.SH DESCRIPTION
.PP
Download files from GNUnet.
+
.TP
\fB\-a \fILEVEL\fR, \fB\-\-anonymity=LEVEL\fR
set desired level of receiver anonymity. Default is 1.
+
.TP
\fB\-c \fIFILENAME\fR, \fB\-\-config=FILENAME\fR
use config file (defaults: ~/.gnunet/gnunet.conf)
+
.TP
\fB\-D, \fB\-\-delete\-incomplete\fR
causes gnunet\-download to delete incomplete downloads when aborted with CTRL\-C. Note that complete files that are part of an incomplete recursive download will not be deleted even with this option. Without this option, terminating gnunet\-download with a signal will cause incomplete downloads to stay on disk. If gnunet\-download runs to (normal) completion finishing the download, this option has no effect.
+
.TP
\fB\-h\fR, \fB\-\-help\fR
print help page
-.TP
-\fB\-H \fIHOSTNAME\fR, \fB\-\-host=\fIHOSTNAME\fR
-on which host is gnunetd running (default: localhost). You can also specify a port using the syntax HOSTNAME:PORT. The default port is 2087.
+
.TP
\fB\-L \fILOGLEVEL\fR, \fB\-\-loglevel=LOGLEVEL\fR
Change the loglevel. Possible values for LOGLEVEL are
ERROR, WARNING, INFO and DEBUG.
+
+.TP
+\fB\-n\fR, \fB\-\-no-network\fR
+Only search locally, do not forward requests to other peers.
+
.TP
\fB\-o \fIFILENAME\fR, \fB\-\-output=FILENAME\fR
write the file to FILENAME. Hint: when recursively downloading a directory, append a '/' to the end of the FILENAME to create a directory of that name. If no FILENAME is specified, gnunet\-download constructs a temporary ID from the URI of the file. The final filename is constructed based on meta\-data extracted using libextractor (if available).
+
.TP
\fB\-p \fIDOWNLOADS\fR, \fB\-\-parallelism=DOWNLOADS\fR
set the maximum number of parallel downloads that is allowed. More parallel downloads can, to some extent, improve the overall time to download content. However, parallel downloads also take more memory (see also option \-r which can be used to limit memory utilization) and more sockets. GNUnet typically only supports 1024 sockets per process, and each parallel download requires a new socket. Hence it is not possible to run more than about 1000 downloads in parallel. This option is used to limit the number of files that are downloaded in parallel (\-r can be used to limit the number of blocks that are concurrently requested). As a result, the value only matters for recursive downloads. The default value is 32. Note that the overall limit of 1000 downloads applies to gnunet\-service\-fs as well and using multiple gnunet\-download processes hence does not increase this limit.
+
.TP
\fB\-r \fIREQUESTS\fR, \fB\-\-request-parallelism=REQUESTS\fR
set the maximum number of parallel requests that is allowed. If multiple files are downloaded, gnunet\-download will not run them in parallel if this would cause the number of pending requests to possibly exceed the given value. This is useful since, for example, downloading dozens of multi\-gigabyte files in parallel could exhaust memory resources and would hardly improve performance. Note that the limit only applies to this specific process and that other download activities by other processes are not included in this limit. Consider raising this limit for large recursive downloads with many large files if memory and network bandwidth are not fully utilized and if the parallelism limit (\-p option) is not reached. This option also only matters for recursive downloads. The default value is 4092.
+
.TP
\fB\-R\fR, \fB\-\-recursive\fR
download directories recursively (and in parallel); note that the URI must belong to a GNUnet directory and that the filename given must end with a '/' \-\- otherwise, only the file corresponding to the URI will be downloaded. Note that in addition to using '-R', you must also specify a filename ending in '.gnd' so that the code realizes that the top-level file is a directory (since we have no meta data).
+
.TP
\fB\-v\fR, \fB\-\-version\fR
print the version number
+
.TP
\fB\-V\fR, \fB\-\-verbose\fR
print progress information
+
.SH NOTES
The GNUNET_URI is typically obtained from gnunet\-search. gnunet\-gtk can also be used instead of gnunet\-download.
If you ever have to abort a download, you can at any time continue it by re\-issuing gnunet\-download with the same filename. In that case GNUnet will not download blocks again that are already present. GNUnets file\-encoding will ensure file integrity, even if the existing file was not downloaded from GNUnet in the first place. Temporary information will be appended to the target file until the download is completed.