Hi.
(Not a bug, but feature I always missed in wget).
--limit-size option support added.
Nothing special: just skip any HTTP/FTP file with size bigger than set
in --limit-size=.
Size can be set as nnnM, nnnk (megabytes, kilobytes).
For those who found it interesting:
Full patched 1.13.4 source tree:
http://yurichev.com/non-wiki-files/wget-1.13.4-limitsize/wget-1.13.4-limitsize-src.zip
Patched and compiled (cygwin, win32):
http://yurichev.com/non-wiki-files/wget-1.13.4-limitsize/wget-1.13.4-limitsize-win32-cygwin.zip
Patched and compiled (linux x86):
http://yurichev.com/non-wiki-files/wget-1.13.4-limitsize/wget-1.13.4-limitsize_linux_x86.tar.bz2
diff -rupN wget-1.13.4.original/src/ftp.c wget-1.13.4-limitsize/src/ftp.c
--- wget-1.13.4.original/src/ftp.c 2011-09-13 11:05:12.000000000 +0300
+++ wget-1.13.4-limitsize/src/ftp.c 2012-03-28 00:50:09.731942900 +0300
@@ -1747,6 +1747,14 @@ ftp_retrieve_list (struct url *u, struct
err = RETROK;
dlthis = true;
+
+ if ((f->type == FT_PLAINFILE) && opt.limit_size && (f->size >
opt.limit_size))
+ {
+ logprintf (LOG_VERBOSE, _("%s -- remote file size (%s) "), quote
(con->target), human_readable(f->size));
+ logprintf (LOG_VERBOSE, _("is bigger than limit-size (%s) option --
not retrieving.\n"), human_readable(opt.limit_size));
+ dlthis = false;
+ }
+
if (opt.timestamping && f->type == FT_PLAINFILE)
{
struct_stat st;
diff -rupN wget-1.13.4.original/src/http.c wget-1.13.4-limitsize/src/http.c
--- wget-1.13.4.original/src/http.c 2011-09-07 13:58:01.000000000 +0300
+++ wget-1.13.4-limitsize/src/http.c 2012-03-28 01:24:23.114389600 +0300
@@ -2377,6 +2377,25 @@ read_header:
}
}
+ if (opt.limit_size && (contlen > opt.limit_size))
+ {
+ logprintf (LOG_VERBOSE, _("%s -- remote file size (%s) "), quote
(u->path), human_readable(contlen));
+ logprintf (LOG_VERBOSE, _("is bigger than limit-size (%s) option -- not
retrieving.\n"), human_readable(opt.limit_size));
+
+ /* this piece of code is copypasted from the code below... */
+
+ /* In case the caller inspects. */
+ hs->len = contlen;
+ hs->res = 0;
+ /* Mark as successfully retrieved. */
+ *dt |= RETROKF;
+ xfree_null (type);
+ CLOSE_INVALIDATE (sock); /* would be CLOSE_FINISH, but there
+ might be more bytes in the body. */
+ xfree (head);
+ return RETRUNNEEDED;
+ }
+
if (statcode == HTTP_STATUS_RANGE_NOT_SATISFIABLE
|| (!opt.timestamping && hs->restval > 0 && statcode == HTTP_STATUS_OK
&& contrange == 0 && contlen >= 0 && hs->restval >= contlen))
diff -rupN wget-1.13.4.original/src/init.c wget-1.13.4-limitsize/src/init.c
--- wget-1.13.4.original/src/init.c 2011-08-19 13:06:20.000000000 +0300
+++ wget-1.13.4-limitsize/src/init.c 2012-03-27 23:57:02.270630500 +0300
@@ -194,6 +194,7 @@ static const struct {
{ "iri", &opt.enable_iri, cmd_boolean },
{ "keepsessioncookies", &opt.keep_session_cookies, cmd_boolean },
{ "limitrate", &opt.limit_rate, cmd_bytes },
+ { "limitsize", &opt.limit_size, cmd_bytes },
{ "loadcookies", &opt.cookies_input, cmd_file },
{ "localencoding", &opt.locale, cmd_string },
{ "logfile", &opt.lfilename, cmd_file },
diff -rupN wget-1.13.4.original/src/main.c wget-1.13.4-limitsize/src/main.c
--- wget-1.13.4.original/src/main.c 2011-09-06 16:50:11.000000000 +0300
+++ wget-1.13.4-limitsize/src/main.c 2012-03-27 23:31:24.067650300 +0300
@@ -226,6 +226,7 @@ static struct cmdline_option option_data
{ "keep-session-cookies", 0, OPT_BOOLEAN, "keepsessioncookies", -1 },
{ "level", 'l', OPT_VALUE, "reclevel", -1 },
{ "limit-rate", 0, OPT_VALUE, "limitrate", -1 },
+ { "limit-size", 0, OPT_VALUE, "limitsize", -1 },
{ "load-cookies", 0, OPT_VALUE, "loadcookies", -1 },
{ "local-encoding", 0, OPT_VALUE, "localencoding", -1 },
{ "max-redirect", 0, OPT_VALUE, "maxredirect", -1 },
@@ -502,6 +503,8 @@ Download:\n"),
N_("\
--limit-rate=RATE limit download rate to RATE.\n"),
N_("\
+ --limit-size=SIZE limit download file size to SIZE.\n"),
+ N_("\
--no-dns-cache disable caching DNS lookups.\n"),
N_("\
--restrict-file-names=OS restrict chars in file names to ones OS
allows.\n"),
diff -rupN wget-1.13.4.original/src/options.h
wget-1.13.4-limitsize/src/options.h
--- wget-1.13.4.original/src/options.h 2011-08-06 13:24:32.000000000 +0300
+++ wget-1.13.4-limitsize/src/options.h 2012-03-27 23:33:39.674406600 +0300
@@ -124,6 +124,9 @@ struct options
wgint limit_rate; /* Limit the download rate to this
many bps. */
+
+ wgint limit_size; /* Limit the download file size. */
+
SUM_SIZE_INT quota; /* Maximum file size to download and
store. */
