }
+ return(1);
+}
+/**
+ * xmlParse3986DecOctet:
+ * @str: the string to analyze
+ *
+ * dec-octet = DIGIT ; 0-9
+ * / %x31-39 DIGIT ; 10-99
+ * / "1" 2DIGIT ; 100-199
+ * / "2" %x30-34 DIGIT ; 200-249
+ * / "25" %x30-35 ; 250-255
+ *
+ * Skip a dec-octet.
+ *
+ * Returns 0 if found and skipped, 1 otherwise
+ */
+static int
+xmlParse3986DecOctet(const char **str) {
+ const char *cur = *str;
- max = 80;
- ret = (xmlChar *) xmlMallocAtomic((max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- len = 0;
+ if (!(ISA_DIGIT(cur)))
+ return(1);
+ if (!ISA_DIGIT(cur+1))
+ cur++;
+ else if ((*cur != '0') && (ISA_DIGIT(cur + 1)) && (!ISA_DIGIT(cur+2)))
+ cur += 2;
+ else if ((*cur == '1') && (ISA_DIGIT(cur + 1)) && (ISA_DIGIT(cur + 2)))
+ cur += 3;
+ else if ((*cur == '2') && (*(cur + 1) >= '0') &&
+ (*(cur + 1) <= '4') && (ISA_DIGIT(cur + 2)))
+ cur += 3;
+ else if ((*cur == '2') && (*(cur + 1) == '5') &&
+ (*(cur + 2) >= '0') && (*(cur + 1) <= '5'))
+ cur += 3;
+ else
+ return(1);
+ *str = cur;
+ return(0);
+}
+/**
+ * xmlParse3986Host:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
+ *
+ * Parse an host part and fills in the appropriate fields
+ * of the @uri structure
+ *
+ * host = IP-literal / IPv4address / reg-name
+ * IP-literal = "[" ( IPv6address / IPvFuture ) "]"
+ * IPv4address = dec-octet "." dec-octet "." dec-octet "." dec-octet
+ * reg-name = *( unreserved / pct-encoded / sub-delims )
+ *
+ * Returns 0 or the error code
+ */
+static int
+xmlParse3986Host(xmlURIPtr uri, const char **str)
+{
+ const char *cur = *str;
+ const char *host;
- if (uri->scheme != NULL) {
- p = uri->scheme;
- while (*p != 0) {
- if (len >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret, (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- ret[len++] = *p++;
- }
- if (len >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret, (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- ret[len++] = ':';
+ host = cur;
+ /*
+ * IPv6 and future adressing scheme are enclosed between brackets
+ */
+ if (*cur == '[') {
+ cur++;
+ while ((*cur != ']') && (*cur != 0))
+ cur++;
+ if (*cur != ']')
+ return(1);
+ cur++;
+ goto found;
}
- if (uri->opaque != NULL) {
- p = uri->opaque;
- while (*p != 0) {
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret, (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- if (IS_RESERVED(*(p)) || IS_UNRESERVED(*(p)))
- ret[len++] = *p++;
- else {
- int val = *(unsigned char *)p++;
- int hi = val / 0x10, lo = val % 0x10;
- ret[len++] = '%';
- ret[len++] = hi + (hi > 9? 'A'-10 : '0');
- ret[len++] = lo + (lo > 9? 'A'-10 : '0');
- }
- }
- } else {
- if (uri->server != NULL) {
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret, (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- ret[len++] = '/';
- ret[len++] = '/';
- if (uri->user != NULL) {
- p = uri->user;
- while (*p != 0) {
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- if ((IS_UNRESERVED(*(p))) ||
- ((*(p) == ';')) || ((*(p) == ':')) ||
- ((*(p) == '&')) || ((*(p) == '=')) ||
- ((*(p) == '+')) || ((*(p) == '$')) ||
- ((*(p) == ',')))
- ret[len++] = *p++;
- else {
- int val = *(unsigned char *)p++;
- int hi = val / 0x10, lo = val % 0x10;
- ret[len++] = '%';
- ret[len++] = hi + (hi > 9? 'A'-10 : '0');
- ret[len++] = lo + (lo > 9? 'A'-10 : '0');
- }
- }
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- ret[len++] = '@';
- }
- p = uri->server;
- while (*p != 0) {
- if (len >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- ret[len++] = *p++;
- }
- if (uri->port > 0) {
- if (len + 10 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- len += snprintf((char *) &ret[len], max - len, ":%d", uri->port);
- }
- } else if (uri->authority != NULL) {
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- ret[len++] = '/';
- ret[len++] = '/';
- p = uri->authority;
- while (*p != 0) {
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- if ((IS_UNRESERVED(*(p))) ||
- ((*(p) == '$')) || ((*(p) == ',')) || ((*(p) == ';')) ||
- ((*(p) == ':')) || ((*(p) == '@')) || ((*(p) == '&')) ||
- ((*(p) == '=')) || ((*(p) == '+')))
- ret[len++] = *p++;
- else {
- int val = *(unsigned char *)p++;
- int hi = val / 0x10, lo = val % 0x10;
- ret[len++] = '%';
- ret[len++] = hi + (hi > 9? 'A'-10 : '0');
- ret[len++] = lo + (lo > 9? 'A'-10 : '0');
- }
- }
- } else if (uri->scheme != NULL) {
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- ret[len++] = '/';
- ret[len++] = '/';
- }
- if (uri->path != NULL) {
- p = uri->path;
- while (*p != 0) {
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- if ((IS_UNRESERVED(*(p))) || ((*(p) == '/')) ||
- ((*(p) == ';')) || ((*(p) == '@')) || ((*(p) == '&')) ||
- ((*(p) == '=')) || ((*(p) == '+')) || ((*(p) == '$')) ||
- ((*(p) == ',')))
- ret[len++] = *p++;
- else {
- int val = *(unsigned char *)p++;
- int hi = val / 0x10, lo = val % 0x10;
- ret[len++] = '%';
- ret[len++] = hi + (hi > 9? 'A'-10 : '0');
- ret[len++] = lo + (lo > 9? 'A'-10 : '0');
- }
- }
- }
- if (uri->query != NULL) {
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- ret[len++] = '?';
- p = uri->query;
- while (*p != 0) {
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- if ((IS_UNRESERVED(*(p))) || (IS_RESERVED(*(p))))
- ret[len++] = *p++;
- else {
- int val = *(unsigned char *)p++;
- int hi = val / 0x10, lo = val % 0x10;
- ret[len++] = '%';
- ret[len++] = hi + (hi > 9? 'A'-10 : '0');
- ret[len++] = lo + (lo > 9? 'A'-10 : '0');
- }
- }
- }
- }
- if (uri->fragment != NULL) {
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- ret[len++] = '#';
- p = uri->fragment;
- while (*p != 0) {
- if (len + 3 >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret,
- (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- if ((IS_UNRESERVED(*(p))) || (IS_RESERVED(*(p))))
- ret[len++] = *p++;
- else {
- int val = *(unsigned char *)p++;
- int hi = val / 0x10, lo = val % 0x10;
- ret[len++] = '%';
- ret[len++] = hi + (hi > 9? 'A'-10 : '0');
- ret[len++] = lo + (lo > 9? 'A'-10 : '0');
- }
- }
- }
- if (len >= max) {
- max *= 2;
- ret = (xmlChar *) xmlRealloc(ret, (max + 1) * sizeof(xmlChar));
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlSaveUri: out of memory\n");
- return(NULL);
- }
- }
- ret[len++] = 0;
- return(ret);
-}
-
-/**
- * xmlPrintURI:
- * @stream: a FILE* for the output
- * @uri: pointer to an xmlURI
- *
- * Prints the URI in the stream @stream.
- */
-void
-xmlPrintURI(FILE *stream, xmlURIPtr uri) {
- xmlChar *out;
-
- out = xmlSaveUri(uri);
- if (out != NULL) {
- fprintf(stream, "%s", (char *) out);
- xmlFree(out);
- }
-}
-
-/**
- * xmlCleanURI:
- * @uri: pointer to an xmlURI
- *
- * Make sure the xmlURI struct is free of content
- */
-static void
-xmlCleanURI(xmlURIPtr uri) {
- if (uri == NULL) return;
-
- if (uri->scheme != NULL) xmlFree(uri->scheme);
- uri->scheme = NULL;
- if (uri->server != NULL) xmlFree(uri->server);
- uri->server = NULL;
- if (uri->user != NULL) xmlFree(uri->user);
- uri->user = NULL;
- if (uri->path != NULL) xmlFree(uri->path);
- uri->path = NULL;
- if (uri->fragment != NULL) xmlFree(uri->fragment);
- uri->fragment = NULL;
- if (uri->opaque != NULL) xmlFree(uri->opaque);
- uri->opaque = NULL;
- if (uri->authority != NULL) xmlFree(uri->authority);
- uri->authority = NULL;
- if (uri->query != NULL) xmlFree(uri->query);
- uri->query = NULL;
-}
-
-/**
- * xmlFreeURI:
- * @uri: pointer to an xmlURI
- *
- * Free up the xmlURI struct
- */
-void
-xmlFreeURI(xmlURIPtr uri) {
- if (uri == NULL) return;
-
- if (uri->scheme != NULL) xmlFree(uri->scheme);
- if (uri->server != NULL) xmlFree(uri->server);
- if (uri->user != NULL) xmlFree(uri->user);
- if (uri->path != NULL) xmlFree(uri->path);
- if (uri->fragment != NULL) xmlFree(uri->fragment);
- if (uri->opaque != NULL) xmlFree(uri->opaque);
- if (uri->authority != NULL) xmlFree(uri->authority);
- if (uri->query != NULL) xmlFree(uri->query);
- xmlFree(uri);
-}
-
-/************************************************************************
- * *
- * Helper functions *
- * *
- ************************************************************************/
-
-/**
- * xmlNormalizeURIPath:
- * @path: pointer to the path string
- *
- * Applies the 5 normalization steps to a path string--that is, RFC 2396
- * Section 5.2, steps 6.c through 6.g.
- *
- * Normalization occurs directly on the string, no new allocation is done
- *
- * Returns 0 or an error code
- */
-int
-xmlNormalizeURIPath(char *path) {
- char *cur, *out;
-
- if (path == NULL)
- return(-1);
-
- /* Skip all initial "/" chars. We want to get to the beginning of the
- * first non-empty segment.
- */
- cur = path;
- while (cur[0] == '/')
- ++cur;
- if (cur[0] == '\0')
- return(0);
-
- /* Keep everything we've seen so far. */
- out = cur;
-
- /*
- * Analyze each segment in sequence for cases (c) and (d).
- */
- while (cur[0] != '\0') {
- /*
- * c) All occurrences of "./", where "." is a complete path segment,
- * are removed from the buffer string.
- */
- if ((cur[0] == '.') && (cur[1] == '/')) {
- cur += 2;
- /* '//' normalization should be done at this point too */
- while (cur[0] == '/')
- cur++;
- continue;
- }
-
- /*
- * d) If the buffer string ends with "." as a complete path segment,
- * that "." is removed.
- */
- if ((cur[0] == '.') && (cur[1] == '\0'))
- break;
-
- /* Otherwise keep the segment. */
- while (cur[0] != '/') {
- if (cur[0] == '\0')
- goto done_cd;
- (out++)[0] = (cur++)[0];
- }
- /* nomalize // */
- while ((cur[0] == '/') && (cur[1] == '/'))
- cur++;
-
- (out++)[0] = (cur++)[0];
- }
- done_cd:
- out[0] = '\0';
-
- /* Reset to the beginning of the first segment for the next sequence. */
- cur = path;
- while (cur[0] == '/')
- ++cur;
- if (cur[0] == '\0')
- return(0);
-
- /*
- * Analyze each segment in sequence for cases (e) and (f).
- *
- * e) All occurrences of "<segment>/../", where <segment> is a
- * complete path segment not equal to "..", are removed from the
- * buffer string. Removal of these path segments is performed
- * iteratively, removing the leftmost matching pattern on each
- * iteration, until no matching pattern remains.
- *
- * f) If the buffer string ends with "<segment>/..", where <segment>
- * is a complete path segment not equal to "..", that
- * "<segment>/.." is removed.
- *
- * To satisfy the "iterative" clause in (e), we need to collapse the
- * string every time we find something that needs to be removed. Thus,
- * we don't need to keep two pointers into the string: we only need a
- * "current position" pointer.
- */
- while (1) {
- char *segp, *tmp;
-
- /* At the beginning of each iteration of this loop, "cur" points to
- * the first character of the segment we want to examine.
- */
-
- /* Find the end of the current segment. */
- segp = cur;
- while ((segp[0] != '/') && (segp[0] != '\0'))
- ++segp;
-
- /* If this is the last segment, we're done (we need at least two
- * segments to meet the criteria for the (e) and (f) cases).
- */
- if (segp[0] == '\0')
- break;
-
- /* If the first segment is "..", or if the next segment _isn't_ "..",
- * keep this segment and try the next one.
- */
- ++segp;
- if (((cur[0] == '.') && (cur[1] == '.') && (segp == cur+3))
- || ((segp[0] != '.') || (segp[1] != '.')
- || ((segp[2] != '/') && (segp[2] != '\0')))) {
- cur = segp;
- continue;
- }
-
- /* If we get here, remove this segment and the next one and back up
- * to the previous segment (if there is one), to implement the
- * "iteratively" clause. It's pretty much impossible to back up
- * while maintaining two pointers into the buffer, so just compact
- * the whole buffer now.
- */
-
- /* If this is the end of the buffer, we're done. */
- if (segp[2] == '\0') {
- cur[0] = '\0';
- break;
- }
- /* Valgrind complained, strcpy(cur, segp + 3); */
- /* string will overlap, do not use strcpy */
- tmp = cur;
- segp += 3;
- while ((*tmp++ = *segp++) != 0);
-
- /* If there are no previous segments, then keep going from here. */
- segp = cur;
- while ((segp > path) && ((--segp)[0] == '/'))
- ;
- if (segp == path)
- continue;
-
- /* "segp" is pointing to the end of a previous segment; find it's
- * start. We need to back up to the previous segment and start
- * over with that to handle things like "foo/bar/../..". If we
- * don't do this, then on the first pass we'll remove the "bar/..",
- * but be pointing at the second ".." so we won't realize we can also
- * remove the "foo/..".
- */
- cur = segp;
- while ((cur > path) && (cur[-1] != '/'))
- --cur;
+ /*
+ * try to parse an IPv4
+ */
+ if (ISA_DIGIT(cur)) {
+ if (xmlParse3986DecOctet(&cur) != 0)
+ goto not_ipv4;
+ if (*cur != '.')
+ goto not_ipv4;
+ cur++;
+ if (xmlParse3986DecOctet(&cur) != 0)
+ goto not_ipv4;
+ if (*cur != '.')
+ goto not_ipv4;
+ if (xmlParse3986DecOctet(&cur) != 0)
+ goto not_ipv4;
+ if (*cur != '.')
+ goto not_ipv4;
+ if (xmlParse3986DecOctet(&cur) != 0)
+ goto not_ipv4;
+ goto found;
+not_ipv4:
+ cur = *str;
}
- out[0] = '\0';
-
/*
- * g) If the resulting buffer string still begins with one or more
- * complete path segments of "..", then the reference is
- * considered to be in error. Implementations may handle this
- * error by retaining these components in the resolved path (i.e.,
- * treating them as part of the final URI), by removing them from
- * the resolved path (i.e., discarding relative levels above the
- * root), or by avoiding traversal of the reference.
- *
- * We discard them from the final path.
+ * then this should be a hostname which can be empty
*/
- if (path[0] == '/') {
- cur = path;
- while ((cur[0] == '/') && (cur[1] == '.') && (cur[2] == '.')
- && ((cur[3] == '/') || (cur[3] == '\0')))
- cur += 3;
-
- if (cur != path) {
- out = path;
- while (cur[0] != '\0')
- (out++)[0] = (cur++)[0];
- out[0] = 0;
- }
+ while (ISA_UNRESERVED(cur) || ISA_PCT_ENCODED(cur) || ISA_SUB_DELIM(cur))
+ NEXT(cur);
+found:
+ if (uri != NULL) {
+ if (uri->authority != NULL) xmlFree(uri->authority);
+ uri->authority = NULL;
+ if (uri->server != NULL) xmlFree(uri->server);
+ if (cur != host) {
+ if (uri->cleanup & 2)
+ uri->server = STRNDUP(host, cur - host);
+ else
+ uri->server = xmlURIUnescapeString(host, cur - host, NULL);
+ } else
+ uri->server = NULL;
}
-
+ *str = cur;
return(0);
}
-static int is_hex(char c) {
- if (((c >= '0') && (c <= '9')) ||
- ((c >= 'a') && (c <= 'f')) ||
- ((c >= 'A') && (c <= 'F')))
- return(1);
+/**
+ * xmlParse3986Authority:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
+ *
+ * Parse an authority part and fills in the appropriate fields
+ * of the @uri structure
+ *
+ * authority = [ userinfo "@" ] host [ ":" port ]
+ *
+ * Returns 0 or the error code
+ */
+static int
+xmlParse3986Authority(xmlURIPtr uri, const char **str)
+{
+ const char *cur;
+ int ret;
+
+ cur = *str;
+ /*
+ * try to parse an userinfo and check for the trailing @
+ */
+ ret = xmlParse3986Userinfo(uri, &cur);
+ if ((ret != 0) || (*cur != '@'))
+ cur = *str;
+ else
+ cur++;
+ ret = xmlParse3986Host(uri, &cur);
+ if (ret != 0) return(ret);
+ if (*cur == ':') {
+ cur++;
+ ret = xmlParse3986Port(uri, &cur);
+ if (ret != 0) return(ret);
+ }
+ *str = cur;
return(0);
}
/**
- * xmlURIUnescapeString:
- * @str: the string to unescape
- * @len: the length in bytes to unescape (or <= 0 to indicate full string)
- * @target: optional destination buffer
+ * xmlParse3986Segment:
+ * @str: the string to analyze
+ * @forbid: an optional forbidden character
+ * @empty: allow an empty segment
+ *
+ * Parse a segment and fills in the appropriate fields
+ * of the @uri structure
*
- * Unescaping routine, does not do validity checks !
- * Output is direct unsigned char translation of %XX values (no encoding)
+ * segment = *pchar
+ * segment-nz = 1*pchar
+ * segment-nz-nc = 1*( unreserved / pct-encoded / sub-delims / "@" )
+ * ; non-zero-length segment without any colon ":"
*
- * Returns an copy of the string, but unescaped
+ * Returns 0 or the error code
*/
-char *
-xmlURIUnescapeString(const char *str, int len, char *target) {
- char *ret, *out;
- const char *in;
-
- if (str == NULL)
- return(NULL);
- if (len <= 0) len = strlen(str);
- if (len < 0) return(NULL);
+static int
+xmlParse3986Segment(const char **str, char forbid, int empty)
+{
+ const char *cur;
- if (target == NULL) {
- ret = (char *) xmlMallocAtomic(len + 1);
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlURIUnescapeString: out of memory\n");
- return(NULL);
- }
- } else
- ret = target;
- in = str;
- out = ret;
- while(len > 0) {
- if ((len > 2) && (*in == '%') && (is_hex(in[1])) && (is_hex(in[2]))) {
- in++;
- if ((*in >= '0') && (*in <= '9'))
- *out = (*in - '0');
- else if ((*in >= 'a') && (*in <= 'f'))
- *out = (*in - 'a') + 10;
- else if ((*in >= 'A') && (*in <= 'F'))
- *out = (*in - 'A') + 10;
- in++;
- if ((*in >= '0') && (*in <= '9'))
- *out = *out * 16 + (*in - '0');
- else if ((*in >= 'a') && (*in <= 'f'))
- *out = *out * 16 + (*in - 'a') + 10;
- else if ((*in >= 'A') && (*in <= 'F'))
- *out = *out * 16 + (*in - 'A') + 10;
- in++;
- len -= 3;
- out++;
- } else {
- *out++ = *in++;
- len--;
- }
+ cur = *str;
+ if (!ISA_PCHAR(cur)) {
+ if (empty)
+ return(0);
+ return(1);
}
- *out = 0;
- return(ret);
+ while (ISA_PCHAR(cur) && (*cur != forbid))
+ NEXT(cur);
+ *str = cur;
+ return (0);
}
/**
- * xmlURIEscapeStr:
- * @str: string to escape
- * @list: exception list string of chars not to escape
+ * xmlParse3986PathAbEmpty:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
*
- * This routine escapes a string to hex, ignoring reserved characters (a-z)
- * and the characters in the exception list.
+ * Parse an path absolute or empty and fills in the appropriate fields
+ * of the @uri structure
*
- * Returns a new escaped string or NULL in case of error.
+ * path-abempty = *( "/" segment )
+ *
+ * Returns 0 or the error code
*/
-xmlChar *
-xmlURIEscapeStr(const xmlChar *str, const xmlChar *list) {
- xmlChar *ret, ch;
- const xmlChar *in;
-
- unsigned int len, out;
+static int
+xmlParse3986PathAbEmpty(xmlURIPtr uri, const char **str)
+{
+ const char *cur;
+ int ret;
- if (str == NULL)
- return(NULL);
- if (str[0] == 0)
- return(xmlStrdup(str));
- len = xmlStrlen(str);
- if (!(len > 0)) return(NULL);
+ cur = *str;
- len += 20;
- ret = (xmlChar *) xmlMallocAtomic(len);
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlURIEscapeStr: out of memory\n");
- return(NULL);
+ while (*cur == '/') {
+ cur++;
+ ret = xmlParse3986Segment(&cur, 0, 1);
+ if (ret != 0) return(ret);
}
- in = (const xmlChar *) str;
- out = 0;
- while(*in != 0) {
- if (len - out <= 3) {
- len += 20;
- ret = (xmlChar *) xmlRealloc(ret, len);
- if (ret == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlURIEscapeStr: out of memory\n");
- return(NULL);
- }
- }
+ if (uri != NULL) {
+ if (uri->path != NULL) xmlFree(uri->path);
+ if (*str != cur) {
+ if (uri->cleanup & 2)
+ uri->path = STRNDUP(*str, cur - *str);
+ else
+ uri->path = xmlURIUnescapeString(*str, cur - *str, NULL);
+ } else {
+ uri->path = NULL;
+ }
+ }
+ *str = cur;
+ return (0);
+}
- ch = *in;
+/**
+ * xmlParse3986PathAbsolute:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
+ *
+ * Parse an path absolute and fills in the appropriate fields
+ * of the @uri structure
+ *
+ * path-absolute = "/" [ segment-nz *( "/" segment ) ]
+ *
+ * Returns 0 or the error code
+ */
+static int
+xmlParse3986PathAbsolute(xmlURIPtr uri, const char **str)
+{
+ const char *cur;
+ int ret;
- if ((ch != '@') && (!IS_UNRESERVED(ch)) && (!xmlStrchr(list, ch))) {
- unsigned char val;
- ret[out++] = '%';
- val = ch >> 4;
- if (val <= 9)
- ret[out++] = '0' + val;
- else
- ret[out++] = 'A' + val - 0xA;
- val = ch & 0xF;
- if (val <= 9)
- ret[out++] = '0' + val;
- else
- ret[out++] = 'A' + val - 0xA;
- in++;
- } else {
- ret[out++] = *in++;
+ cur = *str;
+
+ if (*cur != '/')
+ return(1);
+ cur++;
+ ret = xmlParse3986Segment(&cur, 0, 0);
+ if (ret == 0) {
+ while (*cur == '/') {
+ cur++;
+ ret = xmlParse3986Segment(&cur, 0, 1);
+ if (ret != 0) return(ret);
}
+ }
+ if (uri != NULL) {
+ if (uri->path != NULL) xmlFree(uri->path);
+ if (cur != *str) {
+ if (uri->cleanup & 2)
+ uri->path = STRNDUP(*str, cur - *str);
+ else
+ uri->path = xmlURIUnescapeString(*str, cur - *str, NULL);
+ } else {
+ uri->path = NULL;
+ }
+ }
+ *str = cur;
+ return (0);
+}
+
+/**
+ * xmlParse3986PathRootless:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
+ *
+ * Parse an path without root and fills in the appropriate fields
+ * of the @uri structure
+ *
+ * path-rootless = segment-nz *( "/" segment )
+ *
+ * Returns 0 or the error code
+ */
+static int
+xmlParse3986PathRootless(xmlURIPtr uri, const char **str)
+{
+ const char *cur;
+ int ret;
+ cur = *str;
+
+ ret = xmlParse3986Segment(&cur, 0, 0);
+ if (ret != 0) return(ret);
+ while (*cur == '/') {
+ cur++;
+ ret = xmlParse3986Segment(&cur, 0, 1);
+ if (ret != 0) return(ret);
}
- ret[out] = 0;
- return(ret);
+ if (uri != NULL) {
+ if (uri->path != NULL) xmlFree(uri->path);
+ if (cur != *str) {
+ if (uri->cleanup & 2)
+ uri->path = STRNDUP(*str, cur - *str);
+ else
+ uri->path = xmlURIUnescapeString(*str, cur - *str, NULL);
+ } else {
+ uri->path = NULL;
+ }
+ }
+ *str = cur;
+ return (0);
}
/**
- * xmlURIEscape:
- * @str: the string of the URI to escape
+ * xmlParse3986PathNoScheme:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
*
- * Escaping routine, does not do validity checks !
- * It will try to escape the chars needing this, but this is heuristic
- * based it's impossible to be sure.
+ * Parse an path which is not a scheme and fills in the appropriate fields
+ * of the @uri structure
*
- * Returns an copy of the string, but escaped
+ * path-noscheme = segment-nz-nc *( "/" segment )
*
- * 25 May 2001
- * Uses xmlParseURI and xmlURIEscapeStr to try to escape correctly
- * according to RFC2396.
- * - Carl Douglas
+ * Returns 0 or the error code
*/
-xmlChar *
-xmlURIEscape(const xmlChar * str)
+static int
+xmlParse3986PathNoScheme(xmlURIPtr uri, const char **str)
{
- xmlChar *ret, *segment = NULL;
- xmlURIPtr uri;
- int ret2;
-
-#define NULLCHK(p) if(!p) { \
- xmlGenericError(xmlGenericErrorContext, \
- "xmlURIEscape: out of memory\n"); \
- return NULL; }
+ const char *cur;
+ int ret;
- if (str == NULL)
- return (NULL);
+ cur = *str;
- uri = xmlCreateURI();
+ ret = xmlParse3986Segment(&cur, ':', 0);
+ if (ret != 0) return(ret);
+ while (*cur == '/') {
+ cur++;
+ ret = xmlParse3986Segment(&cur, 0, 1);
+ if (ret != 0) return(ret);
+ }
if (uri != NULL) {
- /*
- * Allow escaping errors in the unescaped form
- */
- uri->cleanup = 1;
- ret2 = xmlParseURIReference(uri, (const char *)str);
- if (ret2) {
- xmlFreeURI(uri);
- return (NULL);
+ if (uri->path != NULL) xmlFree(uri->path);
+ if (cur != *str) {
+ if (uri->cleanup & 2)
+ uri->path = STRNDUP(*str, cur - *str);
+ else
+ uri->path = xmlURIUnescapeString(*str, cur - *str, NULL);
+ } else {
+ uri->path = NULL;
}
}
+ *str = cur;
+ return (0);
+}
- if (!uri)
- return NULL;
+/**
+ * xmlParse3986HierPart:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
+ *
+ * Parse an hierarchical part and fills in the appropriate fields
+ * of the @uri structure
+ *
+ * hier-part = "//" authority path-abempty
+ * / path-absolute
+ * / path-rootless
+ * / path-empty
+ *
+ * Returns 0 or the error code
+ */
+static int
+xmlParse3986HierPart(xmlURIPtr uri, const char **str)
+{
+ const char *cur;
+ int ret;
- ret = NULL;
+ cur = *str;
- if (uri->scheme) {
- segment = xmlURIEscapeStr(BAD_CAST uri->scheme, BAD_CAST "+-.");
- NULLCHK(segment)
- ret = xmlStrcat(ret, segment);
- ret = xmlStrcat(ret, BAD_CAST ":");
- xmlFree(segment);
+ if ((*cur == '/') && (*(cur + 1) == '/')) {
+ cur += 2;
+ ret = xmlParse3986Authority(uri, &cur);
+ if (ret != 0) return(ret);
+ if (uri->server == NULL)
+ uri->port = -1;
+ ret = xmlParse3986PathAbEmpty(uri, &cur);
+ if (ret != 0) return(ret);
+ *str = cur;
+ return(0);
+ } else if (*cur == '/') {
+ ret = xmlParse3986PathAbsolute(uri, &cur);
+ if (ret != 0) return(ret);
+ } else if (ISA_PCHAR(cur)) {
+ ret = xmlParse3986PathRootless(uri, &cur);
+ if (ret != 0) return(ret);
+ } else {
+ /* path-empty is effectively empty */
+ if (uri != NULL) {
+ if (uri->path != NULL) xmlFree(uri->path);
+ uri->path = NULL;
+ }
}
+ *str = cur;
+ return (0);
+}
- if (uri->authority) {
- segment =
- xmlURIEscapeStr(BAD_CAST uri->authority, BAD_CAST "/?;:@");
- NULLCHK(segment)
- ret = xmlStrcat(ret, BAD_CAST "//");
- ret = xmlStrcat(ret, segment);
- xmlFree(segment);
- }
+/**
+ * xmlParse3986RelativeRef:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
+ *
+ * Parse an URI string and fills in the appropriate fields
+ * of the @uri structure
+ *
+ * relative-ref = relative-part [ "?" query ] [ "#" fragment ]
+ * relative-part = "//" authority path-abempty
+ * / path-absolute
+ * / path-noscheme
+ * / path-empty
+ *
+ * Returns 0 or the error code
+ */
+static int
+xmlParse3986RelativeRef(xmlURIPtr uri, const char *str) {
+ int ret;
- if (uri->user) {
- segment = xmlURIEscapeStr(BAD_CAST uri->user, BAD_CAST ";:&=+$,");
- NULLCHK(segment)
- ret = xmlStrcat(ret,BAD_CAST "//");
- ret = xmlStrcat(ret, segment);
- ret = xmlStrcat(ret, BAD_CAST "@");
- xmlFree(segment);
+ if ((*str == '/') && (*(str + 1) == '/')) {
+ str += 2;
+ ret = xmlParse3986Authority(uri, &str);
+ if (ret != 0) return(ret);
+ ret = xmlParse3986PathAbEmpty(uri, &str);
+ if (ret != 0) return(ret);
+ } else if (*str == '/') {
+ ret = xmlParse3986PathAbsolute(uri, &str);
+ if (ret != 0) return(ret);
+ } else if (ISA_PCHAR(str)) {
+ ret = xmlParse3986PathNoScheme(uri, &str);
+ if (ret != 0) return(ret);
+ } else {
+ /* path-empty is effectively empty */
+ if (uri != NULL) {
+ if (uri->path != NULL) xmlFree(uri->path);
+ uri->path = NULL;
+ }
}
- if (uri->server) {
- segment = xmlURIEscapeStr(BAD_CAST uri->server, BAD_CAST "/?;:@");
- NULLCHK(segment)
- if (uri->user == NULL)
- ret = xmlStrcat(ret, BAD_CAST "//");
- ret = xmlStrcat(ret, segment);
- xmlFree(segment);
+ if (*str == '?') {
+ str++;
+ ret = xmlParse3986Query(uri, &str);
+ if (ret != 0) return(ret);
+ }
+ if (*str == '#') {
+ str++;
+ ret = xmlParse3986Fragment(uri, &str);
+ if (ret != 0) return(ret);
+ }
+ if (*str != 0) {
+ xmlCleanURI(uri);
+ return(1);
}
+ return(0);
+}
- if (uri->port) {
- xmlChar port[10];
- snprintf((char *) port, 10, "%d", uri->port);
- ret = xmlStrcat(ret, BAD_CAST ":");
- ret = xmlStrcat(ret, port);
- }
+/**
+ * xmlParse3986URI:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
+ *
+ * Parse an URI string and fills in the appropriate fields
+ * of the @uri structure
+ *
+ * scheme ":" hier-part [ "?" query ] [ "#" fragment ]
+ *
+ * Returns 0 or the error code
+ */
+static int
+xmlParse3986URI(xmlURIPtr uri, const char *str) {
+ int ret;
- if (uri->path) {
- segment =
- xmlURIEscapeStr(BAD_CAST uri->path, BAD_CAST ":@&=+$,/?;");
- NULLCHK(segment)
- ret = xmlStrcat(ret, segment);
- xmlFree(segment);
+ ret = xmlParse3986Scheme(uri, &str);
+ if (ret != 0) return(ret);
+ if (*str != ':') {
+ return(1);
}
-
- if (uri->query) {
- segment =
- xmlURIEscapeStr(BAD_CAST uri->query, BAD_CAST ";/?:@&=+,$");
- NULLCHK(segment)
- ret = xmlStrcat(ret, BAD_CAST "?");
- ret = xmlStrcat(ret, segment);
- xmlFree(segment);
+ str++;
+ ret = xmlParse3986HierPart(uri, &str);
+ if (ret != 0) return(ret);
+ if (*str == '?') {
+ str++;
+ ret = xmlParse3986Query(uri, &str);
+ if (ret != 0) return(ret);
+ }
+ if (*str == '#') {
+ str++;
+ ret = xmlParse3986Fragment(uri, &str);
+ if (ret != 0) return(ret);
+ }
+ if (*str != 0) {
+ xmlCleanURI(uri);
+ return(1);
}
+ return(0);
+}
- if (uri->opaque) {
- segment = xmlURIEscapeStr(BAD_CAST uri->opaque, BAD_CAST "");
- NULLCHK(segment)
- ret = xmlStrcat(ret, segment);
- xmlFree(segment);
+/**
+ * xmlParse3986URIReference:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
+ *
+ * Parse an URI reference string and fills in the appropriate fields
+ * of the @uri structure
+ *
+ * URI-reference = URI / relative-ref
+ *
+ * Returns 0 or the error code
+ */
+static int
+xmlParse3986URIReference(xmlURIPtr uri, const char *str) {
+ int ret;
+
+ if (str == NULL)
+ return(-1);
+ xmlCleanURI(uri);
+
+ /*
+ * Try first to parse absolute refs, then fallback to relative if
+ * it fails.
+ */
+ ret = xmlParse3986URI(uri, str);
+ if (ret != 0) {
+ xmlCleanURI(uri);
+ ret = xmlParse3986RelativeRef(uri, str);
+ if (ret != 0) {
+ xmlCleanURI(uri);
+ return(ret);
+ }
}
+ return(0);
+}
- if (uri->fragment) {
- segment = xmlURIEscapeStr(BAD_CAST uri->fragment, BAD_CAST "#");
- NULLCHK(segment)
- ret = xmlStrcat(ret, BAD_CAST "#");
- ret = xmlStrcat(ret, segment);
- xmlFree(segment);
+/**
+ * xmlParseURI:
+ * @str: the URI string to analyze
+ *
+ * Parse an URI based on RFC 3986
+ *
+ * URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
+ *
+ * Returns a newly built xmlURIPtr or NULL in case of error
+ */
+xmlURIPtr
+xmlParseURI(const char *str) {
+ xmlURIPtr uri;
+ int ret;
+
+ if (str == NULL)
+ return(NULL);
+ uri = xmlCreateURI();
+ if (uri != NULL) {
+ ret = xmlParse3986URIReference(uri, str);
+ if (ret) {
+ xmlFreeURI(uri);
+ return(NULL);
+ }
}
+ return(uri);
+}
+
+/**
+ * xmlParseURIReference:
+ * @uri: pointer to an URI structure
+ * @str: the string to analyze
+ *
+ * Parse an URI reference string based on RFC 3986 and fills in the
+ * appropriate fields of the @uri structure
+ *
+ * URI-reference = URI / relative-ref
+ *
+ * Returns 0 or the error code
+ */
+int
+xmlParseURIReference(xmlURIPtr uri, const char *str) {
+ return(xmlParse3986URIReference(uri, str));
+}
- xmlFreeURI(uri);
-#undef NULLCHK
+/**
+ * xmlParseURIRaw:
+ * @str: the URI string to analyze
+ * @raw: if 1 unescaping of URI pieces are disabled
+ *
+ * Parse an URI but allows to keep intact the original fragments.
+ *
+ * URI-reference = URI / relative-ref
+ *
+ * Returns a newly built xmlURIPtr or NULL in case of error
+ */
+xmlURIPtr
+xmlParseURIRaw(const char *str, int raw) {
+ xmlURIPtr uri;
+ int ret;
- return (ret);
+ if (str == NULL)
+ return(NULL);
+ uri = xmlCreateURI();
+ if (uri != NULL) {
+ if (raw) {
+ uri->cleanup |= 2;
+ }
+ ret = xmlParseURIReference(uri, str);
+ if (ret) {
+ xmlFreeURI(uri);
+ return(NULL);
+ }
+ }
+ return(uri);
}
/************************************************************************
* *
- * Escaped URI parsing *
+ * Generic URI structure functions *
* *
************************************************************************/
/**
- * xmlParseURIFragment:
- * @uri: pointer to an URI structure
- * @str: pointer to the string to analyze
+ * xmlCreateURI:
*
- * Parse an URI fragment string and fills in the appropriate fields
- * of the @uri structure.
- *
- * fragment = *uric
+ * Simply creates an empty xmlURI
*
- * Returns 0 or the error code
+ * Returns the new structure or NULL in case of error
*/
-static int
-xmlParseURIFragment(xmlURIPtr uri, const char **str)
-{
- const char *cur;
-
- if (str == NULL)
- return (-1);
+xmlURIPtr
+xmlCreateURI(void) {
+ xmlURIPtr ret;
- cur = *str;
+ ret = (xmlURIPtr) xmlMalloc(sizeof(xmlURI));
+ if (ret == NULL) {
+ xmlURIErrMemory("creating URI structure\n");
+ return(NULL);
+ }
+ memset(ret, 0, sizeof(xmlURI));
+ return(ret);
+}
- while (IS_URIC(cur) || IS_UNWISE(cur))
- NEXT(cur);
- if (uri != NULL) {
- if (uri->fragment != NULL)
- xmlFree(uri->fragment);
- if (uri->cleanup & 2)
- uri->fragment = STRNDUP(*str, cur - *str);
- else
- uri->fragment = xmlURIUnescapeString(*str, cur - *str, NULL);
+/**
+ * xmlSaveUriRealloc:
+ *
+ * Function to handle properly a reallocation when saving an URI
+ * Also imposes some limit on the length of an URI string output
+ */
+static xmlChar *
+xmlSaveUriRealloc(xmlChar *ret, int *max) {
+ xmlChar *temp;
+ int tmp;
+
+ if (*max > MAX_URI_LENGTH) {
+ xmlURIErrMemory("reaching arbitrary MAX_URI_LENGTH limit\n");
+ return(NULL);
}
- *str = cur;
- return (0);
+ tmp = *max * 2;
+ temp = (xmlChar *) xmlRealloc(ret, (tmp + 1));
+ if (temp == NULL) {
+ xmlURIErrMemory("saving URI\n");
+ return(NULL);
+ }
+ *max = tmp;
+ return(temp);
}
/**
- * xmlParseURIQuery:
- * @uri: pointer to an URI structure
- * @str: pointer to the string to analyze
+ * xmlSaveUri:
+ * @uri: pointer to an xmlURI
*
- * Parse the query part of an URI
- *
- * query = *uric
+ * Save the URI as an escaped string
*
- * Returns 0 or the error code
+ * Returns a new string (to be deallocated by caller)
*/
-static int
-xmlParseURIQuery(xmlURIPtr uri, const char **str)
-{
- const char *cur;
+xmlChar *
+xmlSaveUri(xmlURIPtr uri) {
+ xmlChar *ret = NULL;
+ xmlChar *temp;
+ const char *p;
+ int len;
+ int max;
+
+ if (uri == NULL) return(NULL);
+
+
+ max = 80;
+ ret = (xmlChar *) xmlMallocAtomic((max + 1) * sizeof(xmlChar));
+ if (ret == NULL) {
+ xmlURIErrMemory("saving URI\n");
+ return(NULL);
+ }
+ len = 0;
+
+ if (uri->scheme != NULL) {
+ p = uri->scheme;
+ while (*p != 0) {
+ if (len >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len++] = *p++;
+ }
+ if (len >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len++] = ':';
+ }
+ if (uri->opaque != NULL) {
+ p = uri->opaque;
+ while (*p != 0) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ if (IS_RESERVED(*(p)) || IS_UNRESERVED(*(p)))
+ ret[len++] = *p++;
+ else {
+ int val = *(unsigned char *)p++;
+ int hi = val / 0x10, lo = val % 0x10;
+ ret[len++] = '%';
+ ret[len++] = hi + (hi > 9? 'A'-10 : '0');
+ ret[len++] = lo + (lo > 9? 'A'-10 : '0');
+ }
+ }
+ } else {
+ if ((uri->server != NULL) || (uri->port == -1)) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len++] = '/';
+ ret[len++] = '/';
+ if (uri->user != NULL) {
+ p = uri->user;
+ while (*p != 0) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ if ((IS_UNRESERVED(*(p))) ||
+ ((*(p) == ';')) || ((*(p) == ':')) ||
+ ((*(p) == '&')) || ((*(p) == '=')) ||
+ ((*(p) == '+')) || ((*(p) == '$')) ||
+ ((*(p) == ',')))
+ ret[len++] = *p++;
+ else {
+ int val = *(unsigned char *)p++;
+ int hi = val / 0x10, lo = val % 0x10;
+ ret[len++] = '%';
+ ret[len++] = hi + (hi > 9? 'A'-10 : '0');
+ ret[len++] = lo + (lo > 9? 'A'-10 : '0');
+ }
+ }
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len++] = '@';
+ }
+ if (uri->server != NULL) {
+ p = uri->server;
+ while (*p != 0) {
+ if (len >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len++] = *p++;
+ }
+ if (uri->port > 0) {
+ if (len + 10 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ len += snprintf((char *) &ret[len], max - len, ":%d", uri->port);
+ }
+ }
+ } else if (uri->authority != NULL) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len++] = '/';
+ ret[len++] = '/';
+ p = uri->authority;
+ while (*p != 0) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ if ((IS_UNRESERVED(*(p))) ||
+ ((*(p) == '$')) || ((*(p) == ',')) || ((*(p) == ';')) ||
+ ((*(p) == ':')) || ((*(p) == '@')) || ((*(p) == '&')) ||
+ ((*(p) == '=')) || ((*(p) == '+')))
+ ret[len++] = *p++;
+ else {
+ int val = *(unsigned char *)p++;
+ int hi = val / 0x10, lo = val % 0x10;
+ ret[len++] = '%';
+ ret[len++] = hi + (hi > 9? 'A'-10 : '0');
+ ret[len++] = lo + (lo > 9? 'A'-10 : '0');
+ }
+ }
+ } else if (uri->scheme != NULL) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ }
+ if (uri->path != NULL) {
+ p = uri->path;
+ /*
+ * the colon in file:///d: should not be escaped or
+ * Windows accesses fail later.
+ */
+ if ((uri->scheme != NULL) &&
+ (p[0] == '/') &&
+ (((p[1] >= 'a') && (p[1] <= 'z')) ||
+ ((p[1] >= 'A') && (p[1] <= 'Z'))) &&
+ (p[2] == ':') &&
+ (xmlStrEqual(BAD_CAST uri->scheme, BAD_CAST "file"))) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len++] = *p++;
+ ret[len++] = *p++;
+ ret[len++] = *p++;
+ }
+ while (*p != 0) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ if ((IS_UNRESERVED(*(p))) || ((*(p) == '/')) ||
+ ((*(p) == ';')) || ((*(p) == '@')) || ((*(p) == '&')) ||
+ ((*(p) == '=')) || ((*(p) == '+')) || ((*(p) == '$')) ||
+ ((*(p) == ',')))
+ ret[len++] = *p++;
+ else {
+ int val = *(unsigned char *)p++;
+ int hi = val / 0x10, lo = val % 0x10;
+ ret[len++] = '%';
+ ret[len++] = hi + (hi > 9? 'A'-10 : '0');
+ ret[len++] = lo + (lo > 9? 'A'-10 : '0');
+ }
+ }
+ }
+ if (uri->query_raw != NULL) {
+ if (len + 1 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len++] = '?';
+ p = uri->query_raw;
+ while (*p != 0) {
+ if (len + 1 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len++] = *p++;
+ }
+ } else if (uri->query != NULL) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len++] = '?';
+ p = uri->query;
+ while (*p != 0) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ if ((IS_UNRESERVED(*(p))) || (IS_RESERVED(*(p))))
+ ret[len++] = *p++;
+ else {
+ int val = *(unsigned char *)p++;
+ int hi = val / 0x10, lo = val % 0x10;
+ ret[len++] = '%';
+ ret[len++] = hi + (hi > 9? 'A'-10 : '0');
+ ret[len++] = lo + (lo > 9? 'A'-10 : '0');
+ }
+ }
+ }
+ }
+ if (uri->fragment != NULL) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len++] = '#';
+ p = uri->fragment;
+ while (*p != 0) {
+ if (len + 3 >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ if ((IS_UNRESERVED(*(p))) || (IS_RESERVED(*(p))))
+ ret[len++] = *p++;
+ else {
+ int val = *(unsigned char *)p++;
+ int hi = val / 0x10, lo = val % 0x10;
+ ret[len++] = '%';
+ ret[len++] = hi + (hi > 9? 'A'-10 : '0');
+ ret[len++] = lo + (lo > 9? 'A'-10 : '0');
+ }
+ }
+ }
+ if (len >= max) {
+ temp = xmlSaveUriRealloc(ret, &max);
+ if (temp == NULL) goto mem_error;
+ ret = temp;
+ }
+ ret[len] = 0;
+ return(ret);
- if (str == NULL)
- return (-1);
+mem_error:
+ xmlFree(ret);
+ return(NULL);
+}
- cur = *str;
+/**
+ * xmlPrintURI:
+ * @stream: a FILE* for the output
+ * @uri: pointer to an xmlURI
+ *
+ * Prints the URI in the stream @stream.
+ */
+void
+xmlPrintURI(FILE *stream, xmlURIPtr uri) {
+ xmlChar *out;
- while ((IS_URIC(cur)) ||
- ((uri != NULL) && (uri->cleanup & 1) && (IS_UNWISE(cur))))
- NEXT(cur);
- if (uri != NULL) {
- if (uri->query != NULL)
- xmlFree(uri->query);
- if (uri->cleanup & 2)
- uri->query = STRNDUP(*str, cur - *str);
- else
- uri->query = xmlURIUnescapeString(*str, cur - *str, NULL);
+ out = xmlSaveUri(uri);
+ if (out != NULL) {
+ fprintf(stream, "%s", (char *) out);
+ xmlFree(out);
}
- *str = cur;
- return (0);
}
/**
- * xmlParseURIScheme:
- * @uri: pointer to an URI structure
- * @str: pointer to the string to analyze
- *
- * Parse an URI scheme
- *
- * scheme = alpha *( alpha | digit | "+" | "-" | "." )
+ * xmlCleanURI:
+ * @uri: pointer to an xmlURI
*
- * Returns 0 or the error code
+ * Make sure the xmlURI struct is free of content
*/
-static int
-xmlParseURIScheme(xmlURIPtr uri, const char **str) {
- const char *cur;
+static void
+xmlCleanURI(xmlURIPtr uri) {
+ if (uri == NULL) return;
- if (str == NULL)
- return(-1);
-
- cur = *str;
- if (!IS_ALPHA(*cur))
- return(2);
- cur++;
- while (IS_SCHEME(*cur)) cur++;
- if (uri != NULL) {
- if (uri->scheme != NULL) xmlFree(uri->scheme);
- uri->scheme = STRNDUP(*str, cur - *str);
- }
- *str = cur;
- return(0);
+ if (uri->scheme != NULL) xmlFree(uri->scheme);
+ uri->scheme = NULL;
+ if (uri->server != NULL) xmlFree(uri->server);
+ uri->server = NULL;
+ if (uri->user != NULL) xmlFree(uri->user);
+ uri->user = NULL;
+ if (uri->path != NULL) xmlFree(uri->path);
+ uri->path = NULL;
+ if (uri->fragment != NULL) xmlFree(uri->fragment);
+ uri->fragment = NULL;
+ if (uri->opaque != NULL) xmlFree(uri->opaque);
+ uri->opaque = NULL;
+ if (uri->authority != NULL) xmlFree(uri->authority);
+ uri->authority = NULL;
+ if (uri->query != NULL) xmlFree(uri->query);
+ uri->query = NULL;
+ if (uri->query_raw != NULL) xmlFree(uri->query_raw);
+ uri->query_raw = NULL;
}
/**
- * xmlParseURIOpaquePart:
- * @uri: pointer to an URI structure
- * @str: pointer to the string to analyze
- *
- * Parse an URI opaque part
- *
- * opaque_part = uric_no_slash *uric
+ * xmlFreeURI:
+ * @uri: pointer to an xmlURI
*
- * Returns 0 or the error code
+ * Free up the xmlURI struct
*/
-static int
-xmlParseURIOpaquePart(xmlURIPtr uri, const char **str)
-{
- const char *cur;
-
- if (str == NULL)
- return (-1);
+void
+xmlFreeURI(xmlURIPtr uri) {
+ if (uri == NULL) return;
- cur = *str;
- if (!((IS_URIC_NO_SLASH(cur)) ||
- ((uri != NULL) && (uri->cleanup & 1) && (IS_UNWISE(cur))))) {
- return (3);
- }
- NEXT(cur);
- while ((IS_URIC(cur)) ||
- ((uri != NULL) && (uri->cleanup & 1) && (IS_UNWISE(cur))))
- NEXT(cur);
- if (uri != NULL) {
- if (uri->opaque != NULL)
- xmlFree(uri->opaque);
- if (uri->cleanup & 2)
- uri->opaque = STRNDUP(*str, cur - *str);
- else
- uri->opaque = xmlURIUnescapeString(*str, cur - *str, NULL);
- }
- *str = cur;
- return (0);
+ if (uri->scheme != NULL) xmlFree(uri->scheme);
+ if (uri->server != NULL) xmlFree(uri->server);
+ if (uri->user != NULL) xmlFree(uri->user);
+ if (uri->path != NULL) xmlFree(uri->path);
+ if (uri->fragment != NULL) xmlFree(uri->fragment);
+ if (uri->opaque != NULL) xmlFree(uri->opaque);
+ if (uri->authority != NULL) xmlFree(uri->authority);
+ if (uri->query != NULL) xmlFree(uri->query);
+ if (uri->query_raw != NULL) xmlFree(uri->query_raw);
+ xmlFree(uri);
}
+/************************************************************************
+ * *
+ * Helper functions *
+ * *
+ ************************************************************************/
+
/**
- * xmlParseURIServer:
- * @uri: pointer to an URI structure
- * @str: pointer to the string to analyze
+ * xmlNormalizeURIPath:
+ * @path: pointer to the path string
+ *
+ * Applies the 5 normalization steps to a path string--that is, RFC 2396
+ * Section 5.2, steps 6.c through 6.g.
*
- * Parse a server subpart of an URI, it's a finer grain analysis
- * of the authority part.
- *
- * server = [ [ userinfo "@" ] hostport ]
- * userinfo = *( unreserved | escaped |
- * ";" | ":" | "&" | "=" | "+" | "$" | "," )
- * hostport = host [ ":" port ]
- * host = hostname | IPv4address | IPv6reference
- * hostname = *( domainlabel "." ) toplabel [ "." ]
- * domainlabel = alphanum | alphanum *( alphanum | "-" ) alphanum
- * toplabel = alpha | alpha *( alphanum | "-" ) alphanum
- * IPv6reference = "[" IPv6address "]"
- * IPv6address = hexpart [ ":" IPv4address ]
- * IPv4address = 1*3digit "." 1*3digit "." 1*3digit "." 1*3digit
- * hexpart = hexseq | hexseq "::" [ hexseq ]| "::" [ hexseq ]
- * hexseq = hex4 *( ":" hex4)
- * hex4 = 1*4hexdig
- * port = *digit
+ * Normalization occurs directly on the string, no new allocation is done
*
- * Returns 0 or the error code
+ * Returns 0 or an error code
*/
-static int
-xmlParseURIServer(xmlURIPtr uri, const char **str) {
- const char *cur;
- const char *host, *tmp;
- const int IPV4max = 4;
- const int IPV6max = 8;
- int oct;
+int
+xmlNormalizeURIPath(char *path) {
+ char *cur, *out;
- if (str == NULL)
+ if (path == NULL)
return(-1);
-
- cur = *str;
- /*
- * is there a userinfo ?
- */
- while (IS_USERINFO(cur)) NEXT(cur);
- if (*cur == '@') {
- if (uri != NULL) {
- if (uri->user != NULL) xmlFree(uri->user);
- if (uri->cleanup & 2)
- uri->path = STRNDUP(*str, cur - *str);
- else
- uri->user = xmlURIUnescapeString(*str, cur - *str, NULL);
- }
- cur++;
- } else {
- if (uri != NULL) {
- if (uri->user != NULL) xmlFree(uri->user);
- uri->user = NULL;
- }
- cur = *str;
- }
- /*
- * This can be empty in the case where there is no server
+ /* Skip all initial "/" chars. We want to get to the beginning of the
+ * first non-empty segment.
*/
- host = cur;
- if (*cur == '/') {
- if (uri != NULL) {
- if (uri->authority != NULL) xmlFree(uri->authority);
- uri->authority = NULL;
- if (uri->server != NULL) xmlFree(uri->server);
- uri->server = NULL;
- uri->port = 0;
- }
- return(0);
- }
+ cur = path;
+ while (cur[0] == '/')
+ ++cur;
+ if (cur[0] == '\0')
+ return(0);
+
+ /* Keep everything we've seen so far. */
+ out = cur;
+
/*
- * host part of hostport can denote an IPV4 address, an IPV6 address
- * or an unresolved name. Check the IP first, its easier to detect
- * errors if wrong one.
- * An IPV6 address must start with a '[' and end with a ']'.
+ * Analyze each segment in sequence for cases (c) and (d).
*/
- if (*cur == '[') {
- int compress=0;
- cur++;
- for (oct = 0; oct < IPV6max; ++oct) {
- if (*cur == ':') {
- if (compress)
- return(3); /* multiple compression attempted */
- if (!oct) { /* initial char is compression */
- if (*++cur != ':')
- return(3);
- }
- compress = 1; /* set compression-encountered flag */
- cur++; /* skip over the second ':' */
- continue;
- }
- while(IS_HEX(*cur)) cur++;
- if (oct == (IPV6max-1))
- continue;
- if (*cur != ':')
- break;
- cur++;
- }
- if ((!compress) && (oct != IPV6max))
- return(3);
- if (*cur != ']')
- return(3);
- if (uri != NULL) {
- if (uri->server != NULL) xmlFree(uri->server);
- uri->server = (char *)xmlStrndup((xmlChar *)host+1,
- (cur-host)-1);
- }
- cur++;
- } else {
+ while (cur[0] != '\0') {
/*
- * Not IPV6, maybe IPV4
+ * c) All occurrences of "./", where "." is a complete path segment,
+ * are removed from the buffer string.
*/
- for (oct = 0; oct < IPV4max; ++oct) {
- if (*cur == '.')
- return(3); /* e.g. http://.xml/ or http://18.29..30/ */
- while(IS_DIGIT(*cur)) cur++;
- if (oct == (IPV4max-1))
- continue;
- if (*cur != '.')
- break;
- cur++;
- }
- }
- if ((host[0] != '[') && (oct < IPV4max || (*cur == '.' && cur++) ||
- IS_ALPHA(*cur))) {
- /* maybe host_name */
- if (!IS_ALPHANUM(*cur))
- return(4); /* e.g. http://xml.$oft */
- do {
- do ++cur; while (IS_ALPHANUM(*cur));
- if (*cur == '-') {
- --cur;
- if (*cur == '.')
- return(5); /* e.g. http://xml.-soft */
- ++cur;
- continue;
- }
- if (*cur == '.') {
- --cur;
- if (*cur == '-')
- return(6); /* e.g. http://xml-.soft */
- if (*cur == '.')
- return(7); /* e.g. http://xml..soft */
- ++cur;
- continue;
- }
- break;
- } while (1);
- tmp = cur;
- if (tmp[-1] == '.')
- --tmp; /* e.g. http://xml.$Oft/ */
- do --tmp; while (tmp >= host && IS_ALPHANUM(*tmp));
- if ((++tmp == host || tmp[-1] == '.') && !IS_ALPHA(*tmp))
- return(8); /* e.g. http://xmlsOft.0rg/ */
- }
- if (uri != NULL) {
- if (uri->authority != NULL) xmlFree(uri->authority);
- uri->authority = NULL;
- if (host[0] != '[') { /* it's not an IPV6 addr */
- if (uri->server != NULL) xmlFree(uri->server);
- if (uri->cleanup & 2)
- uri->server = STRNDUP(host, cur - host);
- else
- uri->server = xmlURIUnescapeString(host, cur - host, NULL);
- }
- }
- /*
- * finish by checking for a port presence.
- */
- if (*cur == ':') {
- cur++;
- if (IS_DIGIT(*cur)) {
- if (uri != NULL)
- uri->port = 0;
- while (IS_DIGIT(*cur)) {
- if (uri != NULL)
- uri->port = uri->port * 10 + (*cur - '0');
+ if ((cur[0] == '.') && (cur[1] == '/')) {
+ cur += 2;
+ /* '//' normalization should be done at this point too */
+ while (cur[0] == '/')
cur++;
- }
+ continue;
}
- }
- *str = cur;
- return(0);
-}
-/**
- * xmlParseURIRelSegment:
- * @uri: pointer to an URI structure
- * @str: pointer to the string to analyze
- *
- * Parse an URI relative segment
- *
- * rel_segment = 1*( unreserved | escaped | ";" | "@" | "&" | "=" |
- * "+" | "$" | "," )
- *
- * Returns 0 or the error code
- */
-static int
-xmlParseURIRelSegment(xmlURIPtr uri, const char **str)
-{
- const char *cur;
+ /*
+ * d) If the buffer string ends with "." as a complete path segment,
+ * that "." is removed.
+ */
+ if ((cur[0] == '.') && (cur[1] == '\0'))
+ break;
- if (str == NULL)
- return (-1);
+ /* Otherwise keep the segment. */
+ while (cur[0] != '/') {
+ if (cur[0] == '\0')
+ goto done_cd;
+ (out++)[0] = (cur++)[0];
+ }
+ /* nomalize // */
+ while ((cur[0] == '/') && (cur[1] == '/'))
+ cur++;
- cur = *str;
- if (!((IS_SEGMENT(cur)) ||
- ((uri != NULL) && (uri->cleanup & 1) && (IS_UNWISE(cur))))) {
- return (3);
- }
- NEXT(cur);
- while ((IS_SEGMENT(cur)) ||
- ((uri != NULL) && (uri->cleanup & 1) && (IS_UNWISE(cur))))
- NEXT(cur);
- if (uri != NULL) {
- if (uri->path != NULL)
- xmlFree(uri->path);
- if (uri->cleanup & 2)
- uri->path = STRNDUP(*str, cur - *str);
- else
- uri->path = xmlURIUnescapeString(*str, cur - *str, NULL);
+ (out++)[0] = (cur++)[0];
}
- *str = cur;
- return (0);
-}
+ done_cd:
+ out[0] = '\0';
+
+ /* Reset to the beginning of the first segment for the next sequence. */
+ cur = path;
+ while (cur[0] == '/')
+ ++cur;
+ if (cur[0] == '\0')
+ return(0);
-/**
- * xmlParseURIPathSegments:
- * @uri: pointer to an URI structure
- * @str: pointer to the string to analyze
- * @slash: should we add a leading slash
- *
- * Parse an URI set of path segments
- *
- * path_segments = segment *( "/" segment )
- * segment = *pchar *( ";" param )
- * param = *pchar
- *
- * Returns 0 or the error code
- */
-static int
-xmlParseURIPathSegments(xmlURIPtr uri, const char **str, int slash)
-{
- const char *cur;
+ /*
+ * Analyze each segment in sequence for cases (e) and (f).
+ *
+ * e) All occurrences of "<segment>/../", where <segment> is a
+ * complete path segment not equal to "..", are removed from the
+ * buffer string. Removal of these path segments is performed
+ * iteratively, removing the leftmost matching pattern on each
+ * iteration, until no matching pattern remains.
+ *
+ * f) If the buffer string ends with "<segment>/..", where <segment>
+ * is a complete path segment not equal to "..", that
+ * "<segment>/.." is removed.
+ *
+ * To satisfy the "iterative" clause in (e), we need to collapse the
+ * string every time we find something that needs to be removed. Thus,
+ * we don't need to keep two pointers into the string: we only need a
+ * "current position" pointer.
+ */
+ while (1) {
+ char *segp, *tmp;
- if (str == NULL)
- return (-1);
+ /* At the beginning of each iteration of this loop, "cur" points to
+ * the first character of the segment we want to examine.
+ */
- cur = *str;
+ /* Find the end of the current segment. */
+ segp = cur;
+ while ((segp[0] != '/') && (segp[0] != '\0'))
+ ++segp;
+
+ /* If this is the last segment, we're done (we need at least two
+ * segments to meet the criteria for the (e) and (f) cases).
+ */
+ if (segp[0] == '\0')
+ break;
- do {
- while ((IS_PCHAR(cur)) ||
- ((uri != NULL) && (uri->cleanup & 1) && (IS_UNWISE(cur))))
- NEXT(cur);
- while (*cur == ';') {
- cur++;
- while ((IS_PCHAR(cur)) ||
- ((uri != NULL) && (uri->cleanup & 1) && (IS_UNWISE(cur))))
- NEXT(cur);
+ /* If the first segment is "..", or if the next segment _isn't_ "..",
+ * keep this segment and try the next one.
+ */
+ ++segp;
+ if (((cur[0] == '.') && (cur[1] == '.') && (segp == cur+3))
+ || ((segp[0] != '.') || (segp[1] != '.')
+ || ((segp[2] != '/') && (segp[2] != '\0')))) {
+ cur = segp;
+ continue;
}
- if (*cur != '/')
- break;
- cur++;
- } while (1);
- if (uri != NULL) {
- int len, len2 = 0;
- char *path;
- /*
- * Concat the set of path segments to the current path
+ /* If we get here, remove this segment and the next one and back up
+ * to the previous segment (if there is one), to implement the
+ * "iteratively" clause. It's pretty much impossible to back up
+ * while maintaining two pointers into the buffer, so just compact
+ * the whole buffer now.
*/
- len = cur - *str;
- if (slash)
- len++;
- if (uri->path != NULL) {
- len2 = strlen(uri->path);
- len += len2;
- }
- path = (char *) xmlMallocAtomic(len + 1);
- if (path == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlParseURIPathSegments: out of memory\n");
- *str = cur;
- return (-1);
- }
- if (uri->path != NULL)
- memcpy(path, uri->path, len2);
- if (slash) {
- path[len2] = '/';
- len2++;
+ /* If this is the end of the buffer, we're done. */
+ if (segp[2] == '\0') {
+ cur[0] = '\0';
+ break;
}
- path[len2] = 0;
- if (cur - *str > 0) {
- if (uri->cleanup & 2) {
- memcpy(&path[len2], *str, cur - *str);
- path[len2 + (cur - *str)] = 0;
- } else
- xmlURIUnescapeString(*str, cur - *str, &path[len2]);
- }
- if (uri->path != NULL)
- xmlFree(uri->path);
- uri->path = path;
- }
- *str = cur;
- return (0);
-}
+ /* Valgrind complained, strcpy(cur, segp + 3); */
+ /* string will overlap, do not use strcpy */
+ tmp = cur;
+ segp += 3;
+ while ((*tmp++ = *segp++) != 0)
+ ;
-/**
- * xmlParseURIAuthority:
- * @uri: pointer to an URI structure
- * @str: pointer to the string to analyze
- *
- * Parse the authority part of an URI.
- *
- * authority = server | reg_name
- * server = [ [ userinfo "@" ] hostport ]
- * reg_name = 1*( unreserved | escaped | "$" | "," | ";" | ":" |
- * "@" | "&" | "=" | "+" )
- *
- * Note : this is completely ambiguous since reg_name is allowed to
- * use the full set of chars in use by server:
- *
- * 3.2.1. Registry-based Naming Authority
- *
- * The structure of a registry-based naming authority is specific
- * to the URI scheme, but constrained to the allowed characters
- * for an authority component.
- *
- * Returns 0 or the error code
- */
-static int
-xmlParseURIAuthority(xmlURIPtr uri, const char **str) {
- const char *cur;
- int ret;
+ /* If there are no previous segments, then keep going from here. */
+ segp = cur;
+ while ((segp > path) && ((--segp)[0] == '/'))
+ ;
+ if (segp == path)
+ continue;
- if (str == NULL)
- return(-1);
-
- cur = *str;
+ /* "segp" is pointing to the end of a previous segment; find it's
+ * start. We need to back up to the previous segment and start
+ * over with that to handle things like "foo/bar/../..". If we
+ * don't do this, then on the first pass we'll remove the "bar/..",
+ * but be pointing at the second ".." so we won't realize we can also
+ * remove the "foo/..".
+ */
+ cur = segp;
+ while ((cur > path) && (cur[-1] != '/'))
+ --cur;
+ }
+ out[0] = '\0';
/*
- * try first to parse it as a server string.
+ * g) If the resulting buffer string still begins with one or more
+ * complete path segments of "..", then the reference is
+ * considered to be in error. Implementations may handle this
+ * error by retaining these components in the resolved path (i.e.,
+ * treating them as part of the final URI), by removing them from
+ * the resolved path (i.e., discarding relative levels above the
+ * root), or by avoiding traversal of the reference.
+ *
+ * We discard them from the final path.
*/
- ret = xmlParseURIServer(uri, str);
- if ((ret == 0) && (*str != NULL) &&
- ((**str == 0) || (**str == '/') || (**str == '?')))
- return(0);
- *str = cur;
+ if (path[0] == '/') {
+ cur = path;
+ while ((cur[0] == '/') && (cur[1] == '.') && (cur[2] == '.')
+ && ((cur[3] == '/') || (cur[3] == '\0')))
+ cur += 3;
- /*
- * failed, fallback to reg_name
- */
- if (!IS_REG_NAME(cur)) {
- return(5);
- }
- NEXT(cur);
- while (IS_REG_NAME(cur)) NEXT(cur);
- if (uri != NULL) {
- if (uri->server != NULL) xmlFree(uri->server);
- uri->server = NULL;
- if (uri->user != NULL) xmlFree(uri->user);
- uri->user = NULL;
- if (uri->authority != NULL) xmlFree(uri->authority);
- if (uri->cleanup & 2)
- uri->authority = STRNDUP(*str, cur - *str);
- else
- uri->authority = xmlURIUnescapeString(*str, cur - *str, NULL);
+ if (cur != path) {
+ out = path;
+ while (cur[0] != '\0')
+ (out++)[0] = (cur++)[0];
+ out[0] = 0;
+ }
}
- *str = cur;
- return(0);
-}
-
-/**
- * xmlParseURIHierPart:
- * @uri: pointer to an URI structure
- * @str: pointer to the string to analyze
- *
- * Parse an URI hierarchical part
- *
- * hier_part = ( net_path | abs_path ) [ "?" query ]
- * abs_path = "/" path_segments
- * net_path = "//" authority [ abs_path ]
- *
- * Returns 0 or the error code
- */
-static int
-xmlParseURIHierPart(xmlURIPtr uri, const char **str) {
- int ret;
- const char *cur;
-
- if (str == NULL)
- return(-1);
-
- cur = *str;
- if ((cur[0] == '/') && (cur[1] == '/')) {
- cur += 2;
- ret = xmlParseURIAuthority(uri, &cur);
- if (ret != 0)
- return(ret);
- if (cur[0] == '/') {
- cur++;
- ret = xmlParseURIPathSegments(uri, &cur, 1);
- }
- } else if (cur[0] == '/') {
- cur++;
- ret = xmlParseURIPathSegments(uri, &cur, 1);
- } else {
- return(4);
- }
- if (ret != 0)
- return(ret);
- if (*cur == '?') {
- cur++;
- ret = xmlParseURIQuery(uri, &cur);
- if (ret != 0)
- return(ret);
- }
- *str = cur;
return(0);
}
-/**
- * xmlParseAbsoluteURI:
- * @uri: pointer to an URI structure
- * @str: pointer to the string to analyze
- *
- * Parse an URI reference string and fills in the appropriate fields
- * of the @uri structure
- *
- * absoluteURI = scheme ":" ( hier_part | opaque_part )
- *
- * Returns 0 or the error code
- */
-static int
-xmlParseAbsoluteURI(xmlURIPtr uri, const char **str) {
- int ret;
- const char *cur;
-
- if (str == NULL)
- return(-1);
-
- cur = *str;
-
- ret = xmlParseURIScheme(uri, str);
- if (ret != 0) return(ret);
- if (**str != ':') {
- *str = cur;
+static int is_hex(char c) {
+ if (((c >= '0') && (c <= '9')) ||
+ ((c >= 'a') && (c <= 'f')) ||
+ ((c >= 'A') && (c <= 'F')))
return(1);
- }
- (*str)++;
- if (**str == '/')
- return(xmlParseURIHierPart(uri, str));
- return(xmlParseURIOpaquePart(uri, str));
+ return(0);
}
/**
- * xmlParseRelativeURI:
- * @uri: pointer to an URI structure
- * @str: pointer to the string to analyze
+ * xmlURIUnescapeString:
+ * @str: the string to unescape
+ * @len: the length in bytes to unescape (or <= 0 to indicate full string)
+ * @target: optional destination buffer
*
- * Parse an relative URI string and fills in the appropriate fields
- * of the @uri structure
- *
- * relativeURI = ( net_path | abs_path | rel_path ) [ "?" query ]
- * abs_path = "/" path_segments
- * net_path = "//" authority [ abs_path ]
- * rel_path = rel_segment [ abs_path ]
+ * Unescaping routine, but does not check that the string is an URI. The
+ * output is a direct unsigned char translation of %XX values (no encoding)
+ * Note that the length of the result can only be smaller or same size as
+ * the input string.
*
- * Returns 0 or the error code
+ * Returns a copy of the string, but unescaped, will return NULL only in case
+ * of error
*/
-static int
-xmlParseRelativeURI(xmlURIPtr uri, const char **str) {
- int ret = 0;
- const char *cur;
+char *
+xmlURIUnescapeString(const char *str, int len, char *target) {
+ char *ret, *out;
+ const char *in;
if (str == NULL)
- return(-1);
-
- cur = *str;
- if ((cur[0] == '/') && (cur[1] == '/')) {
- cur += 2;
- ret = xmlParseURIAuthority(uri, &cur);
- if (ret != 0)
- return(ret);
- if (cur[0] == '/') {
- cur++;
- ret = xmlParseURIPathSegments(uri, &cur, 1);
+ return(NULL);
+ if (len <= 0) len = strlen(str);
+ if (len < 0) return(NULL);
+
+ if (target == NULL) {
+ ret = (char *) xmlMallocAtomic(len + 1);
+ if (ret == NULL) {
+ xmlURIErrMemory("unescaping URI value\n");
+ return(NULL);
}
- } else if (cur[0] == '/') {
- cur++;
- ret = xmlParseURIPathSegments(uri, &cur, 1);
- } else if (cur[0] != '#' && cur[0] != '?') {
- ret = xmlParseURIRelSegment(uri, &cur);
- if (ret != 0)
- return(ret);
- if (cur[0] == '/') {
- cur++;
- ret = xmlParseURIPathSegments(uri, &cur, 1);
+ } else
+ ret = target;
+ in = str;
+ out = ret;
+ while(len > 0) {
+ if ((len > 2) && (*in == '%') && (is_hex(in[1])) && (is_hex(in[2]))) {
+ in++;
+ if ((*in >= '0') && (*in <= '9'))
+ *out = (*in - '0');
+ else if ((*in >= 'a') && (*in <= 'f'))
+ *out = (*in - 'a') + 10;
+ else if ((*in >= 'A') && (*in <= 'F'))
+ *out = (*in - 'A') + 10;
+ in++;
+ if ((*in >= '0') && (*in <= '9'))
+ *out = *out * 16 + (*in - '0');
+ else if ((*in >= 'a') && (*in <= 'f'))
+ *out = *out * 16 + (*in - 'a') + 10;
+ else if ((*in >= 'A') && (*in <= 'F'))
+ *out = *out * 16 + (*in - 'A') + 10;
+ in++;
+ len -= 3;
+ out++;
+ } else {
+ *out++ = *in++;
+ len--;
}
}
- if (ret != 0)
- return(ret);
- if (*cur == '?') {
- cur++;
- ret = xmlParseURIQuery(uri, &cur);
- if (ret != 0)
- return(ret);
- }
- *str = cur;
+ *out = 0;
return(ret);
}
/**
- * xmlParseURIReference:
- * @uri: pointer to an URI structure
- * @str: the string to analyze
+ * xmlURIEscapeStr:
+ * @str: string to escape
+ * @list: exception list string of chars not to escape
*
- * Parse an URI reference string and fills in the appropriate fields
- * of the @uri structure
- *
- * URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
+ * This routine escapes a string to hex, ignoring reserved characters (a-z)
+ * and the characters in the exception list.
*
- * Returns 0 or the error code
+ * Returns a new escaped string or NULL in case of error.
*/
-int
-xmlParseURIReference(xmlURIPtr uri, const char *str) {
- int ret;
- const char *tmp = str;
+xmlChar *
+xmlURIEscapeStr(const xmlChar *str, const xmlChar *list) {
+ xmlChar *ret, ch;
+ xmlChar *temp;
+ const xmlChar *in;
+ int len, out;
if (str == NULL)
- return(-1);
- xmlCleanURI(uri);
+ return(NULL);
+ if (str[0] == 0)
+ return(xmlStrdup(str));
+ len = xmlStrlen(str);
+ if (!(len > 0)) return(NULL);
- /*
- * Try first to parse absolute refs, then fallback to relative if
- * it fails.
- */
- ret = xmlParseAbsoluteURI(uri, &str);
- if (ret != 0) {
- xmlCleanURI(uri);
- str = tmp;
- ret = xmlParseRelativeURI(uri, &str);
- }
- if (ret != 0) {
- xmlCleanURI(uri);
- return(ret);
+ len += 20;
+ ret = (xmlChar *) xmlMallocAtomic(len);
+ if (ret == NULL) {
+ xmlURIErrMemory("escaping URI value\n");
+ return(NULL);
}
+ in = (const xmlChar *) str;
+ out = 0;
+ while(*in != 0) {
+ if (len - out <= 3) {
+ temp = xmlSaveUriRealloc(ret, &len);
+ if (temp == NULL) {
+ xmlURIErrMemory("escaping URI value\n");
+ xmlFree(ret);
+ return(NULL);
+ }
+ ret = temp;
+ }
+
+ ch = *in;
+
+ if ((ch != '@') && (!IS_UNRESERVED(ch)) && (!xmlStrchr(list, ch))) {
+ unsigned char val;
+ ret[out++] = '%';
+ val = ch >> 4;
+ if (val <= 9)
+ ret[out++] = '0' + val;
+ else
+ ret[out++] = 'A' + val - 0xA;
+ val = ch & 0xF;
+ if (val <= 9)
+ ret[out++] = '0' + val;
+ else
+ ret[out++] = 'A' + val - 0xA;
+ in++;
+ } else {
+ ret[out++] = *in++;
+ }
- if (*str == '#') {
- str++;
- ret = xmlParseURIFragment(uri, &str);
- if (ret != 0) return(ret);
- }
- if (*str != 0) {
- xmlCleanURI(uri);
- return(1);
}
- return(0);
+ ret[out] = 0;
+ return(ret);
}
/**
- * xmlParseURI:
- * @str: the URI string to analyze
+ * xmlURIEscape:
+ * @str: the string of the URI to escape
*
- * Parse an URI
- *
- * URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
+ * Escaping routine, does not do validity checks !
+ * It will try to escape the chars needing this, but this is heuristic
+ * based it's impossible to be sure.
*
- * Returns a newly built xmlURIPtr or NULL in case of error
+ * Returns an copy of the string, but escaped
+ *
+ * 25 May 2001
+ * Uses xmlParseURI and xmlURIEscapeStr to try to escape correctly
+ * according to RFC2396.
+ * - Carl Douglas
*/
-xmlURIPtr
-xmlParseURI(const char *str) {
+xmlChar *
+xmlURIEscape(const xmlChar * str)
+{
+ xmlChar *ret, *segment = NULL;
xmlURIPtr uri;
- int ret;
+ int ret2;
+
+#define NULLCHK(p) if(!p) { \
+ xmlURIErrMemory("escaping URI value\n"); \
+ xmlFreeURI(uri); \
+ return NULL; } \
if (str == NULL)
- return(NULL);
+ return (NULL);
+
uri = xmlCreateURI();
if (uri != NULL) {
- ret = xmlParseURIReference(uri, str);
- if (ret) {
- xmlFreeURI(uri);
- return(NULL);
- }
+ /*
+ * Allow escaping errors in the unescaped form
+ */
+ uri->cleanup = 1;
+ ret2 = xmlParseURIReference(uri, (const char *)str);
+ if (ret2) {
+ xmlFreeURI(uri);
+ return (NULL);
+ }
}
- return(uri);
-}
-/**
- * xmlParseURIRaw:
- * @str: the URI string to analyze
- * @raw: if 1 unescaping of URI pieces are disabled
- *
- * Parse an URI but allows to keep intact the original fragments.
- *
- * URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
- *
- * Returns a newly built xmlURIPtr or NULL in case of error
- */
-xmlURIPtr
-xmlParseURIRaw(const char *str, int raw) {
- xmlURIPtr uri;
- int ret;
+ if (!uri)
+ return NULL;
- if (str == NULL)
- return(NULL);
- uri = xmlCreateURI();
- if (uri != NULL) {
- if (raw) {
- uri->cleanup |= 2;
- }
- ret = xmlParseURIReference(uri, str);
- if (ret) {
- xmlFreeURI(uri);
- return(NULL);
- }
+ ret = NULL;
+
+ if (uri->scheme) {
+ segment = xmlURIEscapeStr(BAD_CAST uri->scheme, BAD_CAST "+-.");
+ NULLCHK(segment)
+ ret = xmlStrcat(ret, segment);
+ ret = xmlStrcat(ret, BAD_CAST ":");
+ xmlFree(segment);
}
- return(uri);
+
+ if (uri->authority) {
+ segment =
+ xmlURIEscapeStr(BAD_CAST uri->authority, BAD_CAST "/?;:@");
+ NULLCHK(segment)
+ ret = xmlStrcat(ret, BAD_CAST "//");
+ ret = xmlStrcat(ret, segment);
+ xmlFree(segment);
+ }
+
+ if (uri->user) {
+ segment = xmlURIEscapeStr(BAD_CAST uri->user, BAD_CAST ";:&=+$,");
+ NULLCHK(segment)
+ ret = xmlStrcat(ret,BAD_CAST "//");
+ ret = xmlStrcat(ret, segment);
+ ret = xmlStrcat(ret, BAD_CAST "@");
+ xmlFree(segment);
+ }
+
+ if (uri->server) {
+ segment = xmlURIEscapeStr(BAD_CAST uri->server, BAD_CAST "/?;:@");
+ NULLCHK(segment)
+ if (uri->user == NULL)
+ ret = xmlStrcat(ret, BAD_CAST "//");
+ ret = xmlStrcat(ret, segment);
+ xmlFree(segment);
+ }
+
+ if (uri->port) {
+ xmlChar port[10];
+
+ snprintf((char *) port, 10, "%d", uri->port);
+ ret = xmlStrcat(ret, BAD_CAST ":");
+ ret = xmlStrcat(ret, port);
+ }
+
+ if (uri->path) {
+ segment =
+ xmlURIEscapeStr(BAD_CAST uri->path, BAD_CAST ":@&=+$,/?;");
+ NULLCHK(segment)
+ ret = xmlStrcat(ret, segment);
+ xmlFree(segment);
+ }
+
+ if (uri->query_raw) {
+ ret = xmlStrcat(ret, BAD_CAST "?");
+ ret = xmlStrcat(ret, BAD_CAST uri->query_raw);
+ }
+ else if (uri->query) {
+ segment =
+ xmlURIEscapeStr(BAD_CAST uri->query, BAD_CAST ";/?:@&=+,$");
+ NULLCHK(segment)
+ ret = xmlStrcat(ret, BAD_CAST "?");
+ ret = xmlStrcat(ret, segment);
+ xmlFree(segment);
+ }
+
+ if (uri->opaque) {
+ segment = xmlURIEscapeStr(BAD_CAST uri->opaque, BAD_CAST "");
+ NULLCHK(segment)
+ ret = xmlStrcat(ret, segment);
+ xmlFree(segment);
+ }
+
+ if (uri->fragment) {
+ segment = xmlURIEscapeStr(BAD_CAST uri->fragment, BAD_CAST "#");
+ NULLCHK(segment)
+ ret = xmlStrcat(ret, BAD_CAST "#");
+ ret = xmlStrcat(ret, segment);
+ xmlFree(segment);
+ }
+
+ xmlFreeURI(uri);
+#undef NULLCHK
+
+ return (ret);
}
/************************************************************************
*
* Computes he final URI of the reference done by checking that
* the given URI is valid, and building the final URI using the
- * base URI. This is processed according to section 5.2 of the
+ * base URI. This is processed according to section 5.2 of the
* RFC 2396
*
* 5.2. Resolving Relative References to Absolute Form
* as a reference to "." rather than as a synonym for the current
* URI. Should we do that here?
*/
- if (URI == NULL)
+ if (URI == NULL)
ret = -1;
else {
if (*URI) {
res->server = xmlMemStrdup(bas->server);
if (bas->user != NULL)
res->user = xmlMemStrdup(bas->user);
- res->port = bas->port;
+ res->port = bas->port;
}
if (bas->path != NULL)
res->path = xmlMemStrdup(bas->path);
- if (ref->query != NULL)
+ if (ref->query_raw != NULL)
+ res->query_raw = xmlMemStrdup (ref->query_raw);
+ else if (ref->query != NULL)
res->query = xmlMemStrdup(ref->query);
+ else if (bas->query_raw != NULL)
+ res->query_raw = xmlMemStrdup(bas->query_raw);
else if (bas->query != NULL)
res->query = xmlMemStrdup(bas->query);
if (ref->fragment != NULL)
}
if (bas->scheme != NULL)
res->scheme = xmlMemStrdup(bas->scheme);
-
- if (ref->query != NULL)
+
+ if (ref->query_raw != NULL)
+ res->query_raw = xmlMemStrdup(ref->query_raw);
+ else if (ref->query != NULL)
res->query = xmlMemStrdup(ref->query);
if (ref->fragment != NULL)
res->fragment = xmlMemStrdup(ref->fragment);
res->server = xmlMemStrdup(ref->server);
if (ref->user != NULL)
res->user = xmlMemStrdup(ref->user);
- res->port = ref->port;
+ res->port = ref->port;
}
if (ref->path != NULL)
res->path = xmlMemStrdup(ref->path);
res->server = xmlMemStrdup(bas->server);
if (bas->user != NULL)
res->user = xmlMemStrdup(bas->user);
- res->port = bas->port;
+ res->port = bas->port;
}
/*
len += strlen(bas->path);
res->path = (char *) xmlMallocAtomic(len);
if (res->path == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlBuildURI: out of memory\n");
+ xmlURIErrMemory("resolving URI against base\n");
goto done;
}
res->path[0] = 0;
uptr = (xmlChar *) ref->path;
if (*uptr == '/')
uptr++;
- val = xmlStrdup(uptr);
+ /* exception characters from xmlSaveUri */
+ val = xmlURIEscapeStr(uptr, BAD_CAST "/;&=+$,");
}
goto done;
}
}
len = xmlStrlen (uptr) + 1;
}
-
+
if (nbslash == 0) {
if (uptr != NULL)
- val = xmlStrdup (uptr);
+ /* exception characters from xmlSaveUri */
+ val = xmlURIEscapeStr(uptr, BAD_CAST "/;&=+$,");
goto done;
}
*/
val = (xmlChar *) xmlMalloc (len + 3 * nbslash);
if (val == NULL) {
- xmlGenericError(xmlGenericErrorContext,
- "xmlBuildRelativeURI: out of memory\n");
+ xmlURIErrMemory("building relative URI\n");
goto done;
}
vptr = val;
vptr[len - 1] = 0;
}
+ /* escape the freshly-built path */
+ vptr = val;
+ /* exception characters from xmlSaveUri */
+ val = xmlURIEscapeStr(vptr, BAD_CAST "/;&=+$,");
+ xmlFree(vptr);
+
done:
/*
* Free the working variables
* xmlCanonicPath:
* @path: the resource locator in a filesystem notation
*
- * Constructs a canonic path from the specified path.
+ * Constructs a canonic path from the specified path.
*
- * Returns a new canonic path, or a duplicate of the path parameter if the
+ * Returns a new canonic path, or a duplicate of the path parameter if the
* construction fails. The caller is responsible for freeing the memory occupied
- * by the returned string. If there is insufficient memory available, or the
+ * by the returned string. If there is insufficient memory available, or the
* argument is NULL, the function returns NULL.
*/
-#define IS_WINDOWS_PATH(p) \
+#define IS_WINDOWS_PATH(p) \
((p != NULL) && \
(((p[0] >= 'a') && (p[0] <= 'z')) || \
((p[0] >= 'A') && (p[0] <= 'Z'))) && \
xmlChar *
xmlCanonicPath(const xmlChar *path)
{
-#if defined(_WIN32) && !defined(__CYGWIN__)
+/*
+ * For Windows implementations, additional work needs to be done to
+ * replace backslashes in pathnames with "forward slashes"
+ */
+#if defined(_WIN32) && !defined(__CYGWIN__)
int len = 0;
int i = 0;
xmlChar *p = NULL;
if (path == NULL)
return(NULL);
+
+#if defined(_WIN32)
+ /*
+ * We must not change the backslashes to slashes if the the path
+ * starts with \\?\
+ * Those paths can be up to 32k characters long.
+ * Was added specifically for OpenOffice, those paths can't be converted
+ * to URIs anyway.
+ */
+ if ((path[0] == '\\') && (path[1] == '\\') && (path[2] == '?') &&
+ (path[3] == '\\') )
+ return xmlStrdup((const xmlChar *) path);
+#endif
+
+ /* sanitize filename starting with // so it can be used as URI */
+ if ((path[0] == '/') && (path[1] == '/') && (path[2] != '/'))
+ path++;
+
if ((uri = xmlParseURI((const char *) path)) != NULL) {
xmlFreeURI(uri);
return xmlStrdup(path);
}
+ /* Check if this is an "absolute uri" */
absuri = xmlStrstr(path, BAD_CAST "://");
if (absuri != NULL) {
int l, j;
/*
* this looks like an URI where some parts have not been
- * escaped leading to a parsing problem check that the first
+ * escaped leading to a parsing problem. Check that the first
* part matches a protocol.
*/
l = absuri - path;
+ /* Bypass if first part (part before the '://') is > 20 chars */
if ((l <= 0) || (l > 20))
goto path_processing;
+ /* Bypass if any non-alpha characters are present in first part */
for (j = 0;j < l;j++) {
c = path[j];
if (!(((c >= 'a') && (c <= 'z')) || ((c >= 'A') && (c <= 'Z'))))
goto path_processing;
}
+ /* Escape all except the characters specified in the supplied path */
escURI = xmlURIEscapeStr(path, BAD_CAST ":/?_.#&;=");
if (escURI != NULL) {
+ /* Try parsing the escaped path */
uri = xmlParseURI((const char *) escURI);
+ /* If successful, return the escaped string */
if (uri != NULL) {
xmlFreeURI(uri);
return escURI;
}
- xmlFreeURI(uri);
}
}
path_processing:
-#if defined(_WIN32) && !defined(__CYGWIN__)
+/* For Windows implementations, replace backslashes with 'forward slashes' */
+#if defined(_WIN32) && !defined(__CYGWIN__)
/*
- * This really need to be cleaned up by someone with a Windows box
+ * Create a URI structure
*/
uri = xmlCreateURI();
- if (uri == NULL) {
+ if (uri == NULL) { /* Guard against 'out of memory' */
return(NULL);
}
len = xmlStrlen(path);
if ((len > 2) && IS_WINDOWS_PATH(path)) {
+ /* make the scheme 'file' */
uri->scheme = xmlStrdup(BAD_CAST "file");
+ /* allocate space for leading '/' + path + string terminator */
uri->path = xmlMallocAtomic(len + 2);
if (uri->path == NULL) {
- xmlFreeURI(uri);
+ xmlFreeURI(uri); /* Guard agains 'out of memory' */
return(NULL);
}
+ /* Put in leading '/' plus path */
uri->path[0] = '/';
p = uri->path + 1;
strncpy(p, path, len + 1);
}
p = uri->path;
}
+ /* Now change all occurences of '\' to '/' */
while (*p != '\0') {
if (*p == '\\')
*p = '/';
p++;
}
- if (uri->path == NULL) {
- xmlFreeURI(uri);
- return(NULL);
- }
if (uri->scheme == NULL) {
- ret = xmlStrdup((const xmlChar *) path);
+ ret = xmlStrdup((const xmlChar *) uri->path);
} else {
ret = xmlSaveUri(uri);
}
*
* Constructs an URI expressing the existing path
*
- * Returns a new URI, or a duplicate of the path parameter if the
+ * Returns a new URI, or a duplicate of the path parameter if the
* construction fails. The caller is responsible for freeing the memory
* occupied by the returned string. If there is insufficient memory available,
* or the argument is NULL, the function returns NULL.
cal = xmlCanonicPath(path);
if (cal == NULL)
return(NULL);
+#if defined(_WIN32) && !defined(__CYGWIN__)
+ /* xmlCanonicPath can return an URI on Windows (is that the intended behaviour?)
+ If 'cal' is a valid URI allready then we are done here, as continuing would make
+ it invalid. */
+ if ((uri = xmlParseURI((const char *) cal)) != NULL) {
+ xmlFreeURI(uri);
+ return cal;
+ }
+ /* 'cal' can contain a relative path with backslashes. If that is processed
+ by xmlSaveURI, they will be escaped and the external entity loader machinery
+ will fail. So convert them to slashes. Misuse 'ret' for walking. */
+ ret = cal;
+ while (*ret != '\0') {
+ if (*ret == '\\')
+ *ret = '/';
+ ret++;
+ }
+#endif
memset(&temp, 0, sizeof(temp));
temp.path = (char *) cal;
ret = xmlSaveUri(&temp);
projects / platform / upstream / gettext.git / blobdiff