I\'m posting here because I need help with some code relating with libssh.
I read all the official documentation here but still I don\'t understand
The following example uses ifstream in a loop to avoid loading a whole file into a memory (as the accepted answer does):
ifstream fin("C:\\myfile.zip", ios::binary);
while (fin)
{
constexpr size_t max_xfer_buf_size = 10240
char buffer[max_xfer_buf_size];
fin.read(buffer, sizeof(buffer));
if (fin.gcount() > 0)
{
ssize_t nwritten = sftp_write(NULL, buffer, fin.gcount());
if (nwritten != fin.gcount())
{
fprintf(stderr, "Can't write data to file: %s\n", ssh_get_error(ssh_session));
sftp_close(file);
return 1;
}
}
}
Open the file in the usual way (using C++'s fstream or C's stdio.h), read its contents to a buffer, and pass the buffer to sftp_write
.
Something like this:
ifstream fin("file.doc", ios::binary);
if (fin) {
fin.seekg(0, ios::end);
ios::pos_type bufsize = fin.tellg(); // get file size in bytes
fin.seekg(0); // rewind to beginning of file
std::vector<char> buf(bufsize); // allocate buffer
fin.read(buf.data(), bufsize); // read file contents into buffer
sftp_write(file, buf.data(), bufsize); // write buffer to remote file
}
Note that this is a very simple implementation. You should probably open the remote file in append mode, then write the data in chunks instead of sending single huge blob of data.
I used followed the example here.
sftp_read_sync
uses an infinite loop to read a file from a server into /path/to/profile
from server path /etc/profile
.