X-Git-Url: http://sjero.net/git/?p=wget;a=blobdiff_plain;f=src%2Fhtml-url.c;h=3c6c9b924c9a3c9a22648ce0e4a59880cde1c824;hp=97443ea0e20cc9ac7f86bf9007ab173da8cb57c2;hb=HEAD;hpb=e7d78dd2a75d0ea21a63ac4968df6079b8018cd5
diff --git a/src/html-url.c b/src/html-url.c
index 97443ea0..3c6c9b92 100644
--- a/src/html-url.c
+++ b/src/html-url.c
@@ -1,6 +1,6 @@
/* Collect URLs from HTML source.
Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006,
- 2007 Free Software Foundation, Inc.
+ 2007, 2008, 2009, 2010, 2011, 2012 Free Software Foundation, Inc.
This file is part of GNU Wget.
@@ -17,17 +17,18 @@ GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with Wget. If not, see .
-In addition, as a special exception, the Free Software Foundation
-gives permission to link the code of its release of Wget with the
-OpenSSL project's "OpenSSL" library (or with modified versions of it
-that use the same license as the "OpenSSL" library), and distribute
-the linked executables. You must obey the GNU General Public License
-in all respects for all of the code used other than "OpenSSL". If you
-modify this file, you may extend this exception to your version of the
-file, but you are not obligated to do so. If you do not wish to do
-so, delete this exception statement from your version. */
+Additional permission under GNU GPL version 3 section 7
-#include
+If you modify this program, or any covered work, by linking or
+combining it with the OpenSSL project's OpenSSL library (or a
+modified version of that library), containing parts covered by the
+terms of the OpenSSL or SSLeay licenses, the Free Software Foundation
+grants you additional permission to convey the resulting work.
+Corresponding Source for a non-source form of such a combination
+shall include the source code for the parts of OpenSSL used as well
+as that of the covered work. */
+
+#include "wget.h"
#include
#include
@@ -35,15 +36,15 @@ so, delete this exception statement from your version. */
#include
#include
-#include "wget.h"
+#include "exits.h"
#include "html-parse.h"
#include "url.h"
#include "utils.h"
#include "hash.h"
#include "convert.h"
-#include "recur.h" /* declaration of get_urls_html */
-
-struct map_context;
+#include "recur.h"
+#include "html-url.h"
+#include "css-url.h"
typedef void (*tag_handler_t) (int, struct taginfo *, struct map_context *);
@@ -78,7 +79,10 @@ enum {
TAG_SCRIPT,
TAG_TABLE,
TAG_TD,
- TAG_TH
+ TAG_TH,
+ TAG_VIDEO,
+ TAG_AUDIO,
+ TAG_SOURCE
};
/* The list of known tags and functions used for handling them. Most
@@ -109,7 +113,10 @@ static struct known_tag {
{ TAG_SCRIPT, "script", tag_find_urls },
{ TAG_TABLE, "table", tag_find_urls },
{ TAG_TD, "td", tag_find_urls },
- { TAG_TH, "th", tag_find_urls }
+ { TAG_TH, "th", tag_find_urls },
+ { TAG_VIDEO, "video", tag_find_urls },
+ { TAG_AUDIO, "audio", tag_find_urls },
+ { TAG_SOURCE, "source", tag_find_urls }
};
/* tag_url_attributes documents which attributes of which tags contain
@@ -156,23 +163,34 @@ static struct {
{ TAG_SCRIPT, "src", ATTR_INLINE },
{ TAG_TABLE, "background", ATTR_INLINE },
{ TAG_TD, "background", ATTR_INLINE },
- { TAG_TH, "background", ATTR_INLINE }
+ { TAG_TH, "background", ATTR_INLINE },
+ { TAG_VIDEO, "src", ATTR_INLINE },
+ { TAG_VIDEO, "poster", ATTR_INLINE },
+ { TAG_AUDIO, "src", ATTR_INLINE },
+ { TAG_AUDIO, "poster", ATTR_INLINE },
+ { TAG_SOURCE, "src", ATTR_INLINE }
};
/* The lists of interesting tags and attributes are built dynamically,
from the information above. However, some places in the code refer
to the attributes not mentioned here. We add them manually. */
static const char *additional_attributes[] = {
- "rel", /* used by tag_handle_link */
- "http-equiv", /* used by tag_handle_meta */
- "name", /* used by tag_handle_meta */
- "content", /* used by tag_handle_meta */
- "action" /* used by tag_handle_form */
+ "rel", /* used by tag_handle_link */
+ "type", /* used by tag_handle_link */
+ "http-equiv", /* used by tag_handle_meta */
+ "name", /* used by tag_handle_meta */
+ "content", /* used by tag_handle_meta */
+ "action", /* used by tag_handle_form */
+ "style" /* used by check_style_attr */
};
static struct hash_table *interesting_tags;
static struct hash_table *interesting_attributes;
+/* Will contains the (last) charset found in 'http-equiv=content-type'
+ meta tags */
+static char *meta_charset;
+
static void
init_interesting (void)
{
@@ -185,7 +203,7 @@ init_interesting (void)
matches the user's preferences as specified through --ignore-tags
and --follow-tags. */
- int i;
+ size_t i;
interesting_tags = make_nocase_string_hash_table (countof (known_tags));
/* First, add all the tags we know hot to handle, mapped to their
@@ -246,34 +264,30 @@ find_attr (struct taginfo *tag, const char *name, int *attrind)
return NULL;
}
-struct map_context {
- char *text; /* HTML text. */
- char *base; /* Base URI of the document, possibly
- changed through . */
- const char *parent_base; /* Base of the current document. */
- const char *document_file; /* File name of this document. */
- bool nofollow; /* whether NOFOLLOW was specified in a
- tag. */
-
- struct urlpos *head, *tail; /* List of URLs that is being
- built. */
-};
+/* used for calls to append_url */
+#define ATTR_POS(tag, attrind, ctx) \
+ (tag->attrs[attrind].value_raw_beginning - ctx->text)
+#define ATTR_SIZE(tag, attrind) \
+ (tag->attrs[attrind].value_raw_size)
/* Append LINK_URI to the urlpos structure that is being built.
- LINK_URI will be merged with the current document base. TAG and
- ATTRIND are the necessary context to store the position and
- size. */
+ LINK_URI will be merged with the current document base.
+*/
-static struct urlpos *
-append_url (const char *link_uri,
- struct taginfo *tag, int attrind, struct map_context *ctx)
+struct urlpos *
+append_url (const char *link_uri, int position, int size,
+ struct map_context *ctx)
{
int link_has_scheme = url_has_scheme (link_uri);
struct urlpos *newel;
const char *base = ctx->base ? ctx->base : ctx->parent_base;
struct url *url;
+ struct iri *iri = iri_new ();
+ set_uri_encoding (iri, opt.locale, true);
+ iri->utf8_encode = true;
+
if (!base)
{
DEBUGP (("%s: no base, merge will use \"%s\".\n",
@@ -291,7 +305,7 @@ append_url (const char *link_uri,
return NULL;
}
- url = url_parse (link_uri, NULL);
+ url = url_parse (link_uri, NULL, iri, false);
if (!url)
{
DEBUGP (("%s: link \"%s\" doesn't parse.\n",
@@ -307,10 +321,13 @@ append_url (const char *link_uri,
char *complete_uri = uri_merge (base, link_uri);
- DEBUGP (("%s: merge(\"%s\", \"%s\") -> %s\n",
- ctx->document_file, base, link_uri, complete_uri));
+ DEBUGP (("%s: merge(%s, %s) -> %s\n",
+ quotearg_n_style (0, escape_quoting_style, ctx->document_file),
+ quote_n (1, base),
+ quote_n (2, link_uri),
+ quotearg_n_style (3, escape_quoting_style, complete_uri)));
- url = url_parse (complete_uri, NULL);
+ url = url_parse (complete_uri, NULL, iri, false);
if (!url)
{
DEBUGP (("%s: merged link \"%s\" doesn't parse.\n",
@@ -321,12 +338,14 @@ append_url (const char *link_uri,
xfree (complete_uri);
}
- DEBUGP (("appending \"%s\" to urlpos.\n", url->url));
+ iri_free (iri);
+
+ DEBUGP (("appending %s to urlpos.\n", quote (url->url)));
newel = xnew0 (struct urlpos);
newel->url = url;
- newel->pos = tag->attrs[attrind].value_raw_beginning - ctx->text;
- newel->size = tag->attrs[attrind].value_raw_size;
+ newel->pos = position;
+ newel->size = size;
/* A URL is relative if the host is not named, and the name does not
start with `/'. */
@@ -335,17 +354,58 @@ append_url (const char *link_uri,
else if (link_has_scheme)
newel->link_complete_p = 1;
- if (ctx->tail)
+ /* Append the new URL maintaining the order by position. */
+ if (ctx->head == NULL)
+ ctx->head = newel;
+ else
{
- ctx->tail->next = newel;
- ctx->tail = newel;
+ struct urlpos *it, *prev = NULL;
+
+ it = ctx->head;
+ while (it && position > it->pos)
+ {
+ prev = it;
+ it = it->next;
+ }
+
+ newel->next = it;
+
+ if (prev)
+ prev->next = newel;
+ else
+ ctx->head = newel;
}
- else
- ctx->tail = ctx->head = newel;
return newel;
}
+static void
+check_style_attr (struct taginfo *tag, struct map_context *ctx)
+{
+ int attrind;
+ int raw_start;
+ int raw_len;
+ char *style = find_attr (tag, "style", &attrind);
+ if (!style)
+ return;
+
+ /* raw pos and raw size include the quotes, skip them when they are
+ present. */
+ raw_start = ATTR_POS (tag, attrind, ctx);
+ raw_len = ATTR_SIZE (tag, attrind);
+ if( *(char *)(ctx->text + raw_start) == '\''
+ || *(char *)(ctx->text + raw_start) == '"')
+ {
+ raw_start += 1;
+ raw_len -= 2;
+ }
+
+ if(raw_len <= 0)
+ return;
+
+ get_urls_css (ctx, raw_start, raw_len);
+}
+
/* All the tag_* functions are called from collect_tags_mapper, as
specified by KNOWN_TAGS. */
@@ -355,7 +415,8 @@ append_url (const char *link_uri,
static void
tag_find_urls (int tagid, struct taginfo *tag, struct map_context *ctx)
{
- int i, attrind;
+ size_t i;
+ int attrind;
int first = -1;
for (i = 0; i < countof (tag_url_attributes); i++)
@@ -382,7 +443,7 @@ tag_find_urls (int tagid, struct taginfo *tag, struct map_context *ctx)
/* Find whether TAG/ATTRIND is a combination that contains a
URL. */
char *link = tag->attrs[attrind].value;
- const int size = countof (tag_url_attributes);
+ const size_t size = countof (tag_url_attributes);
/* If you're cringing at the inefficiency of the nested loops,
remember that they both iterate over a very small number of
@@ -393,7 +454,8 @@ tag_find_urls (int tagid, struct taginfo *tag, struct map_context *ctx)
if (0 == strcasecmp (tag->attrs[attrind].name,
tag_url_attributes[i].attr_name))
{
- struct urlpos *up = append_url (link, tag, attrind, ctx);
+ struct urlpos *up = append_url (link, ATTR_POS(tag,attrind,ctx),
+ ATTR_SIZE(tag,attrind), ctx);
if (up)
{
int flags = tag_url_attributes[i].flags;
@@ -410,7 +472,7 @@ tag_find_urls (int tagid, struct taginfo *tag, struct map_context *ctx)
/* Handle the BASE tag, for . */
static void
-tag_handle_base (int tagid, struct taginfo *tag, struct map_context *ctx)
+tag_handle_base (int tagid _GL_UNUSED, struct taginfo *tag, struct map_context *ctx)
{
struct urlpos *base_urlpos;
int attrind;
@@ -418,7 +480,8 @@ tag_handle_base (int tagid, struct taginfo *tag, struct map_context *ctx)
if (!newbase)
return;
- base_urlpos = append_url (newbase, tag, attrind, ctx);
+ base_urlpos = append_url (newbase, ATTR_POS(tag,attrind,ctx),
+ ATTR_SIZE(tag,attrind), ctx);
if (!base_urlpos)
return;
base_urlpos->ignore_when_downloading = 1;
@@ -435,13 +498,15 @@ tag_handle_base (int tagid, struct taginfo *tag, struct map_context *ctx)
/* Mark the URL found in