/*
* Process redirections (HTTP 30x answers)
* (This is a work in progress --not finished yet)
*/
static int Cache_redirect(CacheEntry_t *entry, int Flags, BrowserWindow *bw)
{
DilloUrl *NewUrl;
_MSG(" Cache_redirect: redirect_level = %d\n", bw->redirect_level);
/* Don't allow redirection for SpamSafe/local URLs */
if (URL_FLAGS(entry->Url) & URL_SpamSafe) {
a_UIcmd_set_msg(bw, "WARNING: local URL with redirection. Aborting.");
return 0;
}
/* if there's a redirect loop, stop now */
if (bw->redirect_level >= 5)
entry->Flags |= CA_RedirectLoop;
if (entry->Flags & CA_RedirectLoop) {
a_UIcmd_set_msg(bw, "ERROR: redirect loop for: %s", URL_STR_(entry->Url));
bw->redirect_level = 0;
return 0;
}
if ((entry->Flags & CA_Redirect && entry->Location) &&
(entry->Flags & CA_ForceRedirect || entry->Flags & CA_TempRedirect ||
!entry->Data->len || entry->Data->len < 1024)) {
_MSG(">>>> Redirect from: %s\n to %s <<<<\n",
URL_STR_(entry->Url), URL_STR_(entry->Location));
_MSG("%s", entry->Header->str);
if (Flags & WEB_RootUrl) {
/* Redirection of the main page */
NewUrl = a_Url_new(URL_STR_(entry->Location), URL_STR_(entry->Url));
if (entry->Flags & CA_TempRedirect)
a_Url_set_flags(NewUrl, URL_FLAGS(NewUrl) | URL_E2EQuery);
a_Nav_push(bw, NewUrl, entry->Url);
a_Url_free(NewUrl);
} else {
/* Sub entity redirection (most probably an image) */
if (!entry->Data->len) {
_MSG(">>>> Image redirection without entity-content <<<<\n");
} else {
_MSG(">>>> Image redirection with entity-content <<<<\n");
}
}
}
return 0;
}
/*
* Given a URL, find its cache entry, following redirections.
*/
static CacheEntry_t *Cache_entry_search_with_redirect(const DilloUrl *Url)
{
int i;
CacheEntry_t *entry;
for (i = 0; (entry = Cache_entry_search(Url)); ++i) {
/* Test for a redirection loop */
if (entry->Flags & CA_RedirectLoop || i == 3) {
_MSG_WARN("Redirect loop for URL: >%s<\n", URL_STR_(Url));
break;
}
/* Test for a working redirection */
if (entry && entry->Flags & CA_Redirect && entry->Location) {
Url = entry->Location;
} else
break;
}
return entry;
}
static void Cache_finish_msg(CacheEntry_t *entry)
{
if (entry->Flags & CA_GotData) {
/* already finished */
return;
}
if ((entry->ExpectedSize || entry->TransferSize) &&
entry->TypeHdr == NULL) {
MSG_HTTP("Message with a body lacked Content-Type header.\n");
}
if ((entry->Flags & CA_GotLength) &&
(entry->ExpectedSize != entry->TransferSize)) {
MSG_HTTP("Content-Length does NOT match message body at\n"
"%s\n", URL_STR_(entry->Url));
MSG("Expected size: %d, Transfer size: %d\n",
entry->ExpectedSize, entry->TransferSize);
}
entry->Flags |= CA_GotData;
entry->Flags &= ~CA_Stopped; /* it may catch up! */
if (entry->TransferDecoder) {
a_Decode_transfer_free(entry->TransferDecoder);
entry->TransferDecoder = NULL;
}
if (entry->ContentDecoder) {
a_Decode_free(entry->ContentDecoder);
entry->ContentDecoder = NULL;
}
dStr_fit(entry->Data); /* fit buffer size! */
if ((entry = Cache_process_queue(entry))) {
if (entry->Flags & CA_GotHeader) {
Cache_unref_data(entry);
}
}
}
/*
* Receive new data, update the reception buffer (for next read), update the
* cache, and service the client queue.
*
* This function gets called whenever the IO has new data.
* 'Op' is the operation to perform
* 'VPtr' is a (void) pointer to the IO control structure
*/
void a_Cache_process_dbuf(int Op, const char *buf, size_t buf_size,
const DilloUrl *Url)
{
int offset, len;
const char *str;
Dstr *dstr1, *dstr2, *dstr3;
CacheEntry_t *entry = Cache_entry_search(Url);
/* Assert a valid entry (not aborted) */
dReturn_if_fail (entry != NULL);
_MSG("__a_Cache_process_dbuf__\n");
if (Op == IORead) {
/*
* Cache_get_header() will set CA_GotHeader if it has a full header, and
* Cache_parse_header() will unset it if the header ends being
* merely an informational response from the server (i.e., 100 Continue)
*/
for (offset = 0; !(entry->Flags & CA_GotHeader) &&
(len = Cache_get_header(entry, buf + offset, buf_size - offset));
Cache_parse_header(entry) ) {
offset += len;
}
if (entry->Flags & CA_GotHeader) {
str = buf + offset;
len = buf_size - offset;
entry->TransferSize += len;
dstr1 = dstr2 = dstr3 = NULL;
/* Decode arrived data (<= 3 stages) */
if (entry->TransferDecoder) {
dstr1 = a_Decode_process(entry->TransferDecoder, str, len);
str = dstr1->str;
len = dstr1->len;
}
if (entry->ContentDecoder) {
dstr2 = a_Decode_process(entry->ContentDecoder, str, len);
str = dstr2->str;
len = dstr2->len;
}
dStr_append_l(entry->Data, str, len);
if (entry->CharsetDecoder && entry->UTF8Data) {
dstr3 = a_Decode_process(entry->CharsetDecoder, str, len);
dStr_append_l(entry->UTF8Data, dstr3->str, dstr3->len);
}
dStr_free(dstr1, 1);
dStr_free(dstr2, 1);
dStr_free(dstr3, 1);
if (entry->Data->len)
entry->Flags &= ~CA_IsEmpty;
entry = Cache_process_queue(entry);
}
} else if (Op == IOClose) {
if ((entry->ExpectedSize || entry->TransferSize) &&
entry->TypeHdr == NULL) {
MSG_HTTP("Message with a body lacked Content-Type header.\n");
}
if ((entry->Flags & CA_GotLength) &&
(entry->ExpectedSize != entry->TransferSize)) {
MSG_HTTP("Content-Length does NOT match message body,\n"
" at: %s\n", URL_STR_(entry->Url));
MSG("entry->ExpectedSize = %d, entry->TransferSize = %d\n",
entry->ExpectedSize, entry->TransferSize);
}
if (!entry->TransferSize && !(entry->Flags & CA_Redirect) &&
(entry->Flags & WEB_RootUrl)) {
char *eol = strchr(entry->Header->str, '\n');
if (eol) {
char *status_line = dStrndup(entry->Header->str,
eol - entry->Header->str);
MSG_HTTP("Body was empty. Server sent status: %s\n", status_line);
dFree(status_line);
}
}
entry->Flags |= CA_GotData;
entry->Flags &= ~CA_Stopped; /* it may catch up! */
if (entry->TransferDecoder) {
a_Decode_free(entry->TransferDecoder);
entry->TransferDecoder = NULL;
}
if (entry->ContentDecoder) {
a_Decode_free(entry->ContentDecoder);
entry->ContentDecoder = NULL;
}
dStr_fit(entry->Data); /* fit buffer size! */
if ((entry = Cache_process_queue(entry))) {
if (entry->Flags & CA_GotHeader) {
Cache_unref_data(entry);
}
}
} else if (Op == IOAbort) {
/* unused */
MSG("a_Cache_process_dbuf Op = IOAbort; not implemented!\n");
}
}
/*
* Scan, allocate, and set things according to header info.
* (This function needs the whole header to work)
*/
static void Cache_parse_header(CacheEntry_t *entry)
{
char *header = entry->Header->str;
char *Length, *Type, *location_str, *encoding;
#ifndef DISABLE_COOKIES
Dlist *Cookies;
#endif
Dlist *warnings;
void *data;
int i;
_MSG("Cache_parse_header\n");
if (entry->Header->len > 12) {
if (header[9] == '1' && header[10] == '0' && header[11] == '0') {
/* 100: Continue. The "real" header has not come yet. */
MSG("An actual 100 Continue header!\n");
entry->Flags &= ~CA_GotHeader;
dStr_free(entry->Header, 1);
entry->Header = dStr_new("");
return;
}
if (header[9] == '3' && header[10] == '0' &&
(location_str = Cache_parse_field(header, "Location"))) {
/* 30x: URL redirection */
DilloUrl *location_url = a_Url_new(location_str,URL_STR_(entry->Url));
if (prefs.filter_auto_requests == PREFS_FILTER_SAME_DOMAIN &&
!a_Url_same_organization(entry->Url, location_url)) {
/* don't redirect; just show body like usual (if any) */
MSG("Redirection not followed from %s to %s\n",
URL_HOST(entry->Url), URL_STR(location_url));
a_Url_free(location_url);
} else {
entry->Flags |= CA_Redirect;
if (header[11] == '1')
entry->Flags |= CA_ForceRedirect; /* 301 Moved Permanently */
else if (header[11] == '2')
entry->Flags |= CA_TempRedirect; /* 302 Temporary Redirect */
if (URL_FLAGS(location_url) & (URL_Post + URL_Get) &&
dStrAsciiCasecmp(URL_SCHEME(location_url), "dpi") == 0 &&
dStrAsciiCasecmp(URL_SCHEME(entry->Url), "dpi") != 0) {
/* Forbid dpi GET and POST from non dpi-generated urls */
MSG("Redirection Denied! '%s' -> '%s'\n",
URL_STR(entry->Url), URL_STR(location_url));
a_Url_free(location_url);
} else {
entry->Location = location_url;
}
}
dFree(location_str);
} else if (strncmp(header + 9, "401", 3) == 0) {
entry->Auth =
Cache_parse_multiple_fields(header, "WWW-Authenticate");
} else if (strncmp(header + 9, "404", 3) == 0) {
entry->Flags |= CA_NotFound;
}
}
if ((warnings = Cache_parse_multiple_fields(header, "Warning"))) {
for (i = 0; (data = dList_nth_data(warnings, i)); ++i) {
MSG_HTTP("%s\n", (char *)data);
dFree(data);
}
dList_free(warnings);
}
/*
* Get Transfer-Encoding and initialize decoder
*/
encoding = Cache_parse_field(header, "Transfer-Encoding");
entry->TransferDecoder = a_Decode_transfer_init(encoding);
if ((Length = Cache_parse_field(header, "Content-Length")) != NULL) {
if (encoding) {
/*
* If Transfer-Encoding is present, Content-Length must be ignored.
* If the Transfer-Encoding is non-identity, it is an error.
*/
if (dStrAsciiCasecmp(encoding, "identity"))
MSG_HTTP("Content-Length and non-identity Transfer-Encoding "
"headers both present.\n");
} else {
entry->Flags |= CA_GotLength;
entry->ExpectedSize = MAX(strtol(Length, NULL, 10), 0);
}
dFree(Length);
}
dFree(encoding); /* free Transfer-Encoding */
#ifndef DISABLE_COOKIES
if ((Cookies = Cache_parse_multiple_fields(header, "Set-Cookie"))) {
CacheClient_t *client;
for (i = 0; (client = dList_nth_data(ClientQueue, i)); ++i) {
if (client->Url == entry->Url) {
DilloWeb *web = client->Web;
if (!web->requester ||
a_Url_same_organization(entry->Url, web->requester)) {
char *server_date = Cache_parse_field(header, "Date");
a_Cookies_set(Cookies, entry->Url, server_date);
dFree(server_date);
break;
}
}
}
if (i >= dList_length(ClientQueue)) {
MSG("Cache: cookies not accepted from '%s'\n", URL_STR(entry->Url));
}
for (i = 0; (data = dList_nth_data(Cookies, i)); ++i)
dFree(data);
dList_free(Cookies);
}
#endif /* !DISABLE_COOKIES */
/*
* Get Content-Encoding and initialize decoder
*/
encoding = Cache_parse_field(header, "Content-Encoding");
entry->ContentDecoder = a_Decode_content_init(encoding);
dFree(encoding);
if (entry->ExpectedSize > 0) {
if (entry->ExpectedSize > HUGE_FILESIZE) {
entry->Flags |= CA_HugeFile;
}
/* Avoid some reallocs. With MAX_INIT_BUF we avoid a SEGFAULT
* with huge files (e.g. iso files).
* Note: the buffer grows automatically. */
dStr_free(entry->Data, 1);
entry->Data = dStr_sized_new(MIN(entry->ExpectedSize, MAX_INIT_BUF));
}
/* Get Content-Type */
if ((Type = Cache_parse_field(header, "Content-Type"))) {
/* This HTTP Content-Type is not trusted. It's checked against real data
* in Cache_process_queue(); only then CA_GotContentType becomes true. */
a_Cache_set_content_type(entry->Url, Type, "http");
_MSG("TypeHdr {%s} {%s}\n", Type, URL_STR(entry->Url));
_MSG("TypeMeta {%s}\n", entry->TypeMeta);
dFree(Type);
}
Cache_ref_data(entry);
}
