4:19 a.m.
Added files
Index: content/fetchers/fetch_file.c
===================================================================
--- /dev/null 2009-04-16 19:17:07.000000000 +0100
+++ content/fetchers/fetch_file.c 2010-09-09 19:47:22.000000000 +0100
@@ -0,0 +1,651 @@
+/*
+ * Copyright 2010 Vincent Sanders <vince(a)netsurf-browser.org>
+ *
+ * This file is part of NetSurf.
+ *
+ * NetSurf is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * NetSurf is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>;.
+ */
+
+/* file: URL handling. Based on the data fetcher by Rob Kendrik */
+
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <assert.h>
+#include <errno.h>
+#include <stdbool.h>
+#include <inttypes.h>
+#include <string.h>
+#include <strings.h>
+#include <time.h>
+#include <stdio.h>
+#include <dirent.h>
+#include <limits.h>
+#include <stdarg.h>
+
+#include "utils/config.h"
+#include "content/dirlist.h"
+#include "content/fetch.h"
+#include "content/fetchers/fetch_file.h"
+#include "content/urldb.h"
+#include "desktop/netsurf.h"
+#include "desktop/options.h"
+#include "utils/log.h"
+#include "utils/messages.h"
+#include "utils/url.h"
+#include "utils/utils.h"
+#include "utils/ring.h"
+
+/* Maximum size of read buffer */
+#define FETCH_FILE_MAX_BUF_SIZE (1024 * 1024)
+
+/** Context for a fetch */
+struct fetch_file_context {
+ struct fetch_file_context *r_next, *r_prev;
+
+ struct fetch *fetchh; /**< Handle for this fetch */
+
+ bool aborted; /**< Flag indicating fetch has been aborted */
+ bool locked; /**< Flag indicating entry is already entered */
+
+ char *url; /**< The full url the fetch refers to */
+ char *path; /**< The actual path to be used with open() */
+
+ int fd; /**< The file descriptor of the object */
+ struct stat fstat; /**< The objects stat */
+};
+
+static struct fetch_file_context *ring = NULL;
+
+/** issue fetch callbacks with locking */
+static bool fetch_file_send_callback(fetch_msg msg,
+ struct fetch_file_context *ctx, const void *data,
+ unsigned long size, fetch_error_code errorcode)
+{
+ ctx->locked = true;
+ fetch_send_callback(msg, ctx->fetchh, data, size, errorcode);
+ ctx->locked = false;
+
+ return ctx->aborted;
+}
+
+static bool fetch_file_send_header(struct fetch_file_context *ctx, const char *fmt, ...)
+{
+ char header[64];
+ va_list ap;
+
+ va_start(ap, fmt);
+
+ vsnprintf(header, sizeof header, fmt, ap);
+
+ va_end(ap);
+
+ LOG(("%s", header));
+
+ ctx->locked = true;
+ fetch_send_callback(FETCH_HEADER, ctx->fetchh, header, strlen(header),
FETCH_ERROR_NO_ERROR);
+ ctx->locked = false;
+
+ return ctx->aborted;
+}
+
+static bool fetch_file_send_time(struct fetch_file_context *ctx, const char *fmt, const
time_t *val)
+{
+ char header[64];
+ struct tm btm;
+
+ gmtime_r(val, &btm);
+
+ strftime(header, sizeof header, fmt, &btm);
+
+ LOG(("%s", header));
+
+ ctx->locked = true;
+ fetch_send_callback(FETCH_HEADER, ctx->fetchh, header, strlen(header),
FETCH_ERROR_NO_ERROR);
+ ctx->locked = false;
+
+ return ctx->aborted;
+}
+
+/** callback to initialise the file fetcher. */
+static bool fetch_file_initialise(const char *scheme)
+{
+ LOG(("fetch_file_initialise called for %s", scheme));
+ return true;
+}
+
+/** callback to initialise the file fetcher. */
+static void fetch_file_finalise(const char *scheme)
+{
+ LOG(("fetch_file_finalise called for %s", scheme));
+}
+
+/** callback to set up a file fetch context. */
+static void *
+fetch_file_setup(struct fetch *fetchh,
+ const char *url,
+ bool only_2xx,
+ const char *post_urlenc,
+ const struct fetch_multipart_data *post_multipart,
+ const char **headers)
+{
+ struct fetch_file_context *ctx = calloc(1, sizeof(*ctx));
+ char *path;
+
+ if (ctx == NULL)
+ return NULL;
+
+ ctx->fetchh = fetchh;
+
+ ctx->url = strdup(url);
+
+ url_path(url, &path);
+ if (path == NULL) {
+ free(ctx);
+ return NULL;
+ }
+
+ url_unescape(path, &ctx->path);
+ if (ctx->path == NULL) {
+ free(path);
+ free(ctx);
+ return NULL;
+ }
+
+ free(path);
+
+ RING_INSERT(ring, ctx);
+
+ LOG(("new context %p for fetch handle %p, url %s, path %s",
+ ctx, fetchh, ctx->url, ctx->path));
+ LOG(("only_2xx %d, post_urlenc %s, post_multipart %p, headers %p",
+ only_2xx, post_urlenc, post_multipart, headers));
+
+ return ctx;
+}
+
+/** callback to free a file fetch */
+static void fetch_file_free(void *ctx)
+{
+ struct fetch_file_context *c = ctx;
+ LOG(("context %p",ctx));
+
+ free(c->url);
+ free(c->path);
+ RING_REMOVE(ring, c);
+ free(ctx);
+}
+
+/** callback to start a file fetch */
+static bool fetch_file_start(void *ctx)
+{
+ LOG(("context %p",ctx));
+ return true;
+}
+
+/** callback to abort a file fetch */
+static void fetch_file_abort(void *ctx)
+{
+ struct fetch_file_context *c = ctx;
+ LOG(("context %p",ctx));
+
+ /* To avoid the poll loop having to deal with the fetch context
+ * disappearing from under it, we simply flag the abort here.
+ * The poll loop itself will perform the appropriate cleanup.
+ */
+ c->aborted = true;
+}
+
+
+/** Process object as a regular file */
+static void fetch_file_process_plain(struct fetch_file_context *ctx)
+{
+ char *buf;
+ size_t buf_size;
+
+ size_t tot_read = 0;
+ ssize_t res;
+
+ size_t size;
+
+ size = ctx->fstat.st_size;
+
+ /* set buffer size */
+ buf_size = size;
+ if (buf_size > FETCH_FILE_MAX_BUF_SIZE)
+ buf_size = FETCH_FILE_MAX_BUF_SIZE;
+
+ /* allocate the buffer storage */
+ buf = malloc(buf_size);
+ if (buf == NULL) {
+ fetch_file_send_callback(FETCH_ERROR, ctx,
+ "Unable to allocate memory for file data buffer",
+ 0, FETCH_ERROR_MEMORY);
+ return;
+ }
+
+ /* fetch is going to be successful */
+ fetch_set_http_code(ctx->fetchh, 200);
+
+ /* Any callback can result in the fetch being aborted.
+ * Therefore, we _must_ check for this after _every_ call to
+ * fetch_file_send_callback().
+ */
+
+ /* content type */
+ if (fetch_file_send_header(ctx, "Content-Type: %s",
fetch_filetype(ctx->path)))
+ goto fetch_file_process_aborted;
+
+ /* content length */
+ if (fetch_file_send_header(ctx, "Content-Length: %zd", size))
+ goto fetch_file_process_aborted;
+
+ /* Set Last modified header */
+ if (fetch_file_send_time(ctx, "Last-Modified: %a, %d %b %Y %H:%M:%S GMT",
&ctx->fstat.st_mtime))
+ goto fetch_file_process_aborted;
+
+ /* create etag */
+ if (fetch_file_send_header(ctx, "ETag: \"%10" PRId64 "\"",
(int64_t) ctx->fstat.st_mtime))
+ goto fetch_file_process_aborted;
+
+ /* main data loop */
+ do {
+ res = read(ctx->fd, buf, buf_size);
+ if (res == -1) {
+ fetch_file_send_callback(FETCH_ERROR, ctx, "Error reading file", 0,
FETCH_ERROR_PARTIAL_FILE);
+ return;
+ }
+
+
+ if (res == 0) {
+ fetch_file_send_callback(FETCH_ERROR, ctx, "Unexpected EOF reading file", 0,
FETCH_ERROR_PARTIAL_FILE);
+ return;
+ }
+
+ tot_read += res;
+
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buf, res, FETCH_ERROR_NO_ERROR))
+ break;
+
+ } while (tot_read < size);
+
+ if (!ctx->aborted)
+ fetch_file_send_callback(FETCH_FINISHED, ctx, 0, 0, FETCH_ERROR_NO_ERROR);
+
+fetch_file_process_aborted:
+
+ close(ctx->fd);
+ free(buf);
+ return;
+}
+
+static char *gen_nice_title(char *path)
+{
+ char *nice_path, *cnv, *tmp;
+ char *title;
+ int title_length;
+
+ /* Convert path for display */
+ nice_path = malloc(strlen(path) * 4 + 1);
+ if (!nice_path) {
+ return NULL;
+ }
+
+ /* Escape special HTML characters */
+ for (cnv = nice_path, tmp = path; *tmp != '\0'; tmp++) {
+ if (*tmp == '<') {
+ *cnv++ = '&';
+ *cnv++ = 'l';
+ *cnv++ = 't';
+ *cnv++ = ';';
+ } else if (*tmp == '>') {
+ *cnv++ = '&';
+ *cnv++ = 'g';
+ *cnv++ = 't';
+ *cnv++ = ';';
+ } else {
+ *cnv++ = *tmp;
+ }
+ }
+ *cnv = '\0';
+
+ /* Construct a localised title string */
+ title_length = strlen(nice_path) + strlen(messages_get("FileIndex"));
+ title = malloc(title_length);
+
+ if (title == NULL) {
+ return NULL;
+ }
+
+ /* Set title to localised "Index of <nice_path>" */
+ snprintf(title, title_length, messages_get("FileIndex"), nice_path);
+
+ free(nice_path);
+
+ return title;
+}
+
+static void fetch_file_process_error(struct fetch_file_context *ctx, int code)
+{
+ char buffer[1024];
+ const char *title;
+
+ /* content is going to return error code */
+ fetch_set_http_code(ctx->fetchh, code);
+
+ /* content type */
+ if (fetch_file_send_header(ctx, "Content-Type: text/html"))
+ goto fetch_file_process_error_aborted;
+
+ switch (code) {
+ case 401:
+ title = "Permission Denied";
+ break;
+
+ case 404:
+ title = "File not found";
+ break;
+
+ default:
+ title = "Internal Error";
+ }
+
+ snprintf(buffer, sizeof buffer,
"<html><head><title>%s</title></head><body><h1>%s</h1><p>Error
%d while fetching file %s</p></body></html>", title, title,
code,ctx->url);
+
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buffer, strlen(buffer),
FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_error_aborted;
+
+ fetch_file_send_callback(FETCH_FINISHED, ctx, 0, 0, FETCH_ERROR_NO_ERROR);
+
+fetch_file_process_error_aborted:
+ return;
+}
+
+static void fetch_file_process_dir(struct fetch_file_context *ctx)
+{
+ char buffer[1024]; /* Output buffer */
+ bool even = false; /* formatting flag */
+ char *title; /* pretty printed title */
+ url_func_result res; /* result from url routines */
+ char *up; /* url of parent */
+ bool compare; /* result of url compare */
+
+ DIR *scandir; /* handle for enumerating the directory */
+ struct dirent* ent; /* leaf directory entry */
+ struct stat ent_stat; /* stat result of leaf entry */
+ char datebuf[64]; /* buffer for date text */
+ char timebuf[64]; /* buffer for time text */
+ char urlpath[PATH_MAX]; /* buffer for leaf entry path */
+
+#if defined(HAVE_FDOPENDIR)
+ scandir = fdopendir(ctx->fd);
+#else
+ /* this poses the possibility of a race where the directory
+ * has been removed from the namespace or resources for more
+ * fd are now unavailable between the previous open() and this
+ * call.
+ */
+ close(ctx->fd);
+ scandir = opendir(ctx->path);
+#endif
+
+ if (scandir == NULL) {
+ fetch_file_process_error(ctx, 500);
+ return;
+ }
+
+ /* fetch is going to be successful */
+ fetch_set_http_code(ctx->fetchh, 200);
+
+ /* content type */
+ if (fetch_file_send_header(ctx, "Content-Type: text/html"))
+ goto fetch_file_process_dir_aborted;
+
+ /* directory listing top */
+ dirlist_generate_top(buffer, sizeof buffer);
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buffer, strlen(buffer),
FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_dir_aborted;
+
+ /* directory listing title */
+ title = gen_nice_title(ctx->path);
+ dirlist_generate_title(title, buffer, sizeof buffer);
+ free(title);
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buffer, strlen(buffer),
FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_dir_aborted;
+
+ /* Print parent directory link */
+ res = url_parent(ctx->url, &up);
+ if (res == URL_FUNC_OK) {
+ res = url_compare(ctx->url, up, false, &compare);
+ if ((res == URL_FUNC_OK) && !compare) {
+ dirlist_generate_parent_link(up, buffer, sizeof buffer);
+
+ fetch_file_send_callback(FETCH_DATA, ctx,
+ buffer,
+ strlen(buffer),
+ FETCH_ERROR_NO_ERROR);
+
+ }
+ free(up);
+
+ if (ctx->aborted)
+ goto fetch_file_process_dir_aborted;
+
+ }
+
+ /* directory list headings */
+ dirlist_generate_headings(buffer, sizeof buffer);
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buffer, strlen(buffer),
FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_dir_aborted;
+
+ while ((ent = readdir(scandir)) != NULL) {
+
+ if (ent->d_name[0] == '.')
+ continue;
+
+ snprintf(urlpath,
+ sizeof urlpath,
+ "%s%s%s",
+ ctx->path,
+ (ctx->path[strlen(ctx->path) - 1] == '/')?"":"/" ,
+ ent->d_name);
+
+ if (stat(urlpath, &ent_stat) != 0) {
+ ent_stat.st_mode = 0;
+ datebuf[0] = 0;
+ timebuf[0] = 0;
+ } else {
+ /* Get date in output format */
+ if (strftime((char *)&datebuf, sizeof datebuf,
+ "%a %d %b %Y",
+ localtime(&ent_stat.st_mtime)) == 0) {
+ strncpy(datebuf, "-", sizeof datebuf);
+ }
+
+ /* Get time in output format */
+ if (strftime((char *)&timebuf, sizeof timebuf,
+ "%H:%M",
+ localtime(&ent_stat.st_mtime)) == 0) {
+ strncpy(timebuf, "-", sizeof timebuf);
+ }
+ }
+
+ if (S_ISREG(ent_stat.st_mode)) {
+ /* regular file */
+ dirlist_generate_row(even,
+ false,
+ urlpath,
+ ent->d_name,
+ fetch_filetype(urlpath),
+ ent_stat.st_size,
+ datebuf, timebuf,
+ buffer, sizeof(buffer));
+ } else if (S_ISDIR(ent_stat.st_mode)) {
+ /* directory */
+ dirlist_generate_row(even,
+ true,
+ urlpath,
+ ent->d_name,
+ messages_get("FileDirectory"),
+ -1,
+ datebuf, timebuf,
+ buffer, sizeof(buffer));
+ } else {
+ /* something else */
+ dirlist_generate_row(even,
+ false,
+ urlpath,
+ ent->d_name,
+ "",
+ -1,
+ datebuf, timebuf,
+ buffer, sizeof(buffer));
+ }
+
+ if (fetch_file_send_callback(FETCH_DATA, ctx,
+ buffer,
+ strlen(buffer),
+ FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_dir_aborted;
+
+ even = !even;
+ }
+
+ /* directory listing bottom */
+ dirlist_generate_bottom(buffer, sizeof buffer);
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buffer, strlen(buffer),
FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_dir_aborted;
+
+
+ fetch_file_send_callback(FETCH_FINISHED, ctx, 0, 0, FETCH_ERROR_NO_ERROR);
+
+fetch_file_process_dir_aborted:
+
+ closedir(scandir);
+}
+
+
+/* process a file fetch */
+static void fetch_file_process(struct fetch_file_context *ctx)
+{
+ ctx->fd = open(ctx->path, O_RDONLY);
+ if (ctx->fd < 0) {
+ /* process errors as appropriate */
+ switch (errno) {
+ case EACCES:
+ fetch_file_process_error(ctx, 401);
+ break;
+
+ case ENOENT:
+ fetch_file_process_error(ctx, 404);
+ break;
+
+ default:
+ fetch_file_process_error(ctx, 500);
+ break;
+ }
+ return;
+ }
+
+ if (fstat(ctx->fd, &ctx->fstat) != 0) {
+ /* process errors as appropriate */
+ close(ctx->fd);
+ fetch_file_process_error(ctx, 500);
+ return;
+ }
+
+#ifdef fetch_curl_modified_check
+ /* f was fetch_curl context and I cannot determine where f->last_modified was set
*/
+
+ /* Report not modified, if appropriate */
+ if (f->last_modified && f->file_etag &&
+ f->last_modified > s.st_mtime &&
+ f->file_etag == s.st_mtime) {
+ fetch_send_callback(FETCH_NOTMODIFIED,
+ f->fetch_handle, 0, 0,
+ FETCH_ERROR_NO_ERROR);
+ return true;
+ }
+#endif
+
+ if (S_ISDIR(ctx->fstat.st_mode)) {
+ /* directory listing */
+ fetch_file_process_dir(ctx);
+ return;
+ }
+
+ if (S_ISREG(ctx->fstat.st_mode)) {
+ /* not a file respose */
+ fetch_file_process_plain(ctx);
+ return;
+ }
+
+ close(ctx->fd);
+
+ return;
+}
+
+/** callback to poll for additional file fetch contents */
+static void fetch_file_poll(const char *scheme)
+{
+ struct fetch_file_context *c, *next;
+
+ if (ring == NULL) return;
+
+ /* Iterate over ring, processing each pending fetch */
+ c = ring;
+ do {
+ /* Take a copy of the next pointer as we may destroy
+ * the ring item we're currently processing */
+ next = c->r_next;
+
+ /* Ignore fetches that have been flagged as locked.
+ * This allows safe re-entrant calls to this function.
+ * Re-entrancy can occur if, as a result of a callback,
+ * the interested party causes fetch_poll() to be called
+ * again.
+ */
+ if (c->locked == true) {
+ continue;
+ }
+
+ LOG(("polling unlocked context %p",c));
+
+ /* Only process non-aborted fetches */
+ if (!c->aborted) {
+ /* file fetchs can be processed in one go */
+ fetch_file_process(c);
+ }
+
+
+ fetch_remove_from_queues(c->fetchh);
+ fetch_free(c->fetchh);
+
+ /* Advance to next ring entry, exiting if we've reached
+ * the start of the ring or the ring has become empty
+ */
+ } while ( (c = next) != ring && ring != NULL);
+}
+
+void fetch_file_register(void)
+{
+ fetch_add_fetcher("file",
+ fetch_file_initialise,
+ fetch_file_setup,
+ fetch_file_start,
+ fetch_file_abort,
+ fetch_file_free,
+ fetch_file_poll,
+ fetch_file_finalise);
+}
Index: content/fetchers/fetch_file.h
===================================================================
--- /dev/null 2009-04-16 19:17:07.000000000 +0100
+++ content/fetchers/fetch_file.h 2010-09-09 19:47:22.000000000 +0100
@@ -0,0 +1,28 @@
+/*
+ * Copyright 2010 Vincent Sanders <vince(a)netsurf-browser.org>
+ *
+ * This file is part of NetSurf.
+ *
+ * NetSurf is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * NetSurf is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>;.
+ */
+
+/** \file
+ * file: URL method handler
+ */
+
+#ifndef NETSURF_CONTENT_FETCHERS_FETCH_FILE_H
+#define NETSURF_CONTENT_FETCHERS_FETCH_FILE_H
+
+void fetch_file_register(void);
+
+#endif
Changed files
Makefile.sources | 4 +-
amiga/filetype.c | 5 ---
content/content.c | 5 ---
content/content_type.h | 1
content/dirlist.c | 46 +++++++++++++++++++++---------
content/dirlist.h | 8 ++---
content/fetch.c | 2 +
content/fetchers/fetch_curl.c | 64 +++---------------------------------------
render/box.c | 1
utils/config.h | 16 ++++++++++
utils/url.c | 21 +++++++++++++
utils/url.h | 1
12 files changed, 84 insertions(+), 90 deletions(-)
Index: render/box.c
===================================================================
--- render/box.c (revision 10746)
+++ render/box.c (working copy)
@@ -1169,7 +1169,6 @@
#ifdef WITH_PLUGIN
content_get_type(box->object) == CONTENT_PLUGIN ||
#endif
- content_get_type(box->object) == CONTENT_DIRECTORY ||
#ifdef WITH_THEME_INSTALL
content_get_type(box->object) == CONTENT_THEME ||
#endif
Index: Makefile.sources
===================================================================
--- Makefile.sources (revision 10746)
+++ Makefile.sources (working copy)
@@ -6,9 +6,9 @@
#
S_CONTENT := content.c dirlist.c fetch.c hlcache.c llcache.c urldb.c \
- fetchers/fetch_curl.c fetchers/fetch_data.c
+ fetchers/fetch_curl.c fetchers/fetch_data.c fetchers/fetch_file.c
S_CSS := css.c dump.c internal.c select.c utils.c
-S_RENDER := box.c box_construct.c box_normalise.c directory.c favicon.c \
+S_RENDER := box.c box_construct.c box_normalise.c favicon.c \
font.c form.c html.c html_interaction.c html_redraw.c \
hubbub_binding.c imagemap.c layout.c list.c table.c textplain.c
S_UTILS := base64.c filename.c hashtable.c http.c locale.c messages.c \
Index: utils/config.h
===================================================================
--- utils/config.h (revision 10746)
+++ utils/config.h (working copy)
@@ -20,6 +20,7 @@
#define _NETSURF_UTILS_CONFIG_H_
#include <stddef.h>
+#include <dirent.h>
/* Try to detect which features the target OS supports */
@@ -37,6 +38,21 @@
char *strcasestr(const char *haystack, const char *needle);
#endif
+/* fdopendir is actually present on most unix systems but unless
+ * _POSIX_C_SOURCE is set to 2008 it is not declared in the system
+ * headers. It is unavailable on RISC OS which requires fallback code
+ */
+#if (_POSIX_C_SOURCE - 0) >= 200809L
+#define HAVE_FDOPENDIR
+#else
+#if defined(riscos)
+#undef HAVE_FDOPENDIR
+#else
+#define HAVE_FDOPENDIR
+DIR *fdopendir(int fd);
+#endif
+#endif
+
/* For some reason, UnixLib defines this unconditionally.
* Assume we're using UnixLib if building for RISC OS. */
#if (defined(_GNU_SOURCE) || defined(riscos))
Index: utils/url.c
===================================================================
--- utils/url.c (revision 10746)
+++ utils/url.c (working copy)
@@ -902,7 +902,28 @@
return URL_FUNC_FAILED;
}
+/**
+ * Convert an escaped string to plain.
+ * \param result unescaped string owned by caller must be freed with free()
+ * \return URL_FUNC_OK on success
+ */
+url_func_result url_unescape(const char *str, char **result)
+{
+ char *curlstr;
+ char *retstr;
+ curlstr = curl_unescape(str, 0);
+ retstr = strdup(curlstr);
+ curl_free(curlstr);
+
+ if (retstr == NULL) {
+ return URL_FUNC_FAILED;
+ }
+
+ *result = retstr;
+ return URL_FUNC_OK;
+}
+
/**
* Escape a string suitable for inclusion in an URL.
*
Index: utils/url.h
===================================================================
--- utils/url.h (revision 10746)
+++ utils/url.h (working copy)
@@ -53,6 +53,7 @@
bool remove_extensions);
url_func_result url_escape(const char *unescaped, size_t toskip,
bool sptoplus, const char *escexceptions, char **result);
+url_func_result url_unescape(const char *str, char **result);
url_func_result url_canonical_root(const char *url, char **result);
url_func_result url_parent(const char *url, char **result);
url_func_result url_plq(const char *url, char **result);
Index: content/content.c
===================================================================
--- content/content.c (revision 10746)
+++ content/content.c (working copy)
@@ -38,7 +38,6 @@
#include "image/bitmap.h"
#include "desktop/browser.h"
#include "desktop/options.h"
-#include "render/directory.h"
#include "render/html.h"
#include "render/textplain.h"
#ifdef WITH_JPEG
@@ -109,7 +108,6 @@
#ifdef WITH_BMP
{"application/x-ico", CONTENT_ICO},
#endif
- {"application/x-netsurf-directory", CONTENT_DIRECTORY},
#ifdef WITH_THEME_INSTALL
{"application/x-netsurf-theme", CONTENT_THEME},
#endif
@@ -344,9 +342,6 @@
plugin_open, plugin_close, plugin_clone,
true},
#endif
- {directory_create, 0, directory_convert,
- 0, directory_destroy, 0, 0, 0, 0, 0, 0, 0, directory_clone,
- true},
#ifdef WITH_THEME_INSTALL
{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, false},
#endif
Index: content/dirlist.c
===================================================================
--- content/dirlist.c (revision 10746)
+++ content/dirlist.c (working copy)
@@ -51,9 +51,10 @@
* dirlist_generate_bottom()
*/
-const char* dirlist_generate_top(void)
+bool dirlist_generate_top(char *buffer, int buffer_length)
{
- return "<html>\n"
+ int error = snprintf(buffer, buffer_length,
+ "<html>\n"
"<head>\n"
"<style>\n"
"html, body { margin: 0; padding: 0; }\n"
@@ -61,7 +62,7 @@
"h1 { padding: 5mm; margin: 0; "
"border-bottom: 2px solid #bcf; }\n"
"p { padding: 2px 5mm; margin: 0; }\n"
- "div { display: table; width: 94%; margin: 5mm auto 0 auto; "
+ "div { display: table; width: 94%%; margin: 5mm auto 0 auto; "
"padding: 0; }\n"
"a, strong { display: table-row; margin: 0; padding: 0; }\n"
"a.odd { background-color: #bcf; }\n"
@@ -74,7 +75,14 @@
"a.dir > span.type { font-weight: bold; }\n"
"span.size { text-align: right; padding-right: 0.3em; }\n"
"span.size + span.size { text-align: left; "
- "padding-right: 0; }\n";
+ "padding-right: 0; }\n");
+ if (error < 0 || error >= buffer_length)
+ /* Error or buffer too small */
+ return false;
+ else
+ /* OK */
+ return true;
+
}
@@ -142,9 +150,14 @@
* dirlist_generate_bottom()
*/
-bool dirlist_generate_title(char *title, char *buffer, int buffer_length)
+bool dirlist_generate_title(const char *title, char *buffer, int buffer_length)
{
- int error = snprintf(buffer, buffer_length,
+ int error;
+
+ if (title == NULL)
+ title = "";
+
+ error = snprintf(buffer, buffer_length,
"</style>\n"
"<title>%s</title>\n"
"</head>\n"
@@ -245,7 +258,7 @@
* \param directory whether this row is for a directory (or a file)
* \param url url for row entry
* \param name name of row entry
- * \param type MIME type of row entry
+ * \param mimetype MIME type of row entry
* \param size size of row entry. If negative, size is left blank
* \param date date row entry was last modified
* \param time time row entry was last modified
@@ -266,7 +279,7 @@
*/
bool dirlist_generate_row(bool even, bool directory, char *url, char *name,
- char *type, long long size, char *date, char *time,
+ const char *mimetype, long long size, char *date, char *time,
char *buffer, int buffer_length)
{
const char *unit;
@@ -292,7 +305,7 @@
"<span class=\"time\">%s</span></a>\n",
url, even ? "even" : "odd",
directory ? "dir" : "file",
- name, type, size_string, unit, date, time);
+ name, mimetype, size_string, unit, date, time);
if (error < 0 || error >= buffer_length)
/* Error or buffer too small */
return false;
@@ -319,11 +332,18 @@
* dirlist_generate_bottom()
*/
-const char* dirlist_generate_bottom(void)
+bool dirlist_generate_bottom(char *buffer, int buffer_length)
{
- return "</div>\n"
- "</body>\n"
- "</html>\n";
+ int error = snprintf(buffer, buffer_length,
+ "</div>\n"
+ "</body>\n"
+ "</html>\n");
+ if (error < 0 || error >= buffer_length)
+ /* Error or buffer too small */
+ return false;
+ else
+ /* OK */
+ return true;
}
Index: content/dirlist.h
===================================================================
--- content/dirlist.h (revision 10746)
+++ content/dirlist.h (working copy)
@@ -33,15 +33,15 @@
#define DIRLIST_NO_DATE_COLUMN 1 << 3
#define DIRLIST_NO_TIME_COLUMN 1 << 4
-const char* dirlist_generate_top(void);
+bool dirlist_generate_top(char *buffer, int buffer_length);
bool dirlist_generate_hide_columns(int flags, char *buffer, int buffer_length);
-bool dirlist_generate_title(char *title, char *buffer, int buffer_length);
+bool dirlist_generate_title(const char *title, char *buffer, int buffer_length);
bool dirlist_generate_parent_link(char *parent, char *buffer,
int buffer_length);
bool dirlist_generate_headings(char *buffer, int buffer_length);
bool dirlist_generate_row(bool even, bool directory, char *url, char *name,
- char *type, long long size, char *date, char *time,
+ const char *mimetype, long long size, char *date, char *time,
char *buffer, int buffer_length);
-const char* dirlist_generate_bottom(void);
+bool dirlist_generate_bottom(char *buffer, int buffer_length);
#endif
Index: content/content_type.h
===================================================================
--- content/content_type.h (revision 10746)
+++ content/content_type.h (working copy)
@@ -59,7 +59,6 @@
#ifdef WITH_PLUGIN
CONTENT_PLUGIN,
#endif
- CONTENT_DIRECTORY,
#ifdef WITH_THEME_INSTALL
CONTENT_THEME,
#endif
Index: content/fetchers/fetch_curl.c
===================================================================
--- content/fetchers/fetch_curl.c (revision 10746)
+++ content/fetchers/fetch_curl.c (working copy)
@@ -217,7 +217,10 @@
data = curl_version_info(CURLVERSION_NOW);
- for (i = 0; data->protocols[i]; i++)
+ for (i = 0; data->protocols[i]; i++) {
+ if (strcmp(data->protocols[i], "file") == 0)
+ continue; /* do not use curl for file: */
+
if (!fetch_add_fetcher(data->protocols[i],
fetch_curl_initialise,
fetch_curl_setup,
@@ -229,6 +232,7 @@
LOG(("Unable to register cURL fetcher for %s",
data->protocols[i]));
}
+ }
return;
curl_easy_setopt_failed:
@@ -1155,64 +1159,6 @@
return true;
}
- /* find MIME type from filetype for local files */
- if (strncmp(f->url, FILE_SCHEME_PREFIX, FILE_SCHEME_PREFIX_LEN) == 0) {
- struct stat s;
- char *url_path = url_to_path(f->url);
-
- LOG(("Obtaining mime type for file %s", url_path));
-
- if (url_path != NULL && stat(url_path, &s) == 0) {
- /* file: URL and file exists */
- char header[64];
- const char *type;
-
- /* create etag */
- snprintf(header, sizeof header,
- "ETag: \"%10" PRId64 "\"",
- (int64_t) s.st_mtime);
- /* And send it to the header handler */
- fetch_send_callback(FETCH_HEADER, f->fetch_handle,
- header, strlen(header),
- FETCH_ERROR_NO_ERROR);
-
- /* create Content-Type */
- type = fetch_filetype(url_path);
- snprintf(header, sizeof header,
- "Content-Type: %s", type);
- /* Send it to the header handler */
- fetch_send_callback(FETCH_HEADER, f->fetch_handle,
- header, strlen(header),
- FETCH_ERROR_NO_ERROR);
-
- /* create Content-Length */
- snprintf(header, sizeof header,
- "Content-Length: %" PRId64,
- (int64_t) s.st_size);
- /* Send it to the header handler */
- fetch_send_callback(FETCH_HEADER, f->fetch_handle,
- header, strlen(header),
- FETCH_ERROR_NO_ERROR);
-
- /* don't set last modified time so as to ensure that
- * local files are revalidated at all times. */
-
- /* Report not modified, if appropriate */
- if (f->last_modified && f->file_etag &&
- f->last_modified > s.st_mtime &&
- f->file_etag == s.st_mtime) {
- fetch_send_callback(FETCH_NOTMODIFIED,
- f->fetch_handle, 0, 0,
- FETCH_ERROR_NO_ERROR);
- free(url_path);
- return true;
- }
- }
-
- if (url_path != NULL)
- free(url_path);
- }
-
if (f->abort)
return true;
Index: content/fetch.c
===================================================================
--- content/fetch.c (revision 10746)
+++ content/fetch.c (working copy)
@@ -38,6 +38,7 @@
#include "content/fetch.h"
#include "content/fetchers/fetch_curl.h"
#include "content/fetchers/fetch_data.h"
+#include "content/fetchers/fetch_file.h"
#include "content/urldb.h"
#include "desktop/netsurf.h"
#include "desktop/options.h"
@@ -106,6 +107,7 @@
{
fetch_curl_register();
fetch_data_register();
+ fetch_file_register();
fetch_active = false;
}
Index: amiga/filetype.c
===================================================================
--- amiga/filetype.c (revision 10746)
+++ amiga/filetype.c (working copy)
@@ -186,11 +186,6 @@
return "rosprite";
break;
#endif
-
- case CONTENT_DIRECTORY:
- return "drawer";
- break;
-
#if defined(WITH_NS_SVG) || defined(WITH_RSVG)
case CONTENT_SVG:
return "svg";
Conflicted files
Removed files
render/directory.c
render/directory.h
4:59 a.m.
On Thu, 2010-09-09 at 20:19 +0100, John-Mark Bell wrote:
Added files
Index: content/fetchers/fetch_file.c
===================================================================
--- /dev/null 2009-04-16 19:17:07.000000000 +0100
+++ content/fetchers/fetch_file.c 2010-09-09 19:47:22.000000000 +0100
@@ -0,0 +1,651 @@
+/*
+ * Copyright 2010 Vincent Sanders <vince(a)netsurf-browser.org>
+ *
+ * This file is part of NetSurf.
+ *
+ * NetSurf is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * NetSurf is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>;.
+ */
+
+/* file: URL handling. Based on the data fetcher by Rob Kendrik */
Kendrick.
+
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <assert.h>
+#include <errno.h>
+#include <stdbool.h>
+#include <inttypes.h>
+#include <string.h>
+#include <strings.h>
+#include <time.h>
+#include <stdio.h>
+#include <dirent.h>
+#include <limits.h>
+#include <stdarg.h>
+
+#include "utils/config.h"
+#include "content/dirlist.h"
+#include "content/fetch.h"
+#include "content/fetchers/fetch_file.h"
+#include "content/urldb.h"
+#include "desktop/netsurf.h"
+#include "desktop/options.h"
+#include "utils/log.h"
+#include "utils/messages.h"
+#include "utils/url.h"
+#include "utils/utils.h"
+#include "utils/ring.h"
+
+/* Maximum size of read buffer */
+#define FETCH_FILE_MAX_BUF_SIZE (1024 * 1024)
+
+/** Context for a fetch */
+struct fetch_file_context {
+ struct fetch_file_context *r_next, *r_prev;
+
+ struct fetch *fetchh; /**< Handle for this fetch */
+
+ bool aborted; /**< Flag indicating fetch has been aborted */
+ bool locked; /**< Flag indicating entry is already entered */
+
+ char *url; /**< The full url the fetch refers to */
+ char *path; /**< The actual path to be used with open() */
+
+ int fd; /**< The file descriptor of the object */
+ struct stat fstat; /**< The objects stat */
+};
+
+static struct fetch_file_context *ring = NULL;
+
+/** issue fetch callbacks with locking */
+static bool fetch_file_send_callback(fetch_msg msg,
+ struct fetch_file_context *ctx, const void *data,
+ unsigned long size, fetch_error_code errorcode)
+{
+ ctx->locked = true;
+ fetch_send_callback(msg, ctx->fetchh, data, size, errorcode);
+ ctx->locked = false;
+
+ return ctx->aborted;
+}
+
+static bool fetch_file_send_header(struct fetch_file_context *ctx, const char *fmt,
...)
+{
+ char header[64];
+ va_list ap;
+
+ va_start(ap, fmt);
+
+ vsnprintf(header, sizeof header, fmt, ap);
+
+ va_end(ap);
+
+ LOG(("%s", header));
+
+ ctx->locked = true;
+ fetch_send_callback(FETCH_HEADER, ctx->fetchh, header, strlen(header),
FETCH_ERROR_NO_ERROR);
+ ctx->locked = false;
Call fetch_file_send_callback, instead?
+ return ctx->aborted;
+}
+
+static bool fetch_file_send_time(struct fetch_file_context *ctx, const char *fmt, const
time_t *val)
+{
+ char header[64];
+ struct tm btm;
+
+ gmtime_r(val, &btm);
+
+ strftime(header, sizeof header, fmt, &btm);
+
+ LOG(("%s", header));
Presumably redundant now?
+ ctx->locked = true;
+ fetch_send_callback(FETCH_HEADER, ctx->fetchh, header, strlen(header),
FETCH_ERROR_NO_ERROR);
+ ctx->locked = false;
fetch_file_send_callback?
+ return ctx->aborted;
+}
+
+/** callback to initialise the file fetcher. */
+static bool fetch_file_initialise(const char *scheme)
+{
+ LOG(("fetch_file_initialise called for %s", scheme));
+ return true;
+}
+
+/** callback to initialise the file fetcher. */
+static void fetch_file_finalise(const char *scheme)
+{
+ LOG(("fetch_file_finalise called for %s", scheme));
+}
This logging is probably unnecessary now.
+/** callback to set up a file fetch context. */
+static void *
+fetch_file_setup(struct fetch *fetchh,
+ const char *url,
+ bool only_2xx,
+ const char *post_urlenc,
+ const struct fetch_multipart_data *post_multipart,
+ const char **headers)
+{
+ struct fetch_file_context *ctx = calloc(1, sizeof(*ctx));
+ char *path;
+
+ if (ctx == NULL)
+ return NULL;
+
+ ctx->fetchh = fetchh;
+
+ ctx->url = strdup(url);
+
+ url_path(url, &path);
+ if (path == NULL) {
Ensure URL_FUNC_OK here
free(ctx->url);
+ free(ctx);
+ return NULL;
+ }
+
+ url_unescape(path, &ctx->path);
+ if (ctx->path == NULL) {
Ensure URL_FUNC_OK here
+ free(path);
free(ctx->url);
+ free(ctx);
+ return NULL;
+ }
+
+ free(path);
+
+ RING_INSERT(ring, ctx);
+
+ LOG(("new context %p for fetch handle %p, url %s, path %s",
+ ctx, fetchh, ctx->url, ctx->path));
+ LOG(("only_2xx %d, post_urlenc %s, post_multipart %p, headers %p",
+ only_2xx, post_urlenc, post_multipart, headers));
Lose this logging?
+ return ctx;
+}
+
+/** callback to free a file fetch */
+static void fetch_file_free(void *ctx)
+{
+ struct fetch_file_context *c = ctx;
+ LOG(("context %p",ctx));
Ditto
fd?
+ free(c->url);
+ free(c->path);
+ RING_REMOVE(ring, c);
+ free(ctx);
+}
+
+/** callback to start a file fetch */
+static bool fetch_file_start(void *ctx)
+{
+ LOG(("context %p",ctx));
Lose logging
+ return true;
+}
+
+/** callback to abort a file fetch */
+static void fetch_file_abort(void *ctx)
+{
+ struct fetch_file_context *c = ctx;
+ LOG(("context %p",ctx));
Ditto
+ /* To avoid the poll loop having to deal with the fetch context
+ * disappearing from under it, we simply flag the abort here.
+ * The poll loop itself will perform the appropriate cleanup.
+ */
+ c->aborted = true;
+}
+
+
+/** Process object as a regular file */
+static void fetch_file_process_plain(struct fetch_file_context *ctx)
+{
+ char *buf;
+ size_t buf_size;
+
+ size_t tot_read = 0;
+ ssize_t res;
+
+ size_t size;
+
+ size = ctx->fstat.st_size;
+
+ /* set buffer size */
+ buf_size = size;
+ if (buf_size > FETCH_FILE_MAX_BUF_SIZE)
+ buf_size = FETCH_FILE_MAX_BUF_SIZE;
+
+ /* allocate the buffer storage */
+ buf = malloc(buf_size);
+ if (buf == NULL) {
+ fetch_file_send_callback(FETCH_ERROR, ctx,
+ "Unable to allocate memory for file data buffer",
+ 0, FETCH_ERROR_MEMORY);
+ return;
+ }
+
+ /* fetch is going to be successful */
+ fetch_set_http_code(ctx->fetchh, 200);
+
+ /* Any callback can result in the fetch being aborted.
+ * Therefore, we _must_ check for this after _every_ call to
+ * fetch_file_send_callback().
+ */
+
+ /* content type */
+ if (fetch_file_send_header(ctx, "Content-Type: %s",
fetch_filetype(ctx->path)))
+ goto fetch_file_process_aborted;
+
+ /* content length */
+ if (fetch_file_send_header(ctx, "Content-Length: %zd", size))
+ goto fetch_file_process_aborted;
+
+ /* Set Last modified header */
+ if (fetch_file_send_time(ctx, "Last-Modified: %a, %d %b %Y %H:%M:%S GMT",
&ctx->fstat.st_mtime))
+ goto fetch_file_process_aborted;
+
+ /* create etag */
+ if (fetch_file_send_header(ctx, "ETag: \"%10" PRId64
"\"", (int64_t) ctx->fstat.st_mtime))
+ goto fetch_file_process_aborted;
+
+ /* main data loop */
+ do {
+ res = read(ctx->fd, buf, buf_size);
+ if (res == -1) {
+ fetch_file_send_callback(FETCH_ERROR, ctx, "Error reading file", 0,
FETCH_ERROR_PARTIAL_FILE);
free(buf);
+ return;
+ }
+
+
+ if (res == 0) {
+ fetch_file_send_callback(FETCH_ERROR, ctx, "Unexpected EOF reading file",
0, FETCH_ERROR_PARTIAL_FILE);
free(buf);
+ return;
+ }
+
+ tot_read += res;
+
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buf, res, FETCH_ERROR_NO_ERROR))
+ break;
+
+ } while (tot_read < size);
+
+ if (!ctx->aborted)
+ fetch_file_send_callback(FETCH_FINISHED, ctx, 0, 0, FETCH_ERROR_NO_ERROR);
+
+fetch_file_process_aborted:
+
+ close(ctx->fd);
Where is this opened? Does it also need closing in the error cases,
above?
+ free(buf);
+ return;
+}
+
+static char *gen_nice_title(char *path)
+{
+ char *nice_path, *cnv, *tmp;
+ char *title;
+ int title_length;
+
+ /* Convert path for display */
+ nice_path = malloc(strlen(path) * 4 + 1);
+ if (!nice_path) {
nice_path == NULL is preferable
+ return NULL;
+ }
+
+ /* Escape special HTML characters */
+ for (cnv = nice_path, tmp = path; *tmp != '\0'; tmp++) {
+ if (*tmp == '<') {
+ *cnv++ = '&';
+ *cnv++ = 'l';
+ *cnv++ = 't';
+ *cnv++ = ';';
+ } else if (*tmp == '>') {
+ *cnv++ = '&';
+ *cnv++ = 'g';
+ *cnv++ = 't';
+ *cnv++ = ';';
+ } else {
+ *cnv++ = *tmp;
+ }
Also, '&' -> "&", to be safe.
+ }
+ *cnv = '\0';
+
+ /* Construct a localised title string */
+ title_length = strlen(nice_path) + strlen(messages_get("FileIndex"));
strlen(nice_path) == cnv - nice_path
+ title = malloc(title_length);
Add 1 for the trailing NUL?
+ if (title == NULL) {
free(nice_path);
+ return NULL;
+ }
+
+ /* Set title to localised "Index of <nice_path>" */
+ snprintf(title, title_length, messages_get("FileIndex"), nice_path);
+
+ free(nice_path);
+
+ return title;
+}
+
+static void fetch_file_process_error(struct fetch_file_context *ctx, int code)
+{
+ char buffer[1024];
+ const char *title;
+
+ /* content is going to return error code */
+ fetch_set_http_code(ctx->fetchh, code);
+
+ /* content type */
+ if (fetch_file_send_header(ctx, "Content-Type: text/html"))
+ goto fetch_file_process_error_aborted;
+
+ switch (code) {
+ case 401:
+ title = "Permission Denied";
+ break;
+
+ case 404:
+ title = "File not found";
+ break;
+
+ default:
+ title = "Internal Error";
+ }
You can simplify this by doing this:
char key[8];
snprintf(key, sizeof key, "HTTP%03d", code);
title = messages_get(key);
+ snprintf(buffer, sizeof buffer,
"<html><head><title>%s</title></head><body><h1>%s</h1><p>Error
%d while fetching file %s</p></body></html>", title, title,
code,ctx->url);
+
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buffer, strlen(buffer),
FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_error_aborted;
+
+ fetch_file_send_callback(FETCH_FINISHED, ctx, 0, 0, FETCH_ERROR_NO_ERROR);
+
+fetch_file_process_error_aborted:
+ return;
+}
+
+static void fetch_file_process_dir(struct fetch_file_context *ctx)
+{
+ char buffer[1024]; /* Output buffer */
+ bool even = false; /* formatting flag */
+ char *title; /* pretty printed title */
+ url_func_result res; /* result from url routines */
+ char *up; /* url of parent */
+ bool compare; /* result of url compare */
+
+ DIR *scandir; /* handle for enumerating the directory */
+ struct dirent* ent; /* leaf directory entry */
+ struct stat ent_stat; /* stat result of leaf entry */
+ char datebuf[64]; /* buffer for date text */
+ char timebuf[64]; /* buffer for time text */
+ char urlpath[PATH_MAX]; /* buffer for leaf entry path */
+
+#if defined(HAVE_FDOPENDIR)
+ scandir = fdopendir(ctx->fd);
+#else
+ /* this poses the possibility of a race where the directory
+ * has been removed from the namespace or resources for more
+ * fd are now unavailable between the previous open() and this
+ * call.
+ */
+ close(ctx->fd);
+ scandir = opendir(ctx->path);
+#endif
+ if (scandir == NULL) {
+ fetch_file_process_error(ctx, 500);
+ return;
+ }
+ /* fetch is going to be successful */
+ fetch_set_http_code(ctx->fetchh, 200);
+
+ /* content type */
+ if (fetch_file_send_header(ctx, "Content-Type: text/html"))
+ goto fetch_file_process_dir_aborted;
+
+ /* directory listing top */
+ dirlist_generate_top(buffer, sizeof buffer);
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buffer, strlen(buffer),
FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_dir_aborted;
+
+ /* directory listing title */
+ title = gen_nice_title(ctx->path);
+ dirlist_generate_title(title, buffer, sizeof buffer);
+ free(title);
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buffer, strlen(buffer),
FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_dir_aborted;
+
+ /* Print parent directory link */
+ res = url_parent(ctx->url, &up);
+ if (res == URL_FUNC_OK) {
+ res = url_compare(ctx->url, up, false, &compare);
+ if ((res == URL_FUNC_OK) && !compare) {
+ dirlist_generate_parent_link(up, buffer, sizeof buffer);
+
+ fetch_file_send_callback(FETCH_DATA, ctx,
+ buffer,
+ strlen(buffer),
+ FETCH_ERROR_NO_ERROR);
+
+ }
+ free(up);
+
+ if (ctx->aborted)
+ goto fetch_file_process_dir_aborted;
+
+ }
+
+ /* directory list headings */
+ dirlist_generate_headings(buffer, sizeof buffer);
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buffer, strlen(buffer),
FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_dir_aborted;
Lose an indent level here.
+ while ((ent = readdir(scandir)) != NULL) {
+
+ if (ent->d_name[0] == '.')
+ continue;
+
+ snprintf(urlpath,
+ sizeof urlpath,
+ "%s%s%s",
+ ctx->path,
+ (ctx->path[strlen(ctx->path) - 1] == '/')?"":"/" ,
+ ent->d_name);
+
+ if (stat(urlpath, &ent_stat) != 0) {
+ ent_stat.st_mode = 0;
+ datebuf[0] = 0;
+ timebuf[0] = 0;
+ } else {
+ /* Get date in output format */
+ if (strftime((char *)&datebuf, sizeof datebuf,
+ "%a %d %b %Y",
+ localtime(&ent_stat.st_mtime)) == 0) {
+ strncpy(datebuf, "-", sizeof datebuf);
+ }
+
+ /* Get time in output format */
+ if (strftime((char *)&timebuf, sizeof timebuf,
+ "%H:%M",
+ localtime(&ent_stat.st_mtime)) == 0) {
+ strncpy(timebuf, "-", sizeof timebuf);
+ }
+ }
+
+ if (S_ISREG(ent_stat.st_mode)) {
+ /* regular file */
+ dirlist_generate_row(even,
+ false,
+ urlpath,
+ ent->d_name,
+ fetch_filetype(urlpath),
+ ent_stat.st_size,
+ datebuf, timebuf,
+ buffer, sizeof(buffer));
+ } else if (S_ISDIR(ent_stat.st_mode)) {
+ /* directory */
+ dirlist_generate_row(even,
+ true,
+ urlpath,
+ ent->d_name,
+ messages_get("FileDirectory"),
+ -1,
+ datebuf, timebuf,
+ buffer, sizeof(buffer));
+ } else {
+ /* something else */
+ dirlist_generate_row(even,
+ false,
+ urlpath,
+ ent->d_name,
+ "",
+ -1,
+ datebuf, timebuf,
+ buffer, sizeof(buffer));
+ }
+
+ if (fetch_file_send_callback(FETCH_DATA, ctx,
+ buffer,
+ strlen(buffer),
+ FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_dir_aborted;
+
+ even = !even;
+ }
+
+ /* directory listing bottom */
+ dirlist_generate_bottom(buffer, sizeof buffer);
+ if (fetch_file_send_callback(FETCH_DATA, ctx, buffer, strlen(buffer),
FETCH_ERROR_NO_ERROR))
+ goto fetch_file_process_dir_aborted;
+
+
+ fetch_file_send_callback(FETCH_FINISHED, ctx, 0, 0, FETCH_ERROR_NO_ERROR);
+
+fetch_file_process_dir_aborted:
+
+ closedir(scandir);
+}
+
+
+/* process a file fetch */
+static void fetch_file_process(struct fetch_file_context *ctx)
+{
+ ctx->fd = open(ctx->path, O_RDONLY);
+ if (ctx->fd < 0) {
+ /* process errors as appropriate */
+ switch (errno) {
+ case EACCES:
+ fetch_file_process_error(ctx, 401);
+ break;
403 is a better response code for this case
+ case ENOENT:
+ fetch_file_process_error(ctx, 404);
+ break;
+
+ default:
+ fetch_file_process_error(ctx, 500);
+ break;
+ }
+ return;
+ }
+
+ if (fstat(ctx->fd, &ctx->fstat) != 0) {
+ /* process errors as appropriate */
+ close(ctx->fd);
+ fetch_file_process_error(ctx, 500);
+ return;
+ }
+
+#ifdef fetch_curl_modified_check
+ /* f was fetch_curl context and I cannot determine where f->last_modified was set
*/
+
+ /* Report not modified, if appropriate */
+ if (f->last_modified && f->file_etag &&
+ f->last_modified > s.st_mtime &&
+ f->file_etag == s.st_mtime) {
+ fetch_send_callback(FETCH_NOTMODIFIED,
+ f->fetch_handle, 0, 0,
+ FETCH_ERROR_NO_ERROR);
+ return true;
+ }
+#endif
Lose this?
+ if (S_ISDIR(ctx->fstat.st_mode)) {
+ /* directory listing */
+ fetch_file_process_dir(ctx);
+ return;
+ }
+
+ if (S_ISREG(ctx->fstat.st_mode)) {
+ /* not a file respose */
What?
+ fetch_file_process_plain(ctx);
+ return;
+ }
+
+ close(ctx->fd);
+
+ return;
+}
+
+/** callback to poll for additional file fetch contents */
+static void fetch_file_poll(const char *scheme)
+{
+ struct fetch_file_context *c, *next;
+
+ if (ring == NULL) return;
+
+ /* Iterate over ring, processing each pending fetch */
+ c = ring;
+ do {
+ /* Take a copy of the next pointer as we may destroy
+ * the ring item we're currently processing */
+ next = c->r_next;
+
+ /* Ignore fetches that have been flagged as locked.
+ * This allows safe re-entrant calls to this function.
+ * Re-entrancy can occur if, as a result of a callback,
+ * the interested party causes fetch_poll() to be called
+ * again.
+ */
+ if (c->locked == true) {
+ continue;
+ }
+
+ LOG(("polling unlocked context %p",c));
Lose this logging
+ /* Only process non-aborted fetches */
+ if (!c->aborted) {
+ /* file fetchs can be processed in one go */
fetches
+ fetch_file_process(c);
+ }
+
+
+ fetch_remove_from_queues(c->fetchh);
+ fetch_free(c->fetchh);
+
+ /* Advance to next ring entry, exiting if we've reached
+ * the start of the ring or the ring has become empty
+ */
+ } while ( (c = next) != ring && ring != NULL);
+}
+
+void fetch_file_register(void)
+{
+ fetch_add_fetcher("file",
+ fetch_file_initialise,
+ fetch_file_setup,
+ fetch_file_start,
+ fetch_file_abort,
+ fetch_file_free,
+ fetch_file_poll,
+ fetch_file_finalise);
+}
Index: content/fetchers/fetch_file.h
===================================================================
--- /dev/null 2009-04-16 19:17:07.000000000 +0100
+++ content/fetchers/fetch_file.h 2010-09-09 19:47:22.000000000 +0100
@@ -0,0 +1,28 @@
+/*
+ * Copyright 2010 Vincent Sanders <vince(a)netsurf-browser.org>
+ *
+ * This file is part of NetSurf.
+ *
+ * NetSurf is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * NetSurf is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>;.
+ */
+
+/** \file
+ * file: URL method handler
+ */
+
+#ifndef NETSURF_CONTENT_FETCHERS_FETCH_FILE_H
+#define NETSURF_CONTENT_FETCHERS_FETCH_FILE_H
+
+void fetch_file_register(void);
+
+#endif
Changed files
Makefile.sources | 4 +-
amiga/filetype.c | 5 ---
content/content.c | 5 ---
content/content_type.h | 1
content/dirlist.c | 46 +++++++++++++++++++++---------
content/dirlist.h | 8 ++---
content/fetch.c | 2 +
content/fetchers/fetch_curl.c | 64 +++---------------------------------------
render/box.c | 1
utils/config.h | 16 ++++++++++
utils/url.c | 21 +++++++++++++
utils/url.h | 1
12 files changed, 84 insertions(+), 90 deletions(-)
Index: render/box.c
===================================================================
--- render/box.c (revision 10746)
+++ render/box.c (working copy)
@@ -1169,7 +1169,6 @@
#ifdef WITH_PLUGIN
content_get_type(box->object) == CONTENT_PLUGIN ||
#endif
- content_get_type(box->object) == CONTENT_DIRECTORY ||
#ifdef WITH_THEME_INSTALL
content_get_type(box->object) == CONTENT_THEME ||
#endif
Index: Makefile.sources
===================================================================
--- Makefile.sources (revision 10746)
+++ Makefile.sources (working copy)
@@ -6,9 +6,9 @@
#
S_CONTENT := content.c dirlist.c fetch.c hlcache.c llcache.c urldb.c \
- fetchers/fetch_curl.c fetchers/fetch_data.c
+ fetchers/fetch_curl.c fetchers/fetch_data.c fetchers/fetch_file.c
S_CSS := css.c dump.c internal.c select.c utils.c
-S_RENDER := box.c box_construct.c box_normalise.c directory.c favicon.c \
+S_RENDER := box.c box_construct.c box_normalise.c favicon.c \
font.c form.c html.c html_interaction.c html_redraw.c \
hubbub_binding.c imagemap.c layout.c list.c table.c textplain.c
S_UTILS := base64.c filename.c hashtable.c http.c locale.c messages.c \
Index: utils/config.h
===================================================================
--- utils/config.h (revision 10746)
+++ utils/config.h (working copy)
@@ -20,6 +20,7 @@
#define _NETSURF_UTILS_CONFIG_H_
#include <stddef.h>
+#include <dirent.h>
/* Try to detect which features the target OS supports */
@@ -37,6 +38,21 @@
char *strcasestr(const char *haystack, const char *needle);
#endif
+/* fdopendir is actually present on most unix systems but unless
+ * _POSIX_C_SOURCE is set to 2008 it is not declared in the system
+ * headers. It is unavailable on RISC OS which requires fallback code
+ */
+#if (_POSIX_C_SOURCE - 0) >= 200809L
+#define HAVE_FDOPENDIR
+#else
+#if defined(riscos)
+#undef HAVE_FDOPENDIR
+#else
+#define HAVE_FDOPENDIR
+DIR *fdopendir(int fd);
+#endif
We can't provide this ourselves. Therefore, should this non-POSIX 2008
case simply be #undef HAVE_FDOPENDIR?
+#endif
+
/* For some reason, UnixLib defines this unconditionally.
* Assume we're using UnixLib if building for RISC OS. */
#if (defined(_GNU_SOURCE) || defined(riscos))
Index: utils/url.c
===================================================================
--- utils/url.c (revision 10746)
+++ utils/url.c (working copy)
@@ -902,7 +902,28 @@
return URL_FUNC_FAILED;
}
+/**
+ * Convert an escaped string to plain.
+ * \param result unescaped string owned by caller must be freed with free()
+ * \return URL_FUNC_OK on success
+ */
+url_func_result url_unescape(const char *str, char **result)
+{
+ char *curlstr;
+ char *retstr;
+ curlstr = curl_unescape(str, 0);
if (curlstr == NULL)
return URL_FUNC_NOMEM;
+ retstr = strdup(curlstr);
+ curl_free(curlstr);
+
+ if (retstr == NULL) {
+ return URL_FUNC_FAILED;
URL_FUNC_NOMEM
+ }
+
+ *result = retstr;
+ return URL_FUNC_OK;
+}
+
/**
* Escape a string suitable for inclusion in an URL.
*
Index: utils/url.h
===================================================================
--- utils/url.h (revision 10746)
+++ utils/url.h (working copy)
@@ -53,6 +53,7 @@
bool remove_extensions);
url_func_result url_escape(const char *unescaped, size_t toskip,
bool sptoplus, const char *escexceptions, char **result);
+url_func_result url_unescape(const char *str, char **result);
url_func_result url_canonical_root(const char *url, char **result);
url_func_result url_parent(const char *url, char **result);
url_func_result url_plq(const char *url, char **result);
Index: content/content.c
===================================================================
--- content/content.c (revision 10746)
+++ content/content.c (working copy)
@@ -38,7 +38,6 @@
#include "image/bitmap.h"
#include "desktop/browser.h"
#include "desktop/options.h"
-#include "render/directory.h"
#include "render/html.h"
#include "render/textplain.h"
#ifdef WITH_JPEG
@@ -109,7 +108,6 @@
#ifdef WITH_BMP
{"application/x-ico", CONTENT_ICO},
#endif
- {"application/x-netsurf-directory", CONTENT_DIRECTORY},
#ifdef WITH_THEME_INSTALL
{"application/x-netsurf-theme", CONTENT_THEME},
#endif
@@ -344,9 +342,6 @@
plugin_open, plugin_close, plugin_clone,
true},
#endif
- {directory_create, 0, directory_convert,
- 0, directory_destroy, 0, 0, 0, 0, 0, 0, 0, directory_clone,
- true},
#ifdef WITH_THEME_INSTALL
{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, false},
#endif
Index: content/dirlist.c
===================================================================
--- content/dirlist.c (revision 10746)
+++ content/dirlist.c (working copy)
@@ -51,9 +51,10 @@
* dirlist_generate_bottom()
*/
-const char* dirlist_generate_top(void)
+bool dirlist_generate_top(char *buffer, int buffer_length)
{
- return "<html>\n"
+ int error = snprintf(buffer, buffer_length,
+ "<html>\n"
"<head>\n"
"<style>\n"
"html, body { margin: 0; padding: 0; }\n"
@@ -61,7 +62,7 @@
"h1 { padding: 5mm; margin: 0; "
"border-bottom: 2px solid #bcf; }\n"
"p { padding: 2px 5mm; margin: 0; }\n"
- "div { display: table; width: 94%; margin: 5mm auto 0 auto; "
+ "div { display: table; width: 94%%; margin: 5mm auto 0 auto; "
"padding: 0; }\n"
"a, strong { display: table-row; margin: 0; padding: 0; }\n"
"a.odd { background-color: #bcf; }\n"
@@ -74,7 +75,14 @@
"a.dir > span.type { font-weight: bold; }\n"
"span.size { text-align: right; padding-right: 0.3em; }\n"
"span.size + span.size { text-align: left; "
- "padding-right: 0; }\n";
+ "padding-right: 0; }\n");
+ if (error < 0 || error >= buffer_length)
+ /* Error or buffer too small */
+ return false;
+ else
+ /* OK */
+ return true;
+
}
@@ -142,9 +150,14 @@
* dirlist_generate_bottom()
*/
-bool dirlist_generate_title(char *title, char *buffer, int buffer_length)
+bool dirlist_generate_title(const char *title, char *buffer, int buffer_length)
{
- int error = snprintf(buffer, buffer_length,
+ int error;
+
+ if (title == NULL)
+ title = "";
+
+ error = snprintf(buffer, buffer_length,
"</style>\n"
"<title>%s</title>\n"
"</head>\n"
@@ -245,7 +258,7 @@
* \param directory whether this row is for a directory (or a file)
* \param url url for row entry
* \param name name of row entry
- * \param type MIME type of row entry
+ * \param mimetype MIME type of row entry
* \param size size of row entry. If negative, size is left blank
* \param date date row entry was last modified
* \param time time row entry was last modified
@@ -266,7 +279,7 @@
*/
bool dirlist_generate_row(bool even, bool directory, char *url, char *name,
- char *type, long long size, char *date, char *time,
+ const char *mimetype, long long size, char *date, char *time,
char *buffer, int buffer_length)
{
const char *unit;
@@ -292,7 +305,7 @@
"<span class=\"time\">%s</span></a>\n",
url, even ? "even" : "odd",
directory ? "dir" : "file",
- name, type, size_string, unit, date, time);
+ name, mimetype, size_string, unit, date, time);
if (error < 0 || error >= buffer_length)
/* Error or buffer too small */
return false;
@@ -319,11 +332,18 @@
* dirlist_generate_bottom()
*/
-const char* dirlist_generate_bottom(void)
+bool dirlist_generate_bottom(char *buffer, int buffer_length)
{
- return "</div>\n"
- "</body>\n"
- "</html>\n";
+ int error = snprintf(buffer, buffer_length,
+ "</div>\n"
+ "</body>\n"
+ "</html>\n");
+ if (error < 0 || error >= buffer_length)
+ /* Error or buffer too small */
+ return false;
+ else
+ /* OK */
+ return true;
}
Index: content/dirlist.h
===================================================================
--- content/dirlist.h (revision 10746)
+++ content/dirlist.h (working copy)
@@ -33,15 +33,15 @@
#define DIRLIST_NO_DATE_COLUMN 1 << 3
#define DIRLIST_NO_TIME_COLUMN 1 << 4
-const char* dirlist_generate_top(void);
+bool dirlist_generate_top(char *buffer, int buffer_length);
bool dirlist_generate_hide_columns(int flags, char *buffer, int buffer_length);
-bool dirlist_generate_title(char *title, char *buffer, int buffer_length);
+bool dirlist_generate_title(const char *title, char *buffer, int buffer_length);
bool dirlist_generate_parent_link(char *parent, char *buffer,
int buffer_length);
bool dirlist_generate_headings(char *buffer, int buffer_length);
bool dirlist_generate_row(bool even, bool directory, char *url, char *name,
- char *type, long long size, char *date, char *time,
+ const char *mimetype, long long size, char *date, char *time,
char *buffer, int buffer_length);
-const char* dirlist_generate_bottom(void);
+bool dirlist_generate_bottom(char *buffer, int buffer_length);
#endif
Index: content/content_type.h
===================================================================
--- content/content_type.h (revision 10746)
+++ content/content_type.h (working copy)
@@ -59,7 +59,6 @@
#ifdef WITH_PLUGIN
CONTENT_PLUGIN,
#endif
- CONTENT_DIRECTORY,
#ifdef WITH_THEME_INSTALL
CONTENT_THEME,
#endif
Index: content/fetchers/fetch_curl.c
===================================================================
--- content/fetchers/fetch_curl.c (revision 10746)
+++ content/fetchers/fetch_curl.c (working copy)
@@ -217,7 +217,10 @@
data = curl_version_info(CURLVERSION_NOW);
- for (i = 0; data->protocols[i]; i++)
+ for (i = 0; data->protocols[i]; i++) {
+ if (strcmp(data->protocols[i], "file") == 0)
+ continue; /* do not use curl for file: */
+
if (!fetch_add_fetcher(data->protocols[i],
fetch_curl_initialise,
fetch_curl_setup,
@@ -229,6 +232,7 @@
LOG(("Unable to register cURL fetcher for %s",
data->protocols[i]));
}
+ }
return;
curl_easy_setopt_failed:
@@ -1155,64 +1159,6 @@
return true;
}
- /* find MIME type from filetype for local files */
- if (strncmp(f->url, FILE_SCHEME_PREFIX, FILE_SCHEME_PREFIX_LEN) == 0) {
- struct stat s;
- char *url_path = url_to_path(f->url);
-
- LOG(("Obtaining mime type for file %s", url_path));
-
- if (url_path != NULL && stat(url_path, &s) == 0) {
- /* file: URL and file exists */
- char header[64];
- const char *type;
-
- /* create etag */
- snprintf(header, sizeof header,
- "ETag: \"%10" PRId64 "\"",
- (int64_t) s.st_mtime);
- /* And send it to the header handler */
- fetch_send_callback(FETCH_HEADER, f->fetch_handle,
- header, strlen(header),
- FETCH_ERROR_NO_ERROR);
-
- /* create Content-Type */
- type = fetch_filetype(url_path);
- snprintf(header, sizeof header,
- "Content-Type: %s", type);
- /* Send it to the header handler */
- fetch_send_callback(FETCH_HEADER, f->fetch_handle,
- header, strlen(header),
- FETCH_ERROR_NO_ERROR);
-
- /* create Content-Length */
- snprintf(header, sizeof header,
- "Content-Length: %" PRId64,
- (int64_t) s.st_size);
- /* Send it to the header handler */
- fetch_send_callback(FETCH_HEADER, f->fetch_handle,
- header, strlen(header),
- FETCH_ERROR_NO_ERROR);
-
- /* don't set last modified time so as to ensure that
- * local files are revalidated at all times. */
-
- /* Report not modified, if appropriate */
- if (f->last_modified && f->file_etag &&
- f->last_modified > s.st_mtime &&
- f->file_etag == s.st_mtime) {
- fetch_send_callback(FETCH_NOTMODIFIED,
- f->fetch_handle, 0, 0,
- FETCH_ERROR_NO_ERROR);
- free(url_path);
- return true;
- }
- }
-
- if (url_path != NULL)
- free(url_path);
- }
-
if (f->abort)
return true;
Index: content/fetch.c
===================================================================
--- content/fetch.c (revision 10746)
+++ content/fetch.c (working copy)
@@ -38,6 +38,7 @@
#include "content/fetch.h"
#include "content/fetchers/fetch_curl.h"
#include "content/fetchers/fetch_data.h"
+#include "content/fetchers/fetch_file.h"
#include "content/urldb.h"
#include "desktop/netsurf.h"
#include "desktop/options.h"
@@ -106,6 +107,7 @@
{
fetch_curl_register();
fetch_data_register();
+ fetch_file_register();
fetch_active = false;
}
Index: amiga/filetype.c
===================================================================
--- amiga/filetype.c (revision 10746)
+++ amiga/filetype.c (working copy)
@@ -186,11 +186,6 @@
return "rosprite";
break;
#endif
-
- case CONTENT_DIRECTORY:
- return "drawer";
- break;
-
#if defined(WITH_NS_SVG) || defined(WITH_RSVG)
case CONTENT_SVG:
return "svg";
Conflicted files
Removed files
render/directory.c
render/directory.h
6:08 a.m.
New subject: Review: branches/vince/netsurf-file-fetcher
> +
> +/* file: URL handling. Based on the data fetcher by Rob Kendrik */
Kendrick.
Bugger, fixed, sorry Rob
> +
> + ctx->locked = true;
> + fetch_send_callback(FETCH_HEADER, ctx->fetchh, header, strlen(header),
FETCH_ERROR_NO_ERROR);
> + ctx->locked = false;
Call fetch_file_send_callback, instead?
hmm, ok, another function call but yeah its cleaner, changed
> +
> + LOG(("%s", header));
Presumably redundant now?
removed
> + ctx->locked = true;
> + fetch_send_callback(FETCH_HEADER, ctx->fetchh, header, strlen(header),
FETCH_ERROR_NO_ERROR);
> + ctx->locked = false;
fetch_file_send_callback?
changed
> + return ctx->aborted;
> +}
> +
> +/** callback to initialise the file fetcher. */
> +static bool fetch_file_initialise(const char *scheme)
> +{
> + LOG(("fetch_file_initialise called for %s", scheme));
> + return true;
> +}
> +
> +/** callback to initialise the file fetcher. */
> +static void fetch_file_finalise(const char *scheme)
> +{
> + LOG(("fetch_file_finalise called for %s", scheme));
> +}
This logging is probably unnecessary now.
removed
> +/** callback to set up a file fetch context. */
> +static void *
> +fetch_file_setup(struct fetch *fetchh,
> + const char *url,
> + bool only_2xx,
> + const char *post_urlenc,
> + const struct fetch_multipart_data *post_multipart,
> + const char **headers)
> +{
> + struct fetch_file_context *ctx = calloc(1, sizeof(*ctx));
> + char *path;
> +
> + if (ctx == NULL)
> + return NULL;
> +
> + ctx->fetchh = fetchh;
> +
> + ctx->url = strdup(url);
> +
> + url_path(url, &path);
> + if (path == NULL) {
Ensure URL_FUNC_OK here
res = url_path(url, &path);
if (res != URL_FUNC_OK) {
free(ctx);
return NULL;
}
free(ctx->url);
I moved the allocation after these checks
> + free(ctx);
> + return NULL;
> + }
> +
> + url_unescape(path, &ctx->path);
> + if (ctx->path == NULL) {
Ensure URL_FUNC_OK here
res = url_unescape(path, &ctx->path);
free(path);
if (res != URL_FUNC_OK) {
free(ctx);
return NULL;
}
>
free(ctx->url);
>
moved allocation after
> + free(ctx);
> + return NULL;
> + }
> +
> + free(path);
> +
> + RING_INSERT(ring, ctx);
> +
> + LOG(("new context %p for fetch handle %p, url %s, path %s",
> + ctx, fetchh, ctx->url, ctx->path));
> + LOG(("only_2xx %d, post_urlenc %s, post_multipart %p, headers %p",
> + only_2xx, post_urlenc, post_multipart, headers));
Lose this logging?
gone ;-)
> + return ctx;
> +}
> +
> +/** callback to free a file fetch */
> +static void fetch_file_free(void *ctx)
> +{
> + struct fetch_file_context *c = ctx;
> + LOG(("context %p",ctx));
Ditto
gonned
fd?
I took your suggestion and removed fd and fstat from the structure, its now passed
> + free(c->url);
> + free(c->path);
> + RING_REMOVE(ring, c);
> + free(ctx);
> +}
> +
> +/** callback to start a file fetch */
> +static bool fetch_file_start(void *ctx)
> +{
> + LOG(("context %p",ctx));
Lose logging
removed
> + return true;
> +}
> +
> +/** callback to abort a file fetch */
> +static void fetch_file_abort(void *ctx)
> +{
> + struct fetch_file_context *c = ctx;
> + LOG(("context %p",ctx));
Ditto
goned
> +
> +/** Process object as a regular file */
> +static void fetch_file_process_plain(struct fetch_file_context *ctx)
> +{
> + char *buf;
> + size_t buf_size;
> +
> + size_t tot_read = 0;
> + ssize_t res;
> +
> + size_t size;
> +
> + size = ctx->fstat.st_size;
> +
> + /* set buffer size */
> + buf_size = size;
> + if (buf_size > FETCH_FILE_MAX_BUF_SIZE)
> + buf_size = FETCH_FILE_MAX_BUF_SIZE;
> +
> + /* allocate the buffer storage */
> + buf = malloc(buf_size);
> + if (buf == NULL) {
> + fetch_file_send_callback(FETCH_ERROR, ctx,
> + "Unable to allocate memory for file data buffer",
> + 0, FETCH_ERROR_MEMORY);
> + return;
> + }
> +
> + /* fetch is going to be successful */
> + fetch_set_http_code(ctx->fetchh, 200);
> +
> + /* Any callback can result in the fetch being aborted.
> + * Therefore, we _must_ check for this after _every_ call to
> + * fetch_file_send_callback().
> + */
> +
> + /* content type */
> + if (fetch_file_send_header(ctx, "Content-Type: %s",
fetch_filetype(ctx->path)))
> + goto fetch_file_process_aborted;
> +
> + /* content length */
> + if (fetch_file_send_header(ctx, "Content-Length: %zd", size))
> + goto fetch_file_process_aborted;
> +
> + /* Set Last modified header */
> + if (fetch_file_send_time(ctx, "Last-Modified: %a, %d %b %Y %H:%M:%S
GMT", &ctx->fstat.st_mtime))
> + goto fetch_file_process_aborted;
> +
> + /* create etag */
> + if (fetch_file_send_header(ctx, "ETag: \"%10" PRId64
"\"", (int64_t) ctx->fstat.st_mtime))
> + goto fetch_file_process_aborted;
> +
> + /* main data loop */
> + do {
> + res = read(ctx->fd, buf, buf_size);
> + if (res == -1) {
> + fetch_file_send_callback(FETCH_ERROR, ctx, "Error reading file", 0,
FETCH_ERROR_PARTIAL_FILE);
free(buf);
actually these were not closing the fd either! so replaced both
returns with the "goto fetch_file_process_aborted;" method.
> + return;
> + }
> +
> +
> + if (res == 0) {
> + fetch_file_send_callback(FETCH_ERROR, ctx, "Unexpected EOF reading
file", 0, FETCH_ERROR_PARTIAL_FILE);
free(buf);
as previous
> + return;
> + }
> +
> + tot_read += res;
> +
> + if (fetch_file_send_callback(FETCH_DATA, ctx, buf, res, FETCH_ERROR_NO_ERROR))
> + break;
> +
> + } while (tot_read < size);
> +
> + if (!ctx->aborted)
> + fetch_file_send_callback(FETCH_FINISHED, ctx, 0, 0, FETCH_ERROR_NO_ERROR);
> +
> +fetch_file_process_aborted:
> +
> + close(ctx->fd);
Where is this opened? Does it also need closing in the error cases,
above?
its now passed in, opened by the parent and I have altered the error cases
> + free(buf);
> + return;
> +}
> +
this entire function was lifted wholesale from the old directory.c and
I failed to examine it properly, sorry.
> +static char *gen_nice_title(char *path)
> +{
> + char *nice_path, *cnv, *tmp;
> + char *title;
> + int title_length;
> +
> + /* Convert path for display */
> + nice_path = malloc(strlen(path) * 4 + 1);
> + if (!nice_path) {
nice_path == NULL is preferable
changed
> + return NULL;
> + }
> +
> + /* Escape special HTML characters */
> + for (cnv = nice_path, tmp = path; *tmp != '\0'; tmp++) {
> + if (*tmp == '<') {
> + *cnv++ = '&';
> + *cnv++ = 'l';
> + *cnv++ = 't';
> + *cnv++ = ';';
> + } else if (*tmp == '>') {
> + *cnv++ = '&';
> + *cnv++ = 'g';
> + *cnv++ = 't';
> + *cnv++ = ';';
> + } else {
> + *cnv++ = *tmp;
> + }
Also, '&' -> "&", to be safe.
done
> + }
> + *cnv = '\0';
> +
> + /* Construct a localised title string */
> + title_length = strlen(nice_path) + strlen(messages_get("FileIndex"));
strlen(nice_path) == cnv - nice_path
done
> + title = malloc(title_length);
Add 1 for the trailing NUL?
done
> + if (title == NULL) {
free(nice_path);
yup
> +
> +static void fetch_file_process_error(struct fetch_file_context *ctx, int code)
> +{
> + char buffer[1024];
> + const char *title;
> +
> + /* content is going to return error code */
> + fetch_set_http_code(ctx->fetchh, code);
> +
> + /* content type */
> + if (fetch_file_send_header(ctx, "Content-Type: text/html"))
> + goto fetch_file_process_error_aborted;
> +
> + switch (code) {
> + case 401:
> + title = "Permission Denied";
> + break;
> +
> + case 404:
> + title = "File not found";
> + break;
> +
> + default:
> + title = "Internal Error";
> + }
You can simplify this by doing this:
char key[8];
snprintf(key, sizeof key, "HTTP%03d", code);
title = messages_get(key);
sweet! done, thanks
> +
> + /* directory list headings */
> + dirlist_generate_headings(buffer, sizeof buffer);
> + if (fetch_file_send_callback(FETCH_DATA, ctx, buffer, strlen(buffer),
FETCH_ERROR_NO_ERROR))
> + goto fetch_file_process_dir_aborted;
Lose an indent level here.
ta, done
> +/* process a file fetch */
> +static void fetch_file_process(struct fetch_file_context *ctx)
> +{
> + ctx->fd = open(ctx->path, O_RDONLY);
> + if (ctx->fd < 0) {
> + /* process errors as appropriate */
> + switch (errno) {
> + case EACCES:
> + fetch_file_process_error(ctx, 401);
> + break;
403 is a better response code for this case
done
> + case ENOENT:
> + fetch_file_process_error(ctx, 404);
> + break;
> +
> + default:
> + fetch_file_process_error(ctx, 500);
> + break;
> + }
> + return;
> + }
> +
> + if (fstat(ctx->fd, &ctx->fstat) != 0) {
> + /* process errors as appropriate */
> + close(ctx->fd);
> + fetch_file_process_error(ctx, 500);
> + return;
> + }
> +
> +#ifdef fetch_curl_modified_check
> + /* f was fetch_curl context and I cannot determine where f->last_modified was
set */
> +
> + /* Report not modified, if appropriate */
> + if (f->last_modified && f->file_etag &&
> + f->last_modified > s.st_mtime &&
> + f->file_etag == s.st_mtime) {
> + fetch_send_callback(FETCH_NOTMODIFIED,
> + f->fetch_handle, 0, 0,
> + FETCH_ERROR_NO_ERROR);
> + return true;
> + }
> +#endif
Lose this?
gone
> + if (S_ISDIR(ctx->fstat.st_mode)) {
> + /* directory listing */
> + fetch_file_process_dir(ctx);
> + return;
> + }
> +
> + if (S_ISREG(ctx->fstat.st_mode)) {
> + /* not a file respose */
What?
if (S_ISREG(fdstat.st_mode)) {
/* regular file */
fetch_file_process_plain(ctx, fd, &fdstat);
return;
}
/* unhandled type of file */
close(fd);
fetch_file_process_error(ctx, 501);
return;
}
> + fetch_file_process_plain(ctx);
> + return;
> + }
> +
> + close(ctx->fd);
> +
> + return;
> +}
> +
> +/** callback to poll for additional file fetch contents */
> +static void fetch_file_poll(const char *scheme)
> +{
> + struct fetch_file_context *c, *next;
> +
> + if (ring == NULL) return;
> +
> + /* Iterate over ring, processing each pending fetch */
> + c = ring;
> + do {
> + /* Take a copy of the next pointer as we may destroy
> + * the ring item we're currently processing */
> + next = c->r_next;
> +
> + /* Ignore fetches that have been flagged as locked.
> + * This allows safe re-entrant calls to this function.
> + * Re-entrancy can occur if, as a result of a callback,
> + * the interested party causes fetch_poll() to be called
> + * again.
> + */
> + if (c->locked == true) {
> + continue;
> + }
> +
> + LOG(("polling unlocked context %p",c));
Lose this logging
gone
> + /* Only process non-aborted fetches */
> + if (!c->aborted) {
> + /* file fetchs can be processed in one go */
fetches
changed
>
> +/* fdopendir is actually present on most unix systems but unless
> + * _POSIX_C_SOURCE is set to 2008 it is not declared in the system
> + * headers. It is unavailable on RISC OS which requires fallback code
> + */
> +#if (_POSIX_C_SOURCE - 0) >= 200809L
> +#define HAVE_FDOPENDIR
> +#else
> +#if defined(riscos)
> +#undef HAVE_FDOPENDIR
> +#else
> +#define HAVE_FDOPENDIR
> +DIR *fdopendir(int fd);
> +#endif
We can't provide this ourselves. Therefore, should this non-POSIX 2008
case simply be #undef HAVE_FDOPENDIR?
difficult to call, glibc has actually had this since forever, our
build system explicitly sets posix 2001 and most modern headers
promptly remove the declaration. If this really is becoming such a
pita I can simply remove its use and we can put up with the race, tbh
I am completely beyond caring now
> +#endif
> +
> /* For some reason, UnixLib defines this unconditionally.
> * Assume we're using UnixLib if building for RISC OS. */
> #if (defined(_GNU_SOURCE) || defined(riscos))
> Index: utils/url.c
> ===================================================================
> --- utils/url.c (revision 10746)
> +++ utils/url.c (working copy)
> @@ -902,7 +902,28 @@
> return URL_FUNC_FAILED;
> }
>
> +/**
> + * Convert an escaped string to plain.
> + * \param result unescaped string owned by caller must be freed with free()
> + * \return URL_FUNC_OK on success
> + */
> +url_func_result url_unescape(const char *str, char **result)
> +{
> + char *curlstr;
> + char *retstr;
>
> + curlstr = curl_unescape(str, 0);
if (curlstr == NULL)
return URL_FUNC_NOMEM;
done
> + retstr = strdup(curlstr);
> + curl_free(curlstr);
> +
> + if (retstr == NULL) {
> + return URL_FUNC_FAILED;
URL_FUNC_NOMEM
done
> + }
> +
> + *result = retstr;
> + return URL_FUNC_OK;
> +}
> +
> /**
> * Escape a string suitable for inclusion in an URL.
> *
2:44 a.m.
Firstly, this has the same problem as the old libcurl file fetcher, in
that it insists on opening every file prepended with a slash. The fix
for that particular issue is this:
--8<--
Index: content/fetchers/fetch_file.c
===================================================================
--- content/fetchers/fetch_file.c (revision 10754)
+++ content/fetchers/fetch_file.c (working copy)
@@ -135,8 +135,8 @@
if (ctx == NULL)
return NULL;
- res = url_path(url, &path);
- if (res != URL_FUNC_OK) {
+ path = url_to_path(url);
+ if (path == NULL) {
free(ctx);
return NULL;
}
--8<--
I've not committed that change myself as I'm not in a position to test
on other platforms, but it should work as it asks the platform code
what the path is instead of assuming.
Secondly, the following doesn't work as expected on AmigaOS and
probably doesn't work on anything that isn't UNIX-derived:
+/* process a file fetch */
+static void fetch_file_process(struct fetch_file_context *ctx)
+{
+ ctx->fd = open(ctx->path, O_RDONLY);
+ if (ctx->fd < 0) {
+ /* process errors as appropriate */
[...]
+ if (S_ISDIR(ctx->fstat.st_mode)) {
+ /* directory listing */
+ fetch_file_process_dir(ctx);
+ return;
+ }
The issue is that if you open() a directory, it branches off to report
an error (you can't open() directories) before getting to S_ISDIR() -
so that code never executes. S_ISDIR would fail anyway as there would
be no file descriptor to pass to it.
I would guess that anything which doesn't have fdopendir() would also
not be able to open() a directory.
If open() fails, opendir() should be called. Only if that reports an
error should the error branch be taken.
This should also resolve this comment:
/* this poses the possibility of a race where the directory
* has been removed from the namespace or resources for more
* fd are now unavailable between the previous open() and this
* call.
*/
close(fd);
scandir = opendir(ctx->path);
opendir() has already been called if fdopendir() is not available,
just needs the pointer stored somewhere where the directory parser can
get to it (being passed into the function would probably suffice)
Chris
3:55 a.m.
Sigh, ok, ok I give in, commited a less sophisticated version, does
not assume it can use open() on anything except plain files. Only uses
opendir() on directories and specifically avoids using any expected
clib behaviour introduced in the last couple of decades.
I updated the url_to_path of some frontends in the vauge hope using
the url lib is a bit more correct than their previously implementations.
Should fix all your points from below.
On Sat, Sep 11, 2010 at 06:44:27PM +0100, Chris Young wrote:
Firstly, this has the same problem as the old libcurl file fetcher,
in
that it insists on opening every file prepended with a slash. The fix
for that particular issue is this:
--8<--
Index: content/fetchers/fetch_file.c
===================================================================
--- content/fetchers/fetch_file.c (revision 10754)
+++ content/fetchers/fetch_file.c (working copy)
@@ -135,8 +135,8 @@
if (ctx == NULL)
return NULL;
- res = url_path(url, &path);
- if (res != URL_FUNC_OK) {
+ path = url_to_path(url);
+ if (path == NULL) {
free(ctx);
return NULL;
}
--8<--
I've not committed that change myself as I'm not in a position to test
on other platforms, but it should work as it asks the platform code
what the path is instead of assuming.
Secondly, the following doesn't work as expected on AmigaOS and
probably doesn't work on anything that isn't UNIX-derived:
> +/* process a file fetch */
> +static void fetch_file_process(struct fetch_file_context *ctx)
> +{
> + ctx->fd = open(ctx->path, O_RDONLY);
> + if (ctx->fd < 0) {
> + /* process errors as appropriate */
[...]
> + if (S_ISDIR(ctx->fstat.st_mode)) {
> + /* directory listing */
> + fetch_file_process_dir(ctx);
> + return;
> + }
The issue is that if you open() a directory, it branches off to report
an error (you can't open() directories) before getting to S_ISDIR() -
so that code never executes. S_ISDIR would fail anyway as there would
be no file descriptor to pass to it.
I would guess that anything which doesn't have fdopendir() would also
not be able to open() a directory.
If open() fails, opendir() should be called. Only if that reports an
error should the error branch be taken.
This should also resolve this comment:
> /* this poses the possibility of a race where the directory
> * has been removed from the namespace or resources for more
> * fd are now unavailable between the previous open() and this
> * call.
> */
> close(fd);
> scandir = opendir(ctx->path);
opendir() has already been called if fdopendir() is not available,
just needs the pointer stored somewhere where the directory parser can
get to it (being passed into the function would probably suffice)
Chris
--
Regards Vincent
http://www.kyllikki.org/
8:52 p.m.
On Sat, 11 Sep 2010 19:55:45 +0100, Vincent Sanders wrote:
Sigh, ok, ok I give in
Don't be like that :)
Anyway, it's working perfectly now, here, thanks.
I updated the url_to_path of some frontends in the vauge hope using
the url lib is a bit more correct than their previously implementations.
url_path is probably fine for anything that expects UNIX format paths.
I've tweaked the directory listing creation a touch, as it was reading
files using the url-style path, rather than the native one. It now
asks the platform to add the filename to the end of the path, and then
convert back to a url-style path for the listing. This is the same
(IIRC) as the previous directory lister.
Chris
3872
days inactive
3875
days old
netsurf-dev@netsurf-browser.org
5 comments
3 participants
participants (3)
-
Chris Young -
John-Mark Bell -
Vincent Sanders